The Annual Surveillance of Healthcare Associated Infection Report January - December 2010

Transcription

1 The Annual Surveillance of Healthcare Associated Infection Report January - December 2010

2 Health Protection Scotland is a division of NHS National Services Scotland. Health Protection Scotland website: Citation for this document Health Protection Scotland, The Annual Surveillance of Healthcare Associated Infection Report January December Health Protection Scotland, Glasgow, Published by Health Protection Scotland, Clifton House, Clifton Place, Glasgow G3 7LN. First published May 2011 Comment added to this report in September 2013 Health Protection Scotland 2011 Health Protection Scotland has made every effort to trace holders of copyright in original material and to seek permission for its use in this document. Should copyrighted material have been inadvertently used without appropriate attribution or permission, the copyright holders are asked to contact Health Protection Scotland so that suitable acknowledgement can be made at the first opportunity. Health Protection Scotland consents to the photocopying of this document for professional use. All other proposals for reproduction of large extracts should be addressed to: Health Protection Scotland Clifton House Clifton Place Glasgow G3 7LN Tel: +44 (0) NSS.HPSEnquiries@nhs.net Front Cover photos: Clostridium difficile and Meticillin resistant Staphylococcus aureus (MRSA) Annie Cavanagh/Wellcome Images Hospital corridor and operating theatre: istockphoto

3 Table of Contents List of Tables 2 List of Figures 2 Glossary 4 Introduction 5 Key Summary Points 7 Staphylococcus aureus Infection 8 Introduction 8 Methods 9 Results 9 S. aureus bacteraemia cases and incidence 9 Identification of outliers 10 Typing and susceptibility data 12 Discussion 13 Reducing S. aureus bacteraemias in Scotland 14 Clostridium difficile Infection 15 Introduction 15 Methods 16 Results 16 CDI cases and incidence in patients aged 65 and over 16 CDI cases and incidence in patients aged Age distribution of all CDI cases in Scotland aged 15 and over 19 Impact of new testing algorithm on reported rates 20 Identification of outliers 20 Typing and susceptibility data 21 Discussion 22 Reducing CDI in Scotland 24 Surgical Site Infection 25 Introduction 25 Methods 25 Results 26 Analyses presented in this report 26 Participation in surveillance between January - December Incidence of SSI by risk group 30 Characteristics of the surgical site infection 31 Variation in SSI Rate by NHS board 32 Discussion 33 References 35 1

4 List of Tables Table 1: Quarterly numbers and rates of S. aureus bacteraemias (MRSA and MSSA) in Scotland. 10 Table 2.1: Total number of CDI cases in the age group 65 and over by NHS board for Table 2.2: Incidence rates of CDI in the age group 65 and over by NHS board for Table 2.3: Comparison of annual rates of CDI in the age group 65 and over by NHS board. 17 Table 2.4: Total number of CDI cases in the age group by NHS board for Table 2.5: Incidence rates of CDI in the age group by NHS board for Table 2.6: Impact of new testing on Scottish overall rates in patients aged 65 or over. 20 Table 3: SSI rate with 95% confidence interval by hip arthroplasty procedure January to December Table 4: SSI rate with 95% confidence interval by caesarean section procedure for inpatient and PDS to day 10 January to December Table 5: Cumulative incidence of inpatient SSI, by category of surgical procedure by NNIS risk index for 2007 to 2009 and List of Figures Figure 1: Quarterly S. aureus bacteraemia rates per 1000 total AOBDs for Scotland with trend line, April December Figure 2: Funnel Plots of S. aureus bacteraemia rates per 1000 AOBDs by NHS board (showing all S. aureus bacteraemias and MRSA and MSSA bacteraemias). 11 Figure 3: Number of MRSA isolates (n=690) by specimen site. 12 Figure 4: Number of MRSA isolates (n=690) by strain type. 12 Figure 5: MRSA antibiotic resistance (%) with 95% confidence intervals. 13 Figure 6: Overall quarterly rates of CDI per 1000 total OCBDs in patients aged 65 and over for seventeen quarters of the mandatory surveillance covering the period October 2006 to December Figure 7: Annual overall rates of CDI per 1000 total OCBDs between 2009 and 2010 in patients aged 65 and over in acute and non-acute hospitals in 15 NHS boards in Scotland. 18 Figure 8: Overall quarterly rates of CDI per 1000 AOBDs in patients aged for seven quarters of the mandatory surveillance covering the period April 2009 to December Figure 9: Age distribution of all CDI cases aged 15 years or over in Scotland in Figure 10: Funnel plot of rates of CDI for all NHS boards in Scotland in patients aged 65 and over against total occupied bed days (x1000) for the year Figure 11: Funnel plot of rates of CDI for all NHS boards in Scotland in patients aged against acute occupied bed days (x1000) for the year Figure 12: Breakdown of ribotypes (n=572) in Scotland obtained from severe cases and outbreaks of CDI by quarter during Figure 13: Breakdown of ribotypes (n=559) in Scotland obtained in the snapshot ribotyping programme by quarter during Figure 14: SSI rate for both inpatient and readmission to day 30 by hip arthroplasty procedure January to December Figure 15: SSI rate for both inpatient and PDS to day 10 by caesarean section procedure January to December Figure 16: Rate of inpatient SSI per 100 procedures for caesarean section and hip arthroplasty procedures per quarter, 2007 to Figure 17: Incidence density of inpatient SSI per 1000 post operative inpatient days, by category of procedure and year. 29 Figure 18: Rate of inpatient and PDS to day 10 SSI per 100 procedures for caesarean section and inpatient and readmission to day 30 SSI for hip arthroplasty procedures per quarter, 2007 to Figure 19: Proportion of SSI involving superficial incisions or deep and organ space, January 2007 to December 2010, for hip arthroplasty (inpatient and readmission to day 30) and caesarean section (inpatient and PDS to day 10). 31 Figure 20: Cumulative incidence (number of SSI per 100 procedures) of all caesarean section procedures inpatient and PDS until day 10 post operatively, SSI (NNIS = 0) by NHS board in Figure 21: Cumulative incidence (number of SSI per 100 procedures) for hip arthroplasty inpatient and readmission SSI until day 30 post operatively, SSI (NNIS = 0) by NHS board in

5 The Annual Surveillance of Healthcare Associated Infection Report January - December 2010 Health Protection Scotland The Annual Surveillance of Healthcare Associated Infection Report January to December 2010 was prepared by members of the Antimicrobial Resistance (AMR) and Scottish Surveillance of HAI Programme (SSHAIP) teams within Health Protection Scotland (HPS) with contributions from colleagues across the Healthcare Associated Infection and Infection Control group and Statistics team within HPS. The national surveillance programmes are dependent on the continuing contribution of information, infection reports and micro-organism data from Infection Control, Microbiology, Surveillance Co-ordinators and other clinical colleagues throughout the NHS in Scotland. The National Reference Laboratories also contribute greatly to the programme. This makes our surveillance programme in Scotland fully comprehensive and world class and these continuing contributions are greatly appreciated. Professor Jacqui Reilly Head of Group Healthcare Associated Infection Group Health Protection Scotland May

6 Glossary AMR AOBDs ASA CA-MRSA CEL CDC CDI CVC EARS-Net ECDC EMRSA HAI HAI and IC HALT HDL HEAT HIS HPS ICT ICU ISD MRSA MSSA NES NHS NHSQIS NNIS NSS NWTC OCBDs PCR PDS PVC PVL SAPG SEHD SGHD SIGN SMF SMRSARL SPCs SPSP SSHAIP SSI SSSCDRL VAP Antimicrobial Resistance Acute Occupied Bed Days American Society of Anaesthesiologists Community Associated Meticillin Resistant Staphylococcus aureus Chief Executive Letter Centers for Disease Control Clostridium difficile Infection Central Venous Catheter European Antimicrobial Resistance Surveillance Network European Centres for Disease Prevention and Control Epidemic Meticillin Resistant Staphylococcus aureus Healthcare Associated Infection Healthcare Associated Infection and Infection Control Healthcare Associated Infections in European Long Term Care Facilities Health Department Letter Health Improvement, Efficiency, Access and Treatment Healthcare Improvement Scotland Health Protection Scotland Infection Control Team Intensive Care Unit Information Services Division Meticillin Resistant Staphylococcus aureus Meticillin Sensitive Staphylococcus aureus National Education for Scotland National Health Service NHS Quality Improvement Scotland National Nosocomial Infection Surveillance National Services Scotland Golden Jubilee National Hospital NHS National Waiting Time Centre Board Occupied Bed Days Polymerase Chain Reaction Post Discharge Surveillance Peripheral Vascular Catheters Panton-Valentine Leukocidin Scottish Antimicrobial Prescribing Group Scottish Executive Health Department Scottish Government Health Department Scottish Intercollegiate Guidelines Network Scottish Microbiology Forum Scottish MRSA Reference Laboratory Statistical Process Control Charts Scottish Patient Safety Programme Scottish Surveillance of HAI Programme Surgical Site Infection Scottish Salmonella, Shigella and Clostridium difficile Reference Laboratory Ventilator Associated Pneumonia 4

7 Please note CDI rates in this report have been updated due to a subsequent revision of the national figures of hospital activity. Please go to for the newest revised data. Introduction This is the second annual report on the surveillance of Healthcare Associated Infection (HAI) in Scotland. It has been written as an annual stock take of the epidemiology of HAI in Scotland to supplement the routine quarterly reports produced by Health Protection Scotland (HPS). The aim of this report is to identify key information on HAI in Scotland, demonstrating the burden of infection, trends in infection incidence, share supplementary epidemiological and microbiological data from the national reference laboratories, and identify future priorities. The last national prevalence survey in Scotland indicated that around one in ten patients in acute care and one in thirteen in non acute care were affected by HAI. This burden of infection was estimated to cost acute care in the NHS in Scotland 183 million pounds each year. 1 The most prevalent organisms causing these infections were Staphylococcus aureus (S. aureus) and Clostridium difficile (C. difficile). It therefore remains important that these organisms of concern are monitored continuously. There are however, other organisms of HAI concern which are an emerging public health threat in Scotland. These have been reported by HPS for the second time this year 2 and annual reporting of these antimicrobial resistance (AMR) data will continue in supplement to this annual report. S. aureus bacteraemia data have been monitored in Scotland since 2001 and there have been substantial reductions in these infections since this time. Meticillin resistant S. aureus (MRSA) bacteraemias have reduced significantly over the last 5 years in NHS Scotland. The incidence of meticillin sensitive S. aureus (MSSA) bacteraemias has also reduced, although not at the same rate as MRSA. Multiple infection prevention and control interventions have been implemented during this time. These include national policy initiatives from the Scottish Government Health Department (SGHD) HAI task force such as a hand hygiene campaign, model policies for infection control, and targeted interventions such as infection prevention and control care bundles, which were implemented as part of the Scottish Patient Safety Programme (SPSP). Whilst inroads in reducing these infections have been made, there were nevertheless 1843 S. aureus bacteraemias in Scotland in 2010 and as these bacteraemias represent around 10% of all infections due to the organism 3, a significant burden of these infections in healthcare remains. Importantly the proportion of S. aureus bacteraemias which were MRSA has significantly reduced in the last year from 24% to 19%.This compares favourably with other European countries with a similar history of high endemic proportions of this organism. An emerging issue which warrants better epidemiology is MSSA bacteraemia, in order that interventions can be targeted to reduce the burden of infection. The emerging threat of Panton-Valentine Leukocidin (PVL) is also another area of concern and the report highlights the importance of ensuring we have good surveillance data in Scotland. C. difficile infection (CDI) surveillance commenced in 2005, with mandatory surveillance being established in 2006 to monitor CDI in patients aged 65 and above, and changes to reporting to include all cases aged 15 and above two years ago. The numbers of these infections have significantly reduced over the last 2 years. Over the last year a 38% reduction in rates has been observed and the intelligence from the ribotyping, carried out by the national reference laboratory, indicates that the prevalence of the epidemic types (106, 001 and 027) has decreased. This points to a potential temporal association with the implementation of HPS guidance for the management of CDI and the Scottish Antimicrobial Prescribing Group (SAPG) guidance on prudent antimicrobial prescribing. However, more time is required to assess whether this is sustainable in the longer term and this is important as 2911 cases of CDI occurred in Rates for surgical site infection (SSI) under mandatory national surveillance are low by comparison. Rates of infection in the hip arthroplasty and caesarean section surgery categories have significantly reduced since surveillance became mandatory in However reducing length of stay is of concern in this regard as it may be argued that this observed reduction is an artefact, if only inpatient rates are monitored. Scotland is leading the way internationally with post discharge surveillance and was the first country in the world to introduce mandatory surveillance of patients after discharge from hospital. Almost half of all SSI following hip arthroplasty surgery were detected on readmission to hospital in 2010 and the proportion of caesarean section SSI detected by post discharge surveillance to day 10 was 88%. In total, the surveillance system detected 533 cases of SSI following hip arthroplasty surgery and caesarean section surgery in The incidence density of SSI compares favourably with other European data provided by ECDC 4 ; however there is not room for complacency. Implementation of SPSP care bundles for SSI and the SAPG 5

8 Please note CDI rates in this report have been updated due to a subsequent revision of the national figures of hospital activity. Please go to for the newest revised data. work on surgical antibiotic prophylaxis over the next year should continue to contribute to reducing these clinically significant infections. Norovirus is a commonly detected pathogen in healthcare settings and HPS continued to publish data on ward closures to support winter planning in the last year. These data indicate that norovirus can have a major impact in hospitals. Unlike other HAI, norovirus infection, although uncomfortable for those affected, is usually short-lived and not serious. However, it is highly transmissible and outbreaks in hospitals can cause severe and costly disruption to health services. The season officially started this year on 23rd November This season was much milder than 2009/10 where there was a peak of 53 wards closed at any one time. This year the peak was 26 wards closed. This marked reduction was possibly due to a combination of factors including changes in the virus and the weather which prevented people mixing during the peak season. However, it should be noted that NHS Boards increased their preparedness this year and the extent to which this also impacted on the outcome is the subject of further evaluation by HPS and the NHS boards, in order to learn lessons and ensure that the impact of norovirus on the service is at an irreducible low. International comparison of HAI rates is fraught with challenges as the definitions for infection, and approaches to and methods for data collection vary between countries. Differences in ascertainment, reporting compliance, approaches to infection control and the socio-economic status of individual countries all impact on these data. Harmonisation of data collection methods, surveillance criteria and definitions in SSI and Intensive Care Unit (ICU) surveillance has now been developed by European Centre for Disease Prevention and Control (ECDC). Scotland has contributed data to the European Antimicrobial Resistance Surveillance Network (EARS-Net) and SSI systems for several years, and for the first time this year will contribute to the ICU dataset. HAI surveillance in ICUs (Ventilator Associated Pneumonia and Central Venous Catheter bacteraemia) is now carried out in all intensive care units in Scotland. The first report from this national surveillance system was published earlier this year. 5 Data from 2900 patients admitted to 19 Scottish ICUs were collected in the first year and a total of 235 infections were reported from 189 (6.5%) patients. Of the 235 infections reported, just over half of these were pneumonias and the remainder central venous catheter bacteraemias. As these data accumulate and become comprehensive they will be important measures for benchmarking with Europe. For the first time this year Scotland extended surveillance to care home settings with the first prevalence survey in over 100 care homes using the ECDC Healthcare Associated Infections in European Long Term Care Facilities (HALT) protocol. These data are due to be published later this year and will be an important step in understanding the burden of these infections in care home settings. In response to each of the public health challenges presented by the above noted HAI, there is good infection prevention and control practice and innovation in Scotland. HPS has made major contributions to identifying these priorities for action and prepared proposals on how to develop and implement these initiatives, which have been embedded in national strategy and the HAI task force policy agenda. To do this, we need reliable and independent public health evidence on the impact of prevention measures. Having high quality, comparable disease surveillance data is fundamental in this quest. That is what we seek to provide in this report. This report details work on the mandatory surveillance programmes: S. aureus, C. difficile and surgical site infection, in the last year. 6

9 Please note CDI rates in this report have been updated due to a subsequent revision of the national figures of hospital activity. Please go to for the newest revised data. Key Summary Points This annual report on the epidemiology of HAI in NHS Scotland contains the outputs from the HPS run NHS Scotland mandatory surveillance programmes. It details the numbers of cases in the last year and trends in infection rates for: Staphylococcus aureus bacteraemia, Clostridium difficile infection and surgical site infection. In the past year, there were 1843 Staphylococcus aureus bacteraemias in NHSScotland. The majority (1492, 81%) of these were due to MSSA with around one fifth (19%) due to MRSA (n=351). A significant year on year reduction of 6.7% per annum has been observed since MRSA bacteraemias have reduced more rapidly than those caused by MSSA (17.7% vs. 1.5% per annum). The majority of MRSA isolates (82%) typed by the Scottish MRSA reference laboratory, as part of the snapshot programme in the last year, were attributable to the epidemic strain EMRSA-15. CDI surveillance in NHS Scotland identified 2219 cases in the last year in those aged 65 years and over. There was a significant decrease in the incidence rates of CDI (38%) in the last year. Data from the Scottish Salmonella, Shigella, and Clostridium difficile Reference Laboratory (SSSCDRL) in Scotland indicate that in 2010 the epidemic strains (ribotypes 106, 001 and 027) accounted for 36% of all ribotypes compared to 54% in 2009 suggesting that current infection control and to some extent prescribing measures are having greatest impact on these three. Other measures may be necessary to reduce the risk of infection with the non-epidemic ribotypes. SSI surveillance in two categories of surgery (hip arthroplasty and caesarean section) identified 533 infections in the last year (104 and 429 respectively). These data from Scotland include information on infections happening after discharge from hospitals, which is an increasingly important aspect of monitoring, as length of stay is reducing in hospital over time. The introduction of readmission surveillance until day 30 post operatively for hip arthroplasty has resulted in a higher proportion of SSIs being detected following discharge from hospital (45.2%). The proportion of caesarean section SSI detected by post discharge surveillance to day 10, not including in patient infections, was 87.9%. Overall rates of infection in both these categories of surgery have significantly reduced since 2003 but have remained relatively stable over the last two years of reporting, In summary this report demonstrates that real inroads have been made to reducing HAI in NHS Scotland, however there is still a substantial burden of these infections occurring every year in healthcare. Efforts should be continue to focus on preventing and controlling these infections to the irreducible minimum. These surveillance data form part of the HPS health protection programmes, which are focussed on supporting NHS boards to reduce these HAI. HPS will continue to work with the NHS boards, and other key stakeholders, to deliver these programmes in order to minimise the risk of infection in healthcare in Scotland. 7

10 Staphylococcus aureus Infection Introduction S. aureus is a gram positive bacterium which colonises the nasal cavity of about 30% of the healthy population. Although this colonisation is usually harmless, S. aureus may cause serious infection. These infections are commonly associated with healthcare interventions which allow the bacterium to infect normally sterile body sites. Thus interventions which are beneficial to health may expose patients to risks of infection. Both meticillin sensitive and meticillin resistant S. aureus (MSSA and MRSA) remain endemic in many UK hospitals, causing a range of infections. Amongst the most serious of these infections are S. aureus bacteraemias. The mandatory Scottish national meticillin resistant S. aureus (MRSA) bacteraemia surveillance programme was established by the Scottish Executive Health Department (SEHD) in HPS has published quarterly reports on the numbers and rates of MRSA bacteraemias in NHS settings since April 2002, based on data from this surveillance programme. In January 2005, data collection in the MRSA bacteraemia surveillance programme was enhanced by including reports from the Scottish MRSA Reference Laboratory (SMRSARL). These bacteraemias recorded by the SMRSARL are also reported through the European Antimicrobial Resistance Surveillance Network (EARS-Net). In July 2006, the surveillance programme was extended by the SEHD to include all S. aureus bacteraemias in Scotland, including meticillin sensitive S. aureus (MSSA) bacteraemias. 7 While the Scottish surveillance programme includes data on both MRSA and MSSA bacteraemias, it should be noted that most other countries restrict surveillance of S. aureus bacteraemias to those caused only by MRSA. The HPS S. aureus bacteraemia surveillance programme monitors the occurrence of S. aureus bacteraemias in Scotland. It includes S. aureus bacteraemias occurring in patients who have been in contact with the healthcare system (in acute and non-acute hospitals, as well as in primary care settings) and those who have had acquired S. aureus bacteraemias in the community, without any healthcare contacts. In October 2008, a voluntary revised programme for the collection of MRSA representative typing data was introduced (Snapshot Programme).This programme supports the mandatory S. aureus bacteraemia programme by providing additional epidemiological information on strain typing, toxin production and antimicrobial resistance. This is the second annual report on S. aureus bacteraemias in Scotland. It is mainly concerned with the incidence of S. aureus bacteraemias in Scotland and within individual NHS boards. The Scottish Government Health Department (SGHD) has established targets for individual NHS boards for the reduction of S. aureus bacteraemias. This report does not contain analyses and projections of progress towards the achievement of these Health Improvement, Efficiency, Access and Treatment (HEAT) targets as this is being undertaken with the Scottish Government s Health Directorates. This second annual report summarises the trends in all S. aureus bacteraemias from April It also describes some of the activities undertaken in HPS in support of NHS boards towards reduction in rates of S. aureus bacteraemias in Scotland. Data on the antimicrobial susceptibility, strain distribution and toxin production of S. aureus isolates are also reported in this Annual Report. 8

11 Methods In Scotland, all 15 NHS boards report every isolate of MSSA and MRSA from blood cultures to HPS in compliance with the HDL (2006)38 7 according to the standard protocol. 8 Isolates for the representative typing scheme (Snapshot Programme) are submitted to the SMRSARL according to a separate protocol. 9 These reports are generated by routine hospital laboratory management systems. The NHS boards laboratories submit these S. aureus isolates to the SMRSARL for typing and antimicrobial susceptibility testing. The data reported here are based on information from NHS boards laboratories and from the SMRSARL. Cases are allocated to NHS boards based on the location of the diagnostic laboratory where the specimen was first tested. Data on NHS boards acute occupied bed days (AOBDs) are provided by the Information Services Division (ISD) of NHS National Services Scotland (NSS). These AOBDs include some estimated bed occupancy data, due to incomplete data availability at the time of this publication. Details of the surveillance case definitions used in the programme are contained in the Surveillance Programme protocol. 8 The analyses used are described in the quarterly reports. 10 All S. aureus bacteraemia reports produced by HPS are based on interim data for both bed occupancy and incident S. aureus bacteraemias. Results S. aureus bacteraemia cases and incidence A total of 1843 S. aureus bacteraemias were reported to HPS in 2010, 351 MRSA bacteraemias and 1492 MSSA. Table 1 (overleaf) shows the quarterly numbers of S. aureus (both MRSA and MSSA) bacteraemias in Scotland since April This shows that the overall numbers of S. aureus bacteraemias and the rates of these bacteraemias (adjusted for activity levels measured by bed occupancy) have reduced. Figure 1 shows the quarterly trend in S. aureus bacteraemia rates from April 2005 to December A significant year on year reduction of 6.7% (95% CI ) was observed. The MRSA bacteraemia and MSSA bacteraemia rates reduced year on year by 17.7% (95% CI 14.9, 20.3) and 1.5% (95% CI -1.1, 4.0) respectively. Figure 1: Quarterly S. aureus bacteraemia rates per 1000 total AOBDs for Scotland with trend line, April December Rate per 1000 acute occupied bed days S. aureus bacteraemia per 1000 AOBDs Linear (S. aureus bacteraemia per 1000 AOBDs) 0.0 Apr 05-Jun 05 Oct 05-Dec 05 Apr 06-Jun 06 Oct 06-Dec 06 Apr 07-Jun 07 Oct 07-Dec 07 Apr 08-Jun 08 Oct 08-Dec 08 Apr 09-Jun 09 Oct 09-Dec 09 Apr 10-Jun 10 Oct 10-Dec 10 Quarter 9

12 Quarter Table 1: Quarterly numbers and rates of S. aureus bacteraemias (MRSA and MSSA) in Scotland*. MRSA bacteraemias (n) MSSA bacteraemias (n) S. aureus bacteraemias (n) Acute Occupied Bed Days (AOBDs) MRSA bacteraemias per 1000 AOBDs MSSA bacteraemias per 1000 AOBDs S. aureus bacteraemias Apr 05-Jun Jul 05-Sep Oct 05-Dec Jan 06-Mar Apr 06-Jun Jul 06-Sep Oct 06-Dec Jan 07-Mar Apr 07-Jun Jul 07-Sep Oct 07-Dec Jan 08-Mar Apr 08-Jun Jul 08-Sep Oct 08-Dec Jan 09-Mar Apr 09-Jun 09** Jul 09-Sep 09** Oct 09-Dec 09** Jan 10-Mar 10** Apr 10-Jun 10** Jul 10-Sep 10** Oct 10-Dec 10** per 1000 AOBDs * Figures within this report are provisional and may be subject to update from reporting laboratories ** Reporting of Golden Jubilee National Hospital NHS National Waiting Time Centre Board (NWTC) data began in April 2010 however retrospective data for the NWTC for the period April 2009 to March 2010 are now included within the Scottish data. Identification of outliers Figure 2 contains three funnel plots for the fifteen NHS boards S. aureus bacteraemia rates for 2010, plotted against acute occupied bed days. Figure 2 contains data for all S. aureus bacteraemias, with breakdowns for MRSA and MSSA bacteraemias and shows that while there is variation between NHS boards in their individual bacteraemia rates, no Scottish NHS boards reported rates which were above the confidence limits that would be expected from their activity levels, measured by acute occupied bed days. Complete S. aureus bacteraemia data for individual Scottish NHS boards are published by HPS

13 Figure 2: Funnel Plots of S. aureus bacteraemia rates per 1000 AOBDs by NHS board (showing all S. aureus bacteraemias and MRSA and MSSA bacteraemias).* All S. aureus 0.6 Rate per 1000 acute occupied bed days BR NWTC WI OR DG FV FF HG AA TY GR LN LO GGC 0.0 SH Acute Occupied Bed Days (1000s) Meticillin resistant S. aureus 0.30 Rate per 1000 acute occupied bed days WI NWTC SH OR BR DG FV FF HG AA TY LN GR LO GGC Acute Occupied Bed Days (1000s) Meticillin sensitive S. aureus 0.5 Rate per 1000 acute occupied bed days NWTC WI OR BRDG FF FV HG TY GR AA LN LO GGC 0.0 SH Acute Occupied Bed Days (1000s) *Concave lines represent 95% confidence limits and the horizontal line the mean rate of infection. Key to NHS boards AA Ayrshire & Arran FV Forth Valley LN Lanarkshire SH Shetland BR Borders GGC Greater Glasgow & Clyde LO Lothian TY Tayside DG Dumfries & Galloway GR Grampian NWTC National Waiting Times Centre WI Western Isles FF Fife HG Highland OR Orkney 11

14 Typing and susceptibility data The total number of isolates collected in the MRSA snapshot programme in 2010 was 690 with the majority originating from superficial skin and wound swabs (see Figure 3). Eighty two percent (n=567) of these were attributable to the epidemic strain MRSA-15 (EMRSA-15), 9% (n=62) EMRSA-16 and 9% (n=61) fell into the other category (see Figure 4). The most common antibiotic resistance was to ciprofloxacin with 92% of isolates being resistant (n=636). Sixty six per cent of isolates (n=457) were resistant to erythromycin, 17% (n=117) resistant to trimethroprim and 17% (n=114) resistant to clindamycin. MRSA mupirocin high level resistance was 3% (n=22) with 1% (n=7) of isolates displaying low level resistance. Mupirocin resistance was found to be more common in EMRSA-16. There was no resistance to the glycopeptides (teicoplanin and vancomycin) however 0.3% (n=2) of isolates were resistant to linezolid (see Figure 5). Twenty MRSA strains had the PVL gene present (15 of which were sensitive to ciprofloxacin). The majority of PVL positive isolates originated from superficial skin specimens (n=11), seven were from wound isolates and two were from respiratory tract specimens. The isolates were distributed throughout all age groups with the majority occurring in the 65 years and over category. Figure 3: Number of MRSA isolates (n=690) by specimen site Number of Isolates superficial - skin wound respiratory - upper urine other respiratory - lower superficial - eye genital blood alimentary specimen site Figure 4: Number of MRSA isolates (n=690) by strain type Number of isolates EMRSA-15 EMRSA-16 Other Strain type 12

15 Figure 5: MRSA antibiotic resistance (%) with 95% confidence intervals. Percent ciprofloxacin erythromycin trimethoprim clindamycin kanamycin tetracycline tobramycin gentamicin mupirocin fusidic acid rifampicin linezolid Discussion The rates of S. aureus bacteraemias have significantly reduced in Scotland since This has mainly been a result of a significant reduction in MRSA bacteraemias. There is variation in the S. aureus bacteraemia rates between the 15 NHS boards though none were above the 95% confidence intervals of the Scottish mean rate. Reducing S. aureus bacteraemias continues to be a public health priority. Intelligence gathered from the MRSA snapshot programme will facilitate the understanding of the epidemiology and distribution of strain types throughout Scotland. This can be used to provide meaningful data to inform public health policies. The most common strain of MRSA continues to be EMRSA-15 followed by EMRSA-16. These are the two most common strains of MRSA circulating in the United Kingdom and reflect the information we already have from the mandatory HPS S. aureus bacteraemia surveillance system. Over 90% of isolates were resistant to ciprofloxacin. Sensitivity to ciprofloxacin is commonly used to identify community associated MRSA (CA-MRSA). CA-MRSA may encode PVL and can be highly virulent. In recent years there has been an apparent increase in ciprofloxacin resistance among some CA-MRSA strains. Twenty five percent of the PVL positive isolates were resistant to ciprofloxacin. As such, the majority of these isolates would not be detected if a selective media containing ciprofloxacin was used for isolation. Conversely, using ciprofloxacin sensitivity as a marker for further investigation of CA-MRSA would result in 25% of these strains being missed. HPS intends to monitor susceptibility to ciprofloxacin with regards to the appropriateness of this as a putative marker for the identification of CA-MRSA. The PVL positive strains were mainly isolated from superficial skin specimens, wound swabs and respiratory tract specimens and were distributed across all age groups. It is possible that the burden of PVL related disease in Scotland remains underestimated. It is therefore important that all S. aureus (MSSA and MRSA) PVL suspected isolates are referred to the SMRSARL. Improvements in both surveillance and case ascertainment are required to establish the burden of disease, associated risk factors, and provide support for evidence-based guidance for effective patient management and infection control strategies. With the roll out of the Scottish mandatory national MRSA screening programme it is anticipated that an increase in mupirocin resistance may be observed. The baseline proportions observed in this programme for mupirocin high level and low level resistance were 3% and 1% respectively. HPS (in conjunction with SMRSARL) will continue to monitor 13

16 mupirocin resistance and will alert health boards if any significant increases are observed. Further information on S. aureus antimicrobial susceptibility can be found in the Scottish Antimicrobial Prescribing Group (SAPG) Report on Antimicrobial Resistance and Use in Humans in In the absence of resources to type all isolates, this programme has been designed to provide detailed typing of a subset of isolates, applying a consistent methodology, which allows a more reliable assessment of the changing epidemiology of MRSA strains in Scotland. It is a voluntary surveillance system and its success is dependent on continued compliance by Scottish diagnostic microbiology laboratories in order that we can continue to provide meaningful epidemiological information on circulating strains of MRSA. Reducing S. aureus bacteraemias in Scotland HPS has continued to provide support to NHS Boards to enable them to reduce S. aureus bacteraemias which arise as a consequence of healthcare activity. The principles of the actions recommended by HPS to reduce S. aureus bacteraemias remain unchanged: identification of the clinical areas associated with most S. aureus bacteraemias identification of the commonest sources of infection for S. aureus bacteraemias (e.g. peripheral vascular catheters, pneumonias etc) implementing the quality improvement tool designed to reduce the risk of the primary infections. However, this is often not as simple as it appears. In the past year, HPS has focused on working with ICTs to assist their interpretation of data and support the necessary actions. During work with the NHS boards, one key finding was the need for assistance with the simultaneous analysis of both process and outcome data. This may be summarised as follows: for all clinical areas with a high incidence of S. aureus bacteraemias, a key process measure must be identified and monitored. For example, if in a clinical area peripheral vascular catheters (PVCs) were found to be the leading cause of S. aureus bacteraemias, then the use of the PVC bundle process needs to be regularly monitored where such processes have been implemented, monitoring of the processes must be done to ensure compliance at sufficient levels to prevent S. aureus bacteraemias, e.g. sustained 100% PVC bundle compliance when clinical areas have implemented such optimal process measures but have not achieved a substantial reduction in S. aureus bacteraemias, there needs to be a further assessment, including: whether the process measures are themselves valid whether the process measure is the right one whether additional process measures are necessary. For clinical areas where there is both reduction in S. aureus bacteraemias and optimal process measures then the staff in these areas should receive positive feedback and their good practice shared with other clinical areas. In the coming year HPS will work with Healthcare Improvement Scotland (HIS), formerly NHS Quality Improvement Scotland (NHS QIS), in reviewing all existing quality improvement tools and in identifying the need for new tools. 14

17 Clostridium difficile Infection Please note CDI rates in this report have been updated due to a subsequent revision of the national figures of hospital activity. Please go to for the newest revised data. Introduction C. difficile is a Gram-positive spore forming anaerobic bacterium, widely distributed in the environment and normal gut flora of animals. C. difficile infection (CDI) is a major cause of morbidity and mortality, especially as a healthcare associated infection, and generally follows the use of antibiotics. 11 Disease ranges from mild self-limiting diarrhoea to severe diarrhoea, pseudomembranous colitis, toxic megacolon and death. 12 In Scotland, mandatory surveillance of CDI was introduced in following reports of increasing CDI rates and severity of disease around the world and the rise in voluntary laboratory reports to HPS in the period Surveillance initially recorded the incidence of CDI in patients aged 65 and over. In April 2009, the programme was expanded to include patients aged years. 13 HPS commissioned the Scottish C. difficile Reference Service in 2007 (part of the SSSCDRL) to support the surveillance programme by providing a strain typing (ribotyping) service to Scottish hospitals, as well as monitoring antimicrobial resistance in C. difficile. Monitoring of ribotypes falls into two categories: those associated with outbreaks and severe disease (clinical surveillance), and those reflecting a range of disease presentations from mild to severe (representative surveillance, also known as the Snapshot Programme). The former typing scheme has been running since November 2007, whereas the Snapshot Programme was implemented at the start of The CDI surveillance programme examines trends in cases across Scotland and the impact of interventions on these trends. The national data may be used to support outbreak investigations by providing context both nationally and by NHS board that may help with understanding local epidemiology. The CDI team within HPS also supports outbreak investigations when requested by NHS boards. This second annual report presents data for the calendar year CDI case numbers and incidence rates in the two patient groups (15-64 and 65 and over) are presented for each NHS board and overall for Scotland. New for this annual report, cases from the Golden Jubilee National Hospital (NHS National Waiting Time Centre Board - NWTC) are included in the total figures and rates analyses. Results are also provided for the two typing schemes described above as well as a brief overview of antibiotic susceptibilities. Consensus guidance on diagnostic testing the Recommended Protocol for Testing for C. difficile and Subsequent Culture 14 was produced by the Scottish Microbiology Forum (SMF), SSSCDRL and HPS, and implemented during Laboratories not using cell culture cytotoxicity testing as the initial diagnostic test were recommended to perform a second confirmatory test to increase the accuracy of toxin positive results. The impact of the new testing is discussed. HPS activities in support of the NHS boards and a summary of some approaches to reduce CDI in Scotland are given in the discussion. 15

18 Please note CDI rates in this report have been updated due to a subsequent revision of the national figures of hospital activity. Please go to for the newest revised data. Methods In Scotland, all 15 NHS boards report cases of CDI to HPS and submit isolates for severe cases and outbreaks to the SSSCDRL according to a standard protocol. 15 Isolates for the representative typing scheme (Snapshot Programme) are submitted to the SSSCDRL according to a separate protocol. 16 The data reported here are based on information from NHS board diagnostic laboratories and the SSSCDRL (note that bed day data used in this report were from the calendar year 2009). CDI cases and incidence rates are presented by NHS board. Cases are allocated to NHS boards based on the location of the diagnostic laboratory where the specimen was first tested. The surveillance does not distinguish between cases from acute, non-acute hospitals, and the community. Following introduction of the new testing algorithm any results reported to HPS as equivocal (second test is negative following a positive first test) are removed from the final data published. The impact of this on overall rates is assessed. Full details of the methods may be obtained from the HPS quarterly reports. 17 Results CDI cases and incidence in patients aged 65 and over The total number of new cases identified in 2010 was This is a 39% decrease in comparison with the 3634 cases reported in 2009 (note that contrary to last years annual report, the annual totals include cases from NWTC). The annual incidence rate for Scotland in 2010 was 0.44 per 1000 total occupied bed days (OCBDs), which is a decrease of 38% (95% CI 27%, 47%) compared to The overall quarterly rates for Scotland for the period October 2006 to December 2010 are shown in Figure 6. Overall rates began to plateau during the first three quarters of 2010 subsequent to a continuous decline in overall rates since This was followed by a significant decrease in Q4. For details on cases and rates, see Tables Figure 6: Overall quarterly rates of CDI per 1000 total OCBDs in patients aged 65 and over for seventeen quarters of the mandatory surveillance covering the period October 2006 to December Rate per 1000 total OCBDs in patients aged 65 and above QT QT QT QT QT QT QT QT QT QT QT QT QT QT QT QT QT Quarter 16

19 Please note CDI rates in this report have been updated due to a subsequent revision of the national figures of hospital activity. Please go to for the newest revised data. Table 2.1: Total number of CDI cases in the age group 65 and over by NHS board for NHS board Quarter 1 Quarter 2 Quarter 3 Quarter 4 Annual Scotland * Ayrshire & Arran Borders Dumfries & Galloway Fife Forth Valley National Waiting Times Centre Grampian Greater Glasgow & Clyde Highland Lanarkshire 66 56* Lothian Orkney Shetland Tayside Western Isles * This is a revised number from the quarterly report. Table 2.2: Incidence rates of CDI in the age group 65 and over by NHS board for NHS board Quarter 1 Quarter 2 Quarter 3 Quarter 4 Annual Scotland Ayrshire & Arran Borders Dumfries & Galloway Fife Forth Valley * National Waiting Times Centre Grampian Greater Glasgow & Clyde Highland Lanarkshire * Lothian Orkney Shetland Tayside Western Isles * This is a revised number from the quarterly report. Table 2.3: Comparison of annual rates of CDI in the age group 65 and over by NHS board. Total number of CDI cases Annual rates per 1000 total OCBDs >65 years NHS board Scotland Ayrshire & Arran Borders Dumfries & Galloway Fife Forth Valley National Waiting Times Centre Grampian Greater Glasgow & Clyde Highland Lanarkshire Lothian Orkney Shetland Tayside Western Isles

20 Please note CDI rates in this report have been updated due to a subsequent revision of the national figures of hospital activity. Please go to for the newest revised data. Figure 7 shows a comparison of the annual incidence rates between 2009 and 2010 for each NHS board. All of the NHS boards, except NHS Western Isles (which had a non-significant increase), reported a decrease in rates between 2009 and NHS National Waiting Times Centre and NHS Orkney had a non-significant decrease, whereas rates decreased significantly in NHS Ayrshire & Arran, NHS Borders, NHS Dumfries & Galloway, NHS Fife, NHS Forth Valley, NHS Grampian, NHS Greater Glasgow & Clyde, NHS Highland, NHS Lanarkshire, NHS Lothian, and NHS Tayside. NHS Shetland reported no cases during 2010 and has had no cases in seven of the last eight quarters. Figure 7: Annual overall rates of CDI per 1000 total OCBDs between 2009 and 2010 in patients aged 65 and over in acute and non-acute hospitals in 15 NHS boards in Scotland Rate per 1000 total OCBDs in patients aged 65 or over Ayrshire & Arran Borders Dumfries & Galloway Fife Forth Valley National Waiting Times Centre Grampian Greater Glasgow & Clyde Highland Lanarkshire Lothian Orkney Islands Shetland Islands Tayside Western Isles CDI cases and incidence in patients aged The total number of new cases identified in 2010 was 692. No annual data for 2009 was available as surveillance in this age group began in April 2009 (in the three quarters of 2009 there were 873 cases giving an overall rate of 0.76 per 1000 acute occupied bed days (AOBDs)). The annual incidence rate for Scotland in 2010 was 0.45 per 1000 AOBDs. The overall quarterly rates for Scotland for the period April 2009 to December 2010 are shown in Figure 8. Overall rates decreased during Q1 and Q2 of 2010 before increasing in Q3. This was followed by a significant decrease in Q4. A similar pattern was observed in For details on cases and rates, see Tables (due to lack of data there is no comparison of annual incidence rates for 2009 and 2010). Figure 8: Overall quarterly rates of CDI per 1000 AOBDs in patients aged for seven quarters of the mandatory surveillance covering the period April 2009 to December Rate per 1000 AOBDs in patients aged QT QT QT QT QT QT QT Quarter 18