Hull and East Riding. Chronic Obstructive Pulmonary. Disease (COPD) Equity Audit

Transcription

1 Hull and East Riding Chronic Obstructive Pulmonary Disease (COPD) Equity Audit December 2010 Robert Sheikh Iddenden, Andrew Taylor, Jenny Walker

2 Hull and East Riding Chronic Obstructive Pulmonary Disease (COPD) Equity Audit Comments The COPD Equity Audit was informed by work undertaken mainly by NHS Hull. The report includes information on risk factors, prevalence, hospital activity and mortality for COPD for NHS East Riding of Yorkshire, but may not be complete in terms of existing and future plans for services provision for this Primary Care Trust (PCT). Robert Sheikh Iddenden, Public Health Sciences. 2

3 Hull & East Riding Chronic Obstructive Pulmonary Disease (COPD) Equity Audit Summary Health equity was defined as (i) equity in health the absence of systematic disparities in health with all groups entitled to the highest attainable standard of health; and (ii) equity of access or opportunity equal access to services for equal need. The health equity audit examined prevalence, hospital admissions, mortality, risk factors, access to services and quality of care in relation to age, gender, deprivation and any other groups for COPD where inequity could exist, where data was available to examine any differences locally. Programmes currently underway nationally or locally which could reduce inequalities in relation to care of patients with COPD were highlighted. The main findings were noted: The prevalence of COPD is increasing; registered QOF prevalence in Hull and East Riding increased each year between 2004/05 and 2008/09, with an overall increase over that period of 15%, such that in 2008/09 more than 10,000 people were registered as having COPD in Hull and East Riding. Despite these increases, models suggest that between 1,100 and 9,200 people may have undiagnosed COPD, with most of these in Hull (between 1,000 and 7,000). Each year from 2007/8 to 2009/10 1,325 patients were admitted into hospital due to COPD, with almost 1,900 admissions per year. A further 3,200 patients had just over 5,000 admissions for other reasons, but with COPD recorded as a secondary diagnosis. 98% of admissions due to COPD were emergency admissions as were 69% of admissions for other reasons where COPD recorded as a secondary diagnosis. COPD admissions are subject to seasonal variation, peaking in the winter months, with the greatest peak in recent years during winter 2008/09. Each year around 300 residents of Hull and East Riding died from COPD, accounting for around 6% of all deaths in Hull and just over 4% of all deaths in East Riding. In all age mortality rates in Hull were 56% higher than England, while in Hull they were 14% lower than England. Robert Sheikh Iddenden, Public Health Sciences. 3

4 Differences were even greater for under 75 mortality, 78% higher in Hull than England, 25% lower in East Riding than England. Smoking is the key risk factor for COPD, both current and former smoking. From surveys conducted in 2009, 33% of adults in Hull were smokers, as were 16% in East Riding, compared with 21% in England (2008). Almost 60% of men in Hull were current or former smokers, as well as more than 50% of women in Hull. In East Riding the equivalent percentages were almost 50% of men and more than 40% of women. 520 patients registered with Hull GPs were referred to pulmonary rehabilitation between January 2009 and September 2010 inclusive, representing almost 10% of patients on GP QOF registers. Data on pulmonary rehabilitation referrals for patients registered with East Riding GPs were not provided. 696 patients registered with Hull GPs were in receipt of home oxygen as at 22/09/2010, representing 13% of patients on GP QOF registers. Data on home oxygen for patients registered with East Riding GPs were not provided. An equity audit is a continuous cycle of assessing potential inequalities present, implementing programmes which change factors to reduce the inequalities found, auditing and assessing these programmes, and re-assessing inequalities and considering further programmes. Therefore, the intention is to examine COPD further at a later date. Robert Sheikh Iddenden, Public Health Sciences. 4

5 Hull & East Riding Chronic Obstructive Pulmonary Disease (COPD) Equity Audit Contents Summary... 3 List of tables... 9 List of figures Abbreviations Equity Audit Purpose and Outline Health Equity Audit Cycle Definition of Health Equity Audit COPD Equity Audit Context Policy context Cost of COPD on the NHS and wider economy COPD Equity Audit Scope and Framework Scope of COPD Equity Audit COPD Equity Audit Framework Profiling the Pattern of Need With Respect to Inequalities in COPD Equity Audit of Services COPD and Risk Factors for COPD Chronic Obstructive Pulmonary Disease Complications of COPD Risk Factors for COPD Confounders and Combined Risk Factors Causality Issues in Assessing COPD Assessing the True Effect of COPD Applying National Prevalence Data to Local Population COPD and World Class Commissioning Sources of Data, Definitions and Statistical Methods Kingston-upon-Hull and East Riding of Yorkshire Geographical Area Population Deprivation Geodemographic segmentation General Health COPD in Hull and East Riding Prevalence Diagnosed Prevalence Based on General Practice Registers Modelled COPD prevalence estimates at practice level from the Eastern Region Public Health Observatory model Modelled COPD prevalence estimates at practice level applying the Doncaster PCT methodology Robert Sheikh Iddenden, Public Health Sciences. 5

6 Modelled COPD prevalence estimates at local authority level from the Eastern Region Public Health Observatory model, with projections to Hospital Activity Number of Inpatient Admissions and Day-Cases Admission method Seasonal differences in emergency admissions Procedures Admission Rates Relative to Prevalence Mortality Primary Cause of Death Secondary Cause of Death Seasonal differences in COPD mortality Place of death Risk Factors Smoking Access to Services and Quality of Care NICE Clinical Guidelines 2004 & National Chronic Obstructive Pulmonary Disease National Audit Pulmonary rehabilitation Home oxygen / long term oxygen therapy GP Quality and Outcomes Framework Indicators COPD and Potential Inequalities Age and Gender Modelled Prevalence Modelled Prevalence Compared with Prevalence on GP Registers Hospital Activity Mortality Risk Factors Access to Services and Quality of Care Quality of Care- Ongoing Care in Primary Care Ethnicity Modelled Prevalence Hospital Activity Mortality Risk Factors Access to Services and Quality of Care Deprivation Prevalence on GP Registers Modelled Prevalence Compared with Prevalence on GP Registers Hospital Activity Mortality Prevalence, Hospital Admission Rate and Mortality Geodemographic segmentation Risk Factors Access to Services Quality of Care On-going Care in Primary Care Other Groups Where Potential Inequity May Be Present Robert Sheikh Iddenden, Public Health Sciences. 6

7 People with Learning or Physical Disabilities, Mental Health Problems, or Literacy or Numeracy Problems Gypsies and Travellers Housebound Patients Patients Whose First Language is not English Patients with Memory Impairment Prisoners Potential Programmes for Reducing Inequalities in COPD Reducing the Risk of Developing COPD Smoking Cessation Raising awareness of COPD among the general public Staff Training and Education Diagnosis of COPD Signs and Symptoms Spirometry Reversibility testing Severity and Prognosis Identifying early disease Referral to specialist advice Management of COPD Stop Smoking Effective Drug Therapy Pulmonary Rehabilitation Patient Education Non-Invasive Ventilation (NIV) Long Term Oxygen Therapy (LTOT) Managing Exacerbations Multi-Disciplinary Teams (MDTs) Improving the Quality of Self-Care Programme Budgeting Hull summary East Riding Draft quality standards for COPD Summary of Potential Programmes for Reducing Inequalities in COPD References Appendix A Sources of Data National Sources of Data Surveys Quality and Outcomes Framework Definitions and Classifications Used Risk Factors in Health and Lifestyle Survey Disease Definitions Using International Classification of Diseases Statistical Methods and Terms Standardisation Significance Testing Confidence Intervals Robert Sheikh Iddenden, Public Health Sciences. 7

8 Small Number of Events Appendix B Mean Age of Patients and Deprivation Scores for Local General Practices Appendix C Evidence Base for Self-care for COPD Index Robert Sheikh Iddenden, Public Health Sciences. 8

9 List of tables Table 1: Coverage of equity audit Table 2: Estimated registered and resident population as at October Table 3: Domains of the Index of Multiple Deprivation 2007 and their weights Table 4: Index of Multiple Deprivation 2007 scores and ranks of wards in Hull Table 5: Index of Multiple Deprivation 2007 scores and ranks of wards in East Riding of Yorkshire Table 6: ACORN classifications (based on dominant ACORN classifications at output area) for Hull and East Riding of Yorkshire, October 2008 population Table 7: Health ACORN classifications (based on dominant ACORN classifications at output area) for Hull and East Riding of Yorkshire, October 2008 population Table 8: Age-standardised mortality ratio from all causes for those <75 years for Hull wards, pooled Table 9: Age-standardised all cause under 75 mortality ratio for East Riding of Yorkshire, Table 10: Diagnosed prevalence of COPD on Hull s General Practice registers, trends over time Table 11: Diagnosed prevalence of COPD on East Riding of Yorkshire s General Practice registers, trends over time Table 12: Diagnosed prevalence of COPD on General Practice registers, 2004/05 to 2005/ Table 13: Modelled prevalence estimates (using ERPHO model) and register prevalence of COPD by GP practices in Hull 2008/ Table 14: Number of patients with COPD from the ERPHO model and the practice registers by GP practices in Hull 2008/ Table 15: Modelled prevalence estimates (using ERPHO model) and register prevalence of COPD by GP practices in East Riding 2008/ Table 16: Number of patients with COPD from the ERPHO model and the practice registers by GP practices in East Riding 2008/ Table 17: Modelled prevalence estimates (using Doncaster model) and register prevalence of COPD by GP practices in Hull 2008/ Table 18: Number of patients with COPD from the Doncaster model and the practice registers by GP practices in Hull 2008/ Table 19: Modelled prevalence estimates (using Doncaster model) and register prevalence of COPD by GP practices in East Riding 2008/ Table 20: Number of patients with COPD from the Doncaster model and the practice registers by GP practices in East Riding 2008/ Table 21: Estimated number and prevalence of COPD (aged 16+) for , 2015 and 2020 based on Eastern Region Public Health Observatory model (ERPHO, 2008) with register prevalence Table 22: Number of clinician episodes, hospital stays and patients admitted (inpatient and day-case visits) during 2007/08 and 2009/ Table 23: Average annual number of hospital admissions for hospital admissions having a primary or secondary diagnosis of COPD, 2007/08 to 2009/ Robert Sheikh Iddenden, Public Health Sciences. 9

10 Table 24: Average annual number of hospital admissions by main diagnoses for hospital admissions having a primary or secondary diagnosis of COPD, Hull and East Riding residents 2007/08 to 2009/ Table 25: Percentage of hospital admissions by main diagnoses for hospital admissions having a primary or secondary diagnosis of COPD, Hull 2006/08 to 2008/09, East Riding and combined total 2005/06 to 2007/ Table 26: Deaths during with COPD recorded as the primary cause of death91 Table 27: Mortality rate per 100,000 resident populations for deaths during 2004 to 2008 where COPD recorded as the primary cause of death Table 28: Under 75 SMRs to where COPD recorded as the primary cause of death Table 29: Deaths involving a primary or secondary diagnosis of COPD for patients registered with a Hull GP Table 30: Mortality rate per 100,000 population for deaths involving a primary diagnosis of COPD for patients registered with a Hull GP Table 31: Secondary cause of death for deaths where underlying cause is COPD for patients registered with a Hull GP dying during Table 32: Underlying cause of death where COPD is recorded as a secondary cause of death for patients registered with Hull GPs, Table 33: Number and percentage of deaths where COPD was the primary cause by place of death , by place of death as recorded on the death certificate, and percentage of non-copd deaths by place of death Table 34: Age-adjusted 1 smoking prevalence among adults 2 by gender in England 2008, Hull 2009, and East Riding Table 35: Updated NICE recommendations on severity of airflow obstruction Table 36: Number of patients aged 35 years and over, registered with a Hull GP and referred to pulmonary rehabilitation between January 2009 and September 2010 inclusive, by GP practice, and expressed as a percentage of QOF 2008/09 COPD registers Table 37: Number of patients aged 35 years and over, registered with a Hull GP and in receipt of home oxygen as at 22/09/2010, by GP practice, and expressed as a percentage of QOF 2008/09 COPD registers Table 38: 5 th and 95 th percentiles of QOF COPD indicators for Hull and East Riding General Practices, 2008/ Table 39: QOF COPD indicators for Hull and East Riding, 2008/ Table 40: Hull practices within highest 10% of practices for exceptions and within lowest 10% of practices for indicators Table 41: ERoY practices within highest 10% of practices for exceptions and within lowest 10% of practices for indicators Table 42: QOF influenza immunisation indicator 2008/09 for Hull General Practices. 122 Table 43: QOF influenza immunisation indicator 2008/09 for East Riding General Practices Table 44: QOF influenza immunisation indicator 2008/09 for Hull and East Riding PCTs and England Table 45: QOF recorded FeV 1 indicator 2008/09 for Hull General Practices Table 46: QOF recorded FeV 1 indicator 2008/09 for East Riding General Practices Robert Sheikh Iddenden, Public Health Sciences. 10

11 Table 47: QOF recorded FeV 1 indicator 2008/09 for Hull and East Riding PCTs and England Table 48: QOF recorded inhaler technique indicator 08/09 for Hull General Practices 130 Table 49: QOF recorded inhaler technique indicator 2008/09 for East Riding General Practices Table 50: QOF recorded inhaler technique indicator 2008/09 for Hull and East Riding PCTs and England Table 51: QOF spirometry-confirmed diagnosis indicator 2008/09 for Hull General Practices Table 52: QOF spirometry-confirmed diagnosis indicator 2008/09 for East Riding General Practices Table 53: QOF spirometry-confirmed diagnosis indicator 2008/09 for Hull and East Riding PCTs and England Table 54: Modelled expected COPD prevalence (%) among those aged 16 years and over, by age and by gender, produced at local authority level by Eastern Region PHO141 Table 55: Number of admissions with a primary diagnosis of COPD in Hull and East Riding of Yorkshire (2007/08 to 2009/10 pooled) by age and gender Table 56: Age-specific rate of admissions per 100,000 with a primary diagnosis of COPD in Hull and East Riding of Yorkshire (2007/08 to 2009/10 pooled) by gender Table 57: Number of emergency admissions with a primary diagnosis of COPD in Hull and East Riding of Yorkshire (2007/08 to 2009/10 pooled) by age and gender Table 58: Age-specific rate of emergency admissions per 100,000 with a primary diagnosis of COPD in Hull and East Riding of Yorkshire (2007/08 to 2009/10 pooled) by gender Table 59: Percentage difference in admission rates where COPD is the primary diagnosis between Hull and East Riding of Yorkshire of Yorkshire, 2007/08 to 2009/10 pooled. 151 Table 60: Crude age-specific COPD mortality rates, where the primary cause of death was COPD, deaths registered Table 61: Number of deaths and crude COPD mortality rate involving a primary (underlying) or secondary cause of death of COPD by age and gender for patients registered with Hull GPs, Table 62: Crude and adjusted mortality rates per 100,000 females where the primary cause of death was COPD, deaths registered Table 63: Crude and adjusted mortality rates per 100,000 males where the primary cause of death was COPD, deaths registered Table 64: Percentage of men and women who smoke or who used to smoke from Health Survey for England 2008 and local surveys in Hull and East Riding in Table 65: Changes in the prevalence of smoking in men and women from local surveys Table 66: Age-specific numbers of patients aged 35 years and over registered with Hull GP practices referred to pulmonary rehabilitation between January 2009 and September 2010, and rate of pulmonary rehabilitation referrals per 1,000 patients (using October 2009 population) Table 67: Age-specific numbers (n) of patients aged 35 years and over registered with Hull GP practices referred to pulmonary rehabilitation between January 2009 and July 2010 inclusive that did not attend pulmonary rehabilitation, and the percentage (%) out of all referrals (N) Robert Sheikh Iddenden, Public Health Sciences. 11

12 Table 68: Age-specific numbers of patients aged 35 years and over registered with Hull GP practices in receipt of home oxygen in 2010, and rate of home oxygen use per 1,000 patients (using October 2009 population) Table 69: Exceptions and indicator percentages by mean age of practice patients for Hull practices Table 70: Exceptions and indicator percentages by mean age of practice patients for East Riding practices Table 71: Estimated number of people with COPD by ethnic group using ONS 2007 midyear estimated populations (aged 16+) by broad ethnic group and ERPHO modelled prevalence rates (aged 16+) of COPD Table 72: Smoking status by BME group in Hull from 2007 adult health and lifestyle surveys, with percentages of respondents by broad age bands Table 73: Smoking status by broad age bands for BME groups from Hull s 2007 adult health and lifestyle survey where overall percentages smoking are higher than the Hull average and sufficient numbers are available from the survey (N>100) Table 74: Prevalence of diagnosed COPD on practice registers in Hull and East Riding in relation to local deprivation quintiles Table 75: Relationship between practice deprivation quintile, QOF COPD prevalence and ERPHO model prevalence Table 76: Relationship between practice deprivation quintile (Hull practices), QOF COPD prevalence and ERPHO model prevalence Table 77: Relationship between practice deprivation quintile (East Riding practices), QOF COPD prevalence and ERPHO model prevalence Table 78: Relationship between practice deprivation quintile, QOF COPD prevalence and Doncaster model prevalence Table 79: Relationship between practice deprivation quintile (Hull practices), QOF COPD prevalence and Doncaster model prevalence Table 80: Relationship between practice deprivation quintile (East Riding practices), QOF COPD prevalence and Doncaster model prevalence Table 81: Number and rate per 100,000 resident population of hospital admissions where COPD was the primary diagnosis in Hull and East Riding of Yorkshire (2007/08 to 2009/10 pooled), by gender and IMD 2007 deprivation quintile (local quintiles for Hull, national quintiles for East Riding and combined total) Table 82: Number and rate per 100,000 resident population of emergency hospital admissions where COPD was the primary diagnosis in Hull and East Riding of Yorkshire (2007/08 to 2009/10 pooled), by gender and IMD 2007 deprivation quintile (local quintiles for Hull, national quintiles for East Riding and combined total) Table 83: Number, crude and age-adjusted mortality rates per 100,000 persons for deaths registered with COPD the primary cause of death by deprivation quintile*, gender and PCT Table 84: Number and age-gender-adjusted mortality rate per 100,000 patients registered with Hull GPs registered population for deaths with a primary or secondary cause of death of COPD over two year period Table 85: Relationship between prevalence, hospital admission rate and mortality among local deprivation quintiles Robert Sheikh Iddenden, Public Health Sciences. 12

13 Table 86: Average annual number of COPD deaths* (where COPD is recorded as primary cause of death) and crude mortality rates per 100,000 by ACORN geodemographic classifications, deaths registered ** Table 87: Average annual number of COPD deaths* (where COPD is recorded as primary cause of death) and crude mortality rates per 100,000 by ACORN type where the ACORN category is Hard Pressed, deaths registered ** Table 88: Average annual number of COPD deaths* (where COPD is recorded as primary cause of death) and crude mortality rates per 100,000 by Health ACORN geodemographic classifications, deaths registered ** Table 89: Average annual number of emergency admissions where the primary diagnosis was COPD and crude emergency admission rates per 100,000 by ACORN geodemographic classifications, admissions in Hull and East Riding residents 2007/08 to 2009/10 inclusive Table 90: Average annual number of emergency admissions where the primary diagnosis was COPD and crude emergency admission rates per 100,000 by Health ACORN geodemographic classifications, admissions in Hull and East Riding residents 2007/08 to 2009/10 inclusive Table 91: Smoking status of survey respondents from surveys in Hull and East Riding conducted in 2009 by PCT and deprivation quintile (age-gender-adjusted percentages) Table 92: Number of patients aged 35 years and over registered with Hull GP practices referred to pulmonary rehabilitation between January 2009 and September 2010 inclusive by practice deprivation quintile, and percentage of all patients on QOF 2008/09 COPD registers Table 93: Numbers of patients aged 35 years and over registered with Hull GP practices referred to pulmonary rehabilitation between January 2009 and July 2010 inclusive that did not attend pulmonary rehabilitation, and the percentage out of all referrals with 95% confidence intervals, by practice deprivation quintiles Table 94: Age-adjusted percentage of patients (Hull residents only) aged 35 years and over referred to pulmonary rehabilitation between January 2009 and September 2010 inclusive, by local quintiles of IMD Table 95: Numbers of patients aged 35 years and over registered with Hull GP practices referred to pulmonary rehabilitation between January 2009 and July 2010 inclusive that did not attend pulmonary rehabilitation, and the percentage out of all referrals with 96% confidence intervals by local deprivation quintiles Table 96: Number of patients aged 35 years and over registered with Hull GP practices receiving home oxygen in 2010 by practice deprivation quintile, and percentage of all patients on QOF 2008/09 COPD registers Table 97: Age-adjusted percentage of patients aged 35 years and over receiving home oxygen in 2010, by local quintiles of IMD2007, Hull residents only Table 98: COPD QOF exceptions and indicator percentages by deprivation quintile of Hull practices (based on mean IMD2007 score of practice), local (Hull) quintiles defined at practice level Table 99: COPD QOF exceptions and indicator percentages by deprivation quintile of East Riding practices (based on mean IMD2007 score of practice), local (East Riding) quintiles defined at practice level Table 100: Smoking cessation statistics by gender Robert Sheikh Iddenden, Public Health Sciences. 13

14 Table 101: Smoking cessation statistics by age band Table 102: Locations of smoking cessation services in Hull by area Table 103: The NICE CG101 recommendations on symptoms Table 104: The NICE CG101 recommendations on spirometry [NB Recommendations that are new to the 2010 CG update are prefixed with a U ] Table 105: The NICE CG101 recommendations on spirometric reversibility testing Table 106: The NICE CG101 recommendation on severity and prognostic factors Table 107: The NICE CG101 recommendation on assessment of severity of airflow obstruction Table 108: The NICE CG101 recommendations on early identification Table 109: The NICE CG101 reasons for referral for specialist advice Table 110: The NICE CG101 recommendations on effective drug therapy Table 111: The NICE CG101 recommendations on pulmonary rehabilitation Table 112: The NICE CG101 recommendations on patient education Table 113: The NICE CG101 recommendations on long term oxygen therapy Table 114: The NICE CG101 Recommendations on the management of exacerbations Table 115: The NICE CG101 Recommendations on self-management of exacerbations Table 116: The NICE CG101 Recommendations on multi-disciplinary team management Table 117: The NICE CG101 Recommendations on physiotherapy Table 118: The NICE CG101 Recommendations on depression / anxiety Table 119: The NICE CG101 Recommendations on nutrition Table 120: The NICE CG101 Recommendations on palliative care Table 121: The NICE CG101 Recommendations on Occupational Therapy assessments Table 122: The NICE CG101 Recommendations on the role of Social Services Table 123: Programme budgeting data from YHPHO SpOT tool for Hull PCT Table 124: Programme budgeting data from YHPHO SpOT tool for East Riding PCT 263 Table 125: NICE draft quality standards for COPD Table 126: Quality and Outcome Framework indicators for COPD 2008/ Table 127: Definitions and Classifications used in the Health and Lifestyle survey Table 128: International Classification of Diseases: classifications used Table 129: International Classification of Diseases: classifications used Table 130: Mean age of patients and deprivation scores for Hull general practices Table 131: Mean age of patients and deprivation scores for East Riding of Yorkshire general practices Table 132: Evidence base for self-care for COPD (Department of Health) Robert Sheikh Iddenden, Public Health Sciences. 14

15 List of figures Figure 1: Health equity audit cycle Figure 2: COPD equity audit framework Figure 3: Wards within Hull Figure 4: Wards within East Riding of Yorkshire Figure 5: Population pyramid for Hull (bars) relative to England (line), Figure 6: Population pyramid for East Riding (bars) relative to England (line), Figure 7: Index of Multiple Deprivation 2007 national quintiles for East Riding of Yorkshire Figure 8: Index of Multiple Deprivation 2007 national quintiles for Hull Figure 9: Dominant ACORN categories by output areas in Hull Figure 10: Dominant ACORN categories by output areas in East Riding of Yorkshire.. 45 Figure 11: Dominant Health ACORN groups by output areas in Hull Figure 12: Dominant Health ACORN groups by output areas in East Riding of Yorkshire50 Figure 13: Prevalence of COPD on General Practice registers 2008/09 for practices in Hull and East Riding, practices sorted in order of increasing mean age of patients as at October Figure 14: Prevalence of COPD on General Practice registers 2007/08 for practices in Hull and East Riding, practices sorted in order of increasing mean age of patients as at October Figure 15: Prevalence of COPD on General Practice registers 2006/07 for practices in Hull and East Riding, practices sorted in order of increasing mean age of patients as at October Figure 16: Prevalence of COPD on General Practice registers 2005/06 for practices in Hull and East Riding, practices sorted in order of increasing mean age of patients as at October Figure 17: Prevalence of COPD on General Practice registers 2004/05 for practices in Hull and East Riding, practices sorted in order of increasing mean age of patients as at October Figure 18: Prevalence of COPD on General Practice registers for patients aged 17+ years, changes over time 2004/05 to 2008/09, for practices in Hull and East Riding, practices sorted in order of increasing mean age of patients(averaged over each year) 61 Figure 19: Monthly emergency admissions with primary diagnosis COPD in Hull and East Riding of Yorkshire residents 2007/08 to 2009/ Figure 20: Plot of age-gender standardised 2007/08 to 2009/10 (pooled) COPD emergency admission rate per 100,000 practice population and the practice level QOF 2008/09 prevalence (%) Figure 21: Plot of age-gender standardised 2007/08 to 2009/10 (pooled) COPD emergency admission rate per 100,000 practice population and the difference between registered and ERPHO modelled prevalence of COPD at practice level, Hull practices only Figure 22: Trends in indirectly standardised mortality ratio for COPD for all ages Figure 23: Monthly deaths where COPD recorded as the primary cause of death, Hull and East Riding of Yorkshire residents 2006 to Figure 24: Smoking status by gender in survey respondents in Hull 2007, East Riding 2009 and England 2008, age-adjusted percentages Robert Sheikh Iddenden, Public Health Sciences. 15

16 Figure 25: Observed and expected COPD prevalence (%), GP practices in Hull (QOF ) sorted by mean age of practice registered patients Figure 26: Relationship between difference between GP registered and ERPHO modelled expected prevalence of COPD for each practice and mean age of registered patients144 Figure 27: Relationship between difference between GP registered and Doncaster modelled prevalence of COPD for each practice and mean age of registered patients145 Figure 28: Age-specific rate of admissions with a primary diagnosis of COPD per 100,000 residents of Hull and East Riding (2007/08 to 2009/10 pooled), by gender Figure 29: Rate of emergency admissions with a primary diagnosis of COPD per 100,000 resident population for Hull and East Riding of Yorkshire (2007/08 to 2009/10 pooled) by age and gender Figure 30: Percentage of men and women that reported they were current smokers or exsmokers in the adult health and lifestyle surveys in Hull 2007 and East Riding Figure 31: Prevalence of diagnosed COPD on practice registers 2008/09 in relation to practice deprivation score (IMD 2007) Figure 32: GP register (2008/09) and modelled (Doncaster and ERPHO models) prevalence of COPD for each practice in order of increasing mean deprivation score (IMD2007, based on October 2008 GP populations) Figure 33: GP register (2008/09) and modelled (Doncaster and ERPHO models) prevalence of COPD for each practice in order of increasing mean deprivation score IMD2007 within PCT, based on October 2008 GP populations) Figure 34: Relationship between deprivation and percentage difference between QOF COPD prevalence and ERPHO model prevalence, Hull and East Riding practices overall Figure 35: Relationship between deprivation and percentage difference between QOF COPD prevalence and ERPHO model prevalence Figure 36: Relationship between deprivation and percentage difference between QOF COPD prevalence and Doncaster model prevalence Figure 37: ACORN types in Hull within Hard Pressed ACORN category Figure 38: ACORN types in East Riding of Yorkshire within Hard Pressed ACORN category Figure 39: Dominant Health ACORN types where the dominant Health ACORN group is Existing Problems by output areas in Hull Figure 40: Dominant Health ACORN types where the dominant Health ACORN group is Existing Problems by output areas in East Riding of Yorkshire Figure 41: Dominant ACORN types (at output area level) where the annual average emergency admission rate from 2007/08 to 2009/10 for COPD was 300 per 100,000 Hull and East Riding residents or higher, by output areas in Hull Figure 42: Dominant ACORN types (at output area level) where the annual average emergency admission rate from 2007/08 to 2009/10 for COPD was 300 per 100,000 residents of Hull and Easter Riding or higher, by output areas in East Riding of Yorkshire Figure 43: Dominant Health ACORN types (at output area level) where the annual average emergency admission rate from 2007/08 to 2009/10 for COPD was 300 per 100,000 Hull and East Riding residents, by output areas in Hull Figure 44: Dominant Health ACORN types (at output area level) where the annual average emergency admission rate from 2007/08 to 2009/10 for COPD was 300 per Robert Sheikh Iddenden, Public Health Sciences. 16

17 100,000 Hull and East Riding residents or greater, by output areas in East Riding of Yorkshire Figure 45: Smoking status of survey respondents (Hull 2007, East Riding 2009) by PCT and deprivation quintile, age-gender-adjusted percentages Figure 46: Locations of smoking cessation services in Hull Robert Sheikh Iddenden, Public Health Sciences. 17

18 1. Abbreviations BLF British Lung Foundation BME Black and Minority Ethnic BMI Body Mass Index BTS British Thoracic Society CCT Controlled Clinical Trial CI Confidence Interval CRT Cluster Randomised Trial COPD Chronic obstructive pulmonary disease EH Eastern Hull (Primary Care Trust) pre 1/10/2006 ERoY East Riding of Yorkshire (Local Authority and PCT) ERPHO Eastern Region Public Health Observatory EY East Yorkshire (Primary Care Trust) pre 1/10/2006 FEV1 Forced expiratory volume in 1 second FVC Forced vital capacity GHS...General Household Survey GP General Practitioner HSE Health Survey for England HES Hospital Episode Statistics HIA Health Impact Assessment HNA Health Needs Assessment ICD International Classification of Diseases IMD Index of Multiple Deprivation KuH Kingston-upon-Hull (Local Authority) LA Local Authority LLSOA Lower Layer Super Output Area LSP Local Strategic Partnership LTOT NCROP Long term oxygen treatment National COPD Resources and Outcomes Project NICE National Institute for Health and Clinical Excellence NIV Non-invasive ventilation NOS Not Otherwise Specified NRT Nicotine Replacement Therapy NSF National Service Framework N&EY&NL North and East Yorkshire and Northern Lincolnshire SHA pre 1/10/2006 ONS Office for National Statistics (formerly OPCS) OPCS Office of Population Censuses and Surveys (now ONS) PBMA Programme Budgeting and Marginal Analysis PCMD...Primary Care Mortality Database PCO Primary Care Organisation PCT Primary Care Trust PHO Public Health Observatory PMSU Prime Minister s Strategy Unit PSA Public Service Agreements RCP Royal College of Physicians QMAS Quality Management and Analysis System QOF Quality and Outcomes Framework RCT Randomised Controlled Trial SHA Strategic Health Authority SMR Standardised Mortality Ratio SOA Super Output Area WH West Hull (Primary Care Trust) pre 1/10/2006 WCC World Class Commissioning YHPHO Yorkshire and Humber Public Health Observatory YHSHA Yorkshire and the Humber SHA YW&C Yorkshire Wolds and Coast (Primary Care Trust) pre 1/10/2006 Robert Sheikh Iddenden, Public Health Sciences. 18

19 2. Equity Audit 2.1. Purpose and Outline A COPD health equity audit aims to inform the planning and delivery of services for reducing inequalities related to COPD in Hull and East Riding of Yorkshire. Equity is now a dimension of the Government s approach to setting national targets and managing performance on health inequalities. Targets aim to see faster progress compared with the average in the most deprived areas (which includes Hull PCT). With respect to the Public Service Agreements (PSA) target for the 2004 Spending Review there is no specific focus on COPD. There are specific targets focused on reducing the inequalities gap for life expectancy. Early deaths from COPD contribute to keeping Hull s life expectancy at birth below that for England. Therefore any change in COPD mortality will influence life expectancy. The PSA targets for life expectancy are: By 2010 increase life expectancy at birth in England to 78.6 years for men and to 82.5 years for women. Reduce health inequalities by 10% by 2010 as measured by infant mortality and life expectancy at birth. There are other PSA targets set against improving outcomes for people with long-term conditions: To improve health outcomes for people with long-term conditions by offering a personalised care plan for vulnerable people most at risk; and to reduce emergency bed days by 5% by 2008, through improved care in primary care and community settings for people with long-term conditions. COPD is one of the key goal areas for Hull s World Class Commissioning (WCC) agenda, and various initiatives around early diagnosis and getting patients on the correct pathways are being developed by NHS Hull. NHS Hull is also working with the Yorkshire and Humber Public Health Observatory (YHPHO) in a Programme Budgeting and Marginal Analysis (PBMA) project for COPD. It is hoped that these can inform the equity audit. The objectives of the COPD equity audit are:- 1. To support effective deployment of services and resources for diagnosing and treating COPD according to need and potential for reducing inequalities. 2. To assess the quality and usefulness of routinely available information about COPD in the local population. Robert Sheikh Iddenden, Public Health Sciences. 19

20 3. To provide a baseline to enable future audits to evaluate the effectiveness, acceptability and usefulness of the new policies in relation to reducing inequity in COPD. Health equity audit would be expected to: - Influence the agreement of key local objectives between partners. Influence change in investment or delivery. Because health equity audit is an iterative process future work should: - Review progress against local objectives. Inform the selection of future health equity audit topics Health Equity Audit Cycle Figure 1 illustrates the health equity audit cycle, and how the public health tools such as Health Needs Assessment (HNA) and Health Impact Assessment (HIA) are related. Figure 1: Health equity audit cycle Public Health Tools for Change: How are they related? Race Equality Impact Assessment HIA: helps identify positive / negative health effects of proposed policy/ / actions Agree partners, 1 Priorities and topics 6 Health Equity Review progress profile against local objectives 2 The Health equity audit cycle HNA: analysis of need; can contribute to equity profile and priorities for action 5 Influence change in investment / delivery 4 Agree key local objectives with partners 3 Use evidence to prioritise effective actions Robert Sheikh Iddenden, Public Health Sciences. 20

21 The health equity audit cycle is not complete until change occurs which reduces health inequalities; therefore it is likely that there will be repetitions of the former steps in future. Health equity audit then, is an iterative process which fits in the planning cycle and conforms to the above structure. This means that this health equity audit is only the first step as subsequent step(s) are required in the future to assess the progress against local objectives. It would be hoped that local objectives could be set in relation to an evidence-based evaluation of what polices work to reduce inequalities. However, it is possible that there is little evidence base for this, and local objectives need to be set in relation to educated supposition. Therefore, a single health equity audit will not necessarily reduce the inequalities observed. So, even if the changes made were successful, there still may be inequalities present (either different inequalities or the same inequalities at a reduced level) and it may be necessary to repeat the health equity audit to agree new actions to reduce inequity further Definition of Health Equity Audit There are a range of meanings and definitions relating to health equity and inequalities. The following definitions of health equity have been adopted for the purposes of this audit. Equity in health is the absence of systematic disparities in health (or in the major social determinants of health) between groups with different levels of underlying social advantage/disadvantage...it is the right to the highest attainable standard of health, as reflected by the standard of health enjoyed by the most socially advantaged group within a society (Department of Health, 2003b). Equity of access (opportunity) to services for equal need involves ensuring that there are appropriate and accessible services for all; and consequently that services address any barriers to access and specific needs relating to particular groups (Department of Health, 2003c). Two primary benchmarks were therefore used to define health inequalities in Hull and East Riding of Yorkshire and examine the potential for improvement: The level of inequality in COPD between Hull and East Riding of Yorkshire populations in comparison with the national average for England. The inequalities in COPD between the different social groups/areas within Hull and East Riding of Yorkshire. Robert Sheikh Iddenden, Public Health Sciences. 21

22 2.4. COPD Equity Audit Context Policy context In the UK there is an increasing policy focus on long-term conditions. As an incurable chronic lung condition that can severely affect quality of life, COPD is recognised as an important long-term condition. COPD has no single definition. It is an umbrella term for a number of chronic lung disorders, including chronic bronchitis, emphysema, chronic obstructive airway disease and chronic airway flow limitation. It is estimated that over 3 million people in England live with COPD, of whom 2 million are undiagnosed, and that 25,000 people die from it every year. COPD is the second most common cause of emergency admission to hospital and the fifth largest cause of readmission to hospital. It is also one of the most costly diseases, in terms of acute hospital care. Most COPD cases are directly attributable to smoking. In this country, as infectious diseases decline, the population ages and cigarette smoking persists, the public health burden of COPD is certain to rise. In 2001, the British Thoracic Society report The burden of lung disease (BTS 2001) was one of the first documents to point out the high levels of respiratory disease in the UK and its potential impact. The Chief Medical Officer focused on smoking in his annual reports for 2002 and In his annual report for 2004, It takes your breath away: the Impact of COPD (CMO 2004) he made a number of recommendations, including the need for a more accurate diagnosis (through an improvement in the standards of spirometry) and more structured care for people with COPD. At the same time he also commissioned a strategic scoping review of lung and respiratory disease. This then led to the decision to take national action on these conditions. The NHS already has some guidance available through the National Service Framework for long-term conditions (Department of Health 2005). This set out a range of quality requirements and key priorities that can be applied to people with COPD and other respiratory conditions. However, as the 2006 report by the Healthcare Commission, Clearing the air (CHAI 2006) highlighted, there remained a need for PCTs and the NHS in general to: improve diagnosis for COPD develop structured care appropriate to people s needs help people manage their condition themselves reduce the number of people admitted to hospital. address the poor prognosis for people with COPD improve access to end-of-life care for people with COPD. Robert Sheikh Iddenden, Public Health Sciences. 22

23 The 2007 report Invisible Lives by the British Lung Foundation (BLF 2007) identified specific communities throughout the UK where 1.9 million people are at high risk of hospital admission with COPD. The British Lung Foundation have also suggested the best communication channels to reach and engage these people, making it possible for those at risk to be diagnosed and treated A National COPD Strategy for England Since 2005 the Department of Health, the NHS, patients, clinicians and other key stakeholders have been working together to identify the best ways to improve care and outcomes for those with COPD and to reduce the overall prevalence of the disease. The draft strategy (Department of Health 2010) was consulted upon up to April 6 th The strategy intends to complement other national and local initiatives to improve prevention, diagnosis and treatment. These include the clinical guidelines developed by the National Institute for Health and Clinical Excellence (NICE) and other broader health promotion initiatives such as the national tobacco control programme. This strategy is intended to help: reduce the risk of developing COPD by improving prevention early identification of those people with or at risk of developing COPD ensure accurate diagnosis and evidence-based treatment through an integrated care pathway ensure better support for people with COPD and their carers provide the best support and integrated treatment for COPD exacerbations provide access to the best available support for those who are at the end of life and for those who are bereaved. Clinical guidance regarding the management of COPD (CG12) was produced by the National Institute for Health and Clinical Excellence in 2004 (NICE 2004), updated (CG101) in 2010 (NICE 2010); and by the Global Initiative for Chronic Obstructive Lung Disease (GOLD), updated in ( National COPD Audits A comprehensive national COPD audit was carried out in 2003 by the British Thoracic Society and the Royal College of Physicians. Mortality rates were worryingly high. An average of 15% of patients died within three months of being admitted to hospital with acute exacerbations of COPD (CMO 2004). The Royal College of Physicians published a further UK wide COPD audit in December 2008, the National COPD Resources and Outcomes Project (NCROP), which covered Robert Sheikh Iddenden, Public Health Sciences. 23

24 the organisation and delivery of care via five reports, each with their own recommendations: 1. UK General Practices 2. Acute NHS units 3. UK primary care organisations 4. Patient survey 5. COPD exacerbations admitted to NHS acute units The 2008 COPD audit was carried out to assess progress since the 2003 National COPD Audit and the 2004 NICE guideline Cost of COPD on the NHS and wider economy The profile of COPD means that it is an expensive disease for the NHS when it is not identified and treated early. The direct cost of COPD to the UK healthcare system is estimated to be between 810 million and 930 million a year; and without change this impact is set to grow. There is a broader economic cost. The annual cost of COPD- related lost productivity to employers and the economy has been put at 3.8 billion. Some 25% of people with COPD are prevented from working due to the disease. COPD causes at least 20.4 million lost working days among men and 3.5 million days among women every year more than any other respiratory condition. COPD is responsible for 1.4 million general practice consultations and the use of one million in-patient bed days every year, as reported in the Chief Medical Officer s Annual Report for 2004 (CMO 2004). In addition there is a social cost. Quality of life is undermined. Problems with restricted mobility may be compounded by social isolation and the psychological conditions that go with it. A survey by the British Lung Foundation found that 90% of people with severe COPD were unable to participate in socially important activities such as gardening, 66% were unable to take a holiday because of their disease and 33% had disabling breathlessness COPD Equity Audit Scope and Framework Scope of COPD Equity Audit The COPD Equity Audit will concentrate on COPD defined on the basis of the International Classification of Disease version 10 (ICD 10) coding. The relevant codes are presented in Table 1. It is anticipated that throughout this report all chronic obstructive pulmonary diseases will be combined. Robert Sheikh Iddenden, Public Health Sciences. 24

25 Table 1: Coverage of equity audit Disease or medical condition Chronic obstructive pulmonary disease Bronchitis Emphysema Other chronic obstructive pulmonary disease ICD 10 codes J40-J44 J40-J42 J43 J44 Further details of the definitions used in the COPD Equity Audit are given in Appendix A on page COPD Equity Audit Framework The framework for the audit is presented in Figure 2. Figure 2: COPD equity audit framework Performance management/targets Context Local strategy Reducing inequalities: health Need health inequalities: Prevalence Mortality Deprivation Risk factors NSF Equity of outcome Equity of access and opportunity Geographical Social groups Deprivation Gender Age Other groups Audit and evaluation The audit framework involved profiling the pattern of need with respect to inequalities in COPD and an audit of the services currently provided. Robert Sheikh Iddenden, Public Health Sciences. 25

26 Profiling the Pattern of Need With Respect to Inequalities in COPD The prevalence of COPD is the foremost measure of need; however, because of problems in obtaining comprehensive and analysable prevalence data (both overall and with respect to inequalities), other indicators were considered in order to assess need. The factors considered include risk factors for COPD, GP registers for COPD, estimated prevalence of COPD using a model, prescribing, day case and inpatient hospital admission and mortality. However, not all information was available from all of these sources (both overall and with respect to inequalities), and it was not possible to examine all of these in detail. This occurred in particular for prescribing. Different groups of individuals will be examined in relation to inequity, such as differences between the genders, age groups, ethnic groups and deprivation, etc. Deprivation, as measured by the Index of Multiple Deprivation (IMD) 2007, is used as the main measure of deprivation at a small geographical area level. Details of how the IMD scores were calculated, how the scores were analysed within this report and the local scores are given in section 7.3 on page Equity Audit of Services There was an intention to examine evidence that there was differential access of services locally by different groups of individuals by assessing the relationship between different groups and use of services, and an intention to examine the quality of available datasets and evidence. If this assumption was correct it could be assumed that development of the service could contribute to reducing inequalities in COPD for people of Hull and East Riding. It was recognised from the start, however, that there may be challenges brought about by non-availability of data. The quality of care will also be examined. Although a National Services Framework (NSF) for COPD has yet to be published, NICE guidelines were published in 2004 and revised in This has been undertaken primarily within the section Potential Programmes for Reducing Inequalities in COPD starting on page 213. Robert Sheikh Iddenden, Public Health Sciences. 26

27 3. COPD and Risk Factors for COPD 3.1. Chronic Obstructive Pulmonary Disease Chronic obstructive pulmonary disease (generally referred to as COPD throughout this report) is an umbrella term for a range of diseases of the lung, including bronchitis, emphysema and chronic obstructive airways disease. COPD is one of the most common respiratory diseases in the UK. The main symptom of COPD is a restriction on the ability to breathe properly. Much COPD is diagnosed in middle age in smokers or ex-smokers with a persistent chesty cough (especially in the morning), breathlessness on slight exertion or persistent coughs and colds in the winter. It is diagnosed by a simple breathing test. The 2004 NICE guideline from the National Institute for Health used the following definition of COPD: Chronic obstructive pulmonary disease (COPD) is characterised by airflow obstruction. The airflow obstruction is usually progressive, not fully reversible and does not change markedly over several months. The disease is predominantly caused by smoking. Airflow obstruction is defined as a reduced FEV1 (forced expiratory volume in 1 second) and a reduced FEV1/FVC ratio (where FVC is forced vital capacity), such that FEV1 is less than 80% predicted and FEV1/FVC is less than 0.7. The airflow obstruction is due to a combination of airway and parenchymal damage. The damage is the result of chronic inflammation that differs from that seen in asthma and which is usually the result of tobacco smoke. Significant airflow obstruction may be present before the individual is aware of it. COPD produces symptoms, disability and impaired quality of life which may respond to pharmacological and other therapies that have limited or no impact on the airflow obstruction. COPD is now the preferred term for the conditions in patients with airflow obstruction who were previously diagnosed as having chronic bronchitis or emphysema. Robert Sheikh Iddenden, Public Health Sciences. 27

28 Other factors, particularly occupational exposures, may also contribute to the development of COPD. There is no single diagnostic test for COPD. Making a diagnosis relies on clinical judgement based on a combination of history, physical examination and confirmation of the presence of airflow obstruction using spirometry. (NICE guideline No. 12 (NICE 2004) page 1) 3.2. Complications of COPD People with COPD may often be short of breath, have a persistent cough, and may have a build up of phlegm and mucus in the throat. Many patients with severe disease carry a supply of oxygen with them at all times, as exertional breathlessness can be severe. Emergency admissions to hospital are common in those with acute exacerbations, representing a significant worsening of prognosis (NICE 2004) with one study showing 34% were readmitted and 14% died within 3 months (Roberts et al, 2002) while in a Spanish study 63% were readmitted within 1-year (Garcia-Aymerich et al) Risk Factors for COPD The main risk factor for COPD is smoking. Smoking is responsible for around 80% of COPD (NHS Choices). The risk of developing the disease increases the longer that an individual smokes. The reduction of lung function that naturally occurs around the age of is speeded up, by up to three times, in smokers (NHS Choices). COPD tends to develop in middle age due to the cumulative effect of smoking, so is very rare under the age of 40. Less common causes are passive smoking, pollution (this was a more important factor before the clean air act reduced the incidence of smog in the large UK cities), fumes and dust, genetic susceptibility to COPD. While not a risk factor for developing the disease, being overweight or obese can make the condition worse, as well as leading to other conditions such as diabetes and heart disease (NHS Choices) Confounders and Combined Risk Factors When examining potential risk factors it is possible that they could be confounders rather than having a direct influence on COPD. For example, there is a strong association between deprivation and the risk of developing COPD, and if there was a strong association between different general practices and deprivation, then an association could exist between general practices and COPD. However, if such an association was found, Robert Sheikh Iddenden, Public Health Sciences. 28

29 it could be because COPD is indeed related to the particular general practices (a true relationship which is unlikely) or it could be because the risk of developing COPD increased with deprivation (a confounder, i.e. COPD and general practices are only associated indirectly through their association with deprivation). Age is frequently a confounder. The effect of having two or more risk factors is not necessarily additive. For example, singly factor A (for example COPD) doubled the risk of dying from respiratory disease and singly factor B (smoking) doubled the risk of dying from respiratory disease. If the effects were additive then one would expect that having both A and B present would increase the risk by four times. However, having both factors A and B present together may increase the risk of dying of respiratory disease tenfold (i.e. increase by more than expected based on the sum of the single effects) or double the risk of poor health (i.e. increase by less than expected based on the sum of the single effects) Causality Just because an association between two different factors occurs does not mean that one causes the other. If the factors are measured at the same time (for example, in a crosssectional survey), then it could be that both could have influences on each other. For example, people with mental health problems could live in deprived areas because they are more likely to not have a job and have lower incomes or live in supported accommodation, and/or people living in more deprived areas may be more likely to have mental health problems due to additional stress in their lives. However, more importantly, an association could exist through chance or due to a confounder. There are a number of arguments that can strengthen the likelihood of causality, for instance, a dose response (heavy smokers more likely to get lung cancer than light smokers), other studies finding similar results, biological plausibility, etc. Just observing a relationship is not sufficient to deduce causality. Robert Sheikh Iddenden, Public Health Sciences. 29

30 4. Issues in Assessing COPD 4.1. Assessing the True Effect of COPD It may be difficult to assess the true effect of COPD on the National Health Service owing to the coding of diagnoses. It is likely that a number of health contacts and admissions occur for patients who have COPD which would not have occurred if the patient had not had COPD. However, it is likely that not all such health contacts would have been coded as related to COPD. It is hoped that where a patient is admitted for a condition other than COPD that COPD would have been coded as a secondary diagnosis, but this may not always be the case Applying National Prevalence Data to Local Population Where it is not possible to obtain local estimates of the prevalence of COPD, the only way to obtain an estimate of the number of people with COPD in Hull and East Riding is to estimate the number based on the national prevalence and local resident population. This makes the assumption that the national prevalence applies to the local population. This may not be the case, and this needs to be considered when interpreting the results. It is likely that Hull has a higher prevalence compared to the national figures or East Riding of Yorkshire due to its increased deprivation. The Eastern Region Public Health Observatory (PHO) created a model which predicted COPD prevalence at local authority level for English local authorities, and was updated in October The model has been used within section on page 66 to predict the expected number of residents with the disease (whether or not diagnosed) within Hull and East Riding. The model derived estimates based on age, gender, ethnicity (each from 2005 estimates provided by ONS), smoking prevalence (synthetic estimates derived from Health Surveys for England), prevalence of ex-smokers (national prevalence from Health Surveys for England applied to each local authority), rurality and deprivation scores (Index of Multiple Deprivation 2004). Projected COPD prevalence estimates up to 2020 were also made, using the ONS 2006-based population projections, along with projections to The use of ethnic distributions in the projections is best explained in the Information worksheet of the Excel workbook containing the projections: The 2005 ethnic distribution is used for the projections. The distribution of ethnic groups within the population is lagged by 5, 10 and 15 years for the 2010, 2015, and 2020 projections respectively. So, for example, in 2015 the ethnic distribution in the age group is calculated using the ethnic proportions from the age group in ERPHO 2008 Robert Sheikh Iddenden, Public Health Sciences. 30

31 5. COPD and World Class Commissioning As part of the NHS Hull s World Class Commissioning agenda, COPD was chosen as one of the 8 priority goal areas to receive additional funding to help improve outcomes. The over-arching goal of the COPD strategy group was to increase the recorded prevalence of COPD on general practice registers by 30% over the next 5 years, and to ensure that all diagnosed patients are on the COPD care pathway with an active management plan. A number of initiatives were designed to achieve this goal, and so to reduce the risk of admissions into secondary care for these patients. The following was extracted from Hull PCT s World Class Commissioning Strategy Document submitted in Initiative 7: COPD Prevention, Detection and Diagnosis 7a Prevention All patients over 35 who smoke or who have smoked are screened for COPD in primary care and offered smoking cessation 7b Detection A public awareness campaign to target high risks groups with Information about COPD symptoms, risks of smoking and services available 7c Diagnosis All GP practices to perform spirometry as part of the COPD diagnostic pathway in all patients over 35 who smoke or who have smoked and present with exertional breathlessness, chronic cough, regular sputum production and/or frequent winter bronchitis Initiative 8: COPD Treatment and Rehabilitation 8a Treatment Target practices with low prevalence of recorded COPD and high emergency admissions to optimise treatment as per staged approach to care on COPD pathway in line with Year of Care model and care conductor role 8b Implement the COPD pathway introducing staged management of care across primary and secondary care including existing Physio Outreach post Robert Sheikh Iddenden, Public Health Sciences. 31

32 8c Increase access to LTC support OOH episode avoidance schemes with GP practices including patient self management plans 8d Rehabilitation Provide pulmonary rehabilitation that is equitable across Hull at times and venues suitable for the patient within a reasonable time of referral providing programmes that are multi component and multi disciplinary. - Hull PCT World Class Commissioning Strategy These initiatives were refined in the subsequent refresh of the World Class Commissioning Strategy, submitted in February 2010, with the following extracted from the submitted document. Primary Prevention 7a 1 Increase public understanding of COPD and encourage smokers to access the Hull Stop Smoking Services and quit smoking. Diagnosis and Detection 7b 2 All patients over 35 who smoke or who have smoked and have symptoms will be screened using spirometry and referred to Hull stop smoking services. Treatment and Management 8a 8b 3 8c 4 Identify practices with low COPD prevalence Transform the care pathway. Evaluate the increase in LTC support out of hours and increase the number of admission avoidance schemes. 1 This deliverable has been updated following social marketing work carried out in 2008/09 to inform a public awareness campaign scheduled to take place in Hull in Jan The deliverable has been adjusted following feedback from patients and the Hull Stop Smoking Services undertaking the screening in 2009/10. it now includes criteria to assist in making a clinical judgement on which patients to screen and is consistent with NICE guidance and the General Practice note book. 3 This deliverable has been changed to follow on from the implementation of the staged management approach in primary care and to reflect the need to concentrate on the treatment and management of the moderate to severe COPD patient in order to realise efficiencies in the current system and optimise care. 4 This deliverable has been updated to ensure the LTC increase is effective and self management is optimised. The deliverable has also been extended to further develop the role of PBC in relation to the management of COPD. Robert Sheikh Iddenden, Public Health Sciences. 32

33 Rehabilitation 8d Provide pulmonary rehabilitation that is equitable across Hull. - Hull PCT World Class Commissioning Strategy Sources of Data, Definitions and Statistical Methods Details of sources of data, definitions and statistical methods are given in Appendix A starting on page 274. The following is included: Sources of Data includes data from national surveys, various local surveys in Hull and East Riding of Yorkshire and information on the Quality and Outcomes Framework. Robert Sheikh Iddenden, Public Health Sciences. 33

34 Definitions and Classifications Used includes definitions used for risk factors in the local Health and Lifestyle survey and classifications of diseases based on the International Classification of Diseases version 10. Statistical Methods and Terms includes explanations of standardisation, significance testing, confidence intervals and the problem of small numbers. Robert Sheikh Iddenden, Public Health Sciences. 34

35 7. Kingston-upon-Hull and East Riding of Yorkshire 7.1. Geographical Area Kingston-upon-Hull will generally be referred to as Hull, while East Riding of Yorkshire will generally be referred to as East Riding and abbreviated to ERoY within this report. Hull Teaching Primary Care Trust (PCT) and East Riding of Yorkshire PCT are both within the Yorkshire and the Humber Strategic Health Authority (SHA). The two PCT boundaries are coterminous with the two local authorities. Hull is made up of 23 wards which are illustrated in Figure 3, while East Riding of Yorkshire is made up of 26 wards which are illustrated in Figure 4. Figure 3: Wards within Hull Robert Sheikh Iddenden, Public Health Sciences. 35

36 Figure 4: Wards within East Riding of Yorkshire 7.2. Population There are different estimates of population available from different sources. Official estimates by the Office for National Statistics (ONS) are derived from the 2001 Census population updated for births, deaths and migration (updated mid-year population estimates available annually). Alternatively, population estimates can be obtained from the number of patients registered with General Practitioners (GPs). Local population estimates are available from both sources, and the number of residents from the GP registrations is usually slightly higher than the figures from ONS. In this report, population figures and analyses involving the calculation of rates, for instance mortality rates, will generally use the GP registration file (either based on residence of patient or location of GP depending on the context). Based on GP registration data for October 2009 (Table 2), it can be seen that 10% (28,394 out of 286,967) of the patients who are registered with a Hull GP live in East Riding of Yorkshire (ERoY). Fewer patients among those registered with East Riding GPs Robert Sheikh Iddenden, Public Health Sciences. 36

37 lived in Hull (4,179; 1.3%). This is an important consideration with respect to access to services. Table 2: Estimated registered and resident population as at October 2009 Age (yrs) Local Authority of based on GP practice location and residence of patients, 2009 ERoY GP, Hull GP, ERoY GP Hull GP ERoY Hull GP, GP outwith GP outwith resident resident and and GP, Hull ERoY area, ERoY area, Hull outwith outwith resident resident resident resident resident resident area area ,466 16, , ,078 13, , , ,684 15, , , ,268 17, , , ,143 22, , , ,548 20, , , ,814 17, , , ,679 18, , , ,656 18, , , ,206 18, , , ,133 15, , , ,709 14, , , ,834 13, , , ,122 9, , , ,682 8, , , ,211 7, , , ,788 5, , ,040 4, ,485 Total 306, ,530 4,179 28, , ,472 Total Population pyramids for resident populations 2009 are shown in Figure 5 for Hull and Figure 6 for East Riding. The age profile of population of Hull s population is similar to that for England, although people aged in their twenties formed a larger proportion of Hull s population, due to Hull being a university city. The age profile of East Riding s population is older than that for England. Robert Sheikh Iddenden, Public Health Sciences. 37

38 Figure 5: Population pyramid for Hull (bars) relative to England (line), 2009 Males Females 90 and over Under Population (thousands) *England data from Office for National Statistics Population Estimates Unit. Crown Copyright Figure 6: Population pyramid for East Riding (bars) relative to England (line), 2009 Males Females 90 and over Under Population (thousands) *England data from Office for National Statistics Population Estimates Unit. Crown Copyright Robert Sheikh Iddenden, Public Health Sciences. 38

39 7.3. Deprivation The Index of Multiple Deprivation (IMD) 2007 (Communities and Local Government, 2007) has been calculated for small geographical areas, lower layer super output areas 5 (LLSOAs), of around 1,500 people. The IMD 2007 measures multiple deprivation across seven domains and the scores on each domain are weighted and combined to give the final index score for each LLSOA (Table 3). The IMD 2007 is an update of the IMD 2004 (Communities and Local Government, 2004), and the domains and weights are the same for each The IMD, whether 2004 or 2007, is not as good at measuring affluence. Table 3: Domains of the Index of Multiple Deprivation 2007 and their weights Domain of IMD 2007 Domain weight (%) Income deprivation 22.5 Employment deprivation 22.5 Health deprivation and disability 13.5 Education, skills and training deprivation 13.5 Barriers to housing and services 9.3 Living environment deprivation 9.3 Crime 9.3 It is possible to combine LLSOAs into larger geographical areas by weighting the IMD 2007 scores by the population of the LLSOA. IMD 2007 scores have been produced for local authorities, wards and general practice populations. In the IMD 2007, Hull is ranked as the 11 th most deprived local authority of England whereas East Riding of Yorkshire is ranked as the 232 nd most deprived local authority (out of 354). So Hull is in the most deprived 4% of all local authorities in England and East Riding is in the most deprived 66% of local authorities. The IMD 2007 scores for all of England s LLSOAs and wards have been divided into five approximately equal-sized groups ranging from the 20% most deprived areas to the 20% least deprived areas. These groups are referred to as national quintiles 6. The national quintiles give an indication of how the Hull and East Riding s LLSOAs and wards compare with England in terms of deprivation. Figure 7 illustrates the IMD 2007 deprivation national quintiles for each LLSOA within East Riding. The darkest areas in red are within the top 20% most deprived areas nationally and lightest areas in white are within the top 20% least deprived areas nationally. It can be seen that there are pockets of deprivation in East Riding, which include areas within Goole, and coastal areas within Bridlington and 5 LLSOAs are a geographical unit designed to improve the reporting of small area statistics. They were introduced initially for use on the Neighbourhood Statistics (NeSS) website, but with the intention that they would eventually become the standard across National Statistics. 6 The IMD 2007 score for wards can also be estimated from the average score for each LLSOA within the ward (weighting by the population within each LLSOA), and these ward scores can also be grouped into quintiles (either national or local). Robert Sheikh Iddenden, Public Health Sciences. 39

40 Withernsea. Six percent of the 209 LLSOAs within East Riding are in the most deprived 20% of LLSOAs nationally, 13% in the second most deprived quintile, 20% in the middle quintile, 27% in the second least deprived quintile and 34% in the least deprived quintile. As can be seen in Figure 8, there is a much higher level of deprivation within Hull. Fiftythree percent of the 163 LLSOAs in Hull are in the most deprived 20% nationally, 25% in the second most deprived 20% nationally, 15% in the middle deprivation quintile and 7% in the second least deprived quintile. None of Hull s 163 SOAs are within the least deprived 20% nationally. Figure 7: Index of Multiple Deprivation 2007 national quintiles for East Riding of Yorkshire Robert Sheikh Iddenden, Public Health Sciences. 40

41 Figure 8: Index of Multiple Deprivation 2007 national quintiles for Hull Table 4 gives the IMD scores for the wards of Hull. The higher the IMD score, the worse the deprivation. The local ranks are provided with a value of one denoting the most deprived ward locally and 23 denoting the least deprived ward locally. The national ranking is also given, and similarly the lower the rank the greater the level of deprivation. The national rank percentile is provided. For example, if the percentile is 10, it means that the ward is in the worst 10% of deprived wards nationally, and a value of 75 means that the ward is in the top 75% of deprived wards or alternatively the top 25% least deprived wards. Whilst 11% and 6% of Hull s LLSOAs were in the middle deprivation quintile and second least deprived quintile respectively, as illustrated in Figure 8, there were no wards in these deprivation categories when the LLSOAs were combined into wards. The reason for this is that the wards that contain less deprived areas (with wards illustrated by thicker Robert Sheikh Iddenden, Public Health Sciences. 41

42 black lines in Figure 8), also contain areas which are more deprived, so averaging the deprivation score across the entire ward averages out the extremes 7. Table 4: Index of Multiple Deprivation 2007 scores and ranks of wards in Hull Ward Index Score Local Rank (out of 23) National Rank (out of 7,932) National Percentile St Andrew's Orchard Park and Greenwood Myton Southcoates East Marfleet Bransholme East Newington Bransholme West Longhill Drypool University Pickering Newland Avenue , Southcoates West , Ings , Sutton , Derringham , Boothferry , Holderness , Beverley , Bricknell , Kings Park , Table 5 gives the equivalent data for the wards in East Riding Table 5: Index of Multiple Deprivation 2007 scores and ranks of wards in East Riding of Yorkshire Ward Name IMD Score Local Rank (out of 26) National Rank (Out of 7,932) Percentile National Rank Bridlington South Goole South , Bridlington Central and Old Town , South East Holderness , Goole North , For instance, in Boothferry ward, one LLSOA is in the second least deprived quintile, one in the middle quintile, five in the second most deprived quintile and one in the most deprived quintile so overall the ward is in the second most deprived quintile. Robert Sheikh Iddenden, Public Health Sciences. 42

43 North Holderness , Bridlington North , East Wolds and Coastal , Snaith, Airmyn, Rawcliffe & Marshland , Hessle , Driffield and Rural , Cottingham South , Minster and Woodmansey , Mid Holderness , Tranby , Howdenshire , South West Holderness , Wolds Weighton , Howden , St Mary's , Cottingham North , Pocklington Provincial , Beverley Rural , Dale , Willerby and Kirk Ella , South Hunsley , IMD 2007 deprivation scores for general practices are given in Appendix B on page Geodemographic segmentation An alternative method of examining deprivation in small areas is to utilise one of the geodemographic segmentation tools on the market. A number of geodemographic segmentation tools are available, of which Hull PCT currently holds a licence for the CACI Insite software, from which ACORN and Health ACORN classifications may be produced. The dominant ACORN categories (the highest level ACORN classification) at output area level are shown in Figure 9 for Hull and Figure 10 for East Riding of Yorkshire. 44% of Hull residents lived in output areas classified as being Hard Pressed, the most deprived ACORN category, while only 3% of Hull s residents lived in output areas classified as Wealthy Achievers, the least deprived ACORN category. By contrast, 42% of East Riding of Yorkshire residents lived in output areas classified as Wealthy Achievers while just 11% lived in output areas classified as Hard Pressed. Robert Sheikh Iddenden, Public Health Sciences. 43

44 Figure 9: Dominant ACORN categories by output areas in Hull Robert Sheikh Iddenden, Public Health Sciences. 44

45 Figure 10: Dominant ACORN categories by output areas in East Riding of Yorkshire Of course, if we were to look at smaller areas within East Riding we would see pockets with much higher concentrations of Hard Pressed while in Hull there will be areas that are fairly diverse in terms of ACORN categories. To this end the percentage of residents in each ACORN category within each ward are presented in Table 6 for Hull and for East Riding of Yorkshire. Seven wards in Hull have some output areas classified as Wealthy Achievers, although at 29%, Kings Park is the only ward in Hull with more than 10% of its residents living in these output areas. In East Riding of Yorkshire, eight wards had more than one fifth of its residents living in output areas defined as Hard Pressed, with the highest percentages in wards in Bridlington and Goole. Of course, within each of the four headline ACORN categories there are many subgroups, with 17 ACORN groups and 56 ACORN types defined. These are also shown in Table 6, together with the percentage of residents of Hull and East Riding of Yorkshire PCTs overall within each ACORN category, group and type. Within Hull, the most common Robert Sheikh Iddenden, Public Health Sciences. 45

46 ACORN group was 5.N Struggling Families with 28% of Hull residents residing in these output areas, while the most common ACORN type was 5.N.47 Low Income Families, Terraced Estates, with 14% of Hull residents living in these output areas. This was one of three ACORN types that each accounted for more than 10% of Hull residents, the others being 4.M.42 Home Owning Families, Terraces (14%) and 5.O.51 Single Parents and Pensioners, Council Terraces (10%). East Riding has a more heterogeneous population, with no ACORN type accounting for more than 10% of its residents. The most common ACORN type was 1.C.11 Well-Off Managers, Detached Houses with almost 9% of East Riding residents, one of only four ACORN types that accounted for more than 5% of East Riding residents. The most common ACORN groups were 1.B Affluent Greys and 3.H Secure Families each accounting for almost 16% % of East Riding residents. 1,438 residents of Hull and 3,375 residents of East Riding of Yorkshire live in output areas that are unclassified in terms of ACORN so do not feature in Table 6. Table 6: ACORN classifications (based on dominant ACORN classifications at output area) for Hull and East Riding of Yorkshire, October 2008 population Number and percent of residents ACORN classifications (Category, Group and Type) by ACORN classification Hull East Riding N % N % 1.A.1 Wealthy Mature Professionals, Large Houses , A.2 Wealthy Working Families with Mortgages , A.3 Villages with Wealthy Commuters 0-7, A.4 Well-Off Managers, Larger Houses 0-15, A Wealthy Executives , B.5 Older Affluent Professionals 0-7, B.6 Farming Communities 0-20, B.7 Old People, Detached Homes 0-9, B.8 Mature Couples, Smaller Detached Homes 0-15, B Affluent Greys 0-53, C.9 Older Families, Prosperous Suburbs , C.10 Well-Off Working Families with Mortgages 5, , C.11 Well-Off Managers, Detached Houses 0-29, C.12 Large Families and Houses in Rural Areas 0-1, C Flourishing Families 6, , Wealthy Achievers 7, , D.13 Well-Off Professionals, Larger Houses and Converted Flats 1, D.14 Older Professionals in Suburban Houses and Apartments , D Prosperous Professionals 1, , E.15 Affluent Urban Professionals, Flats E.16 Prosperous Young Professionals, Flats E.17 Young Educated Workers, Flats 1, E.18 Multi-Ethnic Young, Converted Flats E.19 Suburban Privately Renting Professionals 1, E Educated Urbanites 2, Robert Sheikh Iddenden, Public Health Sciences. 46

47 Number and percent of residents ACORN classifications (Category, Group and Type) by ACORN classification Hull East Riding N % N % 2.F.20 Student Flats and Cosmopolitan Sharers 3, F.21 Singles and Sharers, Multi-Ethnic Areas F.22 Low Income Singles, Small Rented Flats 4, , F.23 Student Terraces 2, F Aspiring Singles 10, , Urban Prosperity 15, , G.24 Young Couples, Flats and Terraces 1, G.25 White Collar Singles and Sharers, Terraces 4, , G Starting Out 5, , H.26 Younger White Collar Couples with Mortgages 4, , H.27 Middle Income, Home Owning Areas 0-10, H.28 Working Families with Mortgages 6, , H.29 Mature Families in Suburban Semis 5, , H.30 Established Home Owning Workers 13, , H.31 Home Owning Asian Family Areas H Secure Families 29, , I.32 Retired Home Owners 1, , I.33 Middle Income, Older Couples 1, , I.34 Lower Incomes, Older People, Semis 6, , I Settled Suburbia 9, , J.35 Elderly Singles, Purpose Built Flats 0-2, J.36 Older People, Flats , J Prudent Pensioners , Comfortably Off 44, , K.37 Crowded Asian Terraces K.38 Low Income Asian Families K Asian Communities L.39 Skilled Older Families, Terraces 12, , L.40 Young Working Families 11, , L Post Industrial Families 24, , M.41 Skilled Workers, Semis and Terraces 5, , M.42 Home Owning Families, Terraces 36, , M.43 Older People, Rented Terraces 12, , M Blue Collar Roots 54, , Moderate Means 78, , N.44 Low Income Larger Families, Semis 1, , N.45 Low Income, Older people, Smaller Semis 4, , N.46 Low Income, Routine Jobs, Terraces and Flats N.47 Low Income Families, Terraced Estates 37, , N.48 Families and Single Parents, Semis and Terraces 18, , N.49 Large Families and Single Parents, Many Children 13, N Struggling Families 74, , O.50 Single Elderly People, Council Flats 5, , O.51 Single Parents and Pensioners, Council Terraces 26, , Robert Sheikh Iddenden, Public Health Sciences. 47

48 Number and percent of residents ACORN classifications (Category, Group and Type) by ACORN classification Hull East Riding N % N % 5.O.52 Families and Single Parents, Council Flats O Burdened Singles 32, , P.53 Old People, Many High Rise Flats 5, , P.54 Singles and Single Parents, High Rise Estates 2, P High Rise Hardship 8, , Q.55 Multi-Ethnic Purpose Built Estates Q.56 Multi-Ethnic, Crowded Flats Q Inner City Adversity Hard Pressed 115, , Total 262, , While not a measure of deprivation, the Health ACORN geo-demographic profiles indicate where current and future health needs are likely to be greatest. The Health ACORN groups (the top level classification) are shown for Hull in Figure 11 and East Riding in Figure 12. The percentages of residents of Hull and East Riding in each of these Health ACORN groups, as well as in the 25 Health ACORN types, are shown in Table 7. 22% of Hull residents and 12% of East Riding residents were defined as having Existing Problems. These tend to be areas with a higher than average proportion of elderly people (14% in Hull and 15% in East Riding aged 75 years and over). A further 36% of Hull residents and 5% of East Riding residents were classified as Future Problems. These tend to be in areas where risk factors such as smoking, poor diet and sedentary lifestyles are higher than average. 10% of Hull residents and 38% of East Riding residents were classified as Future Possible Concerns These included areas with more affluent elderly people, as well as areas with lifestyle risk factors, including type 3.4 Affluent professionals, high alcohol consumption, dining out. 31% of Hull residents and 44% of East Riding residents were classified as Healthy. These areas generally have a younger than average age profile (particularly in Hull with its university students), or are smaller communities (especially in East Riding of Yorkshire). 1,438 residents of Hull and 3,375 residents of East Riding of Yorkshire live in output areas that are unclassified in terms of Health ACORN so do not feature in Table 7. Robert Sheikh Iddenden, Public Health Sciences. 48

49 Figure 11: Dominant Health ACORN groups by output areas in Hull Robert Sheikh Iddenden, Public Health Sciences. 49

50 Figure 12: Dominant Health ACORN groups by output areas in East Riding of Yorkshire Table 7: Health ACORN classifications (based on dominant ACORN classifications at output area) for Hull and East Riding of Yorkshire, October 2008 population Number and percent of residents Health ACORN classifications (Group and Type) by Health ACORN classification Hull East Riding N % N % 1.1 Older couples, traditional diets, cardiac issues 2, , Disadvantaged elderly, poor diet, chronic health 2, Vulnerable disadvantaged, smokers with high levels of obesity 4, Post industrial pensioners with long term illness 8, , Deprived neighbourhoods with poor diet, smokers 13, , Elderly with associated health issues 7, , Home owning pensioners, traditional diets 2, , Disadvantaged neighbourhoods with poor diet and severe health issues 17, , Robert Sheikh Iddenden, Public Health Sciences. 50

51 Number and percent of residents Health ACORN classifications (Group and Type) by Health ACORN classification Hull East Riding N % N % Existing problems 58, , Poor single parent families with lifestyle related illnesses 16, , Multi-ethnic, high smoking, high fast food consumption 12, , Urban estates with sedentary lifestyle and low fruit and vegetable consumption 37, , Deprived multi-ethnic estates, smokers and overweight 12, , Disadvantaged multi-ethnic younger adults, with high levels of smoking 17, , Future Problems 95, , Less affluent neighbourhoods, high fast food, sedentary lifestyles 5, , Affluent healthy pensioners, dining out 0-15, Home owning older couples, high level of fat and confectionary 1, , Affluent professionals, high alcohol consumption, dining out 1, , Low income families with some smokers 14, , Affluent families with some dietary concerns 3, , Possible Future Concerns 26, , Young mobile population with good health and diet 18, , Younger affluent, healthy professionals 6, , Students and young professionals, living well 5, , Towns and villages with average health and diet 26, , Mixed communities with better than average health 12, , Affluent towns and villages with excellent health and diet 10, , Healthy 80, , Total 262, , General Health Table 8 uses local individual patient-level mortality data and GP registration population estimates to calculate the age standardised mortality ratios (SMR) for all causes for persons aged less than 75 years for each ward in Hull ( pooled). This is a ratio of all deaths observed in a population to all deaths expected in that population, where the deaths expected are those that would have occurred had the age-specific rates in a reference population applied. In this way, the ratios are adjusted to take into account differences in the age and gender structure of the population (see page 279 for further explanation). The reference population in this case is England, and the age-specific reference rates are also from If the SMR for a particular geographical area is the same as England then the SMR will be 100. A value higher than 100 denotes higher all-cause mortality than England and a value lower than 100 denotes lower all-cause mortality than England. The overall mortality rate for Hull was 33% higher than the mortality rate in England, with the difference between Hull and England greater in women (SMR 136) than in men (SMR 130), in each case being statistically significant. Considering males and females combined, the majority of Hull wards had higher mortality Robert Sheikh Iddenden, Public Health Sciences. 51

52 than England, with mortality rates statistically significantly higher than for England among persons in 13 wards, while only 1 ward, Beverley, had mortality rates that were statistically significantly lower than England 8. Rates in Myton and St. Andrews wards were double those seen in England. Amongst males, 11 wards had SMRs that were statistically significantly above 100, with only one statistically significantly below 100. Amongst females 13 wards had statistically significantly high SMRs and none had SMRs that were statistically significantly below 100. Table 8: Age-standardised mortality ratio from all causes for those <75 years for Hull wards, pooled Ward Name Age-standardised mortality ratio for persons aged <75 years (95% CI), pooled Males Females Persons Avenue 97 ( 74,124) 118 ( 85,159) 106 ( 87,128) Beverley 66 ( 47, 91) 94 ( 65,132) 78 ( 60, 98) Boothferry 83 ( 64,108) 80 ( 56,110) 82 ( 67,100) Bransholme East 147 (113,187) 155 (113,208) 150 (123,181) Bransholme West 154 (121,193) 140 (102,187) 149 (124,178) Bricknell 96 ( 70,129) 68 ( 42,104) 85 ( 66,109) Derringham 102 (79,130) 126 ( 95,163) 112 ( 93,133) Drypool 144 (117,176) 91 ( 63,126) 127 (106,150) Holderness 87 ( 67,110) 84 ( 60,115) 86 ( 70,104) Ings 102 ( 81,128) 145 (114,181) 119 (101,140) Kings Park 111 ( 80,150) 68 ( 39,111) 95 ( 72,123) Longhill 132 (105,164) 147 (112,189) 138 (116,163) Marfleet 167 (137,201) 158 (122,200) 163 (140,189) Myton 197 (169,229) 225 (182,276) 214 (189,242) Newington 171 (141,206) 175 (135,223) 175 (151,203) Newland 107 ( 78,143) 115 ( 75,168) 112 ( 87,141) Orchard Park & Greenwood 151 (124,183) 182 (145,225) 164 (141,189) Pickering 144 (117,175) 150 (117,189) 146 (124,169) St Andrew's 210 (170,256) 201 (149,265) 211 (178,248) Southcoates East 168 (132,212) 199 (151,258) 181 (151,215) Southcoates West 105 ( 77,141) 131 ( 91,183) 116 ( 92,145) Sutton 121 ( 97,148) 137 (106,176) 128 (109,150) University 107 ( 80,140) 143 (105,191) 121 ( 99,148) Hull 130 (124,136) 136 (129,145) 133 (128,138) Table 9 gives the equivalent information for East Riding of Yorkshire. The overall under 75 all-cause mortality rate in East Riding of Yorkshire was 8% lower than for England. The difference was greater for men (SMR 89) than for women (SMR 95), with the 8 If the 95% CI does not include 100 then the rate differs significantly than the national rate. However, one would expect 5% of statistical comparisons to be classified statistically significant by chance even when there was no difference between the local rate and the national rate so this must be borne in mind. Robert Sheikh Iddenden, Public Health Sciences. 52

53 differences for men and persons statistically significant. Considering males and females combined, only seven wards had higher mortality rates among men and women combined than England, of which four (Bridlington Central and Old Town, Bridlington South, Goole South and South East Holderness) had mortality rates that were statistically significantly higher than England 9. Eight wards had mortality rates that were statistically significantly lower than England. Six wards had mortality rates in men that were statistically significant below those in England, one ward having a statistically significantly high SMR. Three wards had mortality rates in women statistically significantly lower than England, while a further three wards had statistically significantly high female SMRs. Table 9: Age-standardised all cause under 75 mortality ratio for East Riding of Yorkshire, Ward Name Age-standardised mortality ratio for persons aged <75 years (95% CI), Males Females Persons Beverley Rural 63 ( 48, 81) 86 ( 64,113) 73 ( 60, 87) Bridlington Central and Old Town 118 ( 94,146) 130 (100,166) 123 (104,144) Bridlington North 109 ( 91,129) 101 ( 80,125) 105 ( 92,121) Bridlington South 130 (109,155) 139 (111,171) 134 (117,153) Cottingham North 67 ( 48, 92) 78 ( 52,112) 72 ( 56, 91) Cottingham South 80 ( 59,106) 109 ( 79,145) 92 ( 74,112) Dale 61 ( 47, 78) 67 ( 48, 90) 63 ( 52, 77) Driffield and Rural 100 ( 80,122) 107 ( 83,136) 102 ( 87,120) East Wolds and Coastal 91 ( 74,111) 81 ( 61,105) 88 ( 74,103) Goole North 98 ( 75,126) 80 ( 55,112) 91 ( 74,112) Goole South 125 ( 97,159) 170 (129,219) 143 (119,171) Hessle 82 ( 64,104) 111 ( 85,141) 94 ( 79,111) Howden 67 ( 41,104) 80 ( 45,132) 73 ( 51,101) Howdenshire 83 ( 66,104) 94 ( 71,122) 88 ( 74,104) Mid Holderness 76 ( 60, 95) 80 ( 60,105) 78 ( 65, 93) Minster and Woodmansey 101 ( 81,124) 69 ( 49, 93) 88 ( 73,104) North Holderness 90 ( 70,115) 82 ( 59,112) 88 ( 72,106) Pocklington Provincial 76 ( 60, 95) 72 ( 53, 95) 74 ( 62, 89) St Mary's 80 ( 63, 99) 99 ( 77,125) 87 ( 74,102) Snaith, Airmyn, Rawcliffe & Marshland 115 ( 90,145) 111 ( 80,150) 114 ( 94,137) South East Holderness 121 (102,143) 121 ( 96,150) 122 (107,140) South Hunsley 53 ( 38, 74) 72 ( 49,101) 61 ( 47, 77) South West Holderness 82 ( 65,102) 98 ( 75,126) 89 ( 74,105) Tranby 90 ( 68,117) 101 ( 73,136) 94 ( 77,115) Willerby and Kirk Ella 66 ( 51, 85) 81 ( 61,106) 72 ( 60, 87) Wolds Weighton 78 ( 61, 98) 84 ( 62,110) 81 ( 67, 96) East Riding of Yorkshire 89 ( 85, 93) 95 ( 90,101) 92 ( 89, 95) 9 If the 95% CI does not include 100 then the rate differs significantly than the national rate. However, one would expect 5% of statistical comparisons to be classified statistically significant by chance even when there was no difference between the local rate and the national rate so this must be borne in mind. Robert Sheikh Iddenden, Public Health Sciences. 53

54 8. COPD in Hull and East Riding This section presents information on prevalence, inpatient hospital admissions and activity, mortality data, risk factors, access to services and quality of care where it is available. Section 9 starting on page 140 will relate this information to other groups based on age, gender, deprivation, ethnicity, etc to assess if there are potential inequalities present Prevalence Diagnosed Prevalence Based on General Practice Registers As part of the contract for GPs which commenced on the 1 st April 2004, General Practices obtain points for achievements against a range of indicators. The system is known as the Quality and Outcomes Framework (QOF) and is used for calculating financial payment. One of the achievements is forming registers of patients with specific diseases including producing a register of people with COPD. There is one indicator related to spirometryconfirmed diagnosis of COPD, plus a further three indicators related to on-going monitoring of patients on the COPD registers. Further details on the QOF indicators for COPD are given in Quality and Outcomes Framework on page 276. The information for the QOF is extracted from the GP systems using the Quality Management and Analysis System (QMAS), and is provided at GP level for all practices throughout England via the Health and Social Care Information Centre (Information Centre, 2009). The information is available in the September after the end of the financial year. As it is a financial payment system to allow payments to be made by the end of April following the end of the financial year, the list size has been taken as at 1 st January, the register indicators are extracted from the QMAS on the 14 th February and the ongoing care indicators are extracted on the 31 st March. Patients who have joined the practice within the last three months of the financial year are automatically excluded from the register and on-going care indicators. The prevalence is calculated as at 14 th February (based on the list size on 1 st January). It has been assumed that the difference between the denominator for a specific on-going care indicator and the number of patients on the register are the number of exceptions 10. For discussion regarding the number of exceptions see GP Quality and Outcomes Framework Indicators on page 117. The practice list size and number of people on the COPD register is known for all practices, but the resulting prevalence from QOF is not adjusted for age in any way. Therefore, practices with particularly young patients such as those predominantly serving student populations will have a much lower prevalence rate and those serving older 10 Patients can be made exceptions from a particular indicator if it is medically inappropriate for that particular person or that particular patient did not attend a particular review. This was introduced so that practices are not penalised, as some of the on-going care indicators depend on achieving a specified percentage of patients receiving the designated level of care. Robert Sheikh Iddenden, Public Health Sciences. 54

55 populations and other populations, such as practices with a high number of nursing home patients, may have a much higher prevalence. To make a partial allowance for the differences in age structure among practices, Figure 13 gives the registered 2008/09 COPD prevalence for each General Practice in Hull and East Riding of Yorkshire sorted by the mean age of all the patients in the practice (based on age as at October 2008). It can be seen that the only practice with a register prevalence of below 0.5% is the practice with the youngest population (Dr Nayar in Hull, with a large percentage of students). It can be seen that the practices in East Riding of Yorkshire tend to have older age structures, although in general they do not tend to have a higher prevalence of COPD. There is considerable variability in the prevalence particularly in those practices that tend to have an older population structure (right hand side of figure) or a younger population structure (left hand side of the figure). COPD prevalence is likely also to be associated with deprivation (see page 39) and the figures do not allow for this in any way. For reference, the mean age of practice patients and mean deprivation scores are given in Appendix B on page 282. COPD prevalence was similar for previous years, as illustrated in Figure 14 (2007/08), Figure 15 (2006/07), Figure 16 (2005/06) and Figure 17 (2004/05). The practices are similarly sorted by mean age of patients in these figures, but mean age as at October of the relevant year, so the order of practices may differ slightly. Furthermore, the practices differed slightly due to mergers and closures. In order to make comparisons over time, Figure 18 compares the registered COPD prevalence over the five financial years 2004/05 to 2008/09 with the practices sorted by the mean age of their patients (mean age averaged over each year). One practice, Dr. Josephs in Hull was a particular outlier between 2004/05 and 2007/08. This was due to over-registration on the COPD register by practice staff. This has now been addressed, and in 2008/09 the prevalence rate for this practice, while still high, is in line with other relatively deprived practices. Of course, a study of 3,217 patients randomly sampled from 5,649 patients with COPD in 38 general practices in Rotherham found 12% of patients had FEV1 (% predicted) that did not support the diagnosis of COPD, suggesting an over-diagnosis of COPD (Strong et al 2009). It is likely that similar over-diagnosis might occur in Hull and East Riding practice registers. Robert Sheikh Iddenden, Public Health Sciences. 55

56 Dr J K Nayar Dr R Raut The Quays Medical Centre Dr K V Gopal Dr G L Clayton Dr P C Ghosh Dr N A Poulose Dr J Venugopal Dr A Kumar-Choudhary Riverside Medical Centre Dr G Palooran And Partners Dr A H Tak & Dr E G Dr R K Awan Dr B L Koul Dr S M Hussain And Partners Dr K K Mahendra Dr J C Joseph Dr J D Blow And Partners Dr Witvliet Dr J Musil And P J Queenan Dr A R M Kelly And Dr B L Bawn Dr Y Adhami Dr R Percival And Partners Dr P F Newman And Partners Northpoint Dr J S Parker & Partners Dr S K Ray Dr J N Singh And Partners Dr H S Suri And Partners Dr V A Rawcliffe And Partners Dr G T Hendow Dr Ali & Partners Dr A K Datta Dr Ak Rej Dr G M Chowdhury Dr J Austin Dr A W Hussain And Partners Dr M S Patel And Partners Dr P C Mitchell And Partners Kingston Medical Group The Springhead Medical Centre Faith House Surgery St Andrews Group Practice Dr M Shaikh & Partners Dr M J Varma And Partners Wheeler Street Healthcare Dr J Robson Dr M Maung And Partners Dr K M Tang And Partner Dr W A Hart And Partners Holme-Bubwith Medical Group Dr J E Clark And Partners Dr S Macphie And Koul Dr L Wrightson And Partners Dr R L Pearson & Partners Dr W G T Sande And Partners Dr A M Harley And Partners Dr T Abraham Dr R W Webster And Partners Dr M Foulds & Partner Dr R G Mitchell Dr R W Harrison The Calvert Practice Dr C J Lambert And Partners Dr A A Mather And Partners Dr J P Brooke And Partners Dr R A Ferguson And Partners Dr A N Crawford And Partners Hessle Grange Medical Practice Dr Cook Dr G Dave Dr P English And Partners Dr J W Richardson And Partners Park View Surgery Mizzen Road Surgery Dr G S Malczewski Dr H R P Meldrum And Partners Dr S J Towers And Partners Dr P G Jones And Partners Dr M Morgan And Partners Dr R D Fouracre And Partners Dr R G Clarke Dr P R Mixer And Partners Dr A J Sykes And Partners Dr H K Macnab And Partners Dr R D Yagnik Dr P M Pickering And Partners Dr P A Harris And Partners Dr M E Hancocks Dr P I Collingwood And Partners Dr M E A Moody And Partners Dr Farley And Partners Dr M K Mallik Dr J G Best And Partner GP register prevalence (%) 2008/09 Figure 13: Prevalence of COPD on General Practice registers 2008/09 for practices in Hull and East Riding, practices sorted in order of increasing mean age of patients as at October Hull East Riding of Yorkshire General Practice (sorted in order of increasing mean age of patients) Robert Sheikh Iddenden, Public Health Sciences. 56

57 Dr J K Nayar The Quays Medical Centre Dr R Raut Dr K V Gopal Dr G L Clayton Dr A Kumar-Choudhary Dr P C Ghosh Dr N A Poulose Dr J Venugopal Dr G Palooran And Partners Riverside Medical Centre Dr B L Koul Dr R K Awan Dr K K Mahendra Dr S M Hussain And Partners Dr A H Tak & Dr E G Dr A R M Kelly And Dr B L Bawn Dr J D Blow And Partners Dr J C Joseph Dr J Musil And P J Queenan Dr Witvliet Dr Y Adhami Dr P F Newman And Partners Dr S K Ray Dr R Percival And Partners Dr J S Parker & Partners Dr J N Singh And Partners Dr Ak Rej Dr H S Suri And Partners Dr A K Datta Dr V A Rawcliffe And Partners Dr G T Hendow Dr A W Hussain And Partners Kingston Medical Group Dr Ali & Partners The Springhead Medical Centre Dr G M Chowdhury Dr J Austin Dr M Shaikh & Partners Dr M S Patel And Partners Faith House Surgery Wheeler Street Healthcare St Andrews Group Practice Dr M J Varma And Partners Dr K M Tang And Partner Dr J Robson Dr P C Mitchell And Partners Dr M Maung And Partners Dr S Macphie And Koul Holme-Bubwith Medical Group Dr W A Hart And Partners Dr L Wrightson And Partners Dr J E Clark And Partners Dr M Foulds & Partner Dr T Abraham Dr A M Harley And Partners Dr R W Webster And Partners Dr R L Pearson & Partners Dr W G T Sande And Partners Dr C J Lambert And Partners Dr R G Mitchell Dr R W Harrison Dr A A Mather And Partners Dr J P Brooke And Partners Dr G Dave Dr R A Ferguson And Partners Dr A N Crawford And Partners Park View Surgery Hessle Grange Medical Practice Dr Cook Dr P English And Partners Dr J W Richardson And Partners Dr M Morgan And Partners Dr G S Malczewski Dr P G Jones And Partners Mizzen Road Surgery Dr S J Towers And Partners Dr R D Fouracre And Partners Dr H R P Meldrum And Partners Dr A J Sykes And Partners Dr P R Mixer And Partners The Calvert Practice Dr R G Clarke Dr H K Macnab And Partners Dr R D Yagnik Dr P M Pickering And Partners Dr P I Collingwood And Partners Dr P A Harris And Partners Dr M E A Moody And Partners Dr M E Hancocks Dr M K Mallik Dr Farley And Partners Dr J G Best And Partner GP register prevalence (%) 2007/08 Figure 14: Prevalence of COPD on General Practice registers 2007/08 for practices in Hull and East Riding, practices sorted in order of increasing mean age of patients as at October Hull East Riding of Yorkshire General Practice (sorted in order of increasing mean age of patients) Robert Sheikh Iddenden, Public Health Sciences. 57

58 Dr J K Nayar The Quays Medical Centre Dr R Raut Dr K V Gopal Dr G L Clayton Dr N A Poulose Dr P C Ghosh Dr A Kumar-Choudhary Dr J Venugopal Dr G Palooran And Partners Dr B L Koul Dr K K Mahendra Dr A H Tak & Dr E G Stryjakiewicz Dr R K Awan Dr A R M Kelly And Dr B L Bawn Dr J D Blow And Partners Dr S M Hussain And Partners Dr J C Joseph Dr Witvliet Dr Y Adhami Dr J Musil And P J Queenan Dr S K Ray Dr E G Stryjakiewicz Dr P F Newman And Partners Dr J N Singh And Partners Dr J S Parker & Partners Dr Ak Rej Riverside Medical Centre Dr A K Datta Dr H S Suri And Partners Dr R Percival And Partners Dr A W Hussain And Partners Dr J Austin Dr G T Hendow Dr G M Chowdhury The Springhead Medical Centre Dr V A Rawcliffe And Partners Kingston Medical Group Dr Ali & Partners Faith House Surgery Dr P C Mitchell And Partners Wheeler Street Healthcare Dr K M Tang And Partner Dr M S Patel And Partners Dr M Shaikh & Partners Dr J Robson Dr M Maung And Partners Dr M J Varma And Partners St Andrews Group Practice Dr S Macphie And Koul Holme-Bubwith Medical Group Dr J E Clark And Partners Dr A M Harley And Partners Dr L Wrightson And Partners Dr M Foulds & Partner Dr W A Hart And Partners Dr T Abraham Dr W G T Sande And Partners Dr R G Mitchell Dr C J Lambert And Partners Dr R W Webster And Partners Dr R L Pearson & Partners Dr J P Brooke And Partners Dr R W Harrison Dr G Dave Dr A A Mather And Partners Dr R A Ferguson And Partners Park View Surgery Dr A N Crawford And Partners Dr M Morgan And Partners Dr G S Malczewski Dr P English And Partners Dr J W Richardson And Partners Dr Cook Hessle Grange Medical Practice Dr P G Jones And Partners Dr S J Towers And Partners Mizzen Road Surgery Dr R D Fouracre And Partners Dr H R P Meldrum And Partners Dr A J Sykes And Partners Dr P R Mixer And Partners Dr R G Clarke Dr H K Macnab And Partners The Calvert Practice Dr P M Pickering And Partners Dr P I Collingwood And Partners Dr P A Harris And Partners Dr M K Mallik Dr R D Yagnik Dr M E A Moody And Partners Dr M E Hancocks Dr Farley And Partners Dr J G Best And Partner GP register prevalence (%) 2006/07 Figure 15: Prevalence of COPD on General Practice registers 2006/07 for practices in Hull and East Riding, practices sorted in order of increasing mean age of patients as at October Hull East Riding of Yorkshire General Practice (sorted in order of increasing mean age of patients) Robert Sheikh Iddenden, Public Health Sciences. 58

59 Dr J K Nayar The Quays Medical Centre Dr R Raut Dr G L Clayton Dr K V Gopal Dr N A Poulose Dr A Kumar-Choudhary Dr J Venugopal Dr P C Ghosh Dr G Palooran And Partners Dr K K Mahendra Dr B L Koul Dr R K Awan Dr A R M Kelly And Dr B L Bawn Dr J D Blow And Partners Dr S M Hussain And Partners Dr J Musil And P J Queenan Dr J C Joseph Dr A H Tak & Dr E G Stryjakiewicz Dr Witvliet Dr S K Ray Dr Y Adhami Dr J N Singh And Partners Dr E G Stryjakiewicz Dr Ak Rej Dr A K Datta Dr J Austin Dr H S Suri And Partners Dr A W Hussain And Partners Dr P F Newman And Partners Dr R Percival And Partners Dr J S Parker & Partners Riverside Medical Centre Dr G T Hendow The Springhead Medical Centre Dr G M Chowdhury Dr K M Tang And Partner Dr Ali & Partners Wheeler Street Healthcare Dr M Shaikh & Partners St Andrews Group Practice Dr V A Rawcliffe And Partners Dr P C Mitchell And Partners Faith House Surgery Kingston Medical Group Dr M S Patel And Partners Dr M Maung And Partners Dr J Robson Holme-Bubwith Medical Group Dr M J Varma And Partners Dr A M Harley And Partners Dr L Wrightson And Partners Dr M Foulds & Partner Dr J E Clark And Partners Dr C J Lambert And Partners Dr W A Hart And Partners Dr R G Mitchell Dr T Abraham Dr R W Webster And Partners Dr S Macphie And Koul Dr G Dave Dr J P Brooke And Partners Dr R L Pearson & Partners Dr R W Harrison Park View Surgery Dr R A Ferguson And Partners Dr W G T Sande And Partners Dr A N Crawford And Partners Dr J W Richardson And Partners Dr P English And Partners Dr M Morgan And Partners Dr G S Malczewski Hessle Grange Medical Practice Dr Cook Dr S J Towers And Partners Mizzen Road Surgery Dr P G Jones And Partners Dr A A Mather And Partners Dr A J Sykes And Partners Dr H R P Meldrum And Partners Dr R D Fouracre And Partners Dr P R Mixer And Partners Dr H K Macnab And Partners Dr P M Pickering And Partners The Calvert Practice Dr R G Clarke Dr M K Mallik Dr M E A Moody And Partners Dr P I Collingwood And Partners Dr M E Hancocks Dr P A Harris And Partners Dr R D Yagnik Dr Farley And Partners Dr J G Best And Partner GP register prevalence (%) 2005/06 Figure 16: Prevalence of COPD on General Practice registers 2005/06 for practices in Hull and East Riding, practices sorted in order of increasing mean age of patients as at October Hull East Riding of Yorkshire General Practice (sorted in order of increasing mean age of patients) Robert Sheikh Iddenden, Public Health Sciences. 59

60 Dr J K Nayar Dr PN Jones and Partner The Quays Medical Centre Dr R Raut Dr G L Clayton Dr K V Gopal Dr J Venugopal Dr P C Ghosh Dr N A Poulose Dr A Kumar-Choudhary Dr G Palooran And Partners Dr K K Mahendra Dr B L Koul Dr J D Blow And Partners Dr A R M Kelly And Dr B L Bawn Dr R K Awan Dr A H Tak & Dr E G Stryjakiewicz Dr J C Joseph Dr A K Datta Dr S M Hussain And Partners Dr Witvliet Dr Y Adhami Dr J N Singh And Partners Dr S K Ray Dr H S Suri And Partners Dr E G Stryjakiewicz Dr R Percival And Partners Dr J Austin Dr Ak Rej Dr P F Newman And Partners Dr A W Hussain And Partners Dr J Musil And P J Queenan Dr J S Parker & Partners Dr K M Tang And Partner Dr G T Hendow Dr G M Chowdhury St Andrews Group Practice The Springhead Medical Centre Dr Ali & Partners Wheeler Street Healthcare Dr V A Rawcliffe And Partners Faith House Surgery Dr M Shaikh & Partners Dr P C Mitchell And Partners Kingston Medical Group Dr A M Harley And Partners Dr J Robson Dr M S Patel And Partners Holme-Bubwith Medical Group Dr M Maung And Partners Dr M J Varma And Partners Dr R G Mitchell Dr L Wrightson And Partners Dr C J Lambert And Partners Dr W A Hart And Partners Dr M Foulds & Partner Dr R W Harrison Dr J E Clark And Partners Dr R W Webster And Partners Dr G Dave Dr J P Brooke And Partners Dr T Abraham Park View Surgery Dr S Macphie And Koul Dr R A Ferguson And Partners Dr R L Pearson & Partners Dr A N Crawford And Partners Dr P English And Partners Dr W G T Sande And Partners Dr J W Richardson And Partners Dr P G Jones And Partners Dr S J Towers And Partners Hessle Grange Medical Practice Dr M Morgan And Partners Dr Cook Mizzen Road Surgery Dr G S Malczewski Dr R D Fouracre And Partners Dr P R Mixer And Partners Dr A J Sykes And Partners Dr H R P Meldrum And Partners Dr A A Mather And Partners Dr P M Pickering And Partners Dr R G Clarke Dr M E A Moody And Partners Dr H K Macnab And Partners Dr M E Hancocks Dr P I Collingwood And Partners Dr P A Harris And Partners Dr M K Mallik Dr R D Yagnik Dr Farley And Partners Dr J G Best And Partner GP register prevalence (%) 2004/05 Figure 17: Prevalence of COPD on General Practice registers 2004/05 for practices in Hull and East Riding, practices sorted in order of increasing mean age of patients as at October Hull East Riding of Yorkshire General Practice (sorted in order of increasing mean age of patients) Robert Sheikh Iddenden, Public Health Sciences. 60

61 Dr PN Jones and Partner Dr J K Nayar The Quays Medical Centre Dr R Raut Dr K V Gopal Dr G L Clayton Dr N A Poulose Dr P C Ghosh Dr A Kumar-Choudhary Dr J Venugopal Dr G Palooran And Partners Dr K K Mahendra Dr B L Koul Dr R K Awan Dr A H Tak & Dr E G Stryjakiewicz Dr J D Blow And Partners Dr A R M Kelly And Dr B L Bawn Dr S M Hussain And Partners Riverside Medical Centre Dr J C Joseph Dr Witvliet Dr J Musil And P J Queenan Dr Y Adhami Dr E G Stryjakiewicz Dr S K Ray Dr P F Newman And Partners Dr J N Singh And Partners Dr R Percival And Partners Dr J S Parker & Partners Northpoint Dr A K Datta Dr H S Suri And Partners Dr Ak Rej Dr A W Hussain And Partners Dr J Austin Dr G T Hendow Dr G M Chowdhury Dr V A Rawcliffe And Partners The Springhead Medical Centre Dr Ali & Partners Kingston Medical Group Dr K M Tang And Partner Faith House Surgery St Andrews Group Practice Dr M Shaikh & Partners Wheeler Street Healthcare Dr P C Mitchell And Partners Dr M S Patel And Partners Dr J Robson Dr M Maung And Partners Dr M J Varma And Partners Holme-Bubwith Medical Group Dr A M Harley And Partners Dr L Wrightson And Partners Dr W A Hart And Partners Dr S Macphie And Koul Dr J E Clark And Partners Dr M Foulds & Partner Dr R G Mitchell Dr C J Lambert And Partners Dr T Abraham Dr R W Webster And Partners Dr R L Pearson & Partners Dr R W Harrison Dr W G T Sande And Partners Dr J P Brooke And Partners Dr G Dave Dr R A Ferguson And Partners Park View Surgery Dr A N Crawford And Partners Dr P English And Partners Dr A A Mather And Partners Dr J W Richardson And Partners Hessle Grange Medical Practice Dr Cook Dr M Morgan And Partners Dr G S Malczewski Mizzen Road Surgery Dr P G Jones And Partners Dr S J Towers And Partners Dr H R P Meldrum And Partners Dr R D Fouracre And Partners Dr A J Sykes And Partners Dr P R Mixer And Partners The Calvert Practice Dr R G Clarke Dr H K Macnab And Partners Dr P M Pickering And Partners Dr R D Yagnik Dr P I Collingwood And Partners Dr P A Harris And Partners Dr M E A Moody And Partners Dr M E Hancocks Dr M K Mallik Dr Farley And Partners Dr J G Best And Partner GP register prevalence (%) 2004/05 to 2008/09 Figure 18: Prevalence of COPD on General Practice registers for patients aged 17+ years, changes over time 2004/05 to 2008/09, for practices in Hull and East Riding, practices sorted in order of increasing mean age of patients(averaged over each year) 8 7 Hull 2004/04 Hull 2005/06 Hull 2006/07 Hull 2007/08 Hull 2008/09 ERoY 2004/05 ERoY 2005/06 ERoY 2006/07 ERoY 2007/08 ERoY 2008/ General Practice (sorted in order of increasing mean age of patients) Robert Sheikh Iddenden, Public Health Sciences. 61

62 The data are also presented in tabular form in Table 10 and Table 11 for Hull Teaching PCT and East Riding of Yorkshire PCT respectively. The practices are sorted in increasing order of the mean age 11 of their practice populations so that the practices at the top of the table will tend to have a lower prevalence of COPD than those practices towards the bottom of the table. Dr Jones practice merged with Dr Nayar & Van Maarseveen s practice so there is no register for 2005/2006 onwards. Dr Stryjakiewicz s practice merged with Dr Tak s practice, so there is no register for 2007/08 onwards. The Riverside Medical Centre and The Calvert Practice started after 2004/05 so there is no information for that year. The Northpoint practice started after 2007/08 so there is only data for 2008/09. It can be seen for the most recent period (2008/09), for 18 Hull and 9 East Riding of Yorkshire practices (shaded boxes), the prevalence on the COPD register had decreased, while 14 and 7 practices for Hull and East Riding of Yorkshire respectively saw two or more consecutive decreases in prevalence between 2004/05 and 2008/09. Given that COPD is incurable and is more prevalent in older people, it would be expected that prevalence over time was likely to increase. However, there will be some variation from year to year, particularly if the age structure of a practice changes, and any small changes are probably due to these demographic changes. What we would not expect to see is large year-on-year decreases in prevalence. Although a few practices have decreases in successive years these decreases tend to be relatively small, and may represent demographic changes in the practice populations. One exception to this is Dr Josephs s practice which, as was mentioned earlier, was over-registering COPD patients, so the large decreases in 2007/08 and 2008/09, are welcome, and represent the removal of patients from the register that should not have been there in the first place. Ten practices in Hull and four in East Riding have seen register prevalence of COPD decrease over the period 2004/05 to 2008/09. Table 10: Diagnosed prevalence of COPD on Hull s General Practice registers, trends over time Code Practice Patient mean age (yrs) COPD prevalence on General Practice Registers (%) 04/05 05/06 06/07 07/08 08/09 B81676 Dr PN Jones and Partner B81104 Dr J K Nayar B81692 The Quays Medical Centre B81631 Dr R Raut B81688 Dr K V Gopal B81685 Dr N A Poulose B81112 Dr P C Ghosh B81002 Dr A Kumar-Choudhary B81634 Dr J Venugopal B81119 Dr G Palooran And Partners B81644 Dr K K Mahendra B81683 Dr B L Koul Derived by taking the mean of the mean age as at October 2004, 2005, 2006, 2007 and 2008 Robert Sheikh Iddenden, Public Health Sciences. 62

63 Code Practice Patient mean age (yrs) COPD prevalence on General Practice Registers (%) 04/05 05/06 06/07 07/08 08/09 B81018 Dr R K Awan B81675 Dr A H Tak & Dr E G Stryjakiewicz B81046 Dr J D Blow And Partners B81048 Dr S M Hussain And Partners Y00955 Riverside Medical Centre B81674 Dr J C Joseph B81089 Dr Witvliet B81052 Dr J Musil And P J Queenan B81668 Dr E G Stryjakiewicz B81690 Dr S K Ray B81040 Dr P F Newman And Partners B81047 Dr J N Singh And Partners B81072 Dr R Percival And Partners B81008 Dr J S Parker & Partners Y02344 Northpoint B81094 Dr A K Datta B81074 Dr Ak Rej B81032 Dr A W Hussain And Partners B81646 Dr J Austin B81616 Dr G T Hendow B81066 Dr G M Chowdhury B81049 Dr V A Rawcliffe And Partners B81056 The Springhead Medical Centre B81001 Dr Ali & Partners B81017 Kingston Medical Group B81081 Dr K M Tang And Partner B81021 Faith House Surgery B81027 St Andrews Group Practice B81682 Dr M Shaikh & Partners B81011 Wheeler Street Healthcare B81020 Dr P C Mitchell And Partners B81053 Dr M Maung And Partners B81054 Dr M J Varma And Partners B81057 Dr S Macphie And Koul B81058 Dr M Foulds & Partner B81645 Dr T Abraham B81035 Dr W G T Sande And Partners B81635 Dr G Dave B81038 Dr A A Mather And Partners B81085 Dr J W Richardson And Partners B81095 Dr Cook B81080 Dr G S Malczewski B81662 Mizzen Road Surgery Y01200 The Calvert Practice B81097 Dr R D Yagnik B81075 Dr M K Mallik Robert Sheikh Iddenden, Public Health Sciences. 63

64 Table 11: Diagnosed prevalence of COPD on East Riding of Yorkshire s General Practice registers, trends over time Code Practice Patient mean age (yrs) COPD prevalence on General Practice Registers (%) 04/05 05/06 06/07 07/08 08/09 B81121 Dr G L Clayton B81120 Dr A R M Kelly And Dr B L Bawn B81658 Dr Y Adhami B81101 Dr H S Suri And Partners B81013 Dr M S Patel And Partners B81653 Dr J Robson B81044 Holme-Bubwith Medical Group B81622 Dr A M Harley And Partners B81068 Dr L Wrightson And Partners B81061 Dr W A Hart And Partners B81029 Dr J E Clark And Partners B81666 Dr R G Mitchell B81062 Dr C J Lambert And Partners B81009 Dr R W Webster And Partners B81082 Dr R L Pearson & Partners B81088 Dr R W Harrison B81036 Dr J P Brooke And Partners B81041 Dr R A Ferguson And Partners B81619 Park View Surgery B81092 Dr A N Crawford And Partners B81050 Dr P English And Partners B81010 Hessle Grange Medical Practice B81084 Dr M Morgan And Partners B81042 Dr P G Jones And Partners B81037 Dr S J Towers And Partners B81070 Dr H R P Meldrum And Partners B81025 Dr R D Fouracre And Partners B81034 Dr A J Sykes And Partners B81051 Dr P R Mixer And Partners B81679 Dr R G Clarke B81014 Dr H K Macnab And Partners B81060 Dr P M Pickering And Partners B81004 Dr P I Collingwood And Partners B81069 Dr P A Harris And Partners B81024 Dr M E A Moody And Partners B81602 Dr M E Hancocks B81006 Dr Farley And Partners B81100 Dr J G Best And Partner Table 12 compares the prevalence of COPD for Hull Teaching PCT and East Riding of Yorkshire PCT separately and combined, along with prevalence for England and the Yorkshire and Humber Strategic Health Authority (Y&HSHA). The COPD prevalence on the registers is higher in both Hull and East Riding of Yorkshire compared to England. Hull has a more deprived population compared to England, so one would expect that the Robert Sheikh Iddenden, Public Health Sciences. 64

65 prevalence would be higher, and whilst East Riding is slightly less deprived compared to England the population is on average older, so one would also expect that the prevalence in East Riding would be similar to or higher than England as a whole. Given that smoking is the main risk factor for COPD, and that smoking prevalence in Hull is much higher than in East Riding, we would expect prevalence of COPD to be higher in Hull than East Riding, which indeed it is. However, both Hull and East Riding have register prevalence of COPD lower than the average for Yorkshire and Humber SHA, although very similar in the case of Hull. While we would expect prevalence to be lower in East Riding, as it is one of the least deprived areas in the region, Hull is the most deprived area in the region, so we would expect its register prevalence of COPD to be higher than the regional average. In terms of which PCT is likely to have the largest percentage of residents with undiagnosed COPD, it is likely to be Hull, which is more deprived (and the 11 th most deprived local authority in England). The main risk factor for COPD, smoking, is associated with deprivation, but so is the propensity to report any symptoms of to GPs. It is likely that residents of Hull would be more willing to attribute health problems to old age, or to accept problems due to old age at a younger age than East Riding residents, which means that it is possible that residents of Hull may be more likely to have undiagnosed COPD than East Riding residents. If these assumptions are correct, the prevalence of undiagnosed COPD in Hull would be greater than in East Riding. Annual increases in register prevalence of COPD in East Riding were similar, while for Hull, England and Y&HSHA increases from 2004/05 to 2005/06 were between one quarter and half the increases seen in subsequent years. Over the whole period, 2004/05 to 2008/09, East Riding saw the largest relative increase in COPD prevalence (14%), with the increase in Hull (12%) similar to England, with both higher than the regional average increase of 9%. Table 12: Diagnosed prevalence of COPD on General Practice registers, 2004/05 to 2005/06 Period covered Prevalence of COPD (%) on GP registers and % change England Y&HSHA Hull ERoY Hull & ERoY 2004/ / / / / Percentage changes from: 2004/05 to 2005/ /06 to 2006/ /07 to 2007/ /08 to 2008/ /05 to 2008/ Robert Sheikh Iddenden, Public Health Sciences. 65

66 Modelled COPD prevalence estimates at practice level from the Eastern Region Public Health Observatory model Modelled estimates of COPD prevalence have been produced by Eastern Region Public Health Observatory (ERPHO) on behalf of the Association of Public Health Observatories (APHO), with the model updated in November 2009 (Eastern Region PHO, 2009) to include practice level modelled estimates. The model produces the expected prevalence of COPD within each GP practice. If we were to accept the results of the model, the difference between the register prevalence and the modelled prevalence would represent the amount of undiagnosed COPD among the patients registered with each GP practice (assuming that the modelled prevalence was higher than the register prevalence for each PCT, which may not be the case). Included in the model were the age and sex of practice populations, the distribution of ethnic group, the prevalence of current and former smoking, IMD 2007 score and urbansuburban-rural category. The modelled estimates from ERPHO are based on practice populations as at 31/03/2007, provided by the National Strategic Tracing Service (NSTS), which implies that the modelled estimates, despite being produced in October 2009, are likely to be the estimated prevalence for 2006/07. However, it is not entirely clear whether this is indeed the case. Therefore, the modelled estimates are presented as is, rather than being applied to 2008/09 practice populations. Ethnicity by practice was derived from HES admissions 2005/06 to 2006/07, with the proportion of admissions from each ethnic group out of all admissions for each practice applied to all age groups for that practice. Weighted average IMD scores were produced for each practice. The urban-suburban-rural category of the MSOA where the practice was physically located was assumed to be applicable to all the practice population. Practice level synthetic smoking estimates were produced by taking weighted averages of the synthetic smoking estimates at Middle Layer Super Output Area (MSOA) level for each practice. These were then used to adjust the national smoking estimates from the Health Survey for England ( pooled) to provide smoking prevalence estimates for each practice. National proportions of former smokers (unadjusted) were used for each practice. In general when such models have been produced, the model is based on research undertaken elsewhere in the UK examining the prevalence of diagnosed disease in the community, which has then been modelled and applied to different populations such as those registered with a particular GP practice. Therefore, how accurate the estimates are, depends on the quality of the initial research and the modelling itself. It is possible that the original research did not include very deprived areas, in which case it is very difficult to generalise and apply the model to very deprived areas like Hull. Furthermore, there are Robert Sheikh Iddenden, Public Health Sciences. 66

67 many reasons why the prevalence could differ among practices, which are not captured by these models. Therefore, just because practices have a particularly low prevalence relative to the model, does not necessarily mean that they are performing badly in any way relative to other general practices. Nevertheless, a comparison of the differences between the modelled prevalence and the practice list registers can act as a starting point for investigation. Practices with a low prevalence or a relatively large difference between the model and the register prevalence can be examined further and considered in relation to patient characteristics using local knowledge. Modelled COPD prevalence estimates alongside register prevalence for each practice are shown for Hull practices in Table 13, while the numbers of patients with COPD, both actual (on the practice registers) and modelled, are shown in Table 14 for Hull practices. For each practice in Hull, the modelled prevalence estimates were higher than the register prevalence, which would be expected if the model reflected the level of diagnosed plus undiagnosed COPD in the community. In 39 of the 55 Hull practices where the model produced estimates (71%) modelled prevalence was more than twice as high as register prevalence, in other words suggesting that fewer than half of the patients with COPD in these practices were diagnosed as having COPD. For two practices, the number or patients on the COPD register in 2008/09 was slightly higher than the modelled number, despite the register prevalence being lower, which arises due to the different populations used, and the fact that the differences between model and register prevalence for these two practices were small. Across all Hull practices combined there was a difference of 7,053 between the modelled and register prevalence, with the model suggesting that fewer than half of the people in Hull with COPD were diagnosed and on practice COPD registers in 2008/09. Table 13: Modelled prevalence estimates (using ERPHO model) and register prevalence of COPD by GP practices in Hull 2008/09 Practice Practice name COPD Prevalence (%) Difference code Modelled Register (relative difference, %) B81001 Dr Ali & Partners (-66.2) B81002 Dr A Kumar-Choudhary (-46.4) B81008 Dr J S Parker & Partners (-72.5) B81011 Wheeler Street Healthcare (-64.7) B81017 Kingston Medical Group (-46.6) B81018 Dr R K Awan (-33.7) B81020 Dr P C Mitchell And Partners (-73.2) B81021 Faith House Surgery (-59.5) B81027 St Andrews Group Practice (-36.9) B81032 Dr A W Hussain And Partners (-62.0) B81035 Dr W G T Sande And Partners (-69.4) B81038 Dr A A Mather And Partners (-64.3) Robert Sheikh Iddenden, Public Health Sciences. 67

68 Practice Practice name COPD Prevalence (%) Difference code Modelled Register (relative difference, %) B81040 Dr P F Newman And Partners (-63.1) B81046 Dr J D Blow And Partners (-53.3) B81047 Dr J N Singh And Partners (-63.8) B81048 Dr S M Hussain And Partners (-58.1) B81049 Dr V A Rawcliffe And Partners (-38.2) B81052 Dr J Musil And P J Queenan (-64.1) B81053 Dr M Maung And Partners (-45.8) B81054 Dr M J Varma And Partners (-68.0) B81056 The Springhead Medical Centre (-58.9) B81057 Dr S Macphie And Koul (-77.5) B81058 Dr M Foulds & Partner (-43.2) B81066 Dr G M Chowdhury (-72.5) B81072 Dr R Percival And Partners (-58.7) B81074 Dr Ak Rej (-56.0) B81075 Dr M K Mallik (-87.1) B81080 Dr G S Malczewski (-61.0) B81081 Dr K M Tang And Partner (-61.6) B81085 Dr J W Richardson And Partners (-70.8) B81089 Dr Witvliet (-64.8) B81094 Dr A K Datta (-72.7) B81095 Dr Cook (-58.8) B81097 Dr R D Yagnik (-61.5) B81104 Dr J K Nayar (-82.2) B81112 Dr P C Ghosh (-46.1) B81119 Dr G Palooran And Partners (-65.0) B81616 Dr G T Hendow (-57.3) B81631 Dr R Raut ( -2.6) B81634 Dr J Venugopal (-66.9) B81635 Dr G Dave (-38.1) B81644 Dr K K Mahendra (-51.5) B81645 Dr T Abraham (-60.0) B81646 Dr J Austin (-62.4) B81662 Mizzen Road Surgery (-66.5) B81674 Dr J C Joseph ( -8.4) B81675 Dr A H Tak & Dr E G Stryjakiewicz (-57.0) B81682 Dr M Shaikh & Partners (-58.7) B81683 Dr B L Koul (-39.5) B81685 Dr N A Poulose (-35.7) B81688 Dr K V Gopal (-19.8) B81690 Dr S K Ray (-68.4) B81692 The Quays Medical Centre (-47.6) Y00955 Riverside Medical Centre (-58.1) Robert Sheikh Iddenden, Public Health Sciences. 68

69 Practice Practice name COPD Prevalence (%) Difference code Modelled Register (relative difference, %) Y01200 The Calvert Practice (-62.8) Y02344 Northpoint 2.33 All Hull practices (-57.8) Table 14: Number of patients with COPD from the ERPHO model and the practice registers by GP practices in Hull 2008/09 Practice Practice name Numbers with COPD code Modelled Register Difference B81001 Dr Ali & Partners B81002 Dr A Kumar-Choudhary B81008 Dr J S Parker & Partners B81011 Wheeler Street Healthcare B81017 Kingston Medical Group B81018 Dr R K Awan B81020 Dr P C Mitchell And Partners B81021 Faith House Surgery B81027 St Andrews Group Practice B81032 Dr A W Hussain And Partners B81035 Dr W G T Sande And Partners B81038 Dr A A Mather And Partners B81040 Dr P F Newman And Partners B81046 Dr J D Blow And Partners B81047 Dr J N Singh And Partners B81048 Dr S M Hussain And Partners B81049 Dr V A Rawcliffe And Partners B81052 Dr J Musil And P J Queenan B81053 Dr M Maung And Partners B81054 Dr M J Varma And Partners B81056 The Springhead Medical Centre B81057 Dr S Macphie And Koul B81058 Dr M Foulds & Partner B81066 Dr G M Chowdhury B81072 Dr R Percival And Partners B81074 Dr Ak Rej B81075 Dr M K Mallik B81080 Dr G S Malczewski B81081 Dr K M Tang And Partner B81085 Dr J W Richardson And Partners B81089 Dr Witvliet B81094 Dr A K Datta B81095 Dr Cook Robert Sheikh Iddenden, Public Health Sciences. 69

70 Practice Practice name Numbers with COPD code Modelled Register Difference B81097 Dr R D Yagnik B81104 Dr J K Nayar B81112 Dr P C Ghosh B81119 Dr G Palooran And Partners B81616 Dr G T Hendow B81631 Dr R Raut B81634 Dr J Venugopal B81635 Dr G Dave B81644 Dr K K Mahendra B81645 Dr T Abraham B81646 Dr J Austin B81662 Mizzen Road Surgery B81674 Dr J C Joseph B81675 Dr A H Tak & Dr E G Stryjakiewicz B81682 Dr M Shaikh & Partners B81683 Dr B L Koul B81685 Dr N A Poulose B81688 Dr K V Gopal B81690 Dr S K Ray B81692 The Quays Medical Centre Y00955 Riverside Medical Centre Y01200 The Calvert Practice Y02344 Northpoint 46 All Hull practices 12,355 5,319-7,036 Equivalent information for East Riding of Yorkshire practices is given in Table 15 (modelled COPD prevalence estimates alongside register prevalence) and Table 16 (the numbers of patients with COPD, both modelled and on the practice register). With the exception of two practices, modelled prevalence was higher than registered prevalence. Modelled prevalence estimates were at least double register prevalence in far fewer East Riding practices (8; 21%) than Hull practices. Over all East Riding practices combined, there was a difference of 2,212 between the modelled and register prevalence of COPD, with register prevalence 29% lower than modelled prevalence. Table 15: Modelled prevalence estimates (using ERPHO model) and register prevalence of COPD by GP practices in East Riding 2008/09 COPD Prevalence (%) Difference Practice Practice name (relative code Modelled Register difference, %) B81004 Dr P I Collingwood And Partners (-46.0) B81006 Dr Farley And Partners (-31.5) Robert Sheikh Iddenden, Public Health Sciences. 70

71 Practice code Practice name COPD Prevalence (%) Modelled Register Difference (relative difference, %) B81009 Dr R W Webster And Partners ( 0.2) B81010 Hessle Grange Medical Practice (-29.7) B81013 Dr M S Patel And Partners (-55.7) B81014 Dr H K Macnab And Partners (-27.9) B81024 Dr M E A Moody And Partners (-45.0) B81025 Dr R D Fouracre And Partners ( -9.1) B81029 Dr J E Clark And Partners (-16.1) B81034 Dr A J Sykes And Partners (-49.5) B81036 Dr J P Brooke And Partners (-35.4) B81037 Dr S J Towers And Partners (-24.7) B81041 Dr R A Ferguson And Partners (-36.4) B81042 Dr P G Jones And Partners (-66.0) B81044 Holme-Bubwith Medical Group (-32.6) B81050 Dr P English And Partners (-25.8) B81051 Dr P R Mixer And Partners (-53.8) B81060 Dr P M Pickering And Partners ( 3.6) B81061 Dr W A Hart And Partners ( -6.9) B81062 Dr C J Lambert And Partners (-16.5) B81068 Dr L Wrightson And Partners (-37.9) B81069 Dr P A Harris And Partners ( -4.7) B81070 Dr H R P Meldrum And Partners (-13.6) B81082 Dr R L Pearson & Partners (-57.1) B81084 Dr M Morgan And Partners (-44.3) B81088 Dr R W Harrison (-19.3) B81092 Dr A N Crawford And Partners (-26.0) B81100 Dr J G Best And Partner (-63.0) B81101 Dr H S Suri And Partners ( -3.7) B81120 Dr A R M Kelly And Dr B L Bawn (-10.6) B81121 Dr G L Clayton (-52.7) B81602 Dr M E Hancocks (-29.9) B81619 Park View Surgery (-21.9) B81622 Dr A M Harley And Partners (-49.4) B81653 Dr J Robson (-30.3) B81658 Dr Y Adhami (-71.1) B81666 Dr R G Mitchell (-42.8) B81679 Dr R G Clarke (-57.3) All East Riding general practices (-29.1) Robert Sheikh Iddenden, Public Health Sciences. 71

72 Table 16: Number of patients with COPD from the ERPHO model and the practice registers by GP practices in East Riding 2008/09 Practice Practice name Numbers with COPD code Modelled Register Difference B81004 Dr P I Collingwood And Partners B81006 Dr Farley And Partners B81009 Dr R W Webster And Partners B81010 Hessle Grange Medical Practice B81013 Dr M S Patel And Partners B81014 Dr H K Macnab And Partners B81024 Dr M E A Moody And Partners B81025 Dr R D Fouracre And Partners B81029 Dr J E Clark And Partners B81034 Dr A J Sykes And Partners B81036 Dr J P Brooke And Partners B81037 Dr S J Towers And Partners B81041 Dr R A Ferguson And Partners B81042 Dr P G Jones And Partners B81044 Holme-Bubwith Medical Group B81050 Dr P English And Partners B81051 Dr P R Mixer And Partners B81060 Dr P M Pickering And Partners B81061 Dr W A Hart And Partners B81062 Dr C J Lambert And Partners B81068 Dr L Wrightson And Partners B81069 Dr P A Harris And Partners B81070 Dr H R P Meldrum And Partners B81082 Dr R L Pearson & Partners B81084 Dr M Morgan And Partners B81088 Dr R W Harrison B81092 Dr A N Crawford And Partners B81100 Dr J G Best And Partner B81101 Dr H S Suri And Partners B81120 Dr A R M Kelly And Dr B L Bawn B81121 Dr G L Clayton B81602 Dr M E Hancocks B81619 Park View Surgery B81622 Dr A M Harley And Partners B81653 Dr J Robson B81658 Dr Y Adhami B81666 Dr R G Mitchell B81679 Dr R G Clarke All East Riding general practices 7,740 5,528-2,212 Robert Sheikh Iddenden, Public Health Sciences. 72

73 Modelled COPD prevalence estimates at practice level applying the Doncaster PCT methodology A more conservative model was produced by Doncaster PCT (Doncaster PCT, 2008) which produces estimated COPD prevalence rates by GP practice, taking deprivation into account using a three stage methodology. This model was one of several that were produced by the Public Health Intelligence Unit of Doncaster PCT, full details of which may be downloaded from the PCT website ( and are summarised below. Stage 1 predicts the number of patients with COPD for each practice (assuming national rates apply), on the basis of age-gender-specific rates derived from Morbidity statistics from General Practice: Fourth national study (HMSO 1995) but increased in line with the prevalence estimates produced by Soriano et al (Soriano, 2000). Stage 2 adjusts these numbers for health inequalities between local authorities. The assumption is that increased COPD mortality reflects an equivalent increase in COPD prevalence, thus COPD SMRs are used for the adjustment. For Hull, the numbers for each practice are increased by 55.2% (Hull s SMR was 155.2) while for East Riding, the numbers are decreased by 10.1% (East Riding s SMR was 89.9). Stage 3 adjusts for the inequalities between practices in each local authority, using a census-derived deprivation score (UV67) which is evaluated for each practice. A multiplying factor is then calculated, based on the linear relationship between UV67 and SMRs for COPD. This is then applied to the stage 2 results, which are then divided by the SMR predicted by the local authority UV67 score (to avoid doublecounting the effect of deprivation, as had already adjusted predicted prevalence by COPD SMR, and much of this difference is attributable to deprivation). In general when such models have been produced, the model is based on research undertaken elsewhere in the UK examining the prevalence of diagnosed disease in the community, which has then been modelled and applied to different populations such as those registered with a particular GP practice. Therefore, how accurate the estimates are, depends on the quality of the initial research and the modelling itself. It is possible that the original research did not include very deprived areas, in which case it is very difficult to generalise and apply the model to very deprived areas like Hull. Furthermore, there are many reasons why the prevalence could differ among practices, which are not captured by these models. Therefore, just because practices have a particularly low prevalence relative to the model, does not necessarily mean that they are performing badly in any way relative to other general practices. Nevertheless, a comparison of the differences between the modelled prevalence and the practice list registers can act as a starting point for investigation. Practices with a low prevalence or a relatively large difference between the model and the register prevalence can be examined further and considered in relation to patient characteristics using local knowledge. Robert Sheikh Iddenden, Public Health Sciences. 73

74 The resulting modelled COPD prevalence estimates are shown alongside register prevalence for each practice in Hull in Table 17 while the numbers of patients with COPD, both modelled and on the practice register, are shown in Table 18. Across all practices combined, the register prevalence in Hull was 20% lower than the Doncaster model prevalence estimates. Modelled prevalence was greater than register prevalence for 41 Hull practices (73%) the register prevalence was greater than predicted prevalence, with the difference between the modelled number of patients with COPD and the actual (registered) number was greater than 50 for 10 practices (18%). Overall, the model predicts 1,309 extra patients in Hull with COPD than currently on practice registers. Table 17: Modelled prevalence estimates (using Doncaster model) and register prevalence of COPD by GP practices in Hull 2008/09 Practice code Practice name COPD Prevalence (%) Difference (relative Modelled Register difference, %) B81001 Dr Ali & Partners (-38.0) B81002 Dr A Kumar-Choudhary ( 7.8) B81008 Dr J S Parker & Partners (-43.9) B81011 Wheeler Street Healthcare (-34.4) B81017 Kingston Medical Group ( 5.4) B81018 Dr R K Awan ( 4.7) B81020 Dr P C Mitchell And Partners (-39.0) B81021 Faith House Surgery (-18.5) B81027 St Andrews Group Practice ( 14.5) B81032 Dr A W Hussain And Partners (-26.8) B81035 Dr W G T Sande And Partners (-36.7) B81038 Dr A A Mather And Partners (-33.3) B81040 Dr P F Newman And Partners (-37.4) B81046 Dr J D Blow And Partners (-22.4) B81047 Dr J N Singh And Partners (-24.0) B81048 Dr S M Hussain And Partners (-14.2) B81049 Dr V A Rawcliffe And Partners ( 7.5) B81052 Dr J Musil And P J Queenan (-22.4) B81053 Dr M Maung And Partners (-15.5) B81054 Dr M J Varma And Partners (-34.5) B81056 The Springhead Medical Centre (-29.7) B81057 Dr S Macphie And Koul (-63.2) B81058 Dr M Foulds & Partner ( 12.8) B81066 Dr G M Chowdhury (-47.6) B81072 Dr R Percival And Partners (-16.9) B81074 Dr Ak Rej (-20.5) B81075 Dr M K Mallik (-78.0) B81080 Dr G S Malczewski (-33.7) B81081 Dr K M Tang And Partner (-35.1) Robert Sheikh Iddenden, Public Health Sciences. 74

75 Practice code Practice name COPD Prevalence (%) Difference (relative Modelled Register difference, %) B81085 Dr J W Richardson And Partners (-47.1) B81089 Dr Witvliet (-38.2) B81094 Dr A K Datta (-33.7) B81095 Dr Cook (-20.8) B81097 Dr R D Yagnik (-20.6) B81104 Dr J K Nayar (-39.6) B81112 Dr P C Ghosh ( 14.7) B81119 Dr G Palooran And Partners (-24.7) B81616 Dr G T Hendow ( -8.2) B81631 Dr R Raut (119.7) B81634 Dr J Venugopal (-26.1) B81635 Dr G Dave ( 9.3) B81644 Dr K K Mahendra ( -5.0) B81645 Dr T Abraham (-31.4) B81646 Dr J Austin (-31.4) B81662 Mizzen Road Surgery (-38.4) B81674 Dr J C Joseph ( 39.6) B81675 Dr A H Tak & Dr E G Stryjakiewicz (-14.7) B81682 Dr M Shaikh & Partners (-30.6) B81683 Dr B L Koul ( 3.4) B81685 Dr N A Poulose ( 34.6) B81688 Dr K V Gopal ( 68.5) B81690 Dr S K Ray (-25.2) B81692 The Quays Medical Centre ( 26.6) Y00955 Riverside Medical Centre ( -8.4) Y01200 The Calvert Practice (-28.3) Y02344 Northpoint ( 12.8) All Hull general practices (-19.8) Table 18: Number of patients with COPD from the Doncaster model and the practice registers by GP practices in Hull 2008/09 Practice Practice name Numbers with COPD code Modelled Register Difference B81001 Dr Ali & Partners B81002 Dr A Kumar-Choudhary B81008 Dr J S Parker & Partners B81011 Wheeler Street Healthcare B81017 Kingston Medical Group B81018 Dr R K Awan B81020 Dr P C Mitchell And Partners B81021 Faith House Surgery Robert Sheikh Iddenden, Public Health Sciences. 75

76 Practice Practice name Numbers with COPD code Modelled Register Difference B81027 St Andrews Group Practice B81032 Dr A W Hussain And Partners B81035 Dr W G T Sande And Partners B81038 Dr A A Mather And Partners B81040 Dr P F Newman And Partners B81046 Dr J D Blow And Partners B81047 Dr J N Singh And Partners B81048 Dr S M Hussain And Partners B81049 Dr V A Rawcliffe And Partners B81052 Dr J Musil And P J Queenan B81053 Dr M Maung And Partners B81054 Dr M J Varma And Partners B81056 The Springhead Medical Centre B81057 Dr S Macphie And Koul B81058 Dr M Foulds & Partner B81066 Dr G M Chowdhury B81072 Dr R Percival And Partners B81074 Dr Ak Rej B81075 Dr M K Mallik B81080 Dr G S Malczewski B81081 Dr K M Tang And Partner B81085 Dr J W Richardson And Partners B81089 Dr Witvliet B81094 Dr A K Datta B81095 Dr Cook B81097 Dr R D Yagnik B81104 Dr J K Nayar B81112 Dr P C Ghosh B81119 Dr G Palooran And Partners B81616 Dr G T Hendow B81631 Dr R Raut B81634 Dr J Venugopal B81635 Dr G Dave B81644 Dr K K Mahendra B81645 Dr T Abraham B81646 Dr J Austin B81662 Mizzen Road Surgery B81674 Dr J C Joseph B81675 Dr A H Tak & Dr E G Stryjakiewicz B81682 Dr M Shaikh & Partners B81683 Dr B L Koul B81685 Dr N A Poulose B81688 Dr K V Gopal Robert Sheikh Iddenden, Public Health Sciences. 76

77 Practice Practice name Numbers with COPD code Modelled Register Difference B81690 Dr S K Ray B81692 The Quays Medical Centre Y00955 Riverside Medical Centre Y01200 The Calvert Practice Y02344 Northpoint All Hull practices 6,628 5,319-1,309 Equivalent analyses for East Riding are shown in Table 19 (modelled COPD prevalence estimates alongside register prevalence for each practice) and Table 20 (numbers of patients with COPD, both modelled and on the practice registers). Across all practices combined, the register prevalence in East Riding was very similar to the Doncaster model prevalence estimates, although with a great deal of variability between East Riding practices. For 21 practices (57%) the register prevalence was higher than the predicted prevalence, while for 4 practices (11%) the predicted numbers of COPD patients was more than 50 higher than on the practice registers. Overall, the model predicts just 77 extra patients in East Riding with COPD than currently on practice registers. Table 19: Modelled prevalence estimates (using Doncaster model) and register prevalence of COPD by GP practices in East Riding 2008/09 Practice code Practice name COPD Prevalence (%) Difference (relative Modelled Register difference,%) B81004 Dr P I Collingwood And Partners (-33.5) B81006 Dr Farley And Partners (-15.7) B81009 Dr R W Webster And Partners ( 54.3) B81010 Hessle Grange Medical Practice ( -2.7) B81013 Dr M S Patel And Partners (-32.3) B81014 Dr H K Macnab And Partners ( -9.1) B81024 Dr M E A Moody And Partners (-30.3) B81025 Dr R D Fouracre And Partners ( 18.2) B81029 Dr J E Clark And Partners ( 14.3) B81034 Dr A J Sykes And Partners (-31.8) B81036 Dr J P Brooke And Partners ( 5.6) B81037 Dr S J Towers And Partners ( -0.5) B81041 Dr R A Ferguson And Partners ( -8.3) B81042 Dr P G Jones And Partners (-46.5) B81044 Holme-Bubwith Medical Group ( 3.9) B81050 Dr P English And Partners ( 3.9) B81051 Dr P R Mixer And Partners (-24.8) B81060 Dr P M Pickering And Partners ( 26.7) B81061 Dr W A Hart And Partners ( 54.1) Robert Sheikh Iddenden, Public Health Sciences. 77

78 Practice code Practice name COPD Prevalence (%) Difference (relative Modelled Register difference,%) B81062 Dr C J Lambert And Partners ( 16.3) B81068 Dr L Wrightson And Partners ( -7.5) B81069 Dr P A Harris And Partners ( 19.1) B81070 Dr H R P Meldrum And Partners ( 9.1) B81082 Dr R L Pearson & Partners (-32.7) B81084 Dr M Morgan And Partners (-30.3) B81088 Dr R W Harrison ( 17.7) B81092 Dr A N Crawford And Partners ( 12.3) B81100 Dr J G Best And Partner (-51.3) B81101 Dr H S Suri And Partners ( 48.0) B81120 Dr A R M Kelly And Dr B L Bawn ( 24.9) B81121 Dr G L Clayton <0.01 ( 0.6) B81602 Dr M E Hancocks (-13.3) B81619 Park View Surgery ( 3.3) B81622 Dr A M Harley And Partners (-17.8) B81653 Dr J Robson (-11.3) B81658 Dr Y Adhami (-57.9) B81666 Dr R G Mitchell (-11.6) B81679 Dr R G Clarke (-33.2) All East Riding general practices ( -1.4) Table 20: Number of patients with COPD from the Doncaster model and the practice registers by GP practices in East Riding 2008/09 Practice Practice name Numbers with COPD code Modelled Register Difference B81004 Dr P I Collingwood And Partners B81006 Dr Farley And Partners B81009 Dr R W Webster And Partners B81010 Hessle Grange Medical Practice B81013 Dr M S Patel And Partners B81014 Dr H K Macnab And Partners B81024 Dr M E A Moody And Partners B81025 Dr R D Fouracre And Partners B81029 Dr J E Clark And Partners B81034 Dr A J Sykes And Partners B81036 Dr J P Brooke And Partners B81037 Dr S J Towers And Partners B81041 Dr R A Ferguson And Partners B81042 Dr P G Jones And Partners B81044 Holme-Bubwith Medical Group B81050 Dr P English And Partners Robert Sheikh Iddenden, Public Health Sciences. 78

79 Practice Practice name Numbers with COPD code Modelled Register Difference B81051 Dr P R Mixer And Partners B81060 Dr P M Pickering And Partners B81061 Dr W A Hart And Partners B81062 Dr C J Lambert And Partners B81068 Dr L Wrightson And Partners B81069 Dr P A Harris And Partners B81070 Dr H R P Meldrum And Partners B81082 Dr R L Pearson & Partners B81084 Dr M Morgan And Partners B81088 Dr R W Harrison B81092 Dr A N Crawford And Partners B81100 Dr J G Best And Partner B81101 Dr H S Suri And Partners B81120 Dr A R M Kelly And Dr B L Bawn B81121 Dr G L Clayton B81602 Dr M E Hancocks B81619 Park View Surgery B81622 Dr A M Harley And Partners B81653 Dr J Robson B81658 Dr Y Adhami B81666 Dr R G Mitchell B81679 Dr R G Clarke All East Riding general practices 5,605 5, Modelled COPD prevalence estimates at local authority level from the Eastern Region Public Health Observatory model, with projections to 2020 In an earlier phase of their modelling of COPD prevalence (ERPHO, 2008), ERPHO also produced, at local authority level, projected estimates of the number (and prevalence) of people aged 16 years and over with COPD for 2010, 2015 and The model was developed from data from the Health Survey for England 2001, using multivariate logistic regression. Included in the model were sex, age, ethnicity, rurality, smoking status and deprivation score (Index of Multiple Deprivation 2004). From this model estimates of COPD prevalence in those aged 16+ years were produced. Resident mid-year 2005 ONS population estimates by three broad ethnic groups were used, White, Black and Asian. To produce future prevalence estimates ONS 2006-based population projections were used, but as these were not produced by ethnic group, the 2005 distribution of ethnic groups was applied to these projections, lagged by 5, 10 and 15 years to produce estimates for 2010, 2015 and 2020 respectively. National estimates of smoking prevalence were used, adjusted by the synthetic smoking estimates at local authority level, in order to produce local authority estimates to use in the model. Robert Sheikh Iddenden, Public Health Sciences. 79

80 The proportion of ex-smokers was not adjusted, with national proportion of ex-smokers applied to each local authority. Full details on the methodology used may be downloaded from the Association of Public Health Observatories website, or the Eastern Region Public Health Observatory website, (Walford, 2008). The comments in the preceding sections regarding applicability of modelled prevalence estimates to local areas apply here too. The results from this model are presented in Table 21 alongside register prevalence (also for those aged 16 years and over) from 2005 to As can be seen, there is almost a three-fold difference between register and estimated COPD prevalence in those aged 16+ for Hull (a much greater difference than in the sum of the practice level estimates that were produced later and can be seen in Table 13 (although these were at wholepopulation level). The difference between modelled and register prevalence in those aged 16+ was smaller for East Riding (modelled estimates around 50% higher), although again higher than the difference in the sum of the practice-based estimates presented in Table 15. It can also be seen that the modelled estimates of COPD prevalence are fairly static over time, suggesting these predictions have limited uses. Table 21: Estimated number and prevalence of COPD (aged 16+) 12 for , 2015 and 2020 based on Eastern Region Public Health Observatory model (ERPHO, 2008) with register prevalence Modelled (ERPHO) and register number and prevalence (%) Year of COPD in 16+ population, with projections to 2020 Hull East Riding Hull & East Riding Model Register Model Register Model Register 2005 Number 12,410 4,687 8,302 4,801 20,712 9,488 Prevalence,% Number 12,489 4,893 8,428 5,045 20,917 9,938 Prevalence,% Number 12,552 5,036 8,571 5,248 21,123 10,284 Prevalence,% Number 12,735 5,319 8,778 5,528 21,514 10,847 Prevalence,% Number 12,841 8,966 21,807 Prevalence,% Number 12,940 9,135 22,075 Prevalence,% Number 13,725 10,165 23,890 Prevalence Number 14,673 11,217 25,890 Prevalence Under the ERPHO model assumption that the prevalence of COPD I those under 16 is negligible. Robert Sheikh Iddenden, Public Health Sciences. 80

81 8.2. Hospital Activity Hospital Episode Statistics (HES) information is available for all day-case and inpatient admissions between April 2001 and March 2010 for Hull and East Riding. Inpatient admissions and day-case stays are recorded as clinician episodes. Clinician episodes represent periods of patient care under the care of a particular clinician. Approximately 85% of hospital stays have a single clinician episode, but it is possible for a patient to have more than one clinician episode if they are moved or transferred to another speciality within their hospital stay and fall under the care of another clinician. The number of clinician episodes and number of hospital stays have both been examined. The last clinician episode has been used (arbitrarily), however information from other clinician episodes within the hospital stay have been retained 13. The number of patients seen over a period of time has also been examined. If presenting the number of hospital stays, then this will comprise multiple admissions from some patients and a single admission for other patients. If examining the number of patients treated as a day-case or as an inpatient, then it is necessary to select a single hospital admission for each patient within the time period of interest. Primary and secondary diagnoses of COPD have been examined, as have the total number of admissions in order to assess the proportion of admissions due to COPD Number of Inpatient Admissions and Day-Cases Over the last three financial years (2007/08 to 2009/10), there were a total of 587,184 clinician episodes for Hull and East Riding of Yorkshire patients combined, relating to 500,720 hospital stays (last clinician episode of hospital stay) by 229,002 patients. Only 8,788 (1.5%) of the total clinician episodes and 5,694 (1.0%) of the hospital stays had a primary diagnosis of COPD, while a further 14,177 (2.4%) of the total clinical episodes had a secondary diagnosis of COPD, as did 9,414 (1.7%) of the hospital stays (for any of the clinician episodes within that hospital stay), as shown in Table 22. For each financial year 2007/08 to 2009/10 the total number of hospital episodes and hospital stays in Hull and East Riding of Yorkshire add to the combined figure for both areas. However, this is not the case for patients. Adding up the total number of patients admitted at any point during a financial year for Hull and East Riding of Yorkshire is approximately the same as the total for Hull and East Riding of Yorkshire combined. It is not exactly the same as a small number of patients will have moved between PCTs within the financial year, and each of these patients contributes once to Hull and once to East Riding of Yorkshire, but only once to Hull and East Riding of Yorkshire combined. 13 For example, diagnoses of COPD is not necessarily coded on every single clinician episode within a patient s stay, and if using the diagnoses coding for the last clinician episode only rather than retaining all information from all clinician episodes within the stay, approximately 5% of the diabetes hospital stays would be missed. Robert Sheikh Iddenden, Public Health Sciences. 81

82 Similarly the total number of patients admitted overt the three year period is not the sum of the three financial years, due to the number of patients that are admitted in two or three of the financial years making up the three-year period. Overall, there were 3,172 patients seen as day cases or admitted as inpatients with a primary diagnosis of COPD over the three-year period 2007/08 to 2009/10 in Hull and East Riding of Yorkshire, with a further 4,130 patients with a secondary diagnosis of COPD. The patients with primary diagnoses of COPD had a total of 5,694 hospital stays between them, giving an average of almost 1.8 hospital stays per patient over the threeyear period. The patients with secondary diagnoses of COPD had a total of 9,414 hospital stays between them, giving an average of almost 2.3 hospital stays per patient over the three-year period, similar to the average number of hospital stays per patients with any diagnosis at 2.2. The 5,694 hospital stays in patients with a primary diagnosis of COPD resulted in 8,788 clinician episodes, suggesting that at least two thirds of hospital stays where COPD was the primary diagnosis had a single clinician episode, similar to those with a secondary diagnosis only. For all diagnoses, at least 85% of hospital stays had a single clinician episode. Therefore, patients with COPD (whether the primary or a secondary diagnosis) appear to have moved from the care of one clinician to the care of another clinician more frequently within their hospital stays than other patients. This may be because patients with COPD, which is more prevalent in older people, are likely to have co-morbidities treated by different medical specialities. Robert Sheikh Iddenden, Public Health Sciences. 82

83 Table 22: Hospital activity (inpatient and day-case visits) during 2007/08 and 2009/10 Type of Analysis Area Financial year diagnosis unit Hull ERoY Total 2007/ , , , / , , ,618 Clinician 2009/ , , ,637 episodes Annual average 07/08 to 09/10 90, , ,728 Total 07/08 to 09/10 271, , ,184 All diagnoses All COPD diagnoses 1,2 COPD primary diagnosis only 1 Hospital stays Patients admitted Clinician episodes Hospital stays Patients admitted Clinician episodes Hospital stays 2007/ ,244 89, , / ,245 90, , / ,816 93, ,240 Annual average 07/08 to 09/10 75,768 91, ,907 Total 07/08 to 09/10 227, , , / ,996 54,023 99, / ,793 56, , / ,545 57, ,684 Annual average 07/08 to 09/10 45,445 56, ,263 Total 07/08 to 09/10 100, , , /2008 3,876 3,023 6, /2009 4,272 3,564 7, /2010 4,351 3,879 8,230 Annual average 07/08 to 09/10 4,166 3,489 7,655 Total 07/08 to 09/10 12,499 10,466 22, /2008 2,556 2,111 4, /2009 2,707 2,364 5, /2010 2,736 2,634 5,370 Annual average 07/08 to 09/10 2,666 2,370 5,036 Total 07/08 to 09/10 7,999 7,109 15, /2008 1,560 1,389 2, /2009 1,685 1,568 3, /2010 1,687 1,732 3,412 Annual average 07/08 to 09/10 1,644 1,563 3,200 Total 07/08 to 09/10 3,639 3,697 7, /2008 1,513 1,109 2, /2009 1,769 1,409 3, /2010 1,736 1,252 2,988 Annual average 07/08 to 09/10 1,673 1,257 2,929 Total 07/08 to 09/10 5,018 3,770 8, /2008 1, , /2009 1, , /2010 1, ,814 Annual average 07/08 to 09/10 1, ,898 Total 07/08 to 09/10 3,228 2,466 5, / , / ,467 Patients 2009/ ,286 admitted Annual average 07/08 to 09/ ,325 Total 07/08 to 09/10 1,719 1,470 3,172 1 COPD recorded for any of the clinician episodes. 2 Primary or secondary diagnoses Robert Sheikh Iddenden, Public Health Sciences. 83

84 Admission method Table 23 gives the average annual number of hospital admissions by admission method where at least one clinician episode involved a primary or secondary diagnosis of COPD. Only 5 admissions for Hull 2006/07 to 2008/09 and 64 for East Riding 2005/06 to 2007/08 were recorded with an admission method that was not elective or emergency, and these are not included in this table. It is rare to have an elective admission for COPD, and it can be seen that only 2.4% of admissions where the primary diagnosis was COPD were elective admissions. Table 23: Average annual number of hospital admissions for hospital admissions having a primary or secondary diagnosis of COPD, 2007/08 to 2009/10 Average annual number of hospital COPD Admission method admissions for COPD by PCT diagnosis Hull ERoY Total Primary Emergency admission 1, ,847 diagnosis Elective admission Secondary Emergency admission 1, ,154 diagnosis Elective admission Any COPD Emergency admission 2,235 1,765 4,000 diagnosis Elective admission ,013 Table 24 gives the average annual number of hospital admissions between 2007/08 and 2009/10 for different primary diagnoses where at least one clinician episode involved a primary or secondary diagnosis of COPD. As primary diagnosis codes could differ over the period of stay, e.g. a patient may be admitted as an emergency and diagnosis may not be accurate initially until further tests are completed, the primary diagnosis for the final clinician episode has been reported. Some diagnosis categories have been combined so that the total number within a group is more than five. The number of admissions has been rounded down to the nearest whole number, so totals may not agree. Some patients will have been included more than once as they have contributed to the count in different ways. For instance, a patient could have been admitted on two separate occasions with two different diagnoses, or with differing admission methods. Each time the patient was admitted within the three-year period will be included as the number gives a count of the average number of hospital admissions over the three-year period rather than the average number of patients admitted. Robert Sheikh Iddenden, Public Health Sciences. 84

85 Elective admissions Emergency admissions Admission method Table 24: Average annual number of hospital admissions by main diagnoses for hospital admissions having a primary or secondary diagnosis of COPD, 2007/08 to 2009/10 Primary diagnosis Average annual hospital admissions 2007/ /10 with primary or secondary COPD diagnosis 1 Hull ERoY Total Lung cancer Other cancers, including benign and in situ neoplasms Coronary heart disease Stroke Heart failure Other diseases of the circulatory system Influenza & pneumonia Chronic obstructive pulmonary disease 1, ,846 Other diseases of the respiratory system Injury, poisoning & certain other consequences of external causes Certain infectious & parasitic diseases Endocrine, nutritional & metabolic diseases Diseases of the nervous system, eye, adexa, ear & mastoid process Diseases of the digestive system Diseases of the skin, subcutaneous tissue, musculoskeletal system & connective tissue Diseases of the genitourinary system Symptoms & signs - circulatory and respiratory systems Other causes of admission All emergency admissions 2,235 1,765 4,000 Lung cancer Other cancers, including benign and in situ neoplasms Coronary heart disease Other diseases of the circulatory system Chronic obstructive pulmonary disease Other diseases of the respiratory system Diseases of the nervous system, eye, adexa, ear & mastoid process Diseases of the digestive system Diseases of the skin, subcutaneous tissue, musculoskeletal system & connective tissue Diseases of the genitourinary system Symptoms & signs - circulatory and respiratory systems Other causes of admission All elective admissions , /07 to 2008/09 for Hull, 2005/06 to 2007/08 for East Riding and Total (Hull and East Riding combined) Robert Sheikh Iddenden, Public Health Sciences. 85

86 Emergency admissions Admission method Table 25 gives the percentages so that different primary diagnoses and admission methods can be compared. It can be seen that there were a wide range of primary diagnoses where COPD was the secondary diagnosis. Of all admissions where COPD was one of the diagnoses, COPD was the most common primary diagnosis, almost half of all emergency admissions and 40% of elective and emergency admissions combined in Hull and 35% in East Riding. Other main categories of primary diagnoses were circulatory diseases, cancers, influenza & pneumonia and other respiratory diseases (10%, 6%, 5% and 5% respectively of elective and emergency admissions across Hull and East Riding). Table 25: Percentage of hospital admissions by main diagnoses for hospital admissions having a primary or secondary diagnosis of COPD, Hull 2006/08 to 2008/09, East Riding and combined total 2005/06 to 2007/08 Hospital admissions with primary or secondary Primary diagnosis diagnosis of COPD (%) Hull ERoY Total Lung cancer Other cancers, including benign and in situ neoplasms Coronary heart disease Stroke Heart failure Other diseases of the circulatory system Influenza & pneumonia Chronic obstructive pulmonary disease Other diseases of the respiratory system Injury, poisoning & certain other consequences of external causes Certain infectious & parasitic diseases Endocrine, nutritional & metabolic diseases Diseases of the nervous system, eye, adexa, ear & mastoid process Diseases of the digestive system Diseases of the skin, subcutaneous tissue, musculoskeletal system & connective tissue Diseases of the genitourinary system Symptoms and signs involving the circulatory and respiratory systems Other causes of admission All emergency admissions Robert Sheikh Iddenden, Public Health Sciences. 86

87 All admissions Elective admissions Admission method Primary diagnosis Hospital admissions with primary or secondary diagnosis of COPD (%) Hull ERoY Total Lung cancer Other cancers, including benign and in situ neoplasms Coronary heart disease Other diseases of the circulatory system Chronic obstructive pulmonary disease Other diseases of the respiratory system Diseases of the nervous system, eye, adexa, ear & mastoid process Diseases of the digestive system Diseases of the skin, subcutaneous tissue, musculoskeletal system & connective tissue Diseases of the genitourinary system Symptoms and signs involving the circulatory and respiratory systems Other causes of admission All elective admissions Lung cancer Other cancers, including benign and in situ neoplasms Coronary heart disease Stroke Heart failure Other diseases of the circulatory system Influenza & pneumonia Chronic obstructive pulmonary disease Other diseases of the respiratory system Injury, poisoning & certain other consequences of external causes Certain infectious & parasitic diseases Endocrine, nutritional & metabolic diseases Diseases of the nervous system, eye, adexa, ear & mastoid process Diseases of the digestive system Diseases of the skin, subcutaneous tissue, musculoskeletal system & connective tissue Diseases of the genitourinary system Symptoms and signs involving the circulatory and respiratory systems Other causes of admission All elective or emergency admissions Robert Sheikh Iddenden, Public Health Sciences. 87

88 Seasonal differences in emergency admissions Given the nature of COPD, with the complications arising from respiratory infections in severe COPD patients, emergency hospital admissions tend to be higher in the colder months of winter than in the more clement months. This seasonality can be seen by examining the trends in the number of emergency admissions for COPD by the month of admission. These trends are shown in Figure 19 for Hull and East Riding combined. As can be seen from the graph, the peak numbers of admissions in the 2008/09 winter were almost double those in preceding and succeeding months. Of course, in milder winters the peaks are lower. Figure 19: Monthly emergency admissions with primary diagnosis COPD in Hull and East Riding of Yorkshire residents 2007/08 to 2009/ Monthly emergency admissions with primary diagnosis of COPD in residents of Hull and East Riding from April 2007 to March Procedures Apr07 May07 Jun07 Jul07 Aug07 Sep07 Oct07 Nov07 Dec07 Jan08 Feb08 Mar08 Apr08 May08 Jun08 Jul08 Aug08 Sep08 Month and year Oct08 Nov08 Dec08 Jan00 Feb09 Mar09 Apr09 May09 Jun09 Jul09 Aug09 Sep09 Oct09 Nov09 Dec09 Jan10 Feb10 Mar10 Due to the nature of the disease, few COPD patients require surgical interventions to control their disease or symptoms. Because of this, the majority of hospital admissions with a primary diagnosis of COPD did not have any procedures recorded (95%). Therefore no analyses of procedure rates were feasible. Robert Sheikh Iddenden, Public Health Sciences. 88

89 Admission Rates Relative to Prevalence It is possible to calculate the age-gender standardised emergency hospital admission rates where COPD is the primary diagnosis for each general practice (per 100,000 practice patients) and then compare this to the QOF registered prevalence for each practice (see section 8.1.1). Practices with the higher registered prevalence appeared to have the higher emergency admission rates (Figure 20). While the linear relationship was statistically significant and stronger for Hull practices than for East Riding practices (details not shown here), the prevalence rate was not a strong predictor of emergency admissions. Figure 20: Plot of age-gender standardised 2007/08 to 2009/10 (pooled) COPD emergency admission rate per 100,000 practice population and the practice level QOF 2008/09 prevalence (%) Age-gender standardised emergency hospital admission rates where COPD was the primary diagnosis for each general practice (per 100,000 practice patients) may also be compared with the difference between the QOF registered prevalence and the ERPHO model estimates of COPD prevalence (see section 8.1.2). Here PCT-specific linear relationships are presented in Figure 21, as the direction of any linear relationships between age-gender standardised emergency admissions for COPD and the absolute differences between QOF registered prevalence and ERPHO modelled prevalence may differ by PCT. For Hull practices those practices with the higher undiagnosed prevalence of COPD appeared to have lower emergency admissions. This linear relationship was barely Robert Sheikh Iddenden, Public Health Sciences. 89

90 statistically significant at the conventional 5% level (p=0.044) and the difference between modelled and register prevalence explained little of the variation in emergency admission rates. For each percentage point increase in the absolute difference between QOF and ERPHO modelled prevalence the predicted age-gender standardised COPD emergency admission rate decreased by 0.44 per 100,000. This provides some evidence that the ERPHO model might over-estimate COPD prevalence in Hull, as one would expect higher emergency admission rates in undiagnosed patients, as these patients would not be on the correct treatment regimen to control their disease, and would therefore be more likely to experience the acute exacerbations that commonly cause emergency admissions in patients with COPD. However, were undiagnosed patients more likely to have mild to moderate COPD rather than moderate to severe disease, then emergency admission rates would be likely to be lower among undiagnosed patients. For East Riding practices there was little relationship between emergency admissions for COPD and the absolute difference between QOF and ERPHO modelled prevalence. Practices with higher differences between registered and modelled prevalence had slightly lower age-gender standardised COPD emergency admission rates with a 0.15 per 100,000 decrease in emergency admission rates for each percentage point increase in the absolute difference between QOF registered and ERPHO modelled prevalence. Figure 21: Plot of age-gender standardised 2007/08 to 2009/10 (pooled) COPD emergency admission rate per 100,000 practice population and the difference between registered and ERPHO modelled prevalence of COPD at practice level, Hull practices only Robert Sheikh Iddenden, Public Health Sciences. 90

91 8.3. Mortality Primary Cause of Death Mortality data (primary cause of death only) is available from the Public Health Mortality File (PHMF) produced by the Office for National statistics (ONS) for both Hull and East Riding of Yorkshire. There were an average of around 2,500 deaths among Hull residents and 3,500 deaths among East Riding residents each year between 2004 and Over this time period around 5% of all deaths were due to COPD (Table 26) with higher proportions in Hull (rising from 4.9% in 2004 to 6.6% in 2008) than in East Riding (around 4% per annum). The differences between Hull and East Riding are likely to be due to the higher rates of smoking in Hull. Table 26: Deaths during with COPD recorded as the primary cause of death Year of death Number and percentage of deaths with COPD recorded as the primary cause of death with percentage of all deaths Hull East Riding Hull and East Riding n % n % n % Total , Table 27 gives the mortality rate per 100,000 residents of Hull and East Riding over the period , both crude and adjusted for age and gender. Crude and age-genderadjusted mortality rates in East Riding were relatively stable, whereas in Hull they have increased, although the largest increase was between 2004 and 2005, whence they have been relatively stable. It might be that deaths from COPD in 2004 were particularly low for some reason, and that the long term trend was higher. Indeed when looking at numbers of deaths we see that the 123 COPD deaths seen in Hull in 2004 was indeed the lowest number of COPD deaths throughout However, what we can see from Table 27 is that the age-gender-adjusted rates in Hull are around one quarter higher than the crude rates, whereas in East Riding the age-gender-adjusted rates are between 5% and 12% lower than the crude rates. As part of the age-standardisation process the weight given to deaths occurring at younger ages is greater than the weight given to deaths in older people. Therefore the higher age-adjusted rates relative to crude rates seen in Hull, and the lower age-adjusted rates relative to crude rates in East Riding, tells us that COPD deaths in Hull tend to occur at a younger age than in East Riding. Robert Sheikh Iddenden, Public Health Sciences. 91

92 Table 27: Mortality rate per 100,000 resident populations for deaths during 2004 to 2008 where COPD recorded as the primary cause of death Year of death Mortality rate per 100,000 resident population for deaths with COPD recorded as the primary cause of death Hull East Riding Hull and East Riding Crude Adjusted Crude Adjusted Crude Adjusted Total The Compendium of Clinical and Health Outcomes Knowledge Base ( provides information on standardised mortality rates by geographical areas, including PCTs, across England. Figure 22 gives the indirectly standardised mortality ratio (SMR) for all ages for mortality where COPD was the primary cause of death for England, Hull and East Riding of Yorkshire. An explanation of SMRs is given in Statistical Methods and Terms on page 279. There is some variability in annual SMRs for Hull and East Riding of Yorkshire as the numbers of deaths are relatively small, so the SMRs have been recalculated for each three year period, which provides SMRs that are more stable, but are still subject to some random variation at a PCT level. The mortality rates are compared with those in 2008 for England (SMR=100). The general trend is of decreasing mortality rates for both Hull and East Riding, similar to the trends for the region and for England. SMRs for Hull did increase for the last two periods, but this may be due to random variation, and the steepest increase in is an artefact of the low number of deaths in Mortality rates were highest in Hull (56% higher in than in England in 2008) and lowest in East Riding (14% lower in than in England in 2008). Robert Sheikh Iddenden, Public Health Sciences. 92

93 All age standardised mortality ratio for COPD Figure 22: Trends in indirectly standardised mortality ratio for COPD for all ages Year Hull East Riding of Yorkshire Yorkshire & The Humber SHA England In order to examine premature mortality it is desirable to examine trends in SMRs for those aged under 75 years. However, this information is only available for the most recent periods from the Compendium, and these are shown in Table 28. Compared to England, the under 75 SMR for COPD for was 178 (95% CI 152 to 207) for Hull and 75 (62, 89) for East Riding. Therefore, COPD mortality rates are 25% lower in East Riding of Yorkshire than the England average but 78% higher in Hull, statistically significant in each case. While under 75 SMRs are relatively stable in East Riding, they have increased substantially in the most recent period ( ) for Hull. This might represent a real and sustained increase in COPD mortality in those aged under 75 in Hull, or it may be due to random variation in the annual numbers of COPD deaths at a PCT level, where the number of COPD deaths under 75 in Hull in 2008 was the highest since 1998, and may well be lower next year. Robert Sheikh Iddenden, Public Health Sciences. 93

94 Table 28: Under 75 SMRs to where COPD recorded as the primary cause of death Under 75 SMRs where primary cause of death was recorded as COPD Years (95% confidence intervals) Hull East Riding YHSHA SMR (95% CI) SMR (95% CI) SMR (95% CI) (128, 177) 79 (66, 93) 117 (113, 122) (118, 167) 71 (58, 85) 116 (111, 120) (132, 183) 74 (62, 89) 119 (114, 124) (152, 207) 75 (62, 89) 118 (113, 123) Secondary Cause of Death Although the Public Health Mortality File does not contain secondary causes of death, the Primary Care Mortality Database (PCMD) does include secondary causes of death, but currently it has only been possible to obtain data for patients registered with Hull GPs, for 2005 and The number and percentage of deaths with a primary or secondary diagnosis of COPD has been given in Table 29 for patients registered with Hull GPs. It is not possible to know how important the secondary cause COPD is in the cause of death. For some deaths, it may be a very important secondary cause of death, but for others it could be a condition that was noted on the death certificate that had relatively little influence on the death. Table 30 gives the mortality rate per 100,000 registered population for patients registered with a Hull GP, for 2005 and Around 40% of all deaths where COPD was mentioned on the death certificate were primarily caused by COPD. Table 29: Deaths involving a primary or secondary diagnosis of COPD for patients registered with a Hull GP Year of death Number (percentage) of deaths involving a primary or secondary diagnosis of COPD for patients registered with a Hull GP Primary diagnosis Secondary diagnosis All diagnoses (5.9) 246 (9.2) 403 (15.1) (6.0) 233 (8.5) 397 (14.5) Total 321 (5.9) 479 (8.9) 800 (14.8) Robert Sheikh Iddenden, Public Health Sciences. 94

95 Table 30: Mortality rate per 100,000 population for deaths involving a primary diagnosis of COPD for patients registered with a Hull GP Year of death Mortality rate per 100,000 resident population for deaths involving a primary diagnosis of COPD for patients registered with a Hull GP (95% CI) Primary diagnosis Secondary diagnosis All diagnoses (46.8, 64.5) 86.4 (75.9, 97.9) (128.0, 156.0) (49.2, 67.2) 81.9 (71.7, 93.1) (126.1, 153.9) Total 56.4 (50.4, 62.9) 84.1 (76.8, 92.0) (131.0, 150.6) The (first recorded) secondary causes of death for the 321 deaths where the underlying cause was COPD are given in Table % of deaths where the underlying cause of death was COPD had a further respiratory cause of death recorded as the first recorded secondary cause of death, while a further 11% had circulatory system diseases recorded. In 15% of cases there was no secondary cause of death recorded on the death certificate. Table 31: Secondary cause of death for deaths where underlying cause is COPD for patients registered with a Hull GP dying during Secondary cause (ICD10) for deaths where underlying cause is COPD for patients registered with a Hull GP, N % Bronchopneumonia, unspecified (J18.0) Pneumonia, unspecified (J18.1 J81.9) Respiratory failure, unspecified (J96.9) Other specified respiratory disorders (J98.8) Other respiratory diseases (J-, not included above) Diseases of the circulatory system (I00 I99) Septicaemia, unspecified (A41.9) Other No other cause specified Total The underlying causes of death for the 479 deaths where COPD was recorded as a secondary 14 cause of death are given in Table 32. The largest proportion of deaths where a secondary cause of death was COPD, had circulatory system diseases as the underlying cause (18%), followed by respiratory diseases (17%) and diseases of the digestive system (17%). 15% had cancer as the underlying cause of death, with almost two thirds of cancer deaths due to cancer of the trachea, bronchus and lung (9% of all deaths where COPD recorded as a secondary cause of death). 14 With up to eight secondary causes allowed. Robert Sheikh Iddenden, Public Health Sciences. 95

96 Table 32: Underlying cause of death where COPD is recorded as a secondary cause of death for patients registered with Hull GPs, Underlying cause of death (ICD10) for deaths were COPD is a secondary cause of death for patients registered with a Hull GP, N % Certain infectious and parasitic diseases (A-) Malignant neoplasm of trachea, bronchus and lung (C33-C34) Other malignant neoplasms (C-, not C33-C34) Ischaemic heart disease (I21-I25) Other forms of heart disease (I10-I20, I26-I52) Cerebrovascular diseases (I60-I69) Other diseases of the circulatory system (I-, not mentioned above) Asthma, unspecified (J459) Bronchiectasis (J47) Pneumonitis due to food and vomit (J690) Other respiratory diseases principally affecting the interstitium (J80-84) Other respiratory diseases (J85-J99) Diseases of the digestive system (K-) Diseases of the skin and subcutaneous tissue (L-) Diseases of musculoskeletal system and connective tissue (M-) Disease of the genitourinary system (N-) Congenital malformations, deformations and chromosomal abnormalities (Q-) External causes of morbidity and mortality (V-, W-, X-, Y-) Other Total Seasonal differences in COPD mortality As for emergency hospital admissions, there are seasonal variations in the number of deaths due to COPD. Monthly deaths where the underlying cause of death was COPD are shown for Hull and East Riding combined in Figure 23 for The peaks and troughs are similar to those seen for emergency admissions, suggesting that many of the emergency admissions for COPD come near the end of life for those patients. There is certainly evidence to suggest that emergency admissions for COPD increase in the latter stages of the disease. Robert Sheikh Iddenden, Public Health Sciences. 96

97 Figure 23: Monthly deaths where COPD recorded as the primary cause of death, Hull and East Riding of Yorkshire residents 2006 to Place of death Monthly deaths where COPD recorded as the primary cause in residents of Hull and ERoY from April 2006 to March 2009 Jun07 Jan06 Feb06 Mar06 Apr06 May06 Jun06 Jul06 Aug06 Sep06 Oct06 Nov06 Dec06 Jan07 Feb07 Mar07 Apr07 May07 Month and year Deaths Median Jul07 Aug07 Sep07 Oct07 Nov07 Dec07 Jan07 Feb08 Mar08 Apr08 May08 Jun08 Jul08 Aug08 Sep08 Oct08 Nov08 Dec08 The place of death for deaths registered during where COPD was the primary cause of death are presented in Table 33, along with the percentage of non-copd deaths by place of death. Among males and females combined COPD deaths occurring at home were slightly higher in Hull (19.5%) than East Riding (18.6%), although deaths in hospital were also higher among residents of Hull (70.9%) than East Riding (65.6%). Men in Hull were less likely to die at home from COPD (20%) than men in East Riding (24%) while conversely women in Hull were more likely to die at home from COPD (19%) than women in East Riding (13%). Amongst men in Hull and women in East Riding the percentage of deaths occurring at home were similar for both COPD and non-copd deaths. Amongst Hull women 19% of COPD deaths occurred at home compared with less than 15% of non-copd deaths, whilst in East Riding men dying from COPD were more likely to die at home (24%) than men dying from other cases (22%). Hospital deaths, on the other hand, were more common among COPD deaths than non-copd deaths in men and women in Hull and women in East Riding, while amongst men in East Riding they were only slightly higher. Men and women in East Riding were also more likely to die from COPD in residential homes than their counterparts in Hull. Robert Sheikh Iddenden, Public Health Sciences. 97

98 Other* COPD In Hull an end of life commissioning strategy has been developed that seeks to extend the diagnosis of end of life, and thus to place more patients of an end of life care pathway. It is expected that implementation of this strategy amongst providers will lead to larger percentages of deaths occurring at home, including deaths due to COPD. Table 33: Number and percentage of deaths where COPD was the primary cause by place of death , by place of death as recorded on the death certificate, and percentage of non-copd deaths by place of death. Cause and place of death Deaths by place of death Hull East Riding Hull & East Riding Men Women Men Women Men Women n % n % n % n % n % n % Home Resid. home NHS est Other Home Resid. home NHS est Other *Excludes 55 deaths where cause of death was missing (0.3% of all deaths) 8.4. Risk Factors Smoking Smoking is the main risk factor for COPD, and the prevalence of smoking among adults in Hull and East Riding is known from local surveys, the most recent being the 2009 Prevalence and Social Capital surveys in Hull and the 2009 Health and Lifestyle survey in East Riding. The East Riding 2009 survey was a postal survey, whereas the 2009 surveys in Hull were both conducted by trained interviewers. National prevalence is also known, from the Health Survey for England (HSE, 2009), the most recent survey relating to 2008 and published in Age-adjusted prevalence rates are shown in Table 34. Smoking prevalence in Hull was 54% higher in Hull than England (44% higher among men and 65% higher among women), while smoking prevalence in East Riding was 26% lower than England (25% lower among men and 27% lower among women). Robert Sheikh Iddenden, Public Health Sciences. 98

99 Age-adjusted percentage of respondents Table 34: Age-adjusted 1 smoking prevalence among adults 2 by gender in England 2008, Hull 2009, and East Riding 2009 Area Age-adjusted smoking prevalence among adults 2 (%) Males Females All England Hull East Riding Standardised to the combined Hull and East Riding population Adults defined as 16 years and over The effect of smoking on COPD risk is cumulative, and ex-smokers are also at high risk of developing COPD, therefore Figure 24 shows the age-adjusted percentages of men and women in Hull (2009), East Riding (2009) and England (2008) that reported they smoked or used to smoke. Figure 24: Smoking status by gender in survey respondents in Hull 2007, East Riding 2009 and England 2008, age-adjusted percentages 70 Ex-smokers Current smokers Hull 2009 East Riding 2009 England 2008 Hull 2009 East Riding England Males Females Robert Sheikh Iddenden, Public Health Sciences. 99

100 This shows that while the age-adjusted percentages in Hull that defined themselves as current smokers were more around double the age-adjusted percentages for East Riding, and 44-65% higher than for England, the differences between these areas between these areas were much reduced once ex-smokers were included, with percentages of current or ex-smokers in Hull men 20% higher than East Riding and 14% higher than England, and in Hull women 25% higher than East Riding and 27% higher than for England. Despite East Riding having around one quarter fewer current smokers than England, once exsmokers were included, the percentages for East Riding were very similar to England, 5% lower among men and 2% higher among women, as far more East Riding men and women had stopped smoking (31% and 29% respectively) than had men and women in Hull (25% and 21%) or England (28% and 23%) Access to Services and Quality of Care NICE Clinical Guidelines 2004 & 2010 The National Institute for Health and Clinical Excellence (NICE) produced its original clinical guideline in 2004 and replaced this with an updated guideline in June Although some of the 2004 recommendations have been deleted, and same new ones added, there are still almost 200 specific recommendations for the management of COPD, from which certain key priorities for implementation have been identified (those recommendations that would make the most difference to the largest number of people and the NHS as a whole). N.B. Where a key recommendation has been updated it is marked as such: Diagnose COPD A diagnosis of COPD should be considered in patients over the age of 35 who have a risk factor (generally smoking) and who present with exertional breathlessness, chronic cough, regular sputum production, frequent winter bronchitis or wheeze. Measure post-bronchodilator spirometry to confirm the diagnosis of COPD (updated recommendation). Stop smoking Encouraging patients with COPD to stop smoking is one of the most important components of their management. All COPD patients still smoking, regardless of age, should be encouraged to stop, and offered help to do so, at every opportunity. (Unless contraindicated, NRT plus a support programme should be offered to all COPD patients who are planning to quit - updated recommendation). Robert Sheikh Iddenden, Public Health Sciences. 100

101 Promote effective inhaled therapy In people with stable COPD who remain breathless or have exacerbations despite use of short-acting bronchodilators as required, offer the following as maintenance therapy: - if forced expiratory volume in 1 second (FEV1) 50% predicted: either long-acting beta2 agonist (LABA) or longacting muscarinic antagonist (LAMA) - if FEV1 < 50% predicted: either LABA with an inhaled corticosteroid (ICS) in a combination inhaler, or LAMA. Offer LAMA in addition to LABA + ICS to people with COPD who remain breathless or have exacerbations despite taking LABA + ICS, irrespective of their FEV1. Provide pulmonary rehabilitation for all who need it Pulmonary rehabilitation should be made available to all appropriate people with COPD including those who have had a recent hospitalisation for an acute exacerbation. (updated recommendation) Use non-invasive ventilation Non-invasive ventilation (NIV) should be used as the treatment of choice for persistent hypercapnic ventilatory failure during exacerbations not responding to medical therapy. It should be delivered by staff trained in its application, experienced in its use and aware of its limitations. When patients are started on NIV, there should be a clear plan covering what to do in the event of deterioration and ceilings of therapy should be agreed. Manage exacerbations The frequency of exacerbations should be reduced by appropriate use of inhaled corticosteroids and bronchodilators, and vaccinations. The impact of exacerbations should be minimised by: - giving self-management advice on responding promptly to the symptoms of an exacerbation starting appropriate treatment with oral steroids and/or antibiotics - use of non-invasive ventilation when indicated - use of hospital-at-home or assisted-discharge schemes. Ensure multidisciplinary working COPD care should be delivered by a multidisciplinary team. - NICE (2010A), pages 4-5 Robert Sheikh Iddenden, Public Health Sciences. 101

102 Other updated recommendations include an updated classification of severity of airflow obstruction to align it with other international guidelines. These are shown in Table 35. Table 35: Updated NICE recommendations on severity of airflow obstruction Updated NICE recommendations on severity of airflow obstruction FEV 1 (% predicted) NICE 2004 NICE Not classified Stage 1 : Mild* Mild Stage 2 : Moderate Moderate Stage 3 : Severe <30 Severe Stage 4 : Very severe * Symptoms should be present to diagnose COPD in people with mild airflow obstruction Or when FEV 1 <50% is accompanied by respiratory failure Realigning the severity of airflow obstruction to agree with other international guidelines will lead to some patients having their severity stage reclassified but they will need reassurance that their clinical condition, treatment and prognosis all remain unchanged. Recommendations on management of exacerbations and the use of NIV were not included in the scope of the update so remain unchanged from the 2004 guideline National Chronic Obstructive Pulmonary Disease National Audit 2008 The third national audit of COPD was conducted during 2008, by the Royal College of Physicians (RCP), the British Thoracic Society (BTS) and the British Lung Foundation (BLF). This resulted in five reports, which are briefly outlined below, together with the key findings and recommendation from each report. The overall aims of this audit were as follows: This third round of national COPD audit has been carried out to assess progress since the 2003 National COPD Audit (RCP, 2003) and the 2004 NICE guidance (NICE, 2004). The audit aims to: Enable units to compare their performance against national standards. Identify resource and organisational factors that may account for observed variations in outcome. Facilitate improvement in the quality of care. Identify changes since the 2003 National COPD Audit - Report of the National Chronic Obstructive Pulmonary Disease Audit 2008: clinical audit of COPD exacerbations admitted to acute NHS units across the UK, page 11 (RCP, 2008C). Robert Sheikh Iddenden, Public Health Sciences. 102

103 Thus the 2008 national COPD Audit did not seek to be representative of COPD patients, but rather to look at ongoing management and care of patients with acute exacerbations of COPD. Report of the National Chronic Obstructive Pulmonary Disease Audit 2008: Resources and Organisation of care in Acute NHS units across the UK (RCP, 2008A); a web-based cross-sectional survey to audit resources and organisation of care for people with COPD attending participating NHS acute units. Key findings and recommendations from this report were as follows: Key Findings 1. There has been a significant increase in the provision of some COPD specific services since 2003, such as early discharge programmes (44% to 59%) and the availability of Non-Invasive Ventilation (NIV) for acute respiratory failure (89% to 97%). 2. There has been a significant improvement since 2003 in staffing available to meet the needs of COPD patients but one-third of units report current staffing vacancies; many units still fall below the staffing levels recommended by The Royal College of Physicians (RCP 2008F) There are many examples of self reported good practice and service innovation. 4. Many of the COPD-specific services are of variable quality. Services often fail to meet the indicators included in the audit and none of the participating units achieved all the 46 COPD service quality standards in full. 5. Specific deficiencies in NIV and oxygen services are noteworthy. Many quality indicators for NIV and oxygen assessment are not met in participating units. This is of concern in this group of particularly ill patients. 6. There are serious deficiencies in the provision of information to patients across all COPD services. Few Units provide information to patients and carers about the use of NIV, about end of life care and the availability of local services. 15 Royal College of Physicians, Consultant Physicians working with patients: the duties, responsibilities and practice of physicians in medicine (4th edition) Robert Sheikh Iddenden, Public Health Sciences. 103

104 7. In general there is limited provision of end of life and palliative care services but some excellent examples of good practice as exemplified by participants in the National COPD Resources and Outcomes Project (NCROP) which is reported elsewhere (Roberts et al, 2008). 8. Although engagement with this National Audit has been outstanding, audit of local service provision is not wellimplemented in most Units. Without such key audit data it is difficult to see how Provider Units, Commissioners or patients themselves may judge the quality of the COPD service or the outcomes that are critical to patient care. Recommendations 1. Further investment is necessary to achieve recommended levels in staffing and to provide comprehensive high quality COPD services for patients. 2. Units should consider how they may meet recommended national and local quality indicators and develop agreed service improvement plans with their Commissioners. 3. Units should undertake regular audit of their COPD service against national quality indicators as a mechanism of both identifying areas requiring improvement and demonstrating good practice. 4. There is a specific need to review and improve the current provision of information to patients about specific medical interventions (e.g. NIV) and service provision (e.g. Palliative Care and Early Discharge schemes). 5. Extra resources are required specifically to improve palliative and end-of-life care provision for patients with COPD. 6. There should be a mechanism for disseminating innovation and good clinical practice amongst units and healthcare professionals wishing to develop their COPD services. -Report of the National Chronic Obstructive Pulmonary Disease Audit 2008: Resources and Organisation of care in Acute NHS units across the UK, pages 7-8 (RCP, 2008A). Report of the National Chronic Obstructive Pulmonary Disease Audit 2008, UK Primary Care Organisations: Resources and Organisation of Care (RCP, 2008B); UK-wide crosssectional paper-based questionnaire in participating Primary Care Organisations to survey resources and organisation of care for people with COPD in their locality. 73% of PCOs participated in the survey. 50% of PCOs had a written plan to develop COPD services, while a further 39% were in development. Of these 89% of participating Robert Sheikh Iddenden, Public Health Sciences. 104

105 PCOs, 97% included pulmonary rehabilitation; 90% include early discharge schemes; 93% included admissions avoidance and 78% included palliative care. 46% of PCOs had an agreed care pathway for COPD, while a further 45% had one in development; 70% provided pulmonary rehabilitation. Patients had access to an early discharge scheme in 73% of PCOs, and access to an admission avoidance scheme in 75% of PCOs. 53% of PCOs reported that long term oxygen therapy (LTOT) assessments took place in hospital only, 4% in primary care only with 40% reporting they took place in hospitals and primary care. Only 44% of PCOs had formal arrangements for patients with COPD to receive palliative care. While the report did not present results by individual PCOs, but did highlight areas of good or innovative practice as reported by PCOs. In Hull and East Riding we have finished comprehensive clinical guidelines for COPD. We are using these guidelines to inform the development of an integrated care pathway and commissioning plan. -Report of the National Chronic Obstructive Pulmonary Disease Audit 2008, UK Primary Care Organisations: Resources and Organisation of Care, Appendix D (RCP, 2008B). Recommendations made in this report were as follows: We recommend that further careful prospective audits of outcomes as well as economic benefit are planned and undertaken as Community COPD services develop or emerge. Schemes involving case management and admission avoidance in particular should be subject to high quality evaluation including peer review scrutiny. There is a significant gap in the provision of palliative care for patients with COPD, despite evidence of effective interventions that are not being currently commissioned. We recommend PCO service improvement plans contain a mandatory consideration of end of life care services for patients with COPD. There should be a mechanism to facilitate the sharing of information and evidenced data between PCOs who are developing their COPD services in order to avoid repetition and maximise the benefit of these changes to patients and local health economies. We endorse the good practice reported by those PCOs that are developing multidisciplinary services which cross sector and service boundaries. Robert Sheikh Iddenden, Public Health Sciences. 105

106 -Report of the National Chronic Obstructive Pulmonary Disease Audit 2008, UK Primary Care Organisations: Resources and Organisation of Care, page 6 (RCP, 2008B). Report of the National Chronic Obstructive Pulmonary Disease Audit 2008: clinical audit of COPD exacerbations admitted to acute units across the UK (RCP, 2008C); all participating UK NHS acute units completed an anonymised web-based clinical audit of up to 60 patients admitted to their hospital with an exacerbation during the audit period (3 rd March to 16 th May 2008). Cases were prospectively identified, with records audited 90 days after index admission. Key findings and recommendations from this report were as follows: Key Findings of Clinical Case Audit Outcomes 1. Overall mortality rates at 90 days have shown a reduction from 15.5% (2003) to 13.9% (2008). 2. Cause of death was recorded as COPD in 65% of those who died, a reduction from the 71% observed in Median length of stay has fallen by 1 day (from 6 to 5 days) compared to There has been an increase from 26% to 34% in the proportion of patients having a shorter stay of at most 3 days since There has been a 2% increase (to 33%) in the readmission rate since The median time to readmission was 38 days. 6. There has been an increase from 47% to 53% in patients discharged under the care of a respiratory physician % of patients are discharged with oxygen, prescribed via concentrator in 80% of cases. Patient characteristics 1. There has been an increase in the proportion of admissions that are female so that COPD is now a disease of equal importance in both men and women. 2. The mean age of admissions has increased by one year to 73 years for men and 72 years for women % of patients still live at home, 36% on their own % of patients received some form of personal care at home, whether paid or unpaid. 5. The median % predicted forced expiratory volume in 1 second (FEV1) for those patients with spirometry recorded in the last 5 years was 38%. Robert Sheikh Iddenden, Public Health Sciences. 106

107 6. 67% of recorded Medical Research Council (MRC) dyspnoea scores are Grade 4-5 in the steady state prior to admission. 7. There are fewer current smokers (33%) than in 2003 (41%). 8. There is a very high level of co-morbidity, the association with cardiovascular disease being particularly strong % of the patients had been admitted for COPD within the preceding 24 months. Process of care measures 1. Most are better documented than in Recording of spirometry remains poor (55%) and unchanged since Medical Research Council (MRC) dyspnoea scale (47%) and body mass index (BMI) (29%) were documented infrequently % of patients were seen by a respiratory specialist healthcare worker during their admission, an increase of 8% since Resuscitation status was poorly recorded. 7. The majority of Chest X-Ray (CXR) comments recorded in the notes were made by junior medical staff. Management 1. The great majority of patients received systemic steroids (86%) and antibiotics (81%) within the first 24 hours although only 61% reported a change in colour of sputum. 2. The recording of oxygen prescription on the drug chart within the first 24 hours was very poor (16%). Respiratory failure 1. 20% of patients were acidotic on admission. An additional 6% became acidotic later in their admission. 2. If ph was not at its lowest at admission, the lowest recorded ph was a median of 12 hours into the admission % of patients with ph <7.35 had a repeat blood gas within 24 hours, but only 26% of these were taken within 1 hour of the previous sample and 54% within 2 hours % of patients received high flow oxygen before arterial blood gases (ABGs) were taken compared to 42% in % of patients received ventilatory support at any time during admission. 6. 1% received invasive ventilation. 7. It is reported that 3% of acidotic patients did not receive ventilatory support as appropriate facilities were unavailable. Robert Sheikh Iddenden, Public Health Sciences. 107

108 8. Only 59% of patients had a documented management plan for escalation/withdrawal of treatment in the event of non-invasive ventilation (NIV) failure % of decisions not to escalate to ventilatory support were made by very junior medical staff. Only 15% of decisions were made by respiratory consultants % of patients who received ventilatory support did not do so until at least 3 hours after admission. 11. For 20% of the patients receiving ventilatory support, this was commenced after 24 hours into the admission. Early Discharge Schemes 1. 18% of patients were accepted onto a scheme, an increase of 4% since Patients spent a median of 8 days on a scheme, an increase of 1 day since Of those patients taken on to an Early Discharge Scheme, the proportion discharged within 48 hours has increased from 26% to 35% since Recommendations 1. Greater attention should be given to the recording of information known to predict outcome in COPD exacerbations. 2. In patients without previously documented spirometry, this should be recorded either at discharge or within 90 days of discharge in order to confirm the diagnosis and assist with further management and prognosis. 3. Recording of key process items should be emphasised to admitting teams. 4. All patients with COPD exacerbation should be seen by a respiratory specialist during their admission. 5. The management of acute respiratory failure requires closer adherence to guidelines, particularly within the first 4 hours of admission, and greater input into decision making by respiratory specialists during the crucial first 12 hours. 6. There should be further development of integrated early discharge schemes. -Report of the National Chronic Obstructive Pulmonary Disease Audit 2008: clinical audit of COPD exacerbations admitted to acute units across the UK, pages 8-9 (RCP, 2008C). Robert Sheikh Iddenden, Public Health Sciences. 108

109 Report 4 of the National Chronic Obstructive Pulmonary Disease Audit 2008: Patient Survey (RCP, 2008D); 30 patients from the 60 audited patients from each participating NHS acute unit admitted 3 rd March to 16 th May 2008, as they approached discharge, were asked to complete a short survey and given information, the questionnaire and a replypaid envelope. This was the first time that a patient survey was included in the national COPD audit. Results aggregated across all patients. 12% of patients reported that they did not know what COPD stands for; 21% did not know they had COPD before their recent hospital admission; 74% had been admitted previously with COPD, of whom 82% had been admitted within the past year. 13% of patients reported that no-one had explained the medicines they were taking while 15% said they did not understand what the chest medicines were for and how they would affect the patient. 32% of patients reported they used oxygen at home, of whom 74% reported they had used it to try and avoid coming into hospital. 48% of patients reported they used a nebuliser at home, of whom 75% had used it to try to avoid coming into hospital. 55% of patients reported they received regular check ups for their chest at their GP s surgery; 42% attend a hospital clinic for their chest. 83% of patients reported getting flare-ups a number of times a year; 70% had not been given a written plan for what to do when their chest was getting bad. Recommendations made in this report were as follows: On diagnosis, and upon review, patients should: Receive a personal COPD plan that is modified as the disease progresses. Be given clear instruction as to how the prompt treatment of exacerbation may reduce the need for hospitalisation. Know whom to contact, how and when, in the event of deterioration. Primary and Secondary Healthcare teams should: Provide more patients with access to antibiotics and steroids to help them self-manage exacerbation. Educate and encourage patients with COPD to seek help early on when symptoms of an exacerbation develop. Establish that systems and services are in place to ensure the rapid assessment and treatment of patients presenting with early symptoms of exacerbation. Ensure patients have sufficient domestic support and home aids to help them manage at home. Robert Sheikh Iddenden, Public Health Sciences. 109

110 Commissioners should: Consider commissioning quality parameters for managing this group of vulnerable and frequently exacerbating patients whose care is spread between Primary and Secondary Healthcare teams. -Report 4 of the National Chronic Obstructive Pulmonary Disease Audit 2008: Patient Survey, page 8 (RCP, 2008D) Report 5 of the National Chronic Obstructive Pulmonary Disease Audit 2008: survey of COPD care within UK general practices (RCP, 2008E); the hospital team sent a paper questionnaire (with covering letter and reply paid envelope) to the GP of the first 30 consecutive patients admitted to hospital for an exacerbation of COPD during 3 rd March to 16 th September This was the first time that a GP survey was included in the national COPD audit. Key messages and recommendations from this report were as follows: Key Messages [Readers are reminded that the data are generated from a survey completed by General Practitioners (or a representative within their surgery) of patients recently discharged from hospital following COPD exacerbation. These data must be interpreted in the knowledge that this is a select group of patients, therefore]. Use of healthcare resources COPD patients admitted to hospital are frequent users of primary care in the 12 months prior to their admission. Three quarters (74%) of admitted patients make contact with their general practice in the month before admission and nearly a third (31%) have 3 or more contacts in those 4 weeks. Although patients make a median of 12 contacts with general practice in the 12 months prior to the audited admission, and have a median of 3 exacerbations, 51% have no contact with outof-hours services. Most patients (90%) of those hospitalised with COPD were known to have the diagnosis prior to admission and exacerbate frequently. They appear to have severe COPD judged by reduced forced expiratory volume in 1 second (FEV1) and low oxygen saturation recordings. Robert Sheikh Iddenden, Public Health Sciences. 110

111 58% of patients received three or more antibiotic courses in the year prior to hospital admission, a higher level than the 42% receiving three or more steroid courses. Disease Management Some 31% of patients appear not to have been seen in Practice Airways Clinics within the 12 months prior to hospital admission; the reason for this is unclear, although General Practitioners observe in free-text comment that such patients are often housebound, have significant co-morbidity, may not attend when invited or may be followed by a Hospital Chest Clinic. [Information from the National COPD Audit 2008: patient survey has suggested that 42% of patients are also under the care of a Hospital Chest Clinic]. An appreciable number of patients are using multiple medications, one third also having been prescribed oxygen within the last 12 months. The level of spirometry use is higher in primary care (74%) than reported in the hospital-based national clinical audit (55%). [Many GP s report in free-text comment that patients have difficulty with attendance or their spirometric technique as a reason for nonrecording]. Only 15% of the patient sample had undergone pulmonary rehabilitation in the 12 months prior to hospital admission, the precise reasons being unclear. [Many GP s report in free-text comment that patients have difficulty with attendance due to severity of disease or that there isn t a locally available service]. There is excellent uptake (85%) of influenza vaccination. Communication Recommendations A significant number of General Practitioners appear concerned regarding the quality of discharge information they receive from their local hospitals. Use of Healthcare Resources Contacts with primary care should be seen as opportunities for medical interventions, including optimisation of therapy, that may reduce long term exacerbation rates and prevent admissions. Robert Sheikh Iddenden, Public Health Sciences. 111

112 Unscheduled COPD patient contact with primary care should include an admission prevention strategy in the event of a failure of first line therapy. Patients requiring admission should be encouraged to access out- of-hours services that in turn should be able to access admission prevention services. Disease Management Primary Care Providers and Commissioners should review the local availability and access to Pulmonary Rehabilitation for COPD patients. Healthcare teams should ensure patients receive appropriate clinical and medication review in line with current guidelines, whether this is provided at surgery or at home. Communication Hospitals should review their arrangements for providing accurate and timely discharge information to General Practitioners. There should be improved sharing of patient-specific data (e.g. spirometry) between primary and secondary healthcare teams. Consideration should be given to having a COPD record card for patients that they can take with them to hospital. -Report 5 of the National Chronic Obstructive Pulmonary Disease Audit 2008: survey of COPD care within UK general practices, pages 6-7 (RCP, 2008E) Pulmonary rehabilitation Data on the number of Hull patients (defined as resident in Hull or registered with a Hull GP) referred to pulmonary rehabilitation was available for the period January 2009 to September 2010 inclusive. While around 5% of patients referred to pulmonary rehabilitation are referred for reasons other than COPD, there was no indicator in the dataset to suggest who these patients were, so all were included in these analyses. Analyses were restricted to patients aged 35 years and over. The numbers of patients registered with a Hull GP and referred to pulmonary rehabilitation are shown in Table 36 by GP practice, as well as this number expressed as a percentage of all patients on QOF 2008/09 COPD registers. Overall there were 520 patients referred to pulmonary rehabilitation between January 2009 and September 2010 inclusive, equivalent to 10% of patients on the QOF 2008/09 COPD registers although, as acknowledged above, not all those referred to pulmonary rehabilitation will have COPD. In 20 of the 56 Hull practices (36%) there were fewer than 5 patients recorded as referred Robert Sheikh Iddenden, Public Health Sciences. 112

113 to pulmonary rehabilitation during this period. Of the remaining 36 practices there were 20 (56%) for which the number of patients referred to pulmonary rehabilitation was equivalent to 10% or more of patients on the 2008/09 QOF COPD registers, including 7 practices where the percentage was more than 15% and 3 practices where the percentage was greater than 20%. Table 36: Number of patients aged 35 years and over, registered with a Hull GP and referred to pulmonary rehabilitation between January 2009 and September 2010 inclusive, by GP practice, and expressed as a percentage of QOF 2008/09 COPD registers Number of patients aged 35 years and over, registered with a Hull GP and referred to pulmonary rehabilitation between January 2009 and September 2010, inclusive Practice code Practice name No on COPD register 2008/09 Referred to pulmonary rehab N % B81001 Dr Ali & Partners B81002 Dr A Kumar-Choudhary B81008 Dr J S Parker & Partners B81011 Wheeler Street Healthcare B81017 Kingston Medical Group B81018 Dr R K Awan B81020 Dr P C Mitchell And Partners B81021 Faith House Surgery B81027 St Andrews Group Practice B81032 Dr A W Hussain And Partners 49 <5 B81035 Dr W G T Sande And Partners B81038 Dr A A Mather And Partners B81040 Dr P F Newman And Partners B81046 Dr J D Blow And Partners B81047 Dr J N Singh And Partners B81048 Dr S M Hussain And Partners B81049 Dr V A Rawcliffe And Partners B81052 Dr J Musil And P J Queenan 73 <5 B81053 Dr M Maung And Partners B81054 Dr M J Varma And Partners B81056 The Springhead Medical Centre B81057 Dr S Macphie And Koul B81058 Dr M Foulds & Partner B81066 Dr G M Chowdhury B81072 Dr R Percival And Partners B81074 Dr Ak Rej B81075 Dr M K Mallik 16 <5 B81080 Dr G S Malczewski B81081 Dr K M Tang And Partner Robert Sheikh Iddenden, Public Health Sciences. 113

114 Number of patients aged 35 years and over, registered with a Hull GP and referred to pulmonary rehabilitation between January 2009 and September 2010, inclusive Referred to Practice No on COPD Practice name pulmonary rehab code register 2008/09 N % B81085 Dr J W Richardson And Partners B81089 Dr Witvliet B81094 Dr A K Datta 23 <5 B81095 Dr Cook 79 <5 B81097 Dr R D Yagnik 34 <5 B81104 Dr J K Nayar 25 <5 B81112 Dr P C Ghosh B81119 Dr G Palooran And Partners B81616 Dr G T Hendow 55 <5 B81631 Dr R Raut B81634 Dr J Venugopal 40 <5 B81635 Dr G Dave 72 <5 B81644 Dr K K Mahendra 35 <5 B81645 Dr T Abraham B81646 Dr J Austin B81662 Mizzen Road Surgery 33 <5 B81674 Dr J C Joseph B81675 Dr A H Tak & Dr E G Stryjakiewicz 168 <5 B81682 Dr M Shaikh & Partners B81683 Dr B L Koul 39 <5 B81685 Dr N A Poulose 64 <5 B81688 Dr K V Gopal 64 <5 B81690 Dr S K Ray 22 <5 B81692 The Quays Medical Centre 28 <5 Y00955 Riverside Medical Centre 54 <5 Y01200 The Calvert Practice 26 <5 Y02344 Northpoint All Hull practices 5, Home oxygen / long term oxygen therapy Data on the number of Hull patients using home oxygen was available for the most recent invoice round, dated 22/09/2010. It was assumed that those using home oxygen were on long term oxygen therapy, although this was not explicitly stated. There was no indication of the reason for the home oxygen, so it was not possible to assert that all those on the database had COPD. However, as this is likely to be the most common cause of need for home oxygen, the data are included here. All patients below the age of 35 were removed from the analysis as any COPD requiring home oxygen is very unlikely at this age. Robert Sheikh Iddenden, Public Health Sciences. 114

115 The numbers of patients aged 35 years and over, registered with a Hull GP and in receipt of home oxygen are shown in Table 37 by GP practice, as well as this number expressed as a percentage of all patients on QOF 2008/09 COPD registers. Overall there were 696 patients aged 35 years and over registered with Hull GP practices that were in receipt of home oxygen in 2010, equivalent to 14% of patients on the QOF 2008/09 COPD registers although, as acknowledged above, not all those using home oxygen will have COPD. In 11 of the 56 Hull practices (20%) there were fewer than 5 patients recorded as using home oxygen in Of the remaining 45 practices there were 8 (18%)for which the number of patients aged 35 years and over using home oxygen in 2010 was equivalent to 20% or more of patients on the 2008/09 QOF COPD registers. These practices included two with relatively large QOF COPD registers as well as the practice with the second smallest QOF COPD register. In a further 9 practices (20%) the number of patients receiving home oxygen in 2010 represented less than 10% of patients on the 2008/09 QOF COPD registers. Table 37: Number of patients aged 35 years and over, registered with a Hull GP and in receipt of home oxygen as at 22/09/2010, by GP practice, and expressed as a percentage of QOF 2008/09 COPD registers Number of patients aged 35 years and over, registered with a Hull GP and in receipt of home oxygen as at 22/09/2010 by GP practice, and expressed as a percentage of patients on QOF 2008/09 COPD registers Practice No on COPD Using home oxygen Practice name code register 2008/09 N % B81001 Dr Ali & Partners B81002 Dr A Kumar-Choudhary B81008 Dr J S Parker & Partners B81011 Wheeler Street Healthcare B81017 Kingston Medical Group B81018 Dr R K Awan B81020 Dr P C Mitchell And Partners B81021 Faith House Surgery B81027 St Andrews Group Practice B81032 Dr A W Hussain And Partners B81035 Dr W G T Sande And Partners B81038 Dr A A Mather And Partners B81040 Dr P F Newman And Partners B81046 Dr J D Blow And Partners B81047 Dr J N Singh And Partners B81048 Dr S M Hussain And Partners B81049 Dr V A Rawcliffe And Partners B81052 Dr J Musil And P J Queenan B81053 Dr M Maung And Partners B81054 Dr M J Varma And Partners B81056 The Springhead Medical Centre B81057 Dr S Macphie And Koul Robert Sheikh Iddenden, Public Health Sciences. 115

116 Number of patients aged 35 years and over, registered with a Hull GP and in receipt of home oxygen as at 22/09/2010 by GP practice, and expressed as a percentage of patients on QOF 2008/09 COPD registers Practice No on COPD Using home oxygen Practice name code register 2008/09 N % B81058 Dr M Foulds & Partner B81066 Dr G M Chowdhury B81072 Dr R Percival And Partners B81074 Dr Ak Rej B81075 Dr M K Mallik 16 <5 B81080 Dr G S Malczewski 49 <5 B81081 Dr K M Tang And Partner B81085 Dr J W Richardson And Partners B81089 Dr Witvliet B81094 Dr A K Datta 23 <5 B81095 Dr Cook B81097 Dr R D Yagnik 34 <5 5.9 B81104 Dr J K Nayar 25 <5 B81112 Dr P C Ghosh B81119 Dr G Palooran And Partners B81616 Dr G T Hendow B81631 Dr R Raut 122 <5 B81634 Dr J Venugopal B81635 Dr G Dave B81644 Dr K K Mahendra B81645 Dr T Abraham 41 <5 B81646 Dr J Austin B81662 Mizzen Road Surgery 33 <5 B81674 Dr J C Joseph B81675 Dr A H Tak & Dr E G Stryjakiewicz B81682 Dr M Shaikh & Partners B81683 Dr B L Koul B81685 Dr N A Poulose 64 <5 B81688 Dr K V Gopal B81690 Dr S K Ray B81692 The Quays Medical Centre 28 <5 Y00955 Riverside Medical Centre 54 <5 Y01200 The Calvert Practice 26 <5 Y02344 Northpoint All Hull practices 5, Robert Sheikh Iddenden, Public Health Sciences. 116

117 GP Quality and Outcomes Framework Indicators As part of the Quality and Outcomes Framework (QOF), information was collected on five indicators for COPD during 2008/09. The main clinical indicator is a register of people with COPD from patients on the General Practice lists; three indicators reflect the ongoing management of COPD, while a further indicator records spirometry-confirmed diagnosis where COPD was diagnosed from 01/04/2008 (further details on these indicators are provided on page 276). Excluding the COPD register which is just a count of all patients on the COPD register, the indicators are reported as underlying achievements, derived from the number of patients on the COPD register that have had the indicated diagnostic procedure or on-going management divided by the total number of patients on the COPD register who are deemed eligible to have received the indicated diagnostic procedure or on-going management, expressed as a percentage. Not all patients on the COPD register are eligible to be included for each of the diagnostic procedure or on-going management indicators. This is the case for two indicators: the diagnostic procedure indicator, COPD12, which is defined as the percentage of all patients with COPD diagnosed after 1st April 2008 in whom the diagnosis has been confirmed by post bronchodilator; and the on-going management indicator COPD 11, defined as the percentage of patients with COPD receiving inhaled treatment in whom there is a record that inhaler technique has been checked in the previous 15 months. For the other two on-going management indicators, the number of eligible patients should be the same as the number of patients on the COPD register. For each of the diagnostic procedure or on-going management indicators there are two figures reported, the number of exceptions (along with the exception rate, defined as the number of exceptions divided by all eligible patients); and the number of patients that received the diagnostic procedure or on-going management (along with the underlying achievement rate). There are specified criteria for defining patients as exceptions for the various indicators, and these are outlined in GMS Statement of Financial Entitlements 16 : A) patients who have been recorded as refusing to attend review who have been invited on at least three occasions during the preceding twelve months B) patients for whom it is not appropriate to review the chronic disease parameters due to particular circumstances e.g. terminal illness, extreme frailty C) patients newly diagnosed within the practice or who have recently registered with the practice, who should have measurements made within three months and delivery of clinical standards within nine months e.g. blood pressure or cholesterol measurements within target levels 16 GMS Statement of Financial Entitlements, Annex D Quality and Outcomes Framework Guidance, available from Robert Sheikh Iddenden, Public Health Sciences. 117

118 D) patients who are on maximum tolerated doses of medication whose levels remain sub-optimal E) patients for whom prescribing a medication is not clinically appropriate e.g. those who have an allergy, another contraindication or have experienced an adverse reaction F) where a patient has not tolerated medication G) where a patient does not agree to investigation or treatment (informed dissent), and this has been recorded in their medical records H) where the patient has a supervening condition which makes treatment of their condition inappropriate e.g. cholesterol reduction where the patient has liver disease I) where an investigative service or secondary care service is unavailable. The percentage of exceptions will influence the percentage with the specified indicator, as the more patients that are excluded, the lower the denominator. However, the percentage of exceptions will depend on the number of patients on the register. Practices with elderly populations or in more deprived areas could have a higher percentage of exceptions as people may be more likely to not attend review appointments or be more likely to have comorbidities which could influence on-going COPD care (see Appendix B on page 282). Table 38 gives the 5 th and 95 th percentiles over all General Practices within each PCT for the percentage of exceptions against a given indicator and the percentage of patients with each indicator (after removing the exceptions) for April The range over the practices (minimum and maximum) could have been presented, but it is possible that there could be one practice with an exception rate of 100% and all remaining practices with an exception rate of between 0% and 1%. Therefore, the 5 th and 95 th percentiles have been presented which essentially ignores the most extreme values (5% of practices with highest and lowest values). For example, the 5 th and 95 th percentiles for Hull for being on the COPD register were 0.99% and 3.22%, so 90% of practices within had a register prevalence of between 0.99 and 3.22%, 5% of practices had a register prevalence of 0.99% or less and 5% of practices had a register of more than 3.22%. The minimum register prevalence was 0.35%, so 5% of Hull s practices had a register prevalence of between 0.35% and 0.99%, and the maximum register prevalence was 3.95%, so 5% of Hull s practices had between 3.22% and 3.95% of patients on their COPD registers. It can be seen that the 5 th and 95 th percentiles of exceptions were similar for most indicators for both Hull and East Riding practices, although the 95 th percentiles for exceptions were higher for East Riding practices for COPD10 and COPD11, and higher for Hull practices for COPD12. Indicator 5 th and 95 th percentiles were again similar for both Hull and East Riding practices, although the 5 th percentile was higher for Hull practices for COPD01 and COPD10, while the 95 th percentile was higher for East Riding practices for COPD01 and COPD10. Robert Sheikh Iddenden, Public Health Sciences. 118

119 Table 38: 5 th and 95 th percentiles of QOF COPD indicators for Hull and East Riding General Practices, 2008/09 QOF COPD indicator Percentage of exceptions: range over practices (percentiles) Percentage with indicator: range over practices (percentiles) Hull ERoY Hull ERoY 5 th 95 th 5 th 95 th 5 th 95 th 5 th 95 th COPD01: Register n/a n/a n/a n/a COPD08: flu jab COPD10: FeV COPD11: Inhaler COPD12: Spirometry The overall averages for all Hull practices and all East Riding practices are given in Table 39. For all indicators, Hull practices combined had a higher percentage of patients excepted compared to East Riding practices combined, as well as higher percentages fulfilling the condition for each indicator once exceptions were removed. This higher rate of exceptions for Hull practices is not totally surprising. Due to Hull s increased deprivation, it is more likely that patients will fail to attend review appointments, have more co-morbidities and increased risk factors for poorer health which mean that they are more likely to be excluded. However, Hull practices appear to be better at meeting the specified criteria for the indicators than East Riding practices. Table 39: QOF COPD indicators for Hull and East Riding, 2008/09 QOF COPD indicator All practices combined (%) Exceptions Indicator Hull ERoY Hull ERoY COPD01: Register n/a n/a COPD08: flu jab COPD10: FeV COPD11: Inhaler COPD12: Spirometry Table 40 gives for Hull the six (10%) practice codes with the highest percentage of exceptions and the six practice codes having the lowest percentages for the indicators, with the percentage of exceptions or the indicator shown in brackets. Table 41 gives the equivalent information for the most extreme 10% (four) practices in East Riding of Yorkshire for each indicator. Some of these practices could have a relatively small number of patients on the COPD register and as a result relatively large variability over the diagnostic and on-going care measures. Any investigation of these practices with regard to their on-going care for COPD or use of spirometry to diagnose COPD should be considered in relation to the characteristics of the practices patient population including Robert Sheikh Iddenden, Public Health Sciences. 119

120 Indicator Exceptions Indicator Exceptions the age distribution and deprivation scores (see Appendix B on page 282 for mean age of patients and mean deprivation scores for each practice). The information in the tables may act as a starting point for further investigation and also provides a comparison for practices which are felt to be similar to each other. Table 40: Hull practices within highest 10% of practices for exceptions and within lowest 10% of practices for indicators Practices within highest 10% of practices for exceptions and within lowest 10% of practices for indicator (percentage of exceptions/indicator) COPD01: on COPD08: flu COPD11: COPD12: COPD10: FeV1 register jab Inhaler Spirometry n/a B81017 (25.9%) B81017 (22.8%) B81020 (23.3%) B81035 (66.7%) n/a B81085 (26.0%) B81020 (21.4%) B81038 (21.1%) B81112 (100%) n/a B81631 (27.9%) B81038 (21.9%) B81085 (27.5%) B81634 (66.7%) n/a B81645 (24.4%) B81049 (30.0%) B81662 (23.1%) B81635 (66.7%) n/a B81692 (28.6%) B81662 (21.2%) B81683 (15.2%) B81683 (75.0%) n/a Y00955 (33.3%) B81692 (25.0%) B81692 (13.6%) B81692 (75.0%) B81008 (1.2%) B81001 (87.8%) B81027 (72.8%) B81027 (86.2%) B81040 (33.3%) B81020 (1.1%) B81027 (84.9%) B81047 (70.4%) B81056 (82.8%) B81057 (0.0%) B81057 (1.0%) B81047 (85.5%) B81057 (73.5%) B81057 (80.6%) B81094 (50.0%) B81075 (0.7%) B81057 (81.5%) B81097 (75.0%) B81104 (82.4%) B81112 (0.0%) B81094 (1.1%) B81080 (82.2%) B81104 (48.0%) B81675 (79.2%) B81634 (0.0%) B81104 (0.3%) B81688 (86.0%) B81675 (75.3%) B81688 (86.5%) B81644 (0.0%) Table 41: ERoY practices within highest 10% of practices for exceptions and within lowest 10% of practices for indicators Practices within highest 10% of practices for exceptions and within lowest 10% of practices for indicator (percentage of exceptions/indicator) COPD01: on register COPD08: flu jab COPD10: FeV1 COPD11: Inhaler COPD12: Spirometry n/a B81025 (23.8%) B81088 (31.0%) B81010 (17.2%) B81014 (60.0%) n/a B81084 (22.4%) B81100 (23.1%) B81082 (25.6%) B81088 (60.0%) n/a B81121 (28.6%) B81602 (25.6%) B81084 (17.6%) B81622 (75.0%) n/a B81658 (27.3%) B81619 (22.9%) B81602 (27.0%) B81653 (66.7%) B81042 (0.8%) B81062 (88.2%) B81042 (72.2%) B81036 (88.6%) B81004 (85.7%) B81082 (0.8%) B81120 (84.6%) B81121 (71.4%) B81084 (85.7%) B81069 (86.2%) B81121 (0.5%) B81658 (87.5%) B81666 (70.0%) B81120 (87.5%) B81120 (0.0%) B81658 (0.7%) B81666 (85.7%) B81679 (66.7%) B81679 (75.0%) B81121 (0.0%) It is possible to examine practice performance over some of these indicators in more detail. In the following sections the absolute number and percentage of people classified Robert Sheikh Iddenden, Public Health Sciences. 120

121 as exceptions and the people with the indicator for each practice in Hull and East Riding of Yorkshire are presented. As well as the published exception rates and underlying achievement (labelled in tables as % of remaining) the final column gives the percentage of patients with the indicator out of the denominator for that indicator (i.e. with the excepted patients included in the denominator). If a practice has a particularly high percentage of exceptions and a high percentage on the indicator when the exceptions are excluded (fourth column of indicator), then including the exceptions in the denominator will substantially reduce the percentage achieving the indicator. Within the tables, values in blue refer to those practices with a percentage that is within the highest 20% of practices within Hull or East Riding of Yorkshire, and values in red refer to those practices where the percentage is in the lowest 20% of practices. The results can be examined in more detail if the percentage of exceptions is particularly high or the indictor is particularly low. However, it is very important to recognise that these percentages do not take the characteristics of the patient population into consideration. Some practices will have older or more deprived patient populations than other practices, and it is likely that one would expect a higher percentage of exceptions in practices with such populations as patients may be less likely to attend review sessions or be more likely to have co-morbidities or life circumstances that make monitoring and treatment more complex and difficult to achieve. Therefore, practices should be examined on a practiceby-practice basis using local knowledge of the practice populations in terms of age and deprivation, and taking into account other circumstances that could affect the percentage of exceptions and the percentage with the indicator measured. For reference the mean age and mean deprivation scores of the practice populations are given in Appendix B on page Influenza immunisation Table 42 provides the numbers and percentages of people on COPD registers that had received the influenza immunisation during 1 st September 2008 and 31 st March 2009, as well as the numbers and percentages that were deemed exceptions for this indicator for Hull general practices. Among Hull s 56 practices 44 (79%) had a recorded influenza immunisation rate among patients on the COPD register of greater than 90%. However, when exceptions were included in the denominator, only 5 practices (9%) had an influenza immunisation rate among patients on the COPD register of more than 90%, with one practice, Dr Mallik, immunising all the people on their register, although the register for this practice only contained 16 patients. 4 practices (7%) managed to immunise less than two thirds of patients on their COPD register, while a further 10 (18%) immunised fewer than 75% of patients on their COPD registers. 46 (82%) of Hull s practices had exception rates of more than 10%, including 15 practices (27%) with exception rates above 20% and 5 practices (9%) with exception rates above 25%. Robert Sheikh Iddenden, Public Health Sciences. 121

122 Practice code No. on COPD register Denominator % of remaining % of denominator Table 42: QOF influenza immunisation indicator 2008/09 for Hull Practices COPD08: The percentage of patients with COPD who have had influenza immunisation in the preceding 1 September to 31 March Exceptions Indicator Practice name N % N B81001 Dr Ali & Partners B81002 Dr A Kumar-Choudhary B81008 Dr J S Parker & Partners B81011 Wheeler Street Healthcare B81017 Kingston Medical Group B81018 Dr R K Awan B81020 Dr P C Mitchell And Partners B81021 Faith House Surgery B81027 St Andrews Group Practice B81032 Dr A W Hussain And Partners B81035 Dr W G T Sande And Partners B81038 Dr A A Mather And Partners B81040 Dr P F Newman And Partners B81046 Dr J D Blow And Partners B81047 Dr J N Singh And Partners B81048 Dr S M Hussain And Partners B81049 Dr V A Rawcliffe And Partners B81052 Dr J Musil And P J Queenan B81053 Dr M Maung And Partners B81054 Dr M J Varma And Partners B81056 The Springhead Medical Centre B81057 Dr S Macphie And Koul B81058 Dr M Foulds & Partner B81066 Dr G M Chowdhury B81072 Dr R Percival And Partners B81074 Dr Ak Rej B81075 Dr M K Mallik B81080 Dr G S Malczewski B81081 Dr K M Tang And Partner B81085 Dr J W Richardson And Partners B81089 Dr Witvliet B81094 Dr A K Datta B81095 Dr Cook B81097 Dr R D Yagnik B81104 Dr J K Nayar B81112 Dr P C Ghosh B81119 Dr G Palooran And Partners B81616 Dr G T Hendow Robert Sheikh Iddenden, Public Health Sciences. 122

123 Practice code No. on COPD register Denominator % of remaining % of denominator COPD08: The percentage of patients with COPD who have had influenza immunisation in the preceding 1 September to 31 March Exceptions Indicator Practice name N % N B81631 Dr R Raut B81634 Dr J Venugopal B81635 Dr G Dave B81644 Dr K K Mahendra B81645 Dr T Abraham B81646 Dr J Austin B81662 Mizzen Road Surgery B81674 Dr J C Joseph B81675 Drs A H Tak & E G Stryjakiewicz B81682 Dr M Shaikh & Partners B81683 Dr B L Koul B81685 Dr N A Poulose B81688 Dr K V Gopal B81690 Dr S K Ray B81692 The Quays Medical Centre Y00955 Riverside Medical Centre Y01200 The Calvert Practice Y02344 Northpoint Table 43 provides the numbers and percentages of people on COPD registers that had received the influenza immunisation during 1 st September 2008 and 31 st March 2009, as well as the numbers and percentages that were deemed exceptions for this indicator for East Riding general practices. Among East Riding s 38 practices 29 (76%) had a recorded influenza immunisation rate among patients on the COPD register of 90% or greater. However, when exceptions were included in the denominator, only 2 practices (5%) had an influenza immunisation rate among patients on the COPD register of more than 90%. 2 practices (5%) managed to immunise less than two thirds of patients on their COPD register, while a further 4 (11%) immunised fewer than 75% of patients on their COPD registers. 23 (61%) of East Riding s practices had exception rates of more than 10%, including 6 practices (16%) with exception rates above 20% and 2 practices (5%) with exception rates above 25%. Robert Sheikh Iddenden, Public Health Sciences. 123

124 Practice code No. on COPD register Denominator % of remaining % of denominator Table 43: QOF influenza immunisation indicator 2008/09 for East Riding Practices COPD08: The percentage of patients with COPD who have had influenza immunisation in the preceding 1 September to 31 March Exceptions Indicator Practice name N % N B81004 Dr P I Collingwood And Partners B81006 Dr Farley And Partners B81009 Dr R W Webster And Partners B81010 Hessle Grange Medical Practice B81013 Dr M S Patel And Partners B81014 Dr H K Macnab And Partners B81024 Dr M E A Moody And Partners B81025 Dr R D Fouracre And Partners B81029 Dr J E Clark And Partners B81034 Dr A J Sykes And Partners B81036 Dr J P Brooke And Partners B81037 Dr S J Towers And Partners B81041 Dr R A Ferguson And Partners B81042 Dr P G Jones And Partners B81044 Holme-Bubwith Medical Group B81050 Dr P English And Partners B81051 Dr P R Mixer And Partners B81060 Dr P M Pickering And Partners B81061 Dr W A Hart And Partners B81062 Dr C J Lambert And Partners B81068 Dr L Wrightson And Partners B81069 Dr P A Harris And Partners B81070 Dr H R P Meldrum And Partners B81082 Dr R L Pearson & Partners B81084 Dr M Morgan And Partners B81088 Dr R W Harrison B81092 Dr A N Crawford And Partners B81100 Dr J G Best And Partner B81101 Dr H S Suri And Partners B81120 Dr A R M Kelly And Dr B L Bawn B81121 Dr G L Clayton B81602 Dr M E Hancocks B81619 Park View Surgery B81622 Dr A M Harley And Partners B81653 Dr J Robson B81658 Dr Y Adhami B81666 Dr R G Mitchell B81679 Dr R G Clarke Robert Sheikh Iddenden, Public Health Sciences. 124

125 Area No. on COPD register Denominator % of remaining % of denominator National data, along with whole PCT comparisons are given in Table 44. The rate of exceptions across Hull general practices (16%) was higher than for England, while the rate across East Riding general practices (13%) was similar to England. The reported underlying achievement for influenza immunisations among patients on practice COPD registers was higher for Hull (93.7%) than for East Riding (92.8%) or England (90.5%). However, if exceptions are included in the denominator, the percentage of patients on COPD registers that received influenza immunisations was lower in Hull (78.7%) than East Riding (80.8%) or England (79.2%). Table 44: QOF influenza immunisation indicator 2008/09 for Hull and East Riding PCTs and England COPD08: The percentage of patients with COPD who have had influenza immunisation in the preceding 1 September to 31 March Exceptions Indicator N % N Hull 5,319 5, , ERoY 5,528 5, , England 834, , , , Recorded FeV 1 Table 45 provides the numbers and percentages of people on COPD registers that had their Forced Expiratory Volume (FEV 1 ) measured and recorded in the preceding 15 months, as well as the numbers and percentages that were deemed exceptions for this indicator for Hull general practices. Among Hull s 56 practices 46 (82%) had measured and recorded FEV 1 for at least 80% of patients on the COPD register (reported as underlying achievement, that is after removing exceptions) although only 19 (34%) had measured and recorded FEV 1 in at least 90% of their COPD registered patients. When exceptions were included in the denominator, however, this reduced to 20 practices (36%) that had measured and recorded FEV 1 in the past 15 months for at least 80% of patients on their COPD registers, and only 4 (7%) had done so for at least 90% of their patients. 5 practices (9%) had measured and recorded FEV1 during the past 15 months in less than two thirds of patients on their COPD register, while 1 practice (Dr Nayar) had done so in less than half of registered COPD patients, although this practice, with the youngest mean age of patients in Hull, had one of the smallest COPD registers and recorded none of their COPD patients as exceptions. 5 Robert Sheikh Iddenden, Public Health Sciences. 125

126 Practice code No. on COPD register Denominator % of remaining % of denominator other practices in Hull recorded no exceptions (each of these practices having a small COPD register of less than 60 patients) while at the other end of the scale two practices reported more than 25% exceptions for this indicator, the Quays Medical Centre and Dr Rawcliffe and partners, the former specialising in mental health and drug use, with a deprived (2 nd highest mean IMD2007 score) and relatively young patient population and hence a small COPD register (N=28); the latter with a large COPD register (N=230, the fourth largest COPD register in Hull). Table 45: QOF recorded FeV 1 indicator 2008/09 for Hull General Practices COPD10: The percentage of patients with COPD with a record of FeV 1 in the previous 15 months Exceptions Indicator Practice name N % N B81001 Dr Ali & Partners B81002 Dr A Kumar-Choudhary B81008 Dr J S Parker & Partners B81011 Wheeler Street Healthcare B81017 Kingston Medical Group B81018 Dr R K Awan B81020 Dr P C Mitchell And Partners B81021 Faith House Surgery B81027 St Andrews Group Practice B81032 Dr A W Hussain And Partners B81035 Dr W G T Sande And Partners B81038 Dr A A Mather And Partners B81040 Dr P F Newman And Partners B81046 Dr J D Blow And Partners B81047 Dr J N Singh And Partners B81048 Dr S M Hussain And Partners B81049 Dr V A Rawcliffe And Partners B81052 Dr J Musil And P J Queenan B81053 Dr M Maung And Partners B81054 Dr M J Varma And Partners B81056 The Springhead Medical Centre B81057 Dr S Macphie And Koul B81058 Dr M Foulds & Partner B81066 Dr G M Chowdhury B81072 Dr R Percival And Partners B81074 Dr Ak Rej B81075 Dr M K Mallik B81080 Dr G S Malczewski B81081 Dr K M Tang And Partner Robert Sheikh Iddenden, Public Health Sciences. 126

127 Practice code No. on COPD register Denominator % of remaining % of denominator COPD10: The percentage of patients with COPD with a record of FeV 1 in the previous 15 months Exceptions Indicator Practice name N % N B81085 Dr J W Richardson And Partners B81089 Dr Witvliet B81094 Dr A K Datta B81095 Dr Cook B81097 Dr R D Yagnik B81104 Dr J K Nayar B81112 Dr P C Ghosh B81119 Dr G Palooran And Partners B81616 Dr G T Hendow B81631 Dr R Raut B81634 Dr J Venugopal B81635 Dr G Dave B81644 Dr K K Mahendra B81645 Dr T Abraham B81646 Dr J Austin B81662 Mizzen Road Surgery B81674 Dr J C Joseph B81675 Drs A H Tak & E G Stryjakiewicz B81682 Dr M Shaikh & Partners B81683 Dr B L Koul B81685 Dr N A Poulose B81688 Dr K V Gopal B81690 Dr S K Ray B81692 The Quays Medical Centre Y00955 Riverside Medical Centre Y01200 The Calvert Practice Y02344 Northpoint Table 46 provides the numbers and percentages of people on COPD registers that had their Forced Expiratory Volume (FEV 1 ) measured and recorded in the preceding 15 months, as well as the numbers and percentages that were deemed exceptions for this indicator for East Riding general practices. Among East Riding s 38 practices 28 (74%) had measured and recorded FEV 1 for at least 80% of patients on the COPD register (reported as underlying achievement, that is after removing exceptions) although only 8 (21%) had measured and recorded FEV 1 in at least 90% of their COPD registered patients, including 2 practices with 100% underlying achievement. When exceptions were included in the denominator, however, this reduced Robert Sheikh Iddenden, Public Health Sciences. 127

128 Practice code No. on COPD register Denominator % of remaining % of denominator to 13 practices (34%) that had measured and recorded FEV 1 in the past 15 months for at least 80% of patients on their COPD registers, and only 2 (5%) had done so for at least 90% of their patients. 6 practices (16%) had measured and recorded FEV 1 during the past 15 months in two thirds or fewer of patients on their COPD register. 4 practices (11%) in East Riding recorded no exceptions (each of these practices having a small COPD register of less than 30 patients) while at the other end of the scale two practices reported more than 25% exceptions for this indicator, Dr Hancocks and Dr Harrisons practices, the former with a small COPD register (N=43). Table 46: QOF recorded FeV 1 indicator 2008/09 for East Riding General Practices COPD10: The percentage of patients with COPD with a record of FeV 1 in the previous 15 months Exceptions Indicator Practice name N % N B81004 Dr P I Collingwood And Partners B81006 Dr Farley And Partners B81009 Dr R W Webster And Partners B81010 Hessle Grange Medical Practice B81013 Dr M S Patel And Partners B81014 Dr H K Macnab And Partners B81024 Dr M E A Moody And Partners B81025 Dr R D Fouracre And Partners B81029 Dr J E Clark And Partners B81034 Dr A J Sykes And Partners B81036 Dr J P Brooke And Partners B81037 Dr S J Towers And Partners B81041 Dr R A Ferguson And Partners B81042 Dr P G Jones And Partners B81044 Holme-Bubwith Medical Group B81050 Dr P English And Partners B81051 Dr P R Mixer And Partners B81060 Dr P M Pickering And Partners B81061 Dr W A Hart And Partners B81062 Dr C J Lambert And Partners B81068 Dr L Wrightson And Partners B81069 Dr P A Harris And Partners B81070 Dr H R P Meldrum And Partners B81082 Dr R L Pearson & Partners B81084 Dr M Morgan And Partners B81088 Dr R W Harrison B81092 Dr A N Crawford And Partners B81100 Dr J G Best And Partner B81101 Dr H S Suri And Partners Robert Sheikh Iddenden, Public Health Sciences. 128

129 Area No. on COPD register Denominator % of remaining % of register Practice code No. on COPD register Denominator % of remaining % of denominator COPD10: The percentage of patients with COPD with a record of FeV 1 in the previous 15 months Exceptions Indicator Practice name N % N B81120 Dr A R M Kelly And Dr B L Bawn B81121 Dr G L Clayton B81602 Dr M E Hancocks B81619 Park View Surgery B81622 Dr A M Harley And Partners B81653 Dr J Robson B81658 Dr Y Adhami B81666 Dr R G Mitchell B81679 Dr R G Clarke National data, along with whole PCT comparisons are given in Table 47. The rate of exceptions across Hull general practices (11.9%) was higher than for England while the rate for East Riding practices (8.4%) was similar to, although slightly lower than, that for England. The reported underlying achievement for FEV 1 recorded in the past 15 months for patients on practice COPD registers was slightly higher for East Riding (86.3%) than for Hull (85.8%), while both were higher than for England. When exceptions were included in the denominators, the percentages with FEV 1 recorded were still highest for East Riding practices (79.1%) but Hull practices (75.6%) were similar to, although slightly higher than, England. Table 47: QOF recorded FeV 1 indicator 2008/09 for Hull and East Riding PCTs and England COPD10: The percentage of patients with COPD with a record of FeV 1 in the previous 15 months Exceptions Indicator N % N Hull 5,319 5, , ERoY 5,528 5, , England 834, ,877 77, , Robert Sheikh Iddenden, Public Health Sciences. 129

130 Practice code No. on COPD register Denominator % of remaining % of those on inhaled treatment Inhaler technique Table 48 provides the numbers and percentages of people on COPD registers receiving inhaled treatment for whom there was a record that inhaler technique had been checked in the previous 15 months, as well as the numbers and percentages that were deemed exceptions for this indicator for Hull general practices. Around 85% of patients on COPD registers in Hull practices were receiving inhaled treatment. Among Hull s 56 practices 48 (86%) had an underlying achievement of checking inhaler technique in 90% or more of their COPD patients that were using inhaled treatments, including 22 practices (39%) achieving more than 95% on this indicator. However, once excepted patients were included in the denominators, the percentage of practices where 90% or more of COPD patients using inhaled treatments had their inhaler technique checked more than halved to 19 (34%). 3 Hull practices (5%) checked inhaler technique among less than 75% of COPD patients receiving inhaled treatments. Across all Hull practices the median percentage of patients using inhaled treatments where inhaler technique was checked in the previous 15 months was 86%. The median percentage of exceptions for this indicator among Hull practices was 6%, while 14 practices (25%) recorded 10% or more exceptions, including 4 practices excepting more than 20% of patients using inhaled treatments. 8 Hull practices recorded no exceptions, although each had a relatively small COPD register (fewer than 60 patients each). Table 48: QOF recorded inhaler technique indicator 08/09 for Hull Practices Practice name COPD11: The percentage of patients with COPD receiving inhaled treatment in whom there is a record that inhaler technique has been checked in the previous 15 months Exceptions Indicator N % N B81001 Dr Ali & Partners B81002 Dr A Kumar-Choudhary B81008 Dr J S Parker & Partners B81011 Wheeler Street Healthcare B81017 Kingston Medical Group B81018 Dr R K Awan B81020 Dr P C Mitchell And Partners B81021 Faith House Surgery B81027 St Andrews Group Practice B81032 Dr A W Hussain And Partners B81035 Dr W G T Sande And Partners B81038 Dr A A Mather And Partners B81040 Dr P F Newman And Partners Robert Sheikh Iddenden, Public Health Sciences. 130

131 Practice code No. on COPD register Denominator % of remaining % of those on inhaled treatment Practice name COPD11: The percentage of patients with COPD receiving inhaled treatment in whom there is a record that inhaler technique has been checked in the previous 15 months Exceptions Indicator N % N B81046 Dr J D Blow And Partners B81047 Dr J N Singh And Partners B81048 Dr S M Hussain And Partners B81049 Dr V A Rawcliffe And Partners B81052 Dr J Musil And P J Queenan B81053 Dr M Maung And Partners B81054 Dr M J Varma And Partners B81056 The Springhead Medical Centre B81057 Dr S Macphie And Koul B81058 Dr M Foulds & Partner B81066 Dr G M Chowdhury B81072 Dr R Percival And Partners B81074 Dr Ak Rej B81075 Dr M K Mallik B81080 Dr G S Malczewski B81081 Dr K M Tang And Partner B81085 Dr J W Richardson And Partners B81089 Dr Witvliet B81094 Dr A K Datta B81095 Dr Cook B81097 Dr R D Yagnik B81104 Dr J K Nayar B81112 Dr P C Ghosh B81119 Dr G Palooran And Partners B81616 Dr G T Hendow B81631 Dr R Raut B81634 Dr J Venugopal B81635 Dr G Dave B81644 Dr K K Mahendra B81645 Dr T Abraham B81646 Dr J Austin B81662 Mizzen Road Surgery B81674 Dr J C Joseph B81675 Drs A H Tak Dr E G Stryjakiewicz B81682 Dr M Shaikh & Partners B81683 Dr B L Koul B81685 Dr N A Poulose B81688 Dr K V Gopal Robert Sheikh Iddenden, Public Health Sciences. 131

132 Practice code No. on COPD register Denominator % of remaining % of those on inhaled treatment Practice name COPD11: The percentage of patients with COPD receiving inhaled treatment in whom there is a record that inhaler technique has been checked in the previous 15 months Exceptions Indicator N % N B81690 Dr S K Ray B81692 The Quays Medical Centre Y00955 Riverside Medical Centre Y01200 The Calvert Practice Y02344 Northpoint Table 49 provides the numbers and percentages of people on COPD registers receiving inhaled treatment for whom there was a record that inhaler technique had been checked in the previous 15 months, as well as the numbers and percentages that were deemed exceptions for this indicator for East Riding general practices. Around 88% of patients on COPD registers in East Riding practices were receiving inhaled treatment. Among East Riding s 38 practices 33 (87%) had an underlying achievement of checking inhaler technique in 90% or more of their COPD patients that were using inhaled treatments, including 17 practices (45%) achieving more than 95% on this indicator. However, once excepted patients were included in the denominators, the percentage of practices where 90% or more of COPD patients using inhaled treatments had their inhaler technique checked more than halved to 13 (34%). 4 East Riding practices (11%) checked inhaler technique among less than 75% of COPD patients receiving inhaled treatments. Across all East Riding practices the median percentage of patients using inhaled treatment where inhaler technique was checked in the previous 15 months was 88%. The median percentage of exceptions for this indicator among East Riding practices was 6%, while 11 practices (29%) recorded 10% or more exceptions, including 2 practices excepting more than 25% of patients using inhaled treatments. 3 East Riding practices recorded no exceptions, although each had a small COPD register (fewer than 20 patients each). Robert Sheikh Iddenden, Public Health Sciences. 132

133 Practice code No. on COPD register Denominator % of remaining % of those on inhaled treatment Table 49: QOF recorded inhaler technique indicator 2008/09 for East Riding Practices Practice name COPD11: The percentage of patients with COPD receiving inhaled treatment in whom there is a record that inhaler technique has been checked in the previous 15 months Exceptions Indicator N % N B81004 Dr P I Collingwood And Partners B81006 Dr Farley And Partners B81009 Dr R W Webster And Partners B81010 Hessle Grange Medical Practice B81013 Dr M S Patel And Partners B81014 Dr H K Macnab And Partners B81024 Dr M E A Moody And Partners B81025 Dr R D Fouracre And Partners B81029 Dr J E Clark And Partners B81034 Dr A J Sykes And Partners B81036 Dr J P Brooke And Partners B81037 Dr S J Towers And Partners B81041 Dr R A Ferguson And Partners B81042 Dr P G Jones And Partners B81044 Holme-Bubwith Medical Group B81050 Dr P English And Partners B81051 Dr P R Mixer And Partners B81060 Dr P M Pickering And Partners B81061 Dr W A Hart And Partners B81062 Dr C J Lambert And Partners B81068 Dr L Wrightson And Partners B81069 Dr P A Harris And Partners B81070 Dr H R P Meldrum And Partners B81082 Dr R L Pearson & Partners B81084 Dr M Morgan And Partners B81088 Dr R W Harrison B81092 Dr A N Crawford And Partners B81100 Dr J G Best And Partner B81101 Dr H S Suri And Partners B81120 Dr A R M Kelly And Dr B L Bawn B81121 Dr G L Clayton B81602 Dr M E Hancocks B81619 Park View Surgery B81622 Dr A M Harley And Partners B81653 Dr J Robson B81658 Dr Y Adhami B81666 Dr R G Mitchell B81679 Dr R G Clarke Robert Sheikh Iddenden, Public Health Sciences. 133

134 Area No. on COPD register Denominator % of remaining % of those on inhaled treatment National data, along with whole PCT comparisons are given in Table 50. The rate of exceptions across Hull general practices (8.1%) was higher than for England while the rate for East Riding practices (6.5%) was similar to that for England. The reported underlying achievement for checking inhaler technique in the last 15 months among patients using inhaled treatments was higher for East Riding (94.1%) than for Hull (92.7%), while both were higher than for England. When exceptions were included in the denominators, the percentage for East Riding practices (87.9%) was still higher than Hull and England, but Hull practices (85.2%) were similar to England. Table 50: QOF recorded inhaler technique indicator 2008/09 for Hull and East Riding PCTs and England COPD11: The percentage of patients with COPD receiving inhaled treatment in whom there is a record that inhaler technique has been checked in the previous 15 months Exceptions Indicator N % N Hull 5,319 4, , ERoY 5,528 4, , England 834, ,821 46, , Spirometry-confirmed diagnosis Table 51 provides the numbers and percentages of patients on COPD registers diagnosed since 01/04/2008 in whom the diagnosis has been confirmed by post bronchodilator spirometry, as well as the numbers and percentages that were deemed exceptions for this indicator for Hull general practices. Around 13% of patients on COPD registers in Hull were diagnosed since 01/04/2008. Among Hull s 56 practices 28 (50%) had confirmed the COPD diagnosis by post bronchodilator spirometry in 100% of patients diagnosed with COPD since 01/04/2008, once exceptions were excluded. 20 practices (36%) had confirmed the diagnosis by post bronchodilator spirometry in fewer than 90% of patients diagnosed since 01/04/2008, again once excluding exceptions. This included two practices (Dr Ghosh s and Dr Venugopal s practice) with 0% and 1 (Dr Datta s practice) with 50%, (although each of these three practices had diagnosed 3 or fewer patients since 01/04/2008) and one practice at 33% (Dr Newman and partners, with 31 patients diagnosed since 01/04/2008). Robert Sheikh Iddenden, Public Health Sciences. 134

135 Practice code No. on COPD register Denominator % of remaining % of diagnosed since 01/04/08 A further 2 practices (Drs MacPhie & Koul and Dr Mahendra) had diagnosed no new cases of COPD since 01/04/2008. Once exceptions were included in the denominators, the number of practices that had confirmed diagnosis of COPD by post bronchodilator spirometry in 100% of patients diagnosed since 01/04/2008 decreased to 4 (7%). 47 Hull practices (84%) had confirmed the COPD diagnosis by post bronchodilator spirometry in fewer than 90% of patients diagnosed since 01/04/2008, including 40 practices (71%) with 75% or lower, and 28 practices (50%) had done so in two thirds or less of patients diagnosed since 01/04/ Hull practices (30%) confirmed the COPD diagnosis by post bronchodilator spirometry in half or less of patients diagnosed since 01/04/2008, including 10 practices where the COPD diagnosis had been confirmed by post bronchodilator spirometry in one third or less of patients diagnosed since 01/04/2008. The median percentage of exceptions for this indicator among Hull practices was 27%, with 14 practices (26%) recording 50% or more exceptions. It should be noted, however, that because the denominators are relatively small for all practices, one extra exception has a disproportionate impact on the rate of exceptions, when compared to an indicator which has the full COPD register as the denominator. Table 51: QOF spirometry-confirmed diagnosis indicator 2008/09 for Hull Practices Practice name COPD12: The percentage of all patients with COPD diagnosed after 1st April 2008 in whom the diagnosis has been confirmed by post bronchodilator spirometry Exceptions Indicator N % N B81001 Dr Ali & Partners B81002 Dr A Kumar-Choudhary B81008 Dr J S Parker & Partners B81011 Wheeler Street Healthcare B81017 Kingston Medical Group B81018 Dr R K Awan B81020 Dr P C Mitchell And Partners B81021 Faith House Surgery B81027 St Andrews Group Practice B81032 Dr A W Hussain And Partners B81035 Dr W G T Sande And Partners B81038 Dr A A Mather And Partners B81040 Dr P F Newman And Partners B81046 Dr J D Blow And Partners B81047 Dr J N Singh And Partners B81048 Dr S M Hussain And Partners B81049 Dr V A Rawcliffe And Partners B81052 Dr J Musil And P J Queenan Robert Sheikh Iddenden, Public Health Sciences. 135

136 Practice code No. on COPD register Denominator % of remaining % of diagnosed since 01/04/08 Practice name COPD12: The percentage of all patients with COPD diagnosed after 1st April 2008 in whom the diagnosis has been confirmed by post bronchodilator spirometry Exceptions Indicator N % N B81053 Dr M Maung And Partners B81054 Dr M J Varma And Partners B81056 The Springhead Medical Centre B81057 Dr S Macphie And Koul B81058 Dr M Foulds & Partner B81066 Dr G M Chowdhury B81072 Dr R Percival And Partners B81074 Dr Ak Rej B81075 Dr M K Mallik B81080 Dr G S Malczewski B81081 Dr K M Tang And Partner B81085 Dr J W Richardson And Partners B81089 Dr Witvliet B81094 Dr A K Datta B81095 Dr Cook B81097 Dr R D Yagnik B81104 Dr J K Nayar B81112 Dr P C Ghosh B81119 Dr G Palooran And Partners B81616 Dr G T Hendow B81631 Dr R Raut B81634 Dr J Venugopal B81635 Dr G Dave B81644 Dr K K Mahendra B81645 Dr T Abraham B81646 Dr J Austin B81662 Mizzen Road Surgery B81674 Dr J C Joseph B81675 Drs A H Tak & E G Stryjakiewicz B81682 Dr M Shaikh & Partners B81683 Dr B L Koul B81685 Dr N A Poulose B81688 Dr K V Gopal B81690 Dr S K Ray B81692 The Quays Medical Centre Y00955 Riverside Medical Centre Y01200 The Calvert Practice Y02344 Northpoint Robert Sheikh Iddenden, Public Health Sciences. 136

137 Practice code No. on COPD register Denominator % of remaining % of diagnosed since 01/04/08 Table 52 provides the numbers and percentages of patients on COPD registers diagnosed since 01/04/2008 in whom the diagnosis has been confirmed by post bronchodilator spirometry, as well as the numbers and percentages that were deemed exceptions for this indicator for East Riding general practices. Around 10% of patients on COPD registers in East Riding were diagnosed since 01/04/2008. Among East Riding s 38 practices 25 (66%) had confirmed the COPD diagnosis by post bronchodilator spirometry in 100% of patients diagnosed with COPD since 01/04/2008, once exceptions were excluded. Each practice in East Riding had confirmed the COPD diagnosis by post bronchodilator spirometry in at least 85% of patients diagnosed since 01/04/2008 (excluding excepted patients), with the exception of 2 practices (5%) where none had been confirmed by post bronchodilator spirometry (although both of these practices had diagnosed 3 or fewer patients since 01/04/2008). Once exceptions were included in the denominators, the number of practices that had confirmed diagnosis of COPD by post bronchodilator spirometry in 100% of patients diagnosed since 01/04/2008 decreased to 6 (16%). 28 East Riding practices (74%) had confirmed the COPD diagnosis by post bronchodilator spirometry in fewer than 90% of all patients diagnosed since 01/04/2008, including 20 practices with 75% or fewer, and 15 practices with two thirds or less of their patients diagnosed since 01/04/2008 having their diagnosis confirmed by post bronchodilator spirometry. 9 East Riding practices (24%) confirmed the COPD diagnosis by post bronchodilator spirometry in half or less of patients diagnosed since 01/04/2008, including 4 practices where the COPD diagnosis had been confirmed by post bronchodilator spirometry in one third or less of patients diagnosed since 01/04/2008. The median percentage of exceptions for this indicator among East Riding practices was 16%, with 7 practices (18%) reporting 50% or more exceptions. Table 52: QOF spirometry-confirmed diagnosis indicator 2008/09 for East Riding Practices Practice name COPD12: The percentage of all patients with COPD diagnosed after 1st April 2008 in whom the diagnosis has been confirmed by post bronchodilator spirometry Exceptions Indicator N % N B81004 Dr P I Collingwood And Partners B81006 Dr Farley And Partners B81009 Dr R W Webster And Partners B81010 Hessle Grange Medical Practice B81013 Dr M S Patel And Partners B81014 Dr H K Macnab And Partners Robert Sheikh Iddenden, Public Health Sciences. 137

138 Practice code No. on COPD register Denominator % of remaining % of diagnosed since 01/04/08 Practice name COPD12: The percentage of all patients with COPD diagnosed after 1st April 2008 in whom the diagnosis has been confirmed by post bronchodilator spirometry Exceptions Indicator N % N B81024 Dr M E A Moody And Partners B81025 Dr R D Fouracre And Partners B81029 Dr J E Clark And Partners B81034 Dr A J Sykes And Partners B81036 Dr J P Brooke And Partners B81037 Dr S J Towers And Partners B81041 Dr R A Ferguson And Partners B81042 Dr P G Jones And Partners B81044 Holme-Bubwith Medical Group B81050 Dr P English And Partners B81051 Dr P R Mixer And Partners B81060 Dr P M Pickering And Partners B81061 Dr W A Hart And Partners B81062 Dr C J Lambert And Partners B81068 Dr L Wrightson And Partners B81069 Dr P A Harris And Partners B81070 Dr H R P Meldrum And Partners B81082 Dr R L Pearson & Partners B81084 Dr M Morgan And Partners B81088 Dr R W Harrison B81092 Dr A N Crawford And Partners B81100 Dr J G Best And Partner B81101 Dr H S Suri And Partners B81120 Dr A R M Kelly And Dr B L Bawn B81121 Dr G L Clayton B81602 Dr M E Hancocks B81619 Park View Surgery B81622 Dr A M Harley And Partners B81653 Dr J Robson B81658 Dr Y Adhami B81666 Dr R G Mitchell B81679 Dr R G Clarke National data, along with whole PCT comparisons are given in Table 53. The rate of exceptions across Hull general practices (28.4%) was higher than for England while the rate for East Riding practices (17.9%) was much smaller than for England. The reported underlying achievement for confirming the COPD diagnosis by post bronchodilator Robert Sheikh Iddenden, Public Health Sciences. 138

139 Area No. on COPD register Denominator % of remaining % of diagnosed since 01/04/08 spirometry in patients diagnosed since 01/04/2008 was higher for East Riding (94.1%) than for England, while the percentage for Hull (88.8%) was lower than for England. When exceptions were included in the denominators, the percentage for East Riding practices (77.3%) was still higher than England, while Hull practices (63.5%) were still lower than England. Table 53: QOF spirometry-confirmed diagnosis indicator 2008/09 for Hull and East Riding PCTs and England COPD12: The percentage of all patients with COPD diagnosed after 1st April 2008 in whom the diagnosis has been confirmed by post bronchodilator spirometry Exceptions Indicator N % N Hull 5, ERoY 5, England 834,312 89,599 22, , Robert Sheikh Iddenden, Public Health Sciences. 139

140 9. COPD and Potential Inequalities 9.1. Age and Gender Modelled Prevalence In terms of QOF data, it is not possible to examine differences by age and gender at a local level easily, because COPD prevalence from the General Practices clinical registers only includes total numbers of individuals who are included on the COPD register, with no information on age or gender recorded. Nevertheless, the effect of age on the prevalence of COPD on the GP registers can be examined very broadly by examining the relationship between the mean age of the registered population for each practice alongside its registered prevalence of COPD. Furthermore, the information can be examined in relation to national prevalence from surveys by examining the expected prevalence derived from the ERPHO and Doncaster models. For example, do practices with patients of the highest mean age have the biggest discrepancies between prevalence on the GP registers and modelled prevalence? As the practice-level ERPHO and Doncaster models do not produce gender-specific estimates, these cannot be used to examine differences in prevalence by gender. An earlier model from ERPHO did produce age-gender-specific prevalence, although at local authority level, not at practice level. These are shown in Table 54. Within this table, modelled expected prevalence for 2005 and 2009 are shown, the former as this was the base year for which these modelled estimates were produced, the latter to match the most recent QOF round. Differences in prevalence estimates were found between this model and the later model (also produced by ERPHO) produced at general practice level (described in the section starting on page 66). Differences were greater for Hull, with the latter model producing an expected prevalence of COPD of 5.4% in Hull and 3% in East Riding, around 10% lower than the earlier model. Were the model shown in Table 54 correct, this would mean that across Hull and East Riding overall men were 48.6% more likely to have COPD than women (similar to the national difference from these modelled estimates of 48.3%). In Hull men would be one third (32.7%) more likely to have COPD than women while in East Riding they would be more than two third (69.6%) more likely to have COPD. It is likely that these differences in the modelled female: male ratio of COPD prevalence between Hull and East Riding, and indeed between Hull and England, are due to the high synthetic smoking estimates for Hull, with smoking the key risk factor for smoking and therefore very influential in any model of COPD prevalence. While the synthetic smoking estimates used in this model are not clear, they have tended to over-inflate the true level of smoking in Hull. However, we know from local surveys that female: male ratio of age-adjusted prevalence estimates are close to 1 in Hull (0.96), compared with 0.81 in East Riding and 0.84 in England (see Table 34). If the male: female ratio of synthetic smoking estimates used in the model Robert Sheikh Iddenden, Public Health Sciences. 140

141 were to reflect this, then we would expect to see modelled male and female smoking prevalence to be close in Hull than East Riding or England. However, it is not clear what was the female: male ratio of synthetic smoking estimates used in the model. There is a strong relationship between the modelled expected prevalence of COPD and age. The modelled expected prevalence in Hull is estimated to be 2.5%, 7.9%, 13.7% and 14.3% for those aged 16-44, 45-64, and 75+ years respectively. In East Riding of Yorkshire, the estimated prevalence is less than half that in Hull, at 0.9%, 3.3%, 6.3% and 6.9% respectively. Differences in modelled expected prevalence by age between Hull, East Riding and England are most likely due to differences in the synthetic smoking rates used in the model as outlined above. Table 54: Modelled expected COPD prevalence (%) among those aged 16 years and over, by age and by gender, produced at local authority level by Eastern Region PHO Modelled expected COPD prevalence (%) among those aged 16 years Age and and above, by age and by gender, produced at local authority level by gender Eastern Region PHO Hull ERoY Hull & Hull & England Hull ERoY ERoY ERoY England Males Females Persons Modelled Prevalence Compared with Prevalence on GP Registers Figure 25 illustrates the recorded prevalence from the GP registers , along with the estimated prevalence from the Doncaster model and the ERPHO model, for each practice. The practices are ordered by mean age of their patients. Robert Sheikh Iddenden, Public Health Sciences. 141

142 Figure 25: Observed and expected COPD prevalence (%), GP practices in Hull (QOF ) sorted by mean age of practice registered patients Robert Sheikh Iddenden, Public Health Sciences. 142

143 Figure 26 illustrates the relationship between the mean age of patients on practice lists and the absolute differences between register prevalence and modelled prevalence. There is a tendency of the difference between the registered and modelled prevalence to be greater for practices which have a higher mean patient age for both Hull and East Riding of Yorkshire. That is, as the mean age of the practice population increases there is a higher percentage of patients who have undiagnosed COPD (assuming that the modelled prevalence is a good approximation of the actual underlying prevalence in the population, which given its reliance on synthetic smoking estimates, it may not be). In Hull, for every one-year increase in the mean age of practice patients, there would be a predicted 0.15 percentage points increase in the difference between the prevalence based on the GP registers and the prevalence based on the ERPHO model. For example, if the prevalence on one register was 1.5% and the ERPHO model prevalence model for that practice was 3% ( 1.5 percentage point difference), another practice that had a population that was on average five years older, would have a difference of 2.25 percentage points, e.g. say 2% prevalence on GP registers and 4.25% for modelled prevalence). The relationship was statistically significant (p<0.001). There was little relationship between mean age of patients and difference between register and modelled prevalence for East Riding of Yorkshire practices, with a one-year increase in mean age of patients leading to a predicted increase of percentage points in the difference between the register and modelled prevalence. Robert Sheikh Iddenden, Public Health Sciences. 143

144 Figure 26: Relationship between difference between GP registered and ERPHO modelled expected prevalence of COPD for each practice and mean age of registered patients Robert Sheikh Iddenden, Public Health Sciences. 144

145 Percentage points difference between QOF registered and modelled expected (ERPHO) COPD prevalence 1 0 East Riding GPs Hull GPs Linear (East Riding GPs) Linear (Hull GPs) y = x R 2 = y = x R 2 = Mean age of patients (years) A similar analysis of the linear relationship between mean age of practice patients and the difference between register prevalence and prevalence predicated by the Doncaster model is shown in Figure 27. Again the relationship between mean age of practice patients and the difference between the Doncaster modelled and register prevalence was small, and not statistically significant for East Riding, while for Hull the relationship was statistically significant, each one-year increase in mean age of practice patients predicted a 0.12 percentage point increase in the difference between register and modelled COPD prevalence. Figure 27: Relationship between difference between GP registered and Doncaster modelled prevalence of COPD for each practice and mean age of registered patients Robert Sheikh Iddenden, Public Health Sciences. 145

146 Percentage points difference between QOF registered and modelled (Doncaster) COPD prevalence East Riding GPs Hull GPs Linear (East Riding GPs) Linear (Hull GPs) -2 y = x R 2 = y = x R 2 = Mean age of patients (years) Robert Sheikh Iddenden, Public Health Sciences. 146

147 Hospital Activity Table 55 gives the number of day case or inpatient admissions with a primary diagnosis of COPD (ICD10 J40-J44) over the three financial year periods 2007/08 to 2009/10 for Hull and East Riding PCTs and the two PCTs combined, by age and gender. Table 55: Number of admissions with a primary diagnosis of COPD in Hull and East Riding of Yorkshire (2007/08 to 2009/10 pooled) by age and gender Age (years) Number of admissions with a primary diagnosis of COPD in Hull and East riding of Yorkshire (2007/08 to 2008/10 pooled) by age and gender Hull East Riding Hull & East Riding Men Women Total Men Women Total Men Women Total < Total 1,584 1,644 3,228 1,186 1,280 2,466 2,770 2,924 5,694 Table 56 gives the rate of admissions with a primary diagnosis of COPD per 100,000 residents (from GP registration files October 2007, 2008 and 2009), while Figure 28 presents this information graphically. Not surprisingly as age increases, the admission rate increases. In general, men tend to have a higher admission rate relative to women, particularly for the older age groups as this pattern is apparent in every age group for those aged 65+ years for Hull and 60+ years for East Riding of Yorkshire. However, in some of the younger age groups, rates in women are higher than in men, particularly in those aged years in ERoY and those aged years in Hull. Interestingly the rate of COPD admissions decreases for women age 85+ years, for both Hull and ERoY, while continuing to increase for men. It is possible that this is a cohort effect, due to fewer women in this oldest age group having ever smoked, and therefore having a lower prevalence of COPD with a consequent lower rate of admissions. This hypothesis could be tested by looking at trends in admissions by age, were sufficient years of admissions data available. Robert Sheikh Iddenden, Public Health Sciences. 147

148 COPD admission rate per 100,000 resident population Table 56: Age-specific rate of admissions per 100,000 with a primary diagnosis of COPD in Hull and East Riding of Yorkshire (2007/08 to 2009/10 pooled) by gender Age-specific rate of admissions per 100,000 with a primary diagnosis of COPD in Age Hull and East Riding of Yorkshire (2007/08 to 2009/10 pooled) by gender (years) Hull East Riding Hull & East Riding Men Women Total Men Women Total Men Women Total < ,302 1, ,672 1,737 1, , ,602 1,843 2, ,413 1,080 1, ,560 1,979 2,226 1,226 1,011 1,109 1,687 1,375 1, ,929 2,724 2,803 1,603 1,318 1,437 2,063 1,847 1, ,982 1,438 2,237 1,799 1,000 1,248 2,565 1,152 1,592 Total Figure 28: Age-specific rate of admissions with a primary diagnosis of COPD per 100,000 residents of Hull and East Riding (2007/08 to 2009/10 pooled), by gender 4,500 Hull men Hull women ERoY men ERoY women 4,000 3,500 3,000 2,500 2,000 1,500 1, < Age (years) Table 57 gives the number of emergency admissions with a primary diagnosis of COPD (ICD10 J40-J44) over the three financial year periods 2006/07 to 2008/09 for Hull and 2005/06 to 2007/08 for East Riding. The overwhelming majority of admissions where COPD was the primary diagnosis were emergency admissions, with around 98% of admissions for Hull residents and 97% of admissions for East Riding residents, although there was some variation by age. Robert Sheikh Iddenden, Public Health Sciences. 148

149 Table 57: Number of emergency admissions with a primary diagnosis of COPD in Hull and East Riding of Yorkshire (2007/08 to 2009/10 pooled) by age and gender Number of emergency admissions with a primary diagnosis of COPD in Hull and East Riding (2007/08 to 2009/10) by age and gender Age Hull East Riding Hull & East Riding (years) (2006/07 to 2008/09) (2005/06 to 2007/08) (2005/06 to 2007/08) Men Women Total Men Women Total Men Women Total < Total 1,540 1,619 3,159 1,135 1,246 2,381 2,675 2,865 5,540 Table 58 gives the rate of emergency admissions with a primary diagnosis of COPD per 100,000 resident population (from GP registration file October 2009), while Figure 29 presents these data graphically. The patterns of emergency admissions were the same as for all admissions, with emergency admission rates increasing with age (except for women aged 85 years and over), higher in men than women, and rates for each gender and age higher in Hull than in East Riding. Table 58: Age-specific rate of emergency admissions per 100,000 with a primary diagnosis of COPD in Hull and East Riding of Yorkshire (2007/08 to 2009/10 pooled) by gender Age-specific rate of emergency admissions per 100,000 with a primary diagnosis of COPD in Hull and East Riding of Yorkshire (2007/08 to 2009/10) by gender Age Hull East Riding Hull & East Riding (years) (2006/07 to 2008/09) (2005/06 to 2007/08) (2005/06 to 2007/08) Men Women Total Men Women Total Men Women Total < ,297 1, ,651 1,724 1, , ,532 1,808 2, ,357 1,053 1, ,457 1,956 2,169 1, ,087 1,630 1,358 1, ,866 2,674 2,748 1,578 1,294 1,412 2,024 1,813 1, ,959 1,428 2,222 1, ,205 2,492 1,137 1,559 Total Robert Sheikh Iddenden, Public Health Sciences. 149

150 COPD emergency admission rate per 100,000 resident population Figure 29: Rate of emergency admissions with a primary diagnosis of COPD per 100,000 resident population for Hull and East Riding of Yorkshire (2007/08 to 2009/10 pooled) by age and gender 4,500 4,000 3,500 3,000 2,500 2,000 1,500 1, Hull men Hull women ERoY men ERoY women 0 < Age (years) Table 59 gives the percentage difference between Hull and East Riding of Yorkshire in the rates per 100,000 residents of admissions with a primary diagnosis of COPD during 2006/07 and 2007/08. For each age and gender admission rates were higher in Hull residents than for East Riding residents. Men in Hull in each ageband (excluding <45 years) had admission rates more than double those in East Riding men of equivalent ages, whilst among women those resident in Hull had admission rates more than double those living in East Riding at each age excluding years and those aged 80 years and over. The greatest differences were seen in those aged 45-49, although the numbers of admissions were relatively low in this age group. Differences between Hull and East Riding appeared to reduce with age, although this was not a smooth trend. Across all ages men and women resident in Hull were 81% and 82% respectively more likely to have had a hospital admission for COPD during this period than men and women resident in East Riding. The differences were slightly larger for emergency admission, with men and women in Hull having emergency admissions for COPD during this period 82% and 86% respectively higher than men and women resident in East Riding. Robert Sheikh Iddenden, Public Health Sciences. 150

151 Table 59: Percentage difference in admission rates where COPD is the primary diagnosis between Hull and East Riding of Yorkshire of Yorkshire, 2007/08 to 2009/10 pooled Percentage difference in admission rates where COPD is the primary Age diagnosis between Hull and East Riding (2007/08 to 2009/10 pooled) (years) All admissions Emergency admissions Men Women All Men Women All < Total Mortality From the Public Health Mortality file, between 2006 and 2008 there were 466 deaths registered in Hull residents and 457 in East Riding residents, where COPD was the primary cause of death as recorded on the death certificate. We were unable to look at secondary causes of death using the Public Health Mortality file as only primary cause of death is recorded in this dataset. Table 60 shows the number and age-specific mortality rates for deaths where the primary cause of death was recorded as COPD. Rates of COPD deaths under the age of 60 years were rare, but rates across Hull and East Riding combined increased rapidly after this age, doubling with each 5-year increase in age between and 70-74, thence increasing by two thirds in those aged and 80-84, increasing by more than one quarter in those aged 85+ to reach 616 COPD deaths per 100,000 residents. Rates in Hull were higher than those in East Riding for each age and gender, with the largest differences at age years (more than three times as high in Hull men and almost two and a half times as high in Hull women), disregarding women aged as there were few deaths at that age. COPD mortality rates were higher in men than women at each age in Hull and East Riding with the exception of those aged and in Hull. Between 2006 and % of all men in Hull aged 85 years and over died from COPD. Robert Sheikh Iddenden, Public Health Sciences. 151

152 Table 60: Crude age-specific COPD mortality rates, where the primary cause of death was COPD, deaths registered Age Total number of COPD deaths Average annual COPD mortality rate per 100,000 residents (years) Men Women Persons Men Women Persons Hull < East Riding < Hull & East Riding < From the Primary Care Mortality Database, deaths involving a secondary diagnosis of COPD in patients registered with Hull GPs can also be examined. Table 61 gives the number and average annual mortality rate per 100,000 patients registered with Hull GPs, although in fewer age bands than the previous table due to smaller numbers of deaths as only deaths for 2005 and 2006 were available at the time of writing this report. Of all deaths involving a primary or secondary diagnosis of COPD 38% in men and 42% in women had COPD recorded as the primary cause of death. As age increased, the proportion of deaths where the primary cause was COPD increased in men and women, with the exception of women aged 85 years and older. In 2005 and % of men aged years and 2.6% of men aged 85 years and over died with COPD as the primary or secondary cause of death, as did 1% of women aged years and 1.8% of women aged 85 years and over. Robert Sheikh Iddenden, Public Health Sciences. 152

153 Table 61: Number of deaths and crude COPD mortality rate involving a primary (underlying) or secondary cause of death of COPD by age and gender for patients registered with Hull GPs, Type of diagnosis Primary Secondary All with COPD mentioned Age (yrs) Total number of COPD deaths during (Hull patients) Average annual COPD mortality rate per 100,000 patients registered with Hull GPs Men Women Persons Men Women Persons < , All < ,506 1,196 1,287 All < ,500 1,006 1, ,571 1,817 2,037 All Table 62 and Table 63 give the crude and age-adjusted mortality rates per 100,000 residents for Hull and East Riding over the period for females and males respectively. Crude and age-adjusted mortality rates in East Riding were relatively stable in men, less so in women due to smaller numbers of deaths each year, while in Hull they were less stable in both men and women. Adjusted rates in East Riding were lower than crude rates for each year and gender (between 3% and 14% lower), while in Hull adjusted rates were higher than crude rates, again for each year and gender (almost 20% higher in women and 27% to 34% higher in men). This suggests that men and women in Hull tend to die from COPD at a younger age than do their counterparts in East Riding, and that men in Hull die at a younger age than women in Hull. Robert Sheikh Iddenden, Public Health Sciences. 153

154 Table 62: Crude and adjusted mortality rates per 100,000 females where the primary cause of death was COPD, deaths registered Crude and adjusted mortality rates per 100,000 females for deaths Year death with primary cause of death COPD registered Hull ERoY Overall Crude Adjusted Crude Adjusted Crude Adjusted Total Table 63: Crude and adjusted mortality rates per 100,000 males where the primary cause of death was COPD, deaths registered Crude and adjusted (Adj.) mortality rates per 100,000 males for deaths Year death with primary cause of death COPD registered Hull ERoY Overall Crude Adjusted Crude Adjusted Crude Adjusted Total Risk Factors Risk factors for developing COPD are given on page 28. The key risk factor, responsible for at least 80% of all COPD is smoking, whether current smoking or past smoking. Information on current and past smoking habits is available from recent surveys conducted in Hull in (Prevalence and Social Capital surveys conducted in 2009) and East Riding (Health and Lifestyle survey conducted in 2009). As well as this some trend data is also available locally from previous health and lifestyle surveys. These data are examined in the following sections Current smoking patterns The overall age-adjusted prevalence of smoking among adults is given in section on page 98. The percentages that defined themselves as current smokers in recent surveys, or as former smokers, are presented by age and gender in Table 64 for Hull 2009, East Riding 2009 and England 2008, and in Figure 30 for Hull and East Riding. Robert Sheikh Iddenden, Public Health Sciences. 154

155 Among men the percentages smoking decreased with age from years to 75+ years, with smoking in those aged years slightly lower than in those aged years. Smoking prevalence was higher at each age among Hull men than in East Riding or England. Differences between Hull and England were greatest in men aged 75+ years, amongst whom twice as many Hull men reported smoking. Fewer East Riding men of each age smoked than men in England, with the greatest difference again in men aged 75+ years, amongst whom 41% fewer in East Riding smoked. When current and former smokers were combined differences between Hull, East Riding and England were smaller. The largest difference between Hull and England was in men aged years, with the percentage that had ever smoked twice as high in Hull. In those aged 65 years and over more men in England had ever smoked than in Hull. 56% more men aged in Hull had ever smoked than in East Riding, while more men in aged 75+ years had ever smoked in East Riding than in Hull. Percentages of men that had ever smoked were similar for East Riding and England. Between half and two thirds of men of each age in Hull had ever smoked (highest at 68% in men aged years), whereas in East Riding (and in England) only in men aged 55 years and over had more than half ever smoked, percentages increasing with age in both cases, but higher in East Riding, reaching 72% in men aged 75 years and over. Patterns among women were different. The percentages that reported they smoked were generally higher among the youngest women, then decreasing as age increased. Percentages of women in Hull that smoked were at least twice as high as in East Riding, highest amongst women aged years. Percentages of women that smoked were around two thirds higher in Hull than England for most age groups (ranging from 54% higher in those aged years to 94% higher in women aged years). Percentages in East Riding were lower than for England at each age, ranging from 11% lower in women aged years to 39% lower in women aged years). The percentages of women in East Riding and England that had ever smoked were very similar, for each age. Fewer than half of women of each age in East Riding or England reported they had ever smoked, with percentages generally increasing with age, although lower in the oldest age group. In Hull more than half of women in each age group had ever smoked with little difference by age, excluding those aged 75+ years where 47% had ever smoked. Differences between Hull and East Riding, and Hull and England, where greatest in those aged years, in which age group around 60% more women in Hull had ever smoked than in either East Riding or England. While the percentages of women aged years that had ever smoked were higher than for men of this age, for each other age the percentages of women that had ever smoked were lower than for men, with the one exception of Hull residents aged years. Differences between men and women were greatest in older age groups, reflecting the fact that women took up smoking in large numbers in the 1940s, 1950s and 1960s, so older women are less likely to have ever smoked than younger women. The largest difference between men and women was seen in East Riding residents aged years, amongst whom 31% more women had ever smoked than had men. Robert Sheikh Iddenden, Public Health Sciences. 155

156 Table 64: Percentage of men and women who smoke or who used to smoke from Health Survey for England 2008 and local surveys in Hull and East Riding in 2009 Smoking status in men and women from surveys (%) Gender Age Hull East Riding of (yrs) Yorkshire 2009 England 2008 Current Former Current Former Current Former Males Females Data from two Hull surveys were combined. In the smaller survey the youngest age group was years. For the purpose of this analysis it has been assumed that the rate in this group is a good proxy for smoking in those aged years, and the data from the two surveys have been combined without adjustment. Robert Sheikh Iddenden, Public Health Sciences. 156

157 Percentage of respondents Figure 30: Percentage of men and women that reported they were current smokers or exsmokers in the adult health and lifestyle surveys in Hull 2007 and East Riding Ex-smoker Current smoker Males Females Males Females Hull 2009 East Riding 2009 As part of Hull s World Class Commissioning (WCC) process, COPD has been chosen as a priority area, and one of the initiatives is to improve diagnosis of COPD among those most at risk of the disease. To this end a programme of spirometry testing is offered to patients aged 35 years and above who smoke or who used to smoke, refined in the second WCC round to include a symptom such as persistent cough or shortness of breath. If we were to take the original criteria, and apply this to the 2009 populations of Hull and East Riding, this suggests that 59% of men in Hull aged 35 years and over are potentially at risk of developing COPD, as are 53% of women. In East Riding the equivalent percentages are lower at 56% of men and 45% of women. Based purely on current or former smoking, men in Hull are 11% more at risk of developing COPD than women in Hull, while men in East Riding are 24% more at risk than women in East Riding Changes in smoking patterns over time Changes in the prevalence of smoking derived from local surveys are presented by age and gender in Table 65. Smoking prevalence decreased in each age group and for each gender between and 2009 in Hull and between and 2009 in East Riding, with the exception of women in Hull aged years and years who saw Robert Sheikh Iddenden, Public Health Sciences. 157

158 increases in smoking prevalence of 7.5% and 1.4% respectively. The largest decreases in prevalence were seen in the oldest age groups. In Hull smoking prevalence among men aged 65 years and over, and women aged 75 years and over, decreased by more than 40%. In East Riding smoking prevalence decreased by 73% in men aged 75 years and over, and decreased by more than 40% in women aged 16-24, 45-54, and 75+ years. Decreases in smoking prevalence were greater in East Riding than Hull for each age and gender, with the exception of men aged years where the decrease in Hull was one third more than in East Riding. In Hull relative decreases in prevalence were greater among men than women for each age group, whereas in East Riding decreases were greater amongst men in those aged under 65 years of age. Because of the larger decreases in smoking prevalence in Hull among women than men, in only three age groups in 2009 was smoking prevalence in Hull greater in men than in women, despite each age having higher prevalence among men in Similarly for East Riding, whereas smoking prevalence was greater among men than women in for each age except those aged years, only in those aged years was smoking prevalence in East Riding greater in men than women in The reductions seen in smoking prevalence seen in both Hull and East Riding, whether through fewer people starting to smoke or through more people stopping smoking earlier, should reduce the long-term future burden of COPD, although the impact in the near future will be lower. Table 65: Changes in the prevalence of smoking in men and women from local surveys Changes in smoking prevalence in men and women from local surveys (%) Hull 18 East Riding of Yorkshire 19 Gender Males Age (yrs) Weighted average 2009 Relative change (%) Relative change (%) 2003/ Data from two 2009 Hull surveys were combined. In the smaller survey the youngest age group was years. For the purpose of this analysis it has been assumed that the rate in this group is a good proxy for smoking in those aged years, and the data from the two surveys have been combined without adjustment. A similar assumption has been made in combining data from the 2003 and 2004 surveys, as the 2003 survey also used 18 years as the lower age cut-off. 19 The surveys were conducted separately by old PCTs. Results have been weighted so that they better represent the combined PCT population (More than 3 times as many East Yorkshire (EY) PCT residents took part than did Yorkshire Wolds and Coast (YWC) PCT residents, although EY and YWC constituted 53.9% and 46.1% of the East Riding population in 2004). Robert Sheikh Iddenden, Public Health Sciences. 158

159 Gender Females Age (yrs) Changes in smoking prevalence in men and women from local surveys (%) Hull 18 East Riding of Yorkshire 19 Weighted average 2009 Relative change (%) Relative change (%) 2003/ Other risk factors As previously mentioned on page 28 apart from smoking there are a few less common risk factors, including passive smoking, pollution (more important before the clean air act reduced the incidence of smog in UK cities), fumes and dust, as well as genetic susceptibility to COPD. However, information on these risk factors is not available at a local level and so they are not examined here Access to Services and Quality of Care Pulmonary rehabilitation The age-specific numbers of patients aged 35 years and over, registered with Hull GP practices and referred to pulmonary rehabilitation between January 2009 and September 2010 inclusive are shown by gender and age in Table 66, together with the rate of pulmonary rehabilitation referrals per 1,000 registered patients. Pulmonary rehabilitation referrals are rare among younger patients, but increase rapidly with age, peaking among patients aged years at 9.3 per 1,000, almost 1% of all patients in this age group. Rates in those aged under 65 years were slightly higher among women than men, while rates in men exceeded those in women for older age groups (50% higher in men aged 75 years and over than in women of this age). Robert Sheikh Iddenden, Public Health Sciences. 159

160 Table 66: Age-specific numbers of patients aged 35 years and over registered with Hull GP practices referred to pulmonary rehabilitation between January 2009 and September 2010, and rate of pulmonary rehabilitation referrals per 1,000 patients (using October 2009 population) Age (years) Age-specific numbers of patients aged 35 years and over registered with Hull GP practices referred to pulmonary rehabilitation between January 2009 and September 2010 inclusive, and rate of pulmonary rehabilitation referrals per 1,000 patients (using October 2009 population) Males Females All n Rate n Rate N Rate Of course, not all those referred to pulmonary rehabilitation attended. In fact, around 60% of those referred did not attend. These DNAs are shown in Table 67, together with the percentage out of all referrals, by gender and age. The overall numbers of referrals differs to those presented in Table 66, as it excludes those referred in August and September 2010 whose attendance or otherwise was not yet known. The largest percentage of DNAs was in the youngest patients referred to pulmonary rehabilitation, with 72% of patients aged not attending for pulmonary rehabilitation. Excluding this age group, the percentage of DNAs increased with age for both men and women. Women were more likely than men to not attend for pulmonary rehabilitation for each age group, with the greatest difference in those aged 55-64, where there were 25% more DNAs amongst women than men. Percentage differences in DNAs between men and women decreased as age increased. The reasons for not attending pulmonary rehabilitation are not known. Table 67: Age-specific numbers (n) of patients aged 35 years and over registered with Hull GP practices referred to pulmonary rehabilitation between January 2009 and July 2010 inclusive that did not attend pulmonary rehabilitation, and the percentage (%) out of all referrals (N) Age (years) Age-specific numbers (n) of patients aged 35 years and over registered with Hull GP practices referred to pulmonary rehabilitation between January 2009 and July 2010 inclusive that did not attend pulmonary rehabilitation, and the percentage (%) out of all referrals (N) Males Females All Referred DNA Referred DNA Referred DNA N n % N n % N n % Robert Sheikh Iddenden, Public Health Sciences. 160

161 Home oxygen / long term oxygen therapy The numbers of patients aged 35 years and over registered with Hull GP practices using home oxygen shown by gender and age in Table 68, together with the rate of home oxygen use per 1,000 registered patients. It is assumed that those using home oxygen are on long term oxygen therapy. Home oxygen use is rare among younger patients, but increases rapidly after the age of 55 years. Among those aged 75 years and over 15 out of every 1,000 persons was using home oxygen. Differences by gender were small for each age group. Table 68: Age-specific numbers of patients aged 35 years and over registered with Hull GP practices in receipt of home oxygen in 2010, and rate of home oxygen use per 1,000 patients (using October 2009 population) Age (years) Age-specific numbers of patients aged 35 years and over registered with Hull GP practices in receipt of home oxygen in 2010, and rate of home oxygen use per 1,000 patients (using October 2009 population) Males Females All n Rate n Rate N Rate Quality of Care- Ongoing Care in Primary Care As mentioned in section on page 54, the QOF information is not available separately for males and females or for different age groups. Therefore, it is difficult to easily examine the on-going measures of care by age. The only possible way to do this is by examining the measures by age at a practice level (e.g. mean age of registered patients). This represents a relatively crude measure of age, but might give an indication of any relationship with age. Table 69 gives the percentage of exceptions and the indicator percentages for Hull practices divided into five groups based on the mean age of their practice patients. Not unexpectedly, as the mean age of patients within the practices increased, the percentage of patients on the COPD register increased (although not consistently over each age quintile). While there were no consistent trends in either exceptions or indicators with the mean age of practice patients, Hull practices with the youngest patient mean ages tended to have the lowest percentage of exceptions and indicator values for most indicators, while Hull practices with the highest mean patient ages had the highest percentage of exceptions and indicator values for most indicators. The main exception to this was the percentage Robert Sheikh Iddenden, Public Health Sciences. 161

162 of patients that had received the influenza immunization in the preceding 6 months. Exceptions for this indicator among Hull practices were highest for practices with the lowest mean patient age. Differences in the percentage of patients meeting the indicator (i.e. not including the exceptions) between practices were small. Table 69: Exceptions and indicator percentages by mean age of practice patients for Hull practices Weighted average of QOF 2008/09 exceptions and Indicator Measure indicators by mean age of patients (quintiles of age) Mean age of patients in years (Hull practices) COPD01: On register % COPD08: flu jab % excl % indict r COPD10: FeV1 % excl % indict r COPD11: Inhaler % excl % indict r COPD12: Spirometry % excl % indict r Table 70 gives the percentage of exceptions and the indicator percentages for East Riding practices divided into five groups based on the mean age of their practice patients. Practices with the highest mean age of patients tended to have the highest percentages of exceptions, although the relationship was not clear, with the oldest quintile not having the highest percentage of exceptions for any indicator, and the youngest two quintiles having more exceptions on two indicators. In terms of indicator percentages these were all lowest in the youngest quintile, with three of them highest in the oldest quintile. One might expect that exceptions generally would be higher among practices with the highest mean age, as older patients are more likely to be less mobile, have poorer health and co-morbidities. However, among Hull and East Riding practices, there was no clear relationship between exceptions and mean age, which might suggest that other factors, such as deprivation, are important. Alternatively, it might be that mean age of patients at a practice level is not a good proxy for age of patients with COPD. Robert Sheikh Iddenden, Public Health Sciences. 162

163 Table 70: Exceptions and indicator percentages by mean age of practice patients for East Riding practices Weighted average of QOF 2008/09 exceptions and Indicator Measure indicators by mean age of patients (quintiles of age) Mean age of patients in years (East Riding practices) COPD01: On register % COPD08: flu jab % excl % indict r COPD10: FeV1 % excl % indict r COPD11: Inhaler % excl % indict r COPD12: Spirometry % excl % indict r Ethnicity The numbers of people from Black and Minority Ethnic (BME) groups in Hull and East Riding is relatively low in relation to some other areas in England. However, the percentage of the population from different BME groups has increased considerably in the last few years, and this could have an influence on the prevalence of COPD in the future, if smoking rates in BME populations are markedly different to White British smoking rates, and as the BME population ages. In Hull, the percentage of White British was estimated to be 96.7% at the time of the 2001 Census, with official (Office for National Statistics) estimates reducing to 92.6% for mid-2005 and 91.1% for mid-2007 (slightly lower than the 93.4% estimated from Hull s Health and Lifestyle survey conducted during 2007 (Sheikh Iddenden, 2008)). In East Riding of Yorkshire, the percentage of White British was estimated to be 98.8% in 2001, reducing to 95.8% in mid-2005 and 95.0% in mid Modelled Prevalence The ERPHO local authority model produced estimates of the prevalence of COPD in those aged 16+ by broad ethnic group (White, Black, Asian) for each year from 2005 to 2009, then 2010, 2015 and The estimated prevalence from this model for 2009 for Hull was estimated 6.1%, 4.7% and 2.7% for White, Black and Asian ethnic groups respectively. In East Riding of Yorkshire, the prevalence from the ERPHO local authority model was estimated to be, in 2009, 3.2%, 1.9% and 1.8% for White, Black and Asian ethnic groups respectively. However, there is no local data available, and it is necessary to rely on the estimated modelled prevalence to get an indication of the local prevalence of COPD by different ethnic groups. Because the local authority level model produced estimates for Hull and East Riding around one third higher than the practice level model, the main use of this model here is to look at the relative modelled prevalence by ethnic groups, although one might also look at the expected numbers of people with COPD by ethnic group based on the model. Robert Sheikh Iddenden, Public Health Sciences. 163

164 The most recent population estimates by BME groups at local authority level were for mid- 2007, and were produced by ONS in 2009 (ONS, 2009). By applying these estimates to the ERPHO local authority modelled COPD estimated prevalence, we can estimate the number of people aged 16+ with COPD in Hull and East Riding by broad ethnic group. Because the age structure of BME populations tend to be different to the White British population, with fewer people in the older age groups, and the prevalence of COPD increases with age, the estimated numbers presented in Table 71 for BME groups will over-estimate the numbers of people with COPD. The mid-2007 BME population estimates show 4%, 2%, 3% and 2% of the Mixed, Asian/Asian British, Black/Black British and Chinese/Other groups respectively in Hull to be aged 60+ (women) or 65+ (men), which compares to 16% of Hull s White population, while For East Riding the percentages of the population aged 60+ (women) or 65+ (men) are 4%, 7%, 6%, 6% and 24% respectively for Mixed, Asian/Asian British, Black/Black British, Chinese/Other and British groups. COPD prevalence in Asian/Asian British residents is modelled to be 55% lower than among White residents of Hull, while for East Riding it is modelled to be 44% lower, which suggests the Asian/Asian British residents in East Riding are older than in Hull. The reverse applies to Black/Black British residents, with modelled prevalence 23% lower than among White residents in Hull, 40% lower in East Riding. Table 71: Estimated number of people with COPD by ethnic group using ONS 2007 midyear estimated populations (aged 16+) by broad ethnic group and ERPHO modelled prevalence rates (aged 16+) of COPD BME group Population (16+) by ethnic group (ONS 2007 mid-year estimates, 000s) Estimated COPD prevalence 2009 (16+) from ERPHO model (%) Estimated number aged 16+ with COPD based on modelled prevalence Hull ERoY Hull ERoY Hull ERoY White ,482 Mixed* Asian/British Asian Black/Black British Chinese/Other* *As the ERPHO model did not provide prevalence estimates for these groups it is assumed that their prevalence rates are similar to those in Whites Hospital Activity Ethnicity is not consistently collected on Hospital Episode Statistics records with around 10% of records for patients with a diagnosis of COPD not having ethnicity recorded on the database; therefore it is not possible to examine hospital day case and inpatient admissions for COPD by ethnic group. Even were ethnicity available for all records, the Robert Sheikh Iddenden, Public Health Sciences. 164

165 numbers of admissions involving COPD would be too small to produce meaningful analyses for BME groups. Due to the lower expected prevalence for members of the Asian/Asian British and Black/Black British groups, it is likely that they will also have lower admission rates Mortality Ethnicity is not recorded on the Public Health Mortality File, and therefore it is not possible to examine mortality by ethnicity. Furthermore, even if ethnicity were recorded, with around 300 deaths from COPD per year in Hull and East Riding of Yorkshire the number of deaths for any of the minority ethnic groups would be too small to be examined further Risk Factors Whilst the prevalence of COPD, as estimated by the ERPHO local authority model, did differ by ethnic group, it was not clear whether the modelled differences were due to the different age structures by ethnic group, as BME groups had a younger age profile, and hence would have a lower prevalence of age-related disease, than the White British group. However, if smoking prevalence is higher among BME groups, this would indicate higher future COPD prevalence. Table 72 presents results from the health and lifestyle surveys conducted among adults in Hull in 2007 (Porter, 2008; Sheikh Iddenden, 2008). This survey consisted of several parts: the main survey which, through quota sampling, was broadly representative of Hull s adult population, and a survey of BME groups, which did not have a methodology that ensured a representative sample of Hull s BME population, so the results need to be interpreted with care. Table 72: Smoking status by BME group in Hull from 2007 adult health and lifestyle surveys, with percentages of respondents by broad age bands. Survey respondents by smoking status Group Age (years) Current Former smoker smoker Ever smoked n % n % n % Main survey (Hull) , , , Non-British White Mixed Indian Bangladeshi/Pakistani Other Asian Caribbean African Chinese Robert Sheikh Iddenden, Public Health Sciences. 165

166 The proportion of respondents smoking varied widely by ethnic group. Percentages were higher among non-british White (mostly young migrant workers from EU accession countries) respondents (39%), respondents with a Mixed ethnicity (50%) and Other Asian (largely Iraqi and Kurds) respondents (57%), compared with 32% from the main survey. However, respondents form these ethnic groups had a much younger age profile than respondents in the main survey, so the percentages of smokers is likely to be higher. Because numbers were small for many of the BME groups in the survey it was not possible to age-adjust the rates. However, looking at the broad agebands 18-34, and 65+ years in those groups with higher percentages smoking than the Hull average, and where there were more than 100 respondents overall, we can see from Table 73 that the higher smoking percentages seen in Other Asians were seen for each age group, while in non-british Whites the high smoking percentages were seen in each age band except 65+ years, where the 0% is based on a denominator of 4. Therefore the increased proportions of survey respondents smoking in these two ethnic groups are not artefacts of a younger age profile. These groups will therefore be at increased risk of developing COPD. Table 73: Smoking status by broad age bands for BME groups from Hull s 2007 adult health and lifestyle survey where overall percentages smoking are higher than the Hull average and sufficient numbers are available from the survey (N>100) Group Survey respondents currently smoking by broad ageband years years 65+ years n % n % n % Main survey (Hull) Non-British White Other Asian Access to Services and Quality of Care Locally, there is no information on access to services and the quality of care for different BME groups Deprivation This section examines COPD in relation to the Index of Multiple Deprivation (IMD) 2007 which is described in section 7.3 on page 39, and illustrated in Figure 7 and Figure 8, and Table 4 and Table Prevalence on GP Registers Whilst the QOF data gives the total number of people on the COPD register for each General Practice, it is possible to indirectly examine this information in relation to Robert Sheikh Iddenden, Public Health Sciences. 166

167 Prevalence of COPD on practice registers 2008/09 deprivation, by calculating a deprivation score for each General Practice based on the deprivation score associated with each patient (based on their postcode of residence). While it is difficult to ascertain how accurate such a measure of practice deprivation might be, the practice deprivation should give an indication of the relative deprivation of the practice population. Figure 31 illustrates the prevalence of diagnosed COPD from 2008/09 practice registers in relation to deprivation score calculated at practice level using the October 2008 practice populations. There is a significant linear association between the percentage of patients on the COPD register and deprivation for both Hull and East Riding of Yorkshire (p<0.001 in each case), with a stronger linear relationship for East Riding. As practice deprivation increases in both Hull and East Riding, the prevalence of diagnosed COPD patients on practice registers tends to increase, a 0.7 percentage point increase in East Riding, and a 0.3 percentage point increase in Hull, for each 10 point increase in mean IMD 2007 score. Figure 31: Prevalence of diagnosed COPD on practice registers 2008/09 in relation to practice deprivation score (IMD 2007) y = x R 2 = y = x R 2 = Practice deprivation score (IMD 2007) Hull East Riding Linear (Hull) Linear (East Riding) Robert Sheikh Iddenden, Public Health Sciences. 167

168 Examining the registered practice prevalence of COPD in relation to local deprivation quintiles 20, Table 74 shows that COPD prevalence is generally higher in practices whose patients live in more deprived areas (although registered COPD prevalence was highest in the middle deprivation quintile). Because of the link with smoking, it is anticipated that prevalence would be higher for practices with patients living in more deprived areas. The fact for Hull and East Riding combined the middle quintile had the highest prevalence rate is due to this quintile including all the most deprived practices in East Riding (which group had the highest prevalence of all). Table 74: Prevalence of diagnosed COPD on practice registers in Hull and East Riding in relation to local deprivation quintiles 21 Local deprivation Prevalence (%) of diagnosed COPD on practice registers quintiles calculated Hull East Riding Hull & ERoY at practice level 22 n % n % n % Most deprived 1, , , , , , , , , , Least deprived , Modelled Prevalence Compared with Prevalence on GP Registers Figure 32 shows the prevalence of COPD from the General Practice registers for COPD in relation to the expected prevalence of COPD derived from the Doncaster and Eastern Region Public Health Observatory models, the latter of which is claimed to represent undiagnosed and diagnosed COPD. Data is plotted for Hull and East Riding of Yorkshire practices in order of increasing deprivation, based on the mean IMD2007 score for each practice in the October 2008 GP registered populations. For practices on the left hand side of the figure, those with less deprived patients on average, the differences between modelled and register prevalence are smaller. However, what at first appears to be an effect of deprivation may be related to the PCT where the practice is based, as when the figure is redrawn, with Hull PCTs and East Riding PCTs each grouped together, as shown in Figure 33, there appears little relationship between mean practice deprivation scores and differences between register 20 It is not possible to produce national quintiles for GP practices as practice populations with mean IMD 2007 scores for are not available nationally. 21 As quintiles were produced by ranking practices by deprivation without regard to population size, the quintiles do not contain 20% of the populations each. 22 Separate quintiles were produced for Hull (56 practices), East Riding (38 practices) and Hull & ERoY (94 practices). Robert Sheikh Iddenden, Public Health Sciences. 168

169 prevalence and modelled prevalence estimates beyond the differences between Hull and East Riding overall. Robert Sheikh Iddenden, Public Health Sciences. 169

170 Figure 32: GP register (2008/09) and modelled (Doncaster and ERPHO models) prevalence of COPD for each practice in order of increasing mean deprivation score (IMD2007, based on October 2008 GP populations) Robert Sheikh Iddenden, Public Health Sciences. 170

171 Figure 33: GP register (2008/09) and modelled (Doncaster and ERPHO models) prevalence of COPD for each practice in order of increasing mean deprivation score IMD2007 within PCT, based on October 2008 GP populations) Robert Sheikh Iddenden, Public Health Sciences. 171

172 Percentage points difference between QOF registered and modelled expected COPD prevalence This is confirmed when examining a plot of the percentage point difference between register COPD prevalence and modelled COPD prevalence from the ERPHO model (Figure 34). Here the linear relationship between the absolute difference between practice disease registers and the ERPHO model and practice IMD 2007 scores is statistically significant (p<0.01). In this case for every 10-point increase in practice-based IMD2007 score the absolute difference between QOF registers and the ERPHO model increases by almost 0.4. Figure 34: Relationship between deprivation and percentage difference between QOF COPD prevalence and ERPHO model prevalence, Hull and East Riding practices overall y = x R 2 = Prcatice level mean IMD 2007 But as we saw earlier, instead of looking at all practices across Hull and East Riding if we look at the linear relationships between practice IMD 2007 deprivation score and the absolute difference in prevalence between practice registers and ERPHO modelled prevalence by PCT, as shown in Figure 35, a different pattern emerges. There is no linear relationship for East Riding practices (p=0.75, linear regression line not shown), and only a very weak relationship for Hull practices which suggests that practices in the more deprived areas in Hull tend to have smaller differences between the register and the model than practices in the less deprived areas, but again it was not statistically Robert Sheikh Iddenden, Public Health Sciences. 172

173 Percentage points difference between QOF registered and modelled expected COPD prevalence significant (p=0.15). There was wide variability in absolute difference with deprivation score, with little of the variation in absolute difference in prevalence between QOF registers and the ERPHO model explained by the practice based deprivation score (less than 5%). Figure 35: Relationship between deprivation and percentage difference between QOF COPD prevalence and ERPHO model prevalence 1 0 East Riding GPs Hull GPs Linear (Hull GPs) y = x R 2 = Prcatice level mean IMD 2007 The difference between the Hull and East Riding combined analysis and the local PCTspecific analysis above reflects the fact that most practices in East Riding are relatively less deprived, while in Hull they are relatively deprived. The direction of the linear regression line in the combined analysis (Figure 34) is probably reflecting differences between Hull and East Riding as a whole (regardless of the level of deprivation within each of these PCTs) rather than the differences by practice deprivation scores. A similar analysis looking at the linear relationship between the Index of Multiple Deprivation 2007 and the percentage difference (absolute difference) between QOF registered COPD prevalence and expected prevalence based on the Doncaster model is presented in Figure 36. The linear relationship between the absolute difference between Robert Sheikh Iddenden, Public Health Sciences. 173

174 Percentage points difference between QOF registered and modelled expected COPD prevalence the practice disease register and the Doncaster model and practice deprivation scores is a relatively weak one. The linear relationship for Hull is statistically significant at the conventional 5% level (p=0.02), while the relationship for East Riding of Yorkshire is not statistically significant (p=0.37, linear regression line not illustrated). When the linear regression line was taken through all practices in Hull and East Riding combined there was no linear relationship (p=0.98) between the absolute difference between QOF registered COPD prevalence and Doncaster model prevalence and practice based deprivation scores (not illustrated) unlike was found previously when using ERPHO modelled COPD estimates. Figure 36: Relationship between deprivation and percentage difference between QOF COPD prevalence and Doncaster model prevalence 3 East Riding GPs Hull GPs Linear (Hull GPs) y = x R 2 = Prcatice level mean IMD 2007 Table 75 gives the prevalence on the GP registers and the modelled prevalence by local deprivation quintile calculated at GP level. There were large differences between the practice COPD register and the ERPHO modelled prevalence (both in absolute and relative terms) in the practices with patients living in the two most deprived areas. In these practices ERPHO modelled prevalence was more than twice as high as register prevalence, with the largest difference in the second most deprived quintile. The absolute Robert Sheikh Iddenden, Public Health Sciences. 174

175 and relative differences were largest in the practices in the second most deprived areas and gradually reduced in size with practices in the least deprived areas having the smallest differences. Table 75: Relationship between practice deprivation quintile, QOF COPD prevalence and ERPHO model prevalence Observed Expected Absolute Local IMD prevalence prevalence** (relative) quintile at Population* (QOF register) (ERPHO model) difference in practice level N % N % prevalence Most deprived 89,184 2, , (-50.3) 2*** 102,554 1, , (-61.4) 3 128,117 2, , (-39.0) 4 121,437 1, , (-44.5) Least deprived 158,402 2, , (-34.5) *Practice list size 2008/09 as extracted from QMAS **Uses estimated GP population as at 31/03/07 for denominator ***Excludes Northpoint Surgery as ERPHO model predates its existence However, as mentioned earlier, these differences between modelled and register prevalence may reflect differences between PCTs more than differences by deprivation of practice patients. Table 76 and Table 77 show the same data for Hull and East Riding GP practices respectively. There is no clear pattern in absolute differences between observed and modelled prevalence by deprivation quintiles of Hull GP practices, while the trend in relative difference is for increasing relative difference as deprivation of practice patients decreases. There is no clear pattern in either absolute of relative differences between observed and modelled prevalence by deprivation quintiles of East Riding of Yorkshire GP practices. Table 76: Relationship between practice deprivation quintile (Hull practices), QOF COPD prevalence and ERPHO model prevalence Observed Expected Absolute Local IMD prevalence prevalence** (relative) quintile at Population* (QOF register) (ERPHO model) difference in practice level N % N % prevalence Most deprived 60,796 1, , (-50.0) 2*** 49,662 1, , (-54.1) 3 66,669 1, , (-62.2) 4 65,128 1, , (-59.8) Least deprived 43, , (-65.2) *Practice list size 2008/09 as extracted from QMAS **Uses estimated GP population as at 31/03/07 for denominator ***Excludes Northpoint Surgery as ERPHO model predates its existence Robert Sheikh Iddenden, Public Health Sciences. 175

176 Table 77: Relationship between practice deprivation quintile (East Riding practices), QOF COPD prevalence and ERPHO model prevalence Observed Expected Absolute Local IMD prevalence prevalence** (relative) quintile at Population* (QOF register) (ERPHO model) difference in practice level N % N % prevalence Most deprived 69,613 1, , (-17.7) 2 59, , (-36.6) 3 70,971 1, , (-26.3) 4 66, , (-45.9) Least deprived 47, (-23.2) *Practice list size 2008/09 as extracted from QMAS **Uses estimated GP population as at 31/03/07 for denominator A similar analysis is presented in Table 78 for the differences between prevalence on the GP registers and the Doncaster modelled prevalence by local deprivation quintile calculated at GP level across Hull and East Riding. Here there was no clear pattern by deprivation quintiles, with the second most deprived quintile having the largest difference, followed by the second least deprived quintile. Table 78: Relationship between practice deprivation quintile, QOF COPD prevalence and Doncaster model prevalence Observed Expected Absolute Local IMD prevalence prevalence (relative) quintile at Population (QOF register) (Doncaster model) difference in practice level N % N % prevalence Most deprived 89,184 2, , ( -7.8) 2 104,525 1, , (-26.5) 3 128,117 2, , ( -5.1) 4 121,437 1, , (-13.9) Least deprived 158,402 2, , ( -3.2) *Practice list size 2008/09 as extracted from QMAS Looking next at Hull GP practices (Table 79) and East Riding of Yorkshire GP practices (Table 80) separately, there were no clear patterns by deprivation quintiles for either PCT. For Hull practices the two least deprived quintiles of practices had larger absolute differences between register and Doncaster modelled prevalence than did the two most deprived quintiles, although the largest difference was for practices in the middle deprivation quintile. For East Riding practices, those in the most deprived quintile, the least deprived quintile and the middle quintile had higher registered prevalence than Doncaster modelled prevalence. Robert Sheikh Iddenden, Public Health Sciences. 176

177 Table 79: Relationship between practice deprivation quintile (Hull practices), QOF COPD prevalence and Doncaster model prevalence Observed Expected Absolute Local IMD prevalence prevalence (relative) quintile at Population (QOF register) (Doncaster model) difference in practice level N % N % prevalence Most deprived 60,796 1, , (-10.9) 2 51,633 1, , (-11.0) 3 66,669 1, , (-27.3) 4 65,128 1, , (-22.0) Least deprived 43, (-31.4) *Practice list size 2008/09 as extracted from QMAS Table 80: Relationship between practice deprivation quintile (East Riding practices), QOF COPD prevalence and Doncaster model prevalence Observed Expected Absolute Local IMD prevalence prevalence (relative) quintile at Population (QOF register) (Doncaster model) difference in practice level N % N % prevalence Most deprived 69,613 1, , ( 6.4) 2 59, , (-12.4) 3 70,971 1, , ( 1.4) 4 66, (-18.3) Least deprived 47, ( 18.8) *Practice list size 2008/09 as extracted from QMAS Hospital Activity The number of hospital admissions and crude admission rates per 100,000, for hospital admissions where COPD was the primary diagnosis are presented in Table 81 by gender and deprivation quintile for Hull and East Riding of Yorkshire for 2007/08 to 2009/10 pooled. The crude admission rate was considerably higher for people living in the most deprived areas compared to the least deprived areas, with rates higher in Hull. Crude admission rates among men in Hull were three times higher in the most deprived local quintile than in the least deprived local quintile, and two and a half times higher among women. Among East Riding residents men in the most deprived national quintile had crude admission rates three and a half times higher than men in the least deprived national quintile, and four times higher among women. When the rates were adjusted for age, the differences between the most and least deprived residents of Hull increased, by 17% in men and by 7% in women. Age-standardisation did not change the rates for East Riding residents as much, with the differences between the most and least deprived quintiles increasing by 7% for men and less than 1% for women. The differences Robert Sheikh Iddenden, Public Health Sciences. 177

178 All Females Males between crude and age-adjusted rates were larger for Hull due to the younger age profile of Hull COPD admissions. Table 81: Number and rate per 100,000 resident population of hospital admissions where COPD was the primary diagnosis in Hull and East Riding of Yorkshire (2007/08 to 2009/10 pooled), by gender and IMD 2007 deprivation quintile (local quintiles for Hull, national quintiles for East Riding and combined total) Gender and IMD 2007 deprivation quintile* Number and rate of admissions per 100,000 residents with COPD the primary diagnosis in Hull and East Riding residents (2007/08 to 2009/10 pooled) by deprivation quintile* and gender Number of Admission rate per 100,000 residents admissions Crude Age-adjusted Hull ERoY Total Hull ERoY Total Hull ERoY Total Most deprived , Least deprived Total 1,584 1,186 2, Most deprived , Least deprived Total 1,644 1,280 2, Most deprived , , Least deprived Total 3,228 2,466 5, *Local quintiles for Hull, national quintiles for ERoY and for combined Total, all from IMD 2007 Results for emergency admissions only are shown in Table 82. These are virtually identical to all admission, as emergency admissions accounted for 98% of all admissions with a primary diagnosis of COPD in Hull, and 96% in East Riding of Yorkshire. Robert Sheikh Iddenden, Public Health Sciences. 178

179 All Females Males Table 82: Number and rate per 100,000 resident population of emergency hospital admissions where COPD was the primary diagnosis in Hull and East Riding of Yorkshire (2007/08 to 2009/10 pooled), by gender and IMD 2007 deprivation quintile (local quintiles for Hull, national quintiles for East Riding and combined total) Gender and IMD 2007 deprivation quintile* Number and rate of emergency admissions per 100,000 residents with COPD the primary diagnosis in Hull and East Riding residents (2007/08 to 2009/10 pooled) by deprivation quintile* and gender Emergency admission rate per 100,000 Number of residents admissions Crude Age-adjusted Hull ERoY Total Hull ERoY Total Hull ERoY Total Most deprived , Least deprived Total 1,540 1,135 2, Most deprived , Least deprived Total 1,619 1,246 2, Most deprived , , Least deprived Total 3,159 2,381 5, *Local quintiles for Hull, national quintiles for ERoY and for combined Total, all from IMD Mortality The crude and age-adjusted mortality rates for deaths with a primary cause of death of COPD are shown in Table 83, by deprivation quintiles 23, gender and PCT. Amongst people living in the most deprived areas crude and age-adjusted rates were more than double rates amongst people living in the least deprived areas. Differences between the most and least deprived areas were larger amongst women in both Hull and East Riding, while differences between the most and least deprived areas were greater in East Riding residents than in Hull residents, for both men and women. Despite this crude rates were higher in Hull than East Riding for each deprivation quintile, excluding the second most deprived quintile. Age-adjusted rates were higher in Hull residents than in East Riding residents for each quintile for both men and women, with the differences in age-adjusted rates between Hull and East Riding greater than the difference in crude rates. Among 23 National deprivation quintiles from IMD 2007 were used for East Riding and the combined Total, while local quintiles of IMD 2007 were used for Hull Robert Sheikh Iddenden, Public Health Sciences. 179

180 Persons Females Males Hull residents age-adjusted rates were on average around 25% higher than crude rates, while among East Riding residents age-adjusted rates were on average around 10% lower than crude rates, suggesting that Hull residents die from COPD younger than East Riding residents. Table 83: Number, crude and age-adjusted mortality rates per 100,000 persons for deaths registered with COPD the primary cause of death by deprivation quintile*, gender and PCT Gender and IMD 2007 deprivation quintile* Total deaths with primary cause COPD, plus crude and age-gender-adjusted mortality rates per 100,000 residents Number of deaths Crude rates Adjusted rates Hull ERoY Total Hull ERoY Total Hull ERoY Total Most deprived Least deprived Most deprived Least deprived Most deprived Least deprived *Local quintiles for Hull, national quintiles for ERoY and for ERoY/Hull combined. A similar analysis looking at all deaths where COPD was mentioned on the death certificate, either primary (underlying) or secondary cause of death, can be done using data from the Primary Care Mortality Database, although this data were available only for Hull practices, for deaths occurring in 2005 and Table 84 gives the number and crude mortality rate per 100,000 patients registered with Hull GPs by local deprivation quintiles for deaths involving a primary or secondary cause of death of COPD. It can be seen that the highest mortality rates were in the most deprived quintile whether COPD was the primary or secondary cause of death. The trend is not linear across the five quintiles, with the middle quintile having the second highest rate where COPD was the primary cause of death, and the mortality rate where COPD was a secondary cause of death higher in the least deprived quintile than the second least deprived quintile. Robert Sheikh Iddenden, Public Health Sciences. 180

181 Table 84: Number and age-gender-adjusted mortality rate per 100,000 patients registered with Hull GPs registered population for deaths with a primary or secondary cause of death of COPD over two year period IMD 2004 local deprivation quintile Total number of COPD deaths over two-year period Average annual COPD crude mortality rate per 100,000 registered population Primary Secondary All Primary Secondary All Most deprived Least deprived Total Prevalence, Hospital Admission Rate and Mortality Table 75 illustrated that the prevalence on the practice registers, from the ERPHO model and the differences between these, were higher in the most deprived quintile compared to the least deprived quintile (albeit a non-consistent pattern across the deprivation quintiles). Table 81 illustrated that the crude and age-adjusted hospital admission rates were also higher in the most deprived quintile area and furthermore that the trend in ageadjusted admission rates was consistent over the deprivation quintiles. Table 83 and Table 84 illustrated that mortality rates were also higher for the most deprived quintile. Given that prevalence is higher, it would be expected that admission rates and mortality rates would also be higher. However, are they even higher than we would expect, given the increased prevalence, suggesting excess admissions or mortality? Table 85 compares these different measures for Hull and East Riding combined, as well as separately for Hull and East Riding PCTs. In each case local deprivation quintiles, derived from IMD 2007, were used. Practice quintiles were used for prevalence, as this is measured at GP practice level, and national quintiles of practices are not available. These practice quintiles were also used for hospital admissions, but not for mortality, as the public health mortality file does not include the GP practice of the deceased. In order to provide comparisons for mortality produced using resident quintiles, estimates of hospital admissions by resident quintiles were also produced. While it is not possible to produce age-gender-adjusted prevalence rates, these have nevertheless been produced for hospital admissions and mortality because, as previously mentioned, prevalence, hospital admissions and mortality are all strongly influenced by the age structure of the population. It should be noted that the agegender-adjusted rates presented here are different to the age-gender-adjusted hospital admission rates in Table 81, and the age-gender-adjusted mortality rates presented in Table 83, as in the earlier analyses national quintiles of IMD 2007 were used, at least for East Riding and for Hull and East Riding combined. Robert Sheikh Iddenden, Public Health Sciences. 181

182 An index has also been included in order that the most deprived quintile can be compared to the other deprivation quintiles. The most deprived quintile has been assigned an index of 100 in each case and a relative index has been calculated for the other quintiles. Looking at the data in Table 85, for Hull and East Riding of Yorkshire combined, registered prevalence was 2.25 in the most deprived quintile of practices, while in the middle deprivation quintile prevalence was 2.28, giving an index value of 101 (2.28 divided by 2.25 and multiplied by 100). Apart from this, registered prevalence decreased as practice-level deprivation decreased. The pattern for modelled prevalence was similar, except that the highest value for each of the models was for the second most deprived practice quintile which had modelled prevalence 5% higher than the most deprived practice quintile in the ERPHO model and 3% higher in the Doncaster model. While we might expect prevalence to be highest in the most deprived quintile due to the relationships between smoking and deprivation and smoking and COPD, smoking prevalence is also high in the second most deprived quintile (certainly in Hull), whilst the most deprived quintile has a lower age profile than the other quintiles, with a mean age of 36 years and a median age of 34 years, compared with mean ages of 39 to 42 years in the other quintiles, and median ages between 38 and 44 years. However, it can be seen that in the second most deprived quintile and middle quintile groups, despite the middle quintile having an index value greater than 100 for registered prevalence and the second most deprived quintile having an index value greater than 100 for modelled prevalence, the indices for hospital admissions and mortality are markedly lower than 100. The most deprived practice-level quintile had an age-gender-adjusted admission rate one third higher than the second most deprived quintile, twice as high as the middle quintile, three times higher than the second least deprived quintile and four times higher than the least deprived quintile. There appear to be excess hospital admissions in the most deprived quintile, when comparing the indices for registered prevalence and hospital admissions. Admissions relative to registered prevalence were highest in the most deprived practice quintile, 10% higher than in the second most deprived quintile and more than twice as high as in the other quintiles. If we accept that registered prevalence is a true reflection of overall prevalence, this might represent a serious inequity, were the increased admissions related to worse management of symptoms and less effective treatment among those in the most deprived quintile. The differences in hospital admission rates between the most deprived resident quintile and the other resident quintiles were even greater than for the practice quintiles, with hospital admissions in the most deprived quintile 75% higher than in the second most deprived quintile, almost three times higher than in the middle quintile, four times higher than the second least deprived quintile and five times higher than in the least deprived quintile. Similarly, the most deprived resident quintile had the highest COPD mortality rate, although the differences between the most deprived and the other quintiles were not as high as for admissions. Comparing admissions and mortality we see that in the most deprived resident quintile there were on average 7.7 admissions for each death, decreasing as deprivation decreased to 4.4 admissions for each death in the least deprived quintile. Again, this might reflect a potentially serious inequity. It suggests that either COPD patients in the most deprived quintile have more co-morbidities than COPD Robert Sheikh Iddenden, Public Health Sciences. 182

183 patients in other quintiles requiring more hospital admissions, or that they have less wellmanaged COPD with more acute exacerbations that end up in a hospital admission, whether through continued smoking or through a different treatment regimen than seen in less deprived COPD patients. If the registered prevalence of COPD, or indeed modelled prevalence as the relationships between admissions and modelled prevalence are similar to the relationship between admissions and registered prevalence are good indicators of the level of COPD in the community, the excess hospital admissions and mortality seen in the most deprived quintile represent a serious inequity. The increased admissions relative to prevalence might indicate that COPD patients in the most deprived quintile are more likely to be diagnosed after hospital admission, rather than in the community through primary care detection programmes. Table 85 also presents the indicators by local quintiles derived at the individual PCT level. The differences between deprivation quintiles in the number of hospital admissions COPD to deaths from COPD are reduced when looking at individual PCTs. Across Hull and East Riding there were 75% more admissions per death in the most deprived resident quintile compared to the least deprived resident quintile. For Hull there were 39% more admissions per deaths in the most deprived quintile while for East Riding there were 49% more admissions per death in the most deprived quintile. Comparing quintiles in Hull with the equivalent quintiles in East Riding, we see that the Hull: East Riding ratio of COPD admissions was greater for each quintile than the Hull: East Riding ratio of deaths, reinforcing the point made above about greater admissions in deprived quintiles relative to deaths, with 8 admissions per death in the most deprived Hull quintile compared with 6 for the most deprived East Riding quintile, decreasing to 6 admissions per death in the least deprived Hull quintile and 4 in the least deprived East Riding quintile. The deprivation gradient seen across Hull and East Riding in terms of admissions per patients on the COPD registers (two and a half times higher in the most deprived quintile relative to the least deprived quintile) was lower within Hull practices (72% higher in the most deprived quintile relative to the least deprived quintile) and non-existent within East Riding practices. The deprivation gradient seen across Hull and East Riding in terms of the number of admissions per death from COPD (75% more admissions per death in the most deprived quintile relative to the least deprived quintile) was smaller within PCT, 32% more admissions per deaths among the most deprived Hull quintile and 50% more admissions per death in the most deprived East Riding quintile, compared with the least deprived quintile of each PCT. Robert Sheikh Iddenden, Public Health Sciences. 183

184 Table 85: Relationship between prevalence, hospital admission rate and mortality among local deprivation quintiles 24 PCT and local deprivation quintile Hull and ERoY Actual 25 Prevalence % (index) Modelled Age-gender-standardised rates per 100,000 persons (index) Hospital admission rate 26 COPD mortality rate From PCMD dataset 28 From QOF ERPHO 27 Practice Resident PHMF 29 Doncaster quintiles quintiles Practice Resident Resident quintiles quintiles 30 quintiles Most dep'd 2.25 (100) 4.52 (100) 2.44 (100) 660 (100) 723 (100) 93.7 (100) ( 82) 4.76 (105) 2.51 (103) 494 ( 75) 412 ( 57) 62.2 ( 66) (101) 3.74 ( 83) 2.40 ( 98) 304 ( 46) 251 ( 35) 48.0 ( 51) ( 67) 2.70 ( 60) 1.74 ( 71) 209 ( 32) 181 ( 25) 38.1 ( 41) Least dep'd 1.37 ( 61) 2.10 ( 46) 1.42 ( 58) 163 ( 25) 135 ( 19) 30.6 ( 33) Hull East Riding Most dep'd 2.33 (100) 4.67 (100) 2.62 (100) 642 (100) 854 (100) (100) (100) (100) ( 90) 4.55 ( 97) 2.36 ( 90) 596 ( 93) 651 ( 76) 84.1 ( 83) 71.8 ( 62) 83.0 ( 76) ( 78) 4.82 (103) 2.50 ( 96) 534 ( 83) 508 ( 59) 77.1 ( 76) 65.8 ( 57) 73.5 ( 68) ( 70) 4.07 ( 87) 2.09 ( 80) 335 ( 52) 363 ( 42) 59.6 ( 59) 62.8 ( 54) 56.4 ( 52) Least dep'd 1.25 ( 53) 3.58 ( 77) 1.82 ( 69) 200 ( 31) 283 ( 33) 66.4 ( 66) 38.2 ( 33) 47.5 ( 44) Most dep'd 2.82 (100) 3.43 (100) 2.65 (100) 318 (100) 400 (100) 62.1 (100) ( 57) 2.55 ( 74) 1.84 ( 70) 216 ( 68) 180 ( 45) 38.6 ( 62) ( 57) 2.17 ( 63) 1.58 ( 60) 170 ( 54) 181 ( 45) 39.5 ( 64) ( 43) 2.22 ( 65) 1.47 ( 55) 147 ( 46) 163 ( 41) 30.0 ( 48) Least dep'd 1.40 ( 50) 1.82 ( 53) 1.18 ( 44) 199 ( 63) 129 ( 32) 30.1 ( 48) 24 Deprivation quintiles derived from IMD QOF 2008/ /07 to 2008/09 pooled, all admissions with primary diagnosis COPD 27 Excludes Northpoint practice to 2006 pooled, patients registered with Hull GP practices only to 2008 pooled, not able to produce by practice deprivation quintiles 30 Excludes Hull residents registered with east Riding GP practices Robert Sheikh Iddenden, Public Health Sciences. 184

185 Geodemographic segmentation An alternative way of looking at deprivation is to use a geo-demographic segmentation tool, such as ACORN or Health ACORN. Mortality data for are presented broken down by ACORN classifications in Table 86 (data are suppressed where fewer than 3 deaths from COPD were registered during ). Rates among the Hard Pressed (the most deprived ACORN category) were two and a half times higher than among the Wealthy Achievers (the most affluent ACORN category). Rates among the Hard Pressed were also more than double those for the Comfortably Off and Moderate Means ACORN categories (there were too few deaths in the Urban Prosperity category to evaluate rates). The Hard Pressed and Wealthy Achievers categories each contain around one quarter of the population of Hull and East riding, yet 44% of deaths from COPD in Hull and East Riding during occurred among the Hard Pressed while just 16% occurred among Wealthy Achievers. Within the Hard Pressed category, there was substantial variation in COPD mortality rates between ACORN groups, with rates ranging from 66 deaths per 100,000 residents in group 5.N Struggling Families to 169 deaths per 100,000 residents in group 5.P High Rise Hardship. The mortality rate in the ACORN group Prudent Pensioners was similar to that for the Hard Pressed. If this group really is dominated by pensioners, we would expect all causes of death to be higher than in ACORN groups with younger age profiles, although with only 7 COPD deaths per year, it is not possible to assess whether this is a true reflection of the underlying COPD mortality rate in this group, as it will be subject to wide random variation. Table 86: Average annual number of COPD deaths* (where COPD is recorded as primary cause of death) and crude mortality rates per 100,000 by ACORN geodemographic classifications, deaths registered ** Average annual... ACORN group and category (bold) Crude Deaths Population rate per 100,000 1.A.1 Wealthy Mature Professionals, Large Houses 4 7, A Wealthy Executives 12 39, B Affluent Greys 24 53, C Flourishing Families 16 55, Wealthy Achievers , D Prosperous Professionals <3 4,462-2.E Educated Urbanites <3 2,622-2.F Aspiring Singles 5 14, Urban Prosperity - 21,625-3.G Starting Out 3 11, H Secure Families 27 82, I Settled Suburbia 31 56, J Prudent Pensioners 7 7, Comfortably Off , Robert Sheikh Iddenden, Public Health Sciences. 185

186 Average annual... ACORN group and category (bold) Crude Deaths Population rate per 100,000 4.L Post Industrial Families 13 33, M Blue Collar Roots 37 83, Moderate Means , N Struggling Families , O Burdened Singles 50 42, P High Rise Hardship 16 9, Hard Pressed , *Rounded to nearest integer **Data suppressed if fewer than 3 deaths between 2006-and 2008 inclusive Due to the larger number of COPD deaths in the Hard Pressed category it is possible to look at these in more detail. To this end average annual numbers of deaths and mortality rates are presented in Table 87 for each ACORN type within the Hard Pressed category. Within each ACORN group there is substantial variability between ACORN types which may be partly due to small numbers of COPD deaths in many of these groups, but may also reflect true differences between the groups. COPD mortality rates were highest in ACORN types 5.N.45, 5.O.50, 5.O.51 and 5.P.53, each with pensioners or old people in the descriptor. For each of these ACORN types the COPD mortality rates were greater than 100 per 100,000. As mentioned earlier we would expect higher mortality rates in these groups for most chronic disease deaths. Table 87: Average annual number of COPD deaths* (where COPD is recorded as primary cause of death) and crude mortality rates per 100,000 by ACORN type where the ACORN category is Hard Pressed, deaths registered ** Average annual... ACORN type, group and category Crude Deaths Population rate per 100,000 5.N.44 Low Income Larger Families, Semis 2 7, N.45 Low Income, Older people, Smaller Semis 15 13, N.47 Low Income Families, Terraced Estates 31 45, N.48 Families & Single Parents, Semis & Terraces 12 20, N.49 Large Families & Single Parents, Many Children 6 14, N Struggling Families , O.50 Single Elderly People, Council Flats 16 10, O.51 Single Parents & Pensioners, Council Terraces 34 32, O Burdened Singles 50 42, P.53 Old People, Many High Rise Flats 14 7, P.54 Singles & Single Parents, High Rise Estates 2 2, P High Rise Hardship 16 9, Hard Pressed , *Rounded to nearest integer **Data suppressed if fewer than 3 deaths between 2006-and 2008 inclusive Robert Sheikh Iddenden, Public Health Sciences. 186

187 If we examine where the Hard Pressed sit geographically (Figure 37 for Hull and Figure 38 for East Riding) we see that they are to be found in large concentrations in Hull, especially in the north and east of the city, but also with large pockets in the west of the city, while in East Riding they are found in smaller pockets to the west of Hull, as well as in Beverley, Goole, Bridlington, Driffield and Holderness. These are areas that also tend to have the highest levels of deprivation within each PCT, as measured by the Index of Multiple Deprivation Figure 37: ACORN types in Hull within Hard Pressed ACORN category Robert Sheikh Iddenden, Public Health Sciences. 187

188 Figure 38: ACORN types in East Riding of Yorkshire within Hard Pressed ACORN category Mortality data for are presented in Table 88 broken down by Health ACORN classifications (data are suppressed where fewer than 3 deaths from COPD were registered during ). Rates among those defined as having Existing Problems (the most health-deprived Health ACORN category) were almost three times higher than among the Healthy (the least health-deprived Health ACORN category). Rates among the Existing Problems were two and a half times higher than those for the Future Possible Concerns and twice as high as those in the Future Problems Health ACORN group. The Existing Problems group contains around one sixth of the population of Hull and East riding, yet one third of the deaths from COPD in Hull and East Riding during occurred among the Existing Problems group. Robert Sheikh Iddenden, Public Health Sciences. 188

189 Within the Existing Problems group, there was substantial variation in COPD mortality rates between Health ACORN types, with rates ranging from 55 deaths per 100,000 residents in group 1.8 Disadvantaged neighbourhoods with poor diet and severe health issues to 320 deaths per 100,000 residents in group 1.2 Disadvantaged elderly, poor diet, chronic health. However, it should be noted that the numbers of deaths each year in these more defined classifications are relatively low, and therefore subject to random variation, so we need to treat the rates with a degree of caution. Of course, the last Health ACORN type has the descriptor that perhaps best fits someone suffering from severe COPD, so it is not surprising that the COPD mortality rate for this classification was the highest. Table 88: Average annual number of COPD deaths* (where COPD is recorded as primary cause of death) and crude mortality rates per 100,000 by Health ACORN geodemographic classifications, deaths registered ** Average annual... Health ACORN type and group (bold) Crude Deaths Population rate per 100, Older couples, traditional diets, cardiac issues 7 4, Disadvantaged elderly, poor diet, chronic health 10 3, Vulnerable disadvantaged, smokers with high levels of obesity 8 4, Post industrial pensioners with long term illness 17 14, Deprived neighbourhoods with poor diet, smokers 19 15, Elderly with associated health issues 19 19, Home owning pensioners, traditional diets 11 11, Disadvantaged neighbourhoods with poor diet and severe health issues 15 27, Existing Problems , Poor single parent families with lifestyle related illnesses 7 17, Multi-ethnic, high smoking, high fast food consumption 5 13, Urban estates with sedentary lifestyle and low fruit and vegetable consumption 30 38, Deprived multi-ethnic estates, smokers and overweight 8 19, Disadvantaged multi-ethnic younger adults, with high levels of smoking 7 20, Future Problems , Less affluent neighbourhoods, high fast food, sedentary lifestyles 15 20, Affluent healthy pensioners, dining out 7 15, Home owning older couples, high level of fat and confectionary 12 32, Affluent professionals, high alcohol consumption, 14 41, Robert Sheikh Iddenden, Public Health Sciences. 189

190 Average annual... Health ACORN type and group (bold) Crude Deaths Population rate per 100,000 dining out 3.5 Low income families with some smokers 12 26, Affluent families with some dietary concerns 3 18, Possible Future Concerns , Young mobile population with good health and diet 8 25, Younger affluent, healthy professionals 3 15, Students and young professionals, living well 1 7, Towns and villages with average health and diet 18 60, Mixed communities with better than average health 31 51, Affluent towns and villages with excellent health and diet 21 70, Healthy , *Rounded to nearest integer **Data suppressed if fewer than 3 deaths between 2006-and 2008 inclusive Upon examining where the Existing Problems sit geographically (Figure 39 for Hull and Figure 40 for East Riding) it can be seen that they are to be found in relatively small pockets throughout Hull, but with larger concentrations in the Eastern and Western edges of the city, while in East Riding of Yorkshire the largest concentration are to be found in Bridlington. Robert Sheikh Iddenden, Public Health Sciences. 190

191 Figure 39: Dominant Health ACORN types where the dominant Health ACORN group is Existing Problems by output areas in Hull Robert Sheikh Iddenden, Public Health Sciences. 191

192 Figure 40: Dominant Health ACORN types where the dominant Health ACORN group is Existing Problems by output areas in East Riding of Yorkshire Similar analyses of emergency hospital admission rates can be produced using ACORN and Health ACORN. When looking at emergency admissions we can look at ACORN types, unlike the above analysis of mortality, as there are far more admissions for COPD (around 1,800 per year) than there are deaths from COPD (around 300 per year). Table 89 contains the average annual number of emergency admissions with a primary diagnosis of COPD in Hull and East Riding residents from 2007/08 to 2009/10 inclusive, together with crude emergency admission rates per 100,000 residents, by ACORN classifications. The ACORN category with the highest emergency admission rate is the Hard Pressed category, with an emergency admission rate of 615 per 100,000 residents, more than Robert Sheikh Iddenden, Public Health Sciences. 192

193 twice as high as those in the Moderate Means and Urban Prosperity categories, almost three times as high as the Comfortably Off category, and more than four times as high as the Wealthy Achievers category. The Hard Pressed category contains the three ACORN groups with the highest emergency admission rates, 5.P High Rise Hardship (1,201 per 100,000), 5.O Burdened Singles (836 per 100,000) and 5.N Struggling Families (465 per 100,000). There was more variation at ACORN type level, with 15 ACORN types having emergency admission rates for COPD of 300 per 100,000 or higher, including 1 ACORN type in the Urban Prosperity category, 4 in the Comfortably Off category, 2 in the Moderate Means category and 8 of the 9 ACORN types within the Hard Pressed category. The output areas where these 15 types are located are shown in Figure 41 for Hull and Figure 42 for East Riding of Yorkshire. More than half of Hull residents live in output areas where the dominant ACORN type is one of these fifteen types, with particular concentrations in the northern, eastern and central parts of the city. In East Riding of Yorkshire one fifth of residents live in output areas where the dominant ACORN type is one of these fifteen types, with the largest concentrations in Bridlington and the west of Hull villages. Table 89: Average annual number 31 of emergency admissions where the primary diagnosis was COPD and crude emergency admission rates per 100,000 by ACORN geodemographic classifications, admissions in Hull and East Riding residents 2007/08 to 2009/10 inclusive Average annual... ACORN type, group and category Emergency admissions Population Crude rate per 100,000 1.A.1 Wealthy Mature Professionals, Large Houses 7 7, A.2 Wealthy Working Families with Mortgages 3 9, A.3 Villages with Wealthy Commuters 5 7, A.4 Well-Off Managers, Larger Houses 25 15, A Wealthy Executives 40 39, B.5 Older Affluent Professionals 8 7, B.6 Farming Communities 33 20, B.7 Old People, Detached Homes 15 9, B.8 Mature Couples, Smaller Detached Homes 36 15, B Affluent Greys 94 53, C.9 Older Families, Prosperous Suburbs 20 8, C.10 Well-Off Working Families with Mortgages 7 16, C.11 Well-Off Managers, Detached Houses 37 29, C.12 Large Families and Houses in Rural Areas 2 1, C Flourishing Families 67 56, Wealthy Achievers , D.13 Well-Off Professionals, Larger Houses and 3 1, Average annual numbers of emergency admissions are rounded down to the nearest integer. Data are suppressed if there were fewer than 5 admissions during 2007/08 to 2009/10. Robert Sheikh Iddenden, Public Health Sciences. 193

194 ACORN type, group and category Converted Flats 2.D.14 Older Professionals in Suburban Houses and Apartments Emergency admissions Average annual... Population Crude rate per 100, , D Prosperous Professionals 6 4, E.17 Young Educated Workers, Flats 2 1, E.19 Suburban Privately Renting Professionals 3 1, E Educated Urbanites 6 2, F.20 Student Flats and Cosmopolitan Sharers 8 4, F.22 Low Income Singles, Small Rented Flats 29 7, F.23 Student Terraces 2 3, F Aspiring Singles 39 14, Urban Prosperity 51 21, G.24 Young Couples, Flats and Terraces 4 1, G.25 White Collar Singles and Sharers, Terraces 11 9, G Starting Out 15 11, H.26 Younger White Collar Couples with Mortgages 11 9, H.27 Middle Income, Home Owning Areas 18 10, H.28 Working Families with Mortgages 10 13, H.29 Mature Families in Suburban Semis 27 15, H.30 Established Home Owning Workers 63 34, H Secure Families , I.32 Retired Home Owners 37 11, I.33 Middle Income, Older Couples 41 23, I.34 Lower Incomes, Older People, Semis 79 22, I Settled Suburbia , J.35 Elderly Singles, Purpose Built Flats 9 2, J.36 Older People, Flats 21 5, J Prudent Pensioners 31 7, Comfortably Off , L.39 Skilled Older Families, Terraces 27 19, L.40 Young Working Families 25 13, L Post Industrial Families 53 33, M.41 Skilled Workers, Semis and Terraces 53 15, M.42 Home Owning Families, Terraces , M.43 Older People, Rented Terraces 84 20, M Blue Collar Roots , Moderate Means , N.44 Low Income Larger Families, Semis 21 7, N.45 Low Income, Older people, Smaller Semis 94 13, N.47 Low Income Families, Terraced Estates , N.48 Families and Single Parents, Semis and Terraces 88 20, Robert Sheikh Iddenden, Public Health Sciences. 194

195 ACORN type, group and category Emergency admissions Average annual... Population Crude rate per 100,000 5.N.49 Large Families and Single Parents, Many Children 59 14, N Struggling Families , O.50 Single Elderly People, Council Flats ,437 1,003 5.O.51 Single Parents and Pensioners, Council Terraces , O Burdened Singles , P.53 Old People, Many High Rise Flats 100 7,104 1,412 5.P.54 Singles and Single Parents, High Rise Estates 14 2, P High Rise Hardship 115 9,575 1, Hard Pressed , Figure 41: Dominant ACORN types (at output area level) where the annual average emergency admission rate from 2007/08 to 2009/10 for COPD was 300 per 100,000 Hull and East Riding residents or higher, by output areas in Hull Robert Sheikh Iddenden, Public Health Sciences. 195

196 Figure 42: Dominant ACORN types (at output area level) where the annual average emergency admission rate from 2007/08 to 2009/10 for COPD was 300 per 100,000 residents of Hull and Easter Riding or higher, by output areas in East Riding of Yorkshire A similar analysis of annual average emergency admissions during 2007/08 to 2009/10 in Hull and East Riding residents by Health ACORN classifications is shown in Table 90. Emergency admission rates in output areas classified as Existing Problems, at 670 per 100,000, were 50% than those classified as Future Problems (417 per 100,000), three times higher than those classified as Future Possible Concerns (205 per 100,000) and four times as high as those classified as Healthy (164 per 100,000). Among the Existing Problems group the emergency admission rate ranged from 2,015 per 100,000 in output areas with the Health ACORN type 1.2 Disadvantaged elderly, poor diet, chronic health to 296 per 100,000 in output areas designated as type 1.7 Home owning pensioners, traditional diets. Twelve Health ACORN types had emergency admission rates that were 300 per 100,000 or higher, including all 8 within the Existing Problems group (this includes 1.7 Home owning pensioners, traditional diets which has been included as the emergency admission rate was 296 per 100,000), 3 of the 5 types Robert Sheikh Iddenden, Public Health Sciences. 196

197 within the Future Problems group and 1 type, 3.1 Less affluent neighbourhoods, high fast food, sedentary lifestyles, within the Possible Future Concerns group. The output areas where these 12 types are located are shown in Figure 43 for Hull and Figure 44 for East Riding of Yorkshire. In Hull 49% of residents live in output areas designated by one of these twelve Health ACORN types, with particular concentrations in the northern, eastern and central parts of the city. In East Riding of Yorkshire one fifth of residents live in output areas designated by one of these eleven Health ACORN types, with particular concentrations in Bridlington, Holderness, the west of Hull villages and Goole. Table 90: Average annual number 32 of emergency admissions where the primary diagnosis was COPD and crude emergency admission rates per 100,000 by Health ACORN geodemographic classifications, admissions in Hull and East Riding residents 2007/08 to 2009/10 inclusive Average annual... Health ACORN type and group (bold) Emergency Crude rate Population admissions per 100, Older couples, traditional diets, cardiac issues 49 4,498 1, Disadvantaged elderly, poor diet, chronic health 64 3,210 2, Vulnerable disadvantaged, smokers with high levels of obesity 52 4,534 1, Post industrial pensioners with long term illness , Deprived neighbourhoods with poor diet, smokers Deprived neighbourhoods with poor diet, smokers , Elderly with associated health issues , Home owning pensioners, traditional diets 35 11, Disadvantaged neighbourhoods with poor diet and severe health issues , Existing problems , Poor single parent families with lifestyle related illnesses 78 17, Multi-ethnic, high smoking, high fast food consumption 31 14, Urban estates with sedentary lifestyle and low fruit and vegetable consumption , Deprived multi-ethnic estates, smokers and overweight 71 19, Disadvantaged multi-ethnic younger adults, with high levels of smoking 55 21, Future Problems , Less affluent neighbourhoods, high fast food, sedentary lifestyles , Affluent healthy pensioners, dining out 24 15, Home owning older couples, high level of fat and confectionary 62 32, Affluent professionals, high alcohol consumption, dining out 60 41, Low income families with some smokers 63 27, Average annual numbers of emergency admissions are rounded down to the nearest integer. Data are suppressed if there were fewer than 5 admissions during 2007/08 to 2009/10. Robert Sheikh Iddenden, Public Health Sciences. 197

198 Health ACORN type and group (bold) Emergency admissions Average annual... Population Crude rate per 100, Affluent families with some dietary concerns 6 18, Possible Future Concerns , Young mobile population with good health and diet 35 26, Younger affluent, healthy professionals 22 15, Students and young professionals, living well 3 7, Towns and villages with average health and diet 97 60, Mixed communities with better than average health , Affluent towns and villages with excellent health and diet 94 70, Healthy , Figure 43: Dominant Health ACORN types (at output area level) where the annual average emergency admission rate from 2007/08 to 2009/10 for COPD was 300 per 100,000 Hull and East Riding residents, by output areas in Hull Robert Sheikh Iddenden, Public Health Sciences. 198

199 Figure 44: Dominant Health ACORN types (at output area level) where the annual average emergency admission rate from 2007/08 to 2009/10 for COPD was 300 per 100,000 Hull and East Riding residents or greater, by output areas in East Riding of Yorkshire Risk Factors Smoking is the key risk factor for COPD, whether current smoking or past smoking. The prevalence of smoking, plus the prevalence of former smokers, may be examined by deprivation quintiles from data collected from local surveys conducted in Hull and East Riding in Two surveys were conducted in Hull during 2009; one was a Social Capital survey that asked some risk factor questions, the other was a Prevalence of Risk Factor survey. Both these surveys asked the same question on smoking, so responses for this question were combined to produce Hull estimates of smoking prevalence. Agegender-adjusted percentages of respondents from these surveys that defined themselves as current smokers or as former smokers are presented in Table 91 and in Figure 45 by deprivation quintile. Robert Sheikh Iddenden, Public Health Sciences. 199

200 Ever smoked Former smoker Current smoker There were clear gradients in the percentages currently smoking, and in the percentages that had ever smoked, by deprivation quintile for each PCT. In both Hull and East Riding the most deprived quintiles was statistically significantly more likely to be a current smoker than the three least deprived quintiles. Excluding the most deprived quintile, the percentages within each quintile in Hull were statistically significantly higher than in the corresponding quintiles in East Riding. There were few differences in the percentage of former smokers between deprivation quintiles, although percentages for each quintile 2 were higher in East Riding. The tests for trend for current smoking and ever smoking were statistically significant for both Hull and East Riding (the latter driven mostly by current smokers). Table 91: Smoking status of survey respondents from surveys in Hull and East Riding conducted in 2009 by PCT and deprivation quintile (age-gender-adjusted percentages) Smoking status in 2009 by PCT and deprivation quintile Smoking status and Age-standardised percentages (95% confidence intervals) deprivation quintile Hull (2007) ERoY (2009) Hull & ERoY Most deprived 45.1 (41.1, 49.5) 33.0 (25.4, 42.0) 39.4 (37.3, 41.6) (37.4, 45.2) 22.6 (18.8, 26.9) 23.9 (21.8, 26.1) (29.3, 35.7) 18.9 (15.8, 22.2) 17.2 (15.4, 19.1) (19.5, 24.8) 13.9 (11.9, 16.2) 13.3 (11.6, 15.1) Least deprived 15.7 (13.6, 18.0) 10.4 ( 8.8, 12.3) 10.4 ( 8.8, 12.3) Most deprived 24.9 (21.9, 28.3) 30.9 (24.3, 38.4) 25.4 (23.7, 27.2) (22.1, 28.7) 30.4 (26.5, 34.7) 28.7 (26.5, 31.0) (20.2, 25.8) 28.5 (25.5, 31.7) 27.7 (25.6, 29.9) (25.2, 31.4) 29.5 (27.1, 32.2) 29.5 (27.2, 31.8) Least deprived 25.5 (22.8, 28.5) 30.6 (28.2, 33.2) 30.6 (28.2, 33.2) Most deprived 70.1 (64.9, 75.5) 63.9 (53.7, 75.3) 64.9 (62.1, 67.7) (61.5, 71.7) 53.0 (47.4, 58.9) 52.5 (49.5, 55.7) (51.2, 59.7) 47.4 (43.1, 51.9) 44.9 (42.1, 47.8) (46.2, 54.4) 43.5 (40.2, 46.9) 42.8 (39.9, 45.7) Least deprived 41.2 (37.7, 44.9) 41.1 (38.1, 44.2) 41.1 (38.1, 44.2) Robert Sheikh Iddenden, Public Health Sciences. 200

201 Most depived Least deprived Most depived Least deprived Most depived Least deprived Age-gender-adjusted percentages Figure 45: Smoking status of survey respondents (Hull 2007, East Riding 2009) by PCT and deprivation quintile, age-gender-adjusted percentages 80 Solid bar=current smoker; transparent bar=former smoker Hull East Riding Hull and East Riding Deprivation quintiles and PCT Access to Services It is possible that people living in more deprived areas are less demanding with regard to their health, and in areas where life expectancy is lower and ill health occurs at an earlier age, it could be that poorer health is accepted as a consequence of ageing rather than an issue for further investigation and possible treatment. This could influence take-up of and access to services. The main symptoms of COPD are shortness of breath, a persistent cough and a build up of mucus and phlegm. Some people may perceive these symptoms to be a smokers cough or just part of the ageing process, and therefore not seek medical advice. If there are differential rates of patients not seeking medical help when they have these symptoms this will lead to inequalities in treatment, emergency hospital admissions and the rate of early deaths from COPD. There is a suggestion from the modelled estimates of prevalence, illustrated in Figure 35 and Figure 36 that in Hull the prevalence of undiagnosed COPD may be higher in more deprived areas. However, as mentioned earlier, it is possible that the models are a poorer fit (in relation to the true underlying prevalence of COPD) in areas at the extremes of deprivation (most deprived areas and least deprived areas in the country) Pulmonary rehabilitation The numbers of patients aged 35 years and over and registered with Hull GP practices referred to pulmonary rehabilitation between January 2009 and September 2010 inclusive Robert Sheikh Iddenden, Public Health Sciences. 201

202 are shown in Table 92 by practice deprivation quintiles, together with the percentage out of all patients on practice QOF 2008/09 COPD registers and age-adjusted percentages out of the whole population aged 35 years and over. The number of people referred to pulmonary rehabilitation was equivalent to between 7% and 12% of those patients on QOF 2008/09 COPD registers by practice deprivation quintiles, with the largest percentage in the most deprived quintile of practices and the lowest percentage in the least deprived quintile of practices. The age-adjusted percentages of practice patients aged 35 years referred to pulmonary rehabilitation between January 2009 and September 2010 inclusive in the most deprived quintile (0.63%) was statistically significantly higher than the percentages in each other quintile, almost double the percentage for the second most deprived, middle and second least deprived quintiles and more than three times higher than the percentage for the least deprived quintile. The differences between practice deprivation quintiles were lower than for registered prevalence, suggesting patients in the most deprived practice quintiles were less likely to be referred to pulmonary rehabilitation, which may represent an important inequity. Table 92: Number of patients aged 35 years and over registered with Hull GP practices referred to pulmonary rehabilitation between January 2009 and September 2010 inclusive by practice deprivation quintile 33, and percentage of all patients on QOF 2008/09 COPD registers Practice deprivation quintiles Number of patients aged 35 years and over registered with Hull GP practices referred to pulmonary rehabilitation between January 2009 and September 2010 inclusive by practice deprivation quintile, and percentage of all patients on QOF 2008/09 COPD registers QOF 2008/09 Referred to pulmonary rehabilitation N % n % QOF % (age-adjusted) register population (95% CI) Most deprived 1, (0.54,0.73) 2 1, (0.27,0.42) 3 1, (0.29,0.42) 4 1, (0.26,0.38) Least deprived (0.13,0.24) Given that around 60% of patients did not attend for pulmonary rehabilitation, it is important to examine these non-attendances by deprivation quintiles to assess whether there is any further inequity related to deprivation. The numbers of patients that did not attend for pulmonary rehabilitation having been referred are shown in Table 93 as well as the percentage out of all referrals. The overall numbers of referrals are lower than in Table 92 as follow-up status was not known for patients referred August to September 2010, so these patients are excluded from Table 92. There were no clear trends in the percentage of pulmonary rehabilitation DNAs by practice deprivation quintiles with no statistically significant differences in the percentage of DNAs. 33 As quintiles were produced by ranking practices by deprivation without regard to population size, the quintiles do not contain 20% of the population each Robert Sheikh Iddenden, Public Health Sciences. 202

203 Table 93: Numbers of patients aged 35 years and over registered with Hull GP practices referred to pulmonary rehabilitation between January 2009 and July 2010 inclusive that did not attend pulmonary rehabilitation, and the percentage out of all referrals with 95% confidence intervals, by practice deprivation quintiles Practice deprivation quintiles Numbers of patients aged 35 years and over registered with Hull GP practices referred to pulmonary rehabilitation between January 2009 and July 2010 inclusive that did not attend pulmonary rehabilitation, and the percentage out of all referrals with 95% confidence intervals Referred DNA N n % (95% CI) Most deprived (54.9, 70.0) (43.5, 65.2) (53.8, 71.5) (50.5, 69.9) Least deprived (38.4, 69.0) Differences in the age-adjusted percentages of Hull residents aged 35 years and over and registered with Hull GP practices and referred to pulmonary rehabilitation between January 2009 and September 2010 inclusive are shown by local quintiles of IMD 2007 in Table 94 by gender. Higher age-adjusted percentages of women than men in the three most deprived quintiles were referred to pulmonary rehabilitation, with the position reversed for the two least deprived quintiles. The differences between men and women were greatest in the least deprived quintile (37% higher in men) and the most deprived quintile (18% higher in women), although the differences by gender were not statistically significant. The age-adjusted percentages in the two most deprived quintiles were almost twice as high as among the least deprived quintile, the differences statistically significant and seen for both men and women. Table 94: Age-adjusted percentage of patients (Hull residents only) aged 35 years and over referred to pulmonary rehabilitation between January 2009 and September 2010 inclusive, by local quintiles of IMD2007 Local deprivation quintiles Age-adjusted percentages of patients (Hull residents only) aged 35 years and over referred to pulmonary rehabilitation between January 2009 and September 2010 inclusive by local quintiles of IMD 2007 (95% confidence intervals) Males Females All Most deprived 0.49 (0.37, 0.63) 0.58 (0.45, 0.75) 0.54 (0.44, 0.64) (0.38, 0.66) 0.54 (0.41, 0.69) 0.52 (0.43, 0.62) (0.34, 0.58) 0.48 (0.37, 0.61) 0.46 (0.39, 0.55) (0.22, 0.43) 0.28 (0.20, 0.39) 0.29 (0.23, 0.37) Least deprived 0.26 (0.18, 0.36) 0.19 (0.12, 0.29) 0.22 (0.17, 0.29) Robert Sheikh Iddenden, Public Health Sciences. 203

204 Approximately 60% of all patients referred to pulmonary rehabilitation over this period did not attend. The numbers of pulmonary rehabilitation DNAs are shown in along with the percentage out of all pulmonary rehabilitation referrals in hull residents (with 95% confidence intervals) are shown in Table 95 by local deprivation quintiles for referrals between January 2009 and July 2010 inclusive (follow-up information on patients referred in August and September 2010 was not known and so these referrals were excluded from this analysis). There were no clear trends in the percentage of pulmonary rehabilitation DNAs by local deprivation quintiles with no statistically significant differences in the percentage of DNAs. Table 95: Numbers of patients aged 35 years and over registered with Hull GP practices referred to pulmonary rehabilitation between January 2009 and July 2010 inclusive that did not attend pulmonary rehabilitation, and the percentage out of all referrals with 96% confidence intervals by local deprivation quintiles Local deprivation quintiles Numbers of patients (Hull residents only) aged 35 years and over referred to pulmonary rehabilitation between January 2009 and July 2010 inclusive that did not attend pulmonary rehabilitation, and the percentage out of all referrals with 95% confidence intervals Referred DNA N n % (95% CI) Most deprived (49.6, 67.4) (51.8, 70.0) (57.8, 75.0) (41.1, 64.8) Least deprived (44.2, 69.7) Home oxygen / long term oxygen therapy The numbers of patients aged 35 years and over and registered with Hull GP practices receiving home oxygen in 2010 are shown in Table 96 by practice deprivation quintiles, together with the percentage out of all patients on practice QOF 2008/09 COPD registers and age-adjusted percentages out of the whole population aged 35 years and over. It is assumed that those receiving home oxygen are on long term oxygen therapy. The number of people receiving home oxygen in 2010 was equivalent to between 11% and 15% of those patients on QOF 2008/09 COPD registers by practice deprivation quintiles, with the largest percentage in the second most deprived quintile of practices and the lowest percentage in the two least deprived quintiles of practices. The age-adjusted percentages of practice patients aged 35 years and over receiving home oxygen in 2010 in the two most deprived quintiles (each at 0.66%) were statistically significantly higher than the percentages in the two least deprived quintiles, 83% higher than the second least deprived quintile and 136% higher than the least deprived quintile. The differences between practice deprivation quintiles were greater than for registered prevalence, Robert Sheikh Iddenden, Public Health Sciences. 204

205 suggesting patients in the most deprived practice quintiles had greater need of home oxygen, therefore were likely to have more advanced, and therefore less well controlled symptoms of, COPD than were patients in the least deprived practice quintiles. Table 96: Number of patients aged 35 years and over registered with Hull GP practices receiving home oxygen in 2010 by practice deprivation quintile 34, and percentage of all patients on QOF 2008/09 COPD registers Practice deprivation quintiles Number of patients aged 35 years and over registered with Hull GP practices receiving home oxygen in 2010 by practice deprivation quintile, and percentage of all patients on QOF 2008/09 COPD registers QOF 2008/09 Home oxygen N % n % QOF % (age-adjusted) register population (95% CI) Most deprived 1, (0.57, 0.76) 2 1, (0.56, 0.77) 3 1, (0.42, 0.57) 4 1, (0.30, 0.43) Least deprived (0.22, 0.37) Differences in the age-adjusted percentages of Hull residents aged 35 years and over and registered with Hull GP practices receiving home oxygen in 2010 by local quintiles of IMD 2007 are shown in Table 97 by gender. Higher age-adjusted percentages of women than men in each except the least deprived quintile received home oxygen in 2010, although the differences by gender were not statistically significant. The age-adjusted percentage in the most deprived quintile of men was almost twice as high as among the two least deprived quintiles of men, the difference statistically significant. The age-adjusted percentage among most deprived quintile of women was 58% higher than the middle quintile, 93% higher than the second least deprived quintile and 129% higher than the least deprived quintile, with the difference statistically significant in each case. 34 As quintiles were produced by ranking practices by deprivation without regard to population size, the quintiles do not contain 20% of the population each Robert Sheikh Iddenden, Public Health Sciences. 205

206 Table 97: Age-adjusted percentage of patients aged 35 years and over receiving home oxygen in 2010, by local quintiles of IMD2007, Hull residents only 35 Local deprivation quintiles Age-adjusted percentages of patients aged 35 years and over receiving home oxygen in 2010 by local quintiles of IMD 2007 (95% confidence intervals) Males Females All Most deprived 0.77 (0.62, 0.95) 0.87 (0.71, 1.06) 0.82 (0.71, 0.95) (0.36, 0.63) 0.60 (0.47, 0.76) 0.54 (0.45, 0.65) (0.45, 0.72) 0.55 (0.43, 0.68) 0.56 (0.47, 0.66) (0.28, 0.51) 0.45 (0.34, 0.58) 0.42 (0.34, 0.51) Least deprived 0.39 (0.28, 0.51) 0.38 (0.28, 0.51) 0.38 (0.31, 0.47) Quality of Care On-going Care in Primary Care As mentioned in section starting on page 54, the QOF information is not available for any subgroups so it is not straightforward to examine the on-going measure of care by deprivation. Without individual-level patient records, the only way to examine quality of care and deprivation, is by examining a measure of deprivation at a practice level (e.g. mean Index of Multiple Deprivation 2007 score based on postcodes of each practice s registered patients). This gives a crude measure of deprivation at practice level. The practice level deprivation scores are given in Appendix B starting on page 282. Table 98 gives the percentages of patients who were made exceptions and the percentages of patients with the measured indicator for practices based on local deprivation quintiles derived from IMD 2007 for Hull practices. Practices with the highest mean deprivation scores in Hull had the highest prevalence of COPD on their registers. For Hull practices there were no clear trends by deprivation quintiles for either indicators percentages or exceptions, apart from those receiving influenza vaccinations (COPD08), for which indicator the most deprived practices recorded the highest percentage of exceptions (18%) and the lowest indicator percentage (91.8%), while the least deprived practices recorded the lowest percentage of exceptions (15.2%) and the highest indicator percentage (95.3%). However the differences between the highest and lowest COPD08 indicator percentages, and percentage of exceptions, were small. The percentage of patients diagnosed since 1st April 2008 in whom the diagnosis has been confirmed by post bronchodilator spirometry (COPD12) was lowest for the most deprived practices (80%) and highest for the least deprived practices (97.4%), although both groups had exceptions of around one third for this indicator. In general, the most deprived practices tended to have lower indicator percentages and lower percentages excepted than did the least deprived practices. 35 As data were only provided for patients registered with Hull GP practices, those residents of Hull not registered with Hull GP practices were excluded from this analysis. Robert Sheikh Iddenden, Public Health Sciences. 206

207 Table 98: COPD QOF exceptions and indicator percentages by deprivation quintile of Hull practices (based on mean IMD2007 score of practice), local (Hull) quintiles defined at practice level 36 Weighted average over practices of QOF 2008/09 exceptions and indicators by mean local deprivation Indicator Measure quintiles (IMD2007) defined at practice level Most Least deprived deprived COPD01: On register % COPD08: flu jab % excl % indict r COPD10: FeV1 % excl % indict r COPD11: Inhaler % excl % indict r COPD12: Spirometry % excl % indict r Table 99 gives the percentages of patients who were made exceptions and the percentages of patients with the measured indicator for practices based on local deprivation quintiles derived from IMD 2007 for East Riding practices. Practices with the highest mean deprivation scores in East Riding had the highest prevalence of COPD on their registers. For East Riding practices there were no discernible patterns with regard to deprivation quintiles of practices. The most deprived practices recorded the highest percentage of exceptions for COPD08 influenza vaccination (15.5%) with the least deprived practices recording the lowest percentage of exceptions (7.5%), although the least deprived practices also had the lowest indicator percentage (91.7%). The differences in the percentages achieving this indicator were small. In general, the most deprived practices recorded fewer exceptions and achieved higher indicator percentages than the least deprived practices, although both exceptions and indicator percentages tended to be highest among practices in the middle and second least deprived quintiles. Of course, even in those cases where there did appear to be a trend in percentages across deprivation quintiles, the trend may occur by chance. Were we use statistical tests to assess these trends, then we would expect to find one statistically significant trend purely by chance in this table, even if no underlying relationship exists between on-going care and deprivation for any of the indicators. 36 Quintiles derived for Hull practices by ranking the mean IMD 2007 score for each of the 56 practices (based on patient postcodes). Robert Sheikh Iddenden, Public Health Sciences. 207

208 Table 99: COPD QOF exceptions and indicator percentages by deprivation quintile of East Riding practices (based on mean IMD2007 score of practice), local (East Riding) quintiles defined at practice level 37 Weighted average over practices of QOF 2008/09 exceptions and indicators by mean local deprivation Indicator Measure quintiles (IMD2007) defined at practice level Most Least deprived deprived COPD01: On register % COPD08: flu jab % excl % indict r COPD10: FeV1 % excl % indict r COPD11: Inhaler % excl % indict r COPD12: Spirometry % excl % indict r Other Groups Where Potential Inequity May Be Present There are a number of other groups where potential inequity may be present in relation to health and access to services for COPD. Most of the groups where an inequity in regards to COPD exists will also experience inequity for a number of health issues. These groups may include gypsies and travellers, prisoners, and people with learning disabilities, mental health issues, vision and hearing problems, etc. It is very rare, however, to have any information in relation to prevalence, hospital admissions, mortality, risk factors, access to services and quality of care for these subgroups. Therefore, in general, issues specific to these groups of people living in Hull and East Riding cannot be considered. In some cases, it may be possible to obtain relevant information from research, surveys and audits completed in other geographical areas of the country. However, these results would not necessarily be transferable to Hull or East Riding of Yorkshire as the populations or people surveyed may differ from the local population in terms of deprivation, age structure, rurality, etc. If there is a suspicion or anecdotal evidence that there could be inequity for a particular group with regard to COPD diagnosis, treatment or access to services, then this would need to be investigated locally for that specific group. 37 Quintiles derived separately for East Riding practices by ranking the mean IMD 2007 score for each of the 38 practices (based on patient postcodes). Robert Sheikh Iddenden, Public Health Sciences. 208

209 People with Learning or Physical Disabilities, Mental Health Problems, or Literacy or Numeracy Problems It is likely that there will be some patients with COPD who have other conditions that will result in problems accessing services due to other disabilities such as learning disabilities, physical disabilities, mental health problems, etc. Furthermore, some of these groups may have literacy or numeracy problems which could affect their ability to understand and correctly use prescribed medications. This is likely to affect more people in Hull with its increased deprivation and subsequent higher percentages of people with no qualifications and lower literacy and numeracy skills. Smoking rates among patients with mental health problems may be higher than in the general population, therefore the prevalence of COPD, and the complications arising from COPD, are likely to be higher among this group of patients. Atypical anti-psychotics such as amisulpride, aripiprazole, clozapine, alanzapine, quetiapine, risperidone and zotepine, include the side-effects of weight gain, which can worsen the symptoms of COPD ((British Medical Association and Royal Pharmaceutical Society of Great Britain, 2007); BNF, page 192). People with mental health problems will take up proportionately more resource as they need lot more support. On-going care is also more difficult for these patients, and there are no specific measures in place for patients with mental health problems who also have COPD. Generally people with learning disabilities will have some form of help in place. This will vary with the specific physical and mental health problems Gypsies and Travellers Gypsies and travellers will also fall within the previous category in relation to literacy skill and physical health problems, but also have problems of isolation and access to service. A small local survey of gypsies and travellers was conducted during 2007 in Hull (for further information, see section on Local Surveys starting on page 275). Whilst the numbers surveyed were small only 100 survey respondents and it is not known if they are representative of Hull s Gypsy and Traveller population, the survey does provide invaluable information on this hard-to-reach group. In the survey, 78% were sited, 20% were housed and 2% were highly mobile or transient Gypsies and Travellers, and 60% of the survey responders were female. Anecdotally, Gypsies and Travellers tend to have a high prevalence of some risk factors for poor health such as smoking, poor diet and obesity. Furthermore, gypsies and travellers tend to have problems accessing services due to stigma and possibly literacy problems. Results from the main Health and Lifestyle survey conducted in Hull during 2007 involving 4,086 survey responders who were representative of Hull s age, gender, geographical and employment structure can be compared with the survey involving the gypsies and travellers. The following results relevant to COPD were found in the local surveys: Robert Sheikh Iddenden, Public Health Sciences. 209

210 68% of gypsies and travellers stated that they currently smoked and a further 32% reported that they used to smoke, compared to 32% of main survey respondents who currently smoked and 26% who used to smoke; the prevalence of obesity was twice as high in the small number of gypsies and travellers compared to the main survey responders, with 27% and 17% of gypsies and travellers being obese (BMI 30-39) and morbidly obese (BMI 40+) respectively compared to 18% and 3% for main survey responders respectively; 99% of gypsy and travellers in the survey were registered with a Hull GP. Therefore, with more than twice the prevalence of smoking than the general Hull population, and with all Gypsy and Traveller respondents being either a current or former smoker, Gypsies and Travellers are at a much higher risk of developing COPD than the general population. And given that obesity can make symptoms more severe, Gypsies and Travellers are likely to have more severe symptoms of COPD that then general Hull population. Given the lower literacy levels of Gypsies and Travellers, there may be issues in terms of understanding how and when to use treatments to alleviate COPD symptoms. Whilst clear verbal instructions can be provided, it is clearly more difficult to provide information for future reference where details have to be retained using memory Housebound Patients Housebound patients fall to a certain extent under the previous category having physical disabilities, but also have problems of isolation and access to service. These people would have prescriptions delivered, but no access to in-pharmacy services. Some patients will be visited at home, but this might not always be the case, and it is possible that inequity of care could occur. Also, it may not be possible to use spirometry to diagnose these patients, while once diagnosed they may not have regular check-ups to ensure the correct ongoing management of their COPD Patients Whose First Language is not English Patients whose first language is not English may often like to give the impression they have understood even when they have not. They will often bring a younger family member with them to translate. It is possible that they do not access clinics proportionately, but they may see practice nurses or use their own health practitioners and medicines within their own communities. Translation services are available Patients with Memory Impairment Patient with memory impairment could have problems taking their medications regularly. For these patients large print labels and reminder charts are available. Where regular Robert Sheikh Iddenden, Public Health Sciences. 210

211 medication is taken, monitored dosage systems can also be provided where tablets are put into compartments with days of the week and time of the day to aid compliance. These are standard aids and are offered across all pharmacies. Furthermore, there is a service where patients can be visited in their homes by a pharmacist to assess their medication needs. This service is also available prior to discharge from hospital. However, this is currently available in Hull and not East Riding of Yorkshire Prisoners In general, prisoners will tend to be younger so be less likely to have COPD than the population in general, although the prevalence of smoking among this group is likely to be higher than the general population, and therefore their risk of developing COPD will also be higher. There are four prisons in Hull and East Riding of Yorkshire all holding male prisoners only. Details are given below on the type of prison and the operational capacity 38 from the Her Majesty s Prison Service website 39. Everthorpe is a category C training prison which opened in 1958 as a borstal. It was converted to its present role in 1991 and now holds convicted male prisoners. In 2005 Everthorpe underwent a significant expansion programme, and now has operational capacity for 689 prisoners. Wolds was opened in 1992 as a remand prison and in 1993 was changed to a local category B prison holding sentenced prisoners, but Wolds is now for mid-term category C male prisoners including second stage lifers. There is an operational capacity of 395 prisoners. It is a privately-run prison and as such has not gone through the transfer of healthcare services from the Home Office to the Department of Health. Hull is a Victorian Prison opened in It became a maximum security dispersal prison n 1969 after extensive security work. In 1986 it became a local prison / remand centre, with four new wings added in It has an operational capacity of 1,044 and holds remand, sentenced and convicted adult males (except Category A) and young offenders. Full Sutton in East Riding of Yorkshire is a modern, purpose-build, maximum security prison for men in category A and category B with an operational capacity of 38 Operational capacity (OC) is the maximum safe capacity of the prison population which is higher than the certified normal accommodation (CNA) It is not known if the capacity figures and other information from the website are entirely up-to-date. Robert Sheikh Iddenden, Public Health Sciences. 211

212 608 prisoners. The prison's primary function is to hold, in conditions of high security, some of the most difficult and dangerous criminals in the country. While it is not possible to accurately estimate the likely numbers of prisoners with COPD, we can make a crude estimated range by applying registered QOF prevalence rates and expected (modelled) prevalence rates from the ERPHO model, under the assumption that the prevalence of COPD in the prisons will be similar to that in the general Hull and East Riding population. Of course, as mentioned earlier, while smoking rates are higher among prisoners, their mean age is younger, so this will reduce current levels of COPD. However, as the registered and modelled prevalence rates are both based on total populations (including children who will not have COPD and women, who have a lower prevalence than men, these prevalence rates will underestimate the prevalence in adult males. These two factors are likely to counterbalance each other to some degree. Therefore, under the assumption that the prevalence rates in the general population of Hull and East Riding are applicable to the prison populations there would be between 12 and 23 in Everthorpe, between 7 and 13 in Wolds, between 19 and 35 in Hull, between 11 and 21 in Full Sutton, with the lower values based on registered QOF prevalence and the higher values based on the ERPHO model. Robert Sheikh Iddenden, Public Health Sciences. 212

213 10. Potential Programmes for Reducing Inequalities in COPD Reducing the Risk of Developing COPD Smoking Cessation Smoking cessation statistics by financial year are published each year by the NHS Information Centre for Health and Social Care 40. PCT level summaries are provided, together with breakdowns at PCT level by broad age band and by gender, although not age band within gender. Data are also produced by ethnicity, but not at a PCT level, so are not presented here. Data that are published at a PCT level includes the number and rate per 100,000 setting quit dates, the number and rate per 100,000 of successful quitters (4-week quits), where A client is counted as a self-reported 4-week quitter if when assessed 4 weeks after the designated quit date, they declare that they have not smoked, even a single puff on a cigarette, in the past two weeks. - Statistics on NHS Stop Smoking Services England, April 2009 March 2010 (Information Centre, 2010) Also published by PCT, were the numbers lost to follow-up and the numbers of 4-week quits that were confirmed by CO validation. Because the Hull and East Riding Smoking Cessation service covers both PCTs, the numbers the numbers published by the Information Centre have been combined, and rates produced using the 2008 mid-year resident population estimates from the Office for National Statistics (currently the mid-year estimates for 2009 were not yet available). Table 100 contains the smoking cessation statistics for broken down by gender, for Hull, East Riding, the Hull and East Riding total, plus regional and national comparisons. Despite having a smaller population Hull had almost twice as many people setting quit dates, and successfully quitting than East Riding, although this is to be expected given that the level of need in Hull is far higher than in East Riding, with the prevalence of smoking in Hull more than double that in East Riding (see Table 34 on page 99). In fact, the rate of quit dates set was two and a half times higher for Hull than East Riding, with the difference greater among women. Rates of quit dates set among Hull residents were higher in women (17% higher than men) while among East Riding residents rates were slightly higher among men (7% higher than women). The differences in 4-week quits between Hull and East Riding were smaller, with rates in Hull just over double those in East Riding, again with a larger difference among women. The 40 Data may be downloaded from Robert Sheikh Iddenden, Public Health Sciences. 213

214 percentages of quit dates set converted to 4-week quits were higher in East Riding than Hull, at least among women. Looking next at the combined Hull and East Riding data, the rates both of quit dates set and 4-week quits achieved were higher than the regional or national averages. The Hull and East Riding smoking cessation service was more successful at converting quits dates set into 4-week quits, with 63% of clients that initially set a quit date successfully quitting by the 4-week follow-up, compared with 53% in the Yorkshire and Humber region and 49% across England as a whole. The Hull and East Riding smoking cessation service was also much more successful at following up clients at 4 weeks, with only 10% lost to follow-up at 4-weeks, compared with 25% across the Yorkshire and Humber region as well as across England as a whole. However the percentages of successful quits that were confirmed by CO validation were lower (at 53%) than either the regional (74%) or national (69%) averages. A similar breakdown of smoking cessation statistics for by broad age bands is presented in Table 101. Rates of quite dates set were higher in Hull than East Riding for each age band, with the differences increasing with age, with rates in those aged 16-4 years 85% higher in Hull while in the aged 60+ years rates in Hull were three times higher than rates in East Riding. Differences between Hull and East Riding were similar for the rates of 4-week quits per 100,000, although slightly lower. The increasing differences with age probably reflects the widening differences in smoking prevalence by age between the two PCTs, with the level of need in Hull relative to East Riding increasing with age. Looking now at combined Hull and East Riding data, to reflect the common smoking cessation service that operates across the two PCTs, the rates of quit dates set and successful quits achieved were greater for Hull and East Riding than the average rates across the Yorkshire and Humber region, the rate of quit dates set around 6% higher in Hull and East Riding, the rate of 4-week quits around 15% higher (and 24% higher in those aged years). Rates of quit dates set in Hull and East Riding were around 5% higher than for England overall in those aged under 45 years, while rates of 4-week quits were higher than for England for each age band (by around 25%). The greater success experienced by Hull and East Riding smoking cessation service in converting quit dates set to 4-week quits was seen for each age band, with differences greatest in those aged years (24% higher than Yorkshire and Humber, 30% higher than England). The percentage of successful quits that were confirmed by CO validation were similar for most age bands in Hull and East Riding (at around 55%) although lower amongst those aged years (47%). This was substantially lower than the 75% seen at most age bands across the Yorkshire and Humber region (again lowest in the youngest group at 71%) and lower than the 70% seen for most ages across the whole of England percentages (again lowest in the youngest group at 64%). Robert Sheikh Iddenden, Public Health Sciences. 214

215 Table 100: Smoking cessation statistics by gender Smoking cessation statistics 2008/09 by gender for Hull, East Riding, Yorkshire and Humber, and England Area and age band (years) Numbers Rates per 100,000 Percentages Setting a quit date 4-week quits 1 Lost to follow up / not known Confirmed 4-week quits 2 Quit dates set 4-week quits 1 Successful quits 3 Confirmed successful quits 4 Hull Males 2,606 1, ,464 1, Females 3,031 1, ,027 2,884 1, All 5,637 3, ,856 2,673 1, East Riding Males 1, , Females 1, , All 3,022 1, ,046 1, Hull and East Riding Males 4,135 2, ,357 1,714 1, Females 4,524 2, ,545 1,830 1, All 8,659 5, ,902 1,773 1, Yorkshire and Humber Males 35,209 19,012 8,520 13,961 1, Females 39,908 20,582 9,886 15,256 1, All 75,117 39,594 18,406 29,217 1, England Males 363, ,066 89, ,438 1, Females 393, ,888 95, ,275 1, All 757, , , ,713 1, Self reported successful quits at the 4-week follow-up (client reported he/she not smoked at all since two weeks after the quit date) 2 Self reported successful quits confirmed by CO validation 3 Percentage of all 4-week quits out of all those setting a quit date 4 Percentage of 4-week quits that were confirmed by CO validation Robert Sheikh Iddenden, Public Health Sciences. 215

216 Table 101: Smoking cessation statistics by age band Smoking cessation statistics 2008/09 by age band for Hull, East Riding, Yorkshire and Humber, and England Area and age band (years) Numbers Rates per 100,000 Percentages Setting a quit date 4-week quits 1 Lost to follow up / not known Confirmed 4-week quits 2 Quit dates set 4-week quits 1 Successful quits 3 Confirmed successful quits 4 Hull ,863 1, ,356 1, , ,755 2, , ,289 2, ,800 1, East Riding , , , Hull and East Riding ,766 1, ,920 1, ,089 1, ,481 1, ,359 1, ,995 1, ,445 1, , Yorkshire and Humber ,245 12,221 8,523 8,664 2, ,613 9,596 4,230 7,092 2,348 1, ,238 10,910 4,059 8,230 1,931 1, ,021 6,867 1,594 5, England , ,286 78,335 71,682 2, ,578 93,943 44,700 65,272 2,392 1, , ,814 43,036 72,508 2,012 1, ,409 65,911 18,809 47,251 1, Self reported successful quits at the 4-week follow-up (client reported he/she not smoked at all since two weeks after the quit date) 2 Self reported successful quits confirmed by CO validation 3 Percentage of all 4-week quits out of all those setting a quit date 4 Percentage of 4-week quits that were confirmed by CO validation Robert Sheikh Iddenden, Public Health Sciences. 216

217 Smoking cessation services are offered at various locations across Hull and East Riding, at GP practices, health centres, pharmacies and community settings. Figure 46 and Table 102 show the locations of smoking cessation services across Hull. Figure 46: Locations of smoking cessation services in Hull Green star=primary care; red star=pharmacy; blue star=other setting. Robert Sheikh Iddenden, Public Health Sciences. 217