Ethnicity, age and incidence rates for renal replacement therapy (RRT) in Birmingham, UK:

Transcription

1 Nephrol Dial Transplant (28) 23: doi: 1.193/ndt/gfn366 Advance Access publication 2 July 28 Original Article Ethnicity, age and incidence rates for renal replacement therapy (RRT) in Birmingham, UK: Mark Lambie 1, Nick Richards 2 and Steve Smith 3 1 Renal Unit, Walsgrave Hospital, 2 Fresenius Medical Care and 3 Department of Renal Medicine, Birmingham Heartlands Hospital, UK Abstract Background. Birmingham is a densely populated, industrial city with a high social deprivation index and large black (B) and Indo-Asian (I) populations. End stage renal failure is known to be more common in ethnic minorities and urban communities. Workforce planning requires accurate epidemiological data but most studies of the effect of ethnicity in the UK are from the early 199s. Methods. RRT acceptance rates for adults (>16 years) in Birmingham were calculated for the 5-year periods and using the 1991 and 21 UK population census datasets and local programmes data. Results. The adult population of Birmingham Health Authority increased slightly (961,41 in 1991 v 977,99 in 21) but the proportion of W fell (82.7% to 74.3%) while B (5.3% to 7.%) and I (1.4% to 17.2%) both increased. Median age was lower for I ( , ) than B ( , ) and W ( , ). Numbers of new patients increased by 29% in W, 98% in B and 19% in I. There was also a substantial increase in acceptance rates (W 92 to 129 pmp, I 175 to 243 pmp, B 191 to 278 pmp) but the proportional increase in I (26.9%) was less than in W (41.5%) or B (48.2%). This is because almost all the increase in RRT acceptance rates for all ethnic groups was seen in the over 55 age group (256 pmp 1991, 481 pmp 21) but 85% of the population growth for I was in the under 55 age group. In all ethnic groups there was a striking increase in acceptance rates for the over 7 s (W 177 to 44 pmp, I 536 to 1711 pmp, B 31 to 1858 pmp). Conclusions. All acceptance rates were equivalent to the highest previously described in the UK. This may be due to local factors including social deprivation, availability of care and physicians attitudes. The increase in patient numbers was due to rising ethnic minority populations and increasing acceptance rates, especially in the elderly. The take-on rate is likely to rise disproportionately for I as the population ages over the next 1 years. This indicates that the future need for RRT in UK inner city areas, especially Correspondence and offprint requests to: Dr. Mark Lambie, Renal Unit, University Hospital, Coventry. United Kingdom. CV2 2DX. Tel: ; Fax: ; lambiem@doctors.org.uk those with a large elderly ethnic population, will be greater than previously estimated. Keywords: end-stage renal disease; ethnicity; incidence Introduction The incidence of end-stage renal failure (ESRF) patients accepted onto renal replacement therapy (RRT) has been shown to vary with ethnicity (higher in Indo-Asians and Blacks), age (increasing with age) and geographical location [1 4]. There is also evidence that the incidence is higher in urban populations and in areas with high social deprivation [3,5]. These variations have significant implications for haemodialysis capacity and nephrology workforce planning, particularly as the English national service framework for renal services (24) clearly recommends that RRT services should be available to all who can benefit [6]. The UK Renal Registry, which provides data on the current dialysis population, conducted a National Renal Review in 22 that showed a steady increase in incidence rates in England from 67 pmp in 1991 to 98 pmp in 22 [7]. In 22 the Renal Association suggested planning assumptions for acceptance rates onto RRT of pmp [8]. However, the data used for estimating the need for RRT in ethnic populations come from an analysis of data from [9]. Since then, there has been a substantial increase in incidence rates in the total population with increasing numbers of elderly patients commenced on dialysis [3]. There is an urgent need to review age-specific incidence rates in different ethnic populations. Although the Birmingham Health Authority (BHA) ceased to be a functional NHS administrative unit in the reorganizations of 22, it is a distinct geopolitical unit that clearly defines a main industrial part of the West Midlands with a substantial ethnic minority population and a high level of social deprivation. Data from the 1991 and 21 censuses are available to define the population and ethnic mix of the BHA area. There are no published data C The Author [28]. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. For Permissions, please journals.permissions@oxfordjournals.org

2 3984 M. Lambie et al. Table 1. Ethnicity coding 12 Study 1991 census 21 census White White White: British White: Irish White: Other Black Black African Black or Black British: Caribbean Black Caribbean Black or Black British: African Black Other Black or Black British: Other Mixed: White and Black African Mixed: White and Black Caribbean Indo-Asian Indian Asian or Asian British: Indian Other Bangladeshi Pakistani Other groups: Asian Other groups: Other Chinese Asian or Asian British: Bangladeshi Asian or Asian British: Pakistani Asian or Asian British: Other Asian Mixed: White and Asian Mixed: Other Mixed Chinese and other ethnic groups: Chinese Chinese and other ethnic groups: Other ethnic group regarding current or historical incidence rates for the BHA population. Two renal units provide dialysis for all patients living in this area, there is little cross-boundary flow and both have complete computerized records dating back to 199. By correlating the numbers of incident RRT patients from the BHA area with the populations recorded in the 1991 and 21 censuses, we were able to investigate the associations between ethnicity and age on RRT incidence rates between 199 and 24. Methods We identified all new RRT patients over 16 years of age accepted by both units (Birmingham Heartlands and Queen Elizabeth Hospitals) from 1 January 199 to 31 December 23. We excluded patients who were already on dialysis and transferred in from other areas, transfers from the paediatric unit, non-residents, failed transplants and patients requiring dialysis during a holiday in the area. Patients with acute renal failure were excluded unless they failed to recover function, or if they died within 9 days. The date they were deemed to have irreversible renal failure and thus were regarded as having ESRF was determined by the clinicians at the time and was routinely recorded on the computerized record. We included patients who received a pre-emptive transplant prior to starting on dialysis. Although the surrounding dialysis units are a substantial distance away from the BHA area and cross-boundary migration was likely to be minimal, nearby units were contacted to identify any BHA patients accepted for dialysis during this period. None were identified. The ethnicity was ascertained by clinicians in both units as White (W), Black (B), Indo-Asian (I) and Other (O) and we kept this classification in the analysis. The Office of National Statistics provided the populations for the BHA area from both the 1991 and 21 censuses, along with a breakdown by age and ethnicity. The coding for ethnicity between censuses was different, but we re-coded them into W, B, I and O as shown in Table 1. As the O group was Number of incident patients Fig. 1. Incident RRT population Black Indo-Asian White small, these patients were only included in the analyses for the total population. The age bands were derived by splitting the dialysis population into approximate thirds and rounding into convenient groups (16 54, 55 7, >7 years). The incidence rates were calculated for two cohorts comprising those patients starting dialysis between 199 and 1995 and between 1999 and 24. These dates were chosen to cross the 1991 and 21 census dates. We chose to examine data averaged over 5-year periods to eliminate year-onyear variation and improve the precision of the estimates. Confidence intervals were obtained from Geigy Scientific Tables, based on the Poisson distribution. Statistical analysis was performed using the commercially available SPSS v1 statistical software package. Results A total of 181 new ESRF patients started RRT between 199 and 24. The ethnic breakdown of numbers of new patients starting each year is shown in Figure 1. There was an increase in the numbers of patients starting RRT in all ethnic groups throughout the study period. The change in the Birmingham population between the 1991 and 21 censuses is shown in Figure 2. The total population remained similar (966,41 in 1991 versus 977,99 in 21) but the ethnic mix changed with W falling from 78.5% to 7% whilst I and B increased from 13.5% to 2% and 5.9% to 7.9%, respectively. The annualized RRT incidence rates for the two time periods are shown in Table 2. There was a consistent increase of 4% in all ethnic groups between the 1991 and 21 cohorts. However, the incidence rate for the total population increased by 55%. This is because the B and I populations which had much higher incidence rates constituted a greater proportion of the total population in 21 (27.9%) than in 1991 (19.4%, Figure 2). In , the incidence rate for I was 1.88 and for B was 2.16 times greater than W. Similar proportional differences (I, 1.87 and B, 2.15) were seen in Changes in the mean age of the incident RRT patients of different ethnic groups over the last 14 years are shown

3 Ethnicity, age and incidence rates for RRT in Birmingham black 25 1 indoasian white Population number Percentage change >7 2 Fig. 2. Birmingham demographics. Table 2. Annualized RRT incidence rates a a % change Total 17 (99 118) 166 ( ) +55 White 92 (83 13) 129 (116 14) +4 Indo-Asian 175 (139 24) 243 ( ) +39 Black 191 ( ) 278 ( ) +46 a 95% confidence intervals are shown in parentheses. Mean age Black White Indo-Asian Fig. 3. Mean average age of incident RRT patients in Figure 3. W was always the oldest, but the mean age of incident RRT patients in all ethnic groups increased. The proportional changes in the age profiles of the different ethnic groups between 1991 and 21 are shown in Figure 4. The population decreased in all age groups for W but the fall was smallest in the >7s. Thus, the median age increased from 45.1 to 45.9 years. Numbers of I increased in all age groups but proportionally much more in the >7s (1864 to 5844, %). However, the smaller percentage increase in numbers in the younger groups ( to in the 16 54s, +65%) occurred in a proportionately much larger young population so the median age fell from 33.5 to 32.9 years. For B, there was a similar percentage increase -5 White Indo-Asian Black Fig. 4. Changes in age profile of Birmingham population from 1991 to 21. Annualised incidence rate pmp >7 White IA Black White IA Black Fig. 5. Annualized incidence rates for RRT by age and ethnicity. in the >7s (1327 to 3874, +192%), but the increase in the younger groups was less than for I and the median age increased from 33.5 to 37.5 years. Annualized RRT incidence rates for the different age and ethnic groups are shown in Figure 5. There was little change in the incidence rates in 16 54s for all three ethnic groups between the 1991 and 21 cohorts, and in the White 55 7 age group ( pmp). There was a significant increase in B ( pmp) and I ( pmp) in the 55 7 age group. The biggest differences were seen in the >7 s. For W, the increase was 248% ( pmp), for I 319% ( pmp) and for B 617% ( pmp). Discussion The National Renal Review in 22 [3] reported an incidence rate for RRT of 98 pmp for England, 118 pmp for Wales and 12 pmp for Scotland. The rate reported for the West Midlands was 113 pmp. The BHA incidence rates of 166 pmp for reported in this study are comparable with the highest reported in the UK to date. Other areas

4 3986 M. Lambie et al. with high incidence rates in 22 [3] include Wolverhampton (169 pmp) and Ealing Hammersmith and Hounslow (17 pmp). Birmingham s incidence rate has been consistently high, being 17 pmp in compared with the rate for England of 82 pmp in [9]. From to incidence rates for the different ethnic populations in Birmingham rose by 39 46%. This is similar in proportion to the increase in the English incidence rates of 46% between 1991 and 22 (67 98 pmp). Most of the increase is due to a greatly increased incidence in the elderly (Figure 5). The lack of change in incidence rates in all younger Birmingham populations suggests that there was no unmet need for RRT. Previous UK studies examining incidence rates in agespecific ethnic groups include Roderick et al. on data from [9], where rates were estimated for the whole of England, and Higgins et al. [1], where rates were estimated for the Coventry population, a region similar to the one in this study. Our results were similar to, or higher than, both studies in all ethnic groups and most age groups. The incidence rates described in this study are reliable as we studied a large number of patients with a complete record of age and ethnicity over a prolonged period of time. One of the common problems affecting calculation of incidence rates is cross-boundary flow, particularly in large urban areas. This was not a problem in our study as the surrounding units are a significant distance away and we confirmed that no patients from BHA were accepted by adjacent units. There are several possible reasons for Birmingham s high incidence rates. These include the large ethnic populations, poor primary care provision, social deprivation, comparatively easy access to renal services and physician practice. Furthermore, census underenumeration may also contribute, although this occurs mainly in younger Blacks and so is unlikely to be important to our data. Unfortunately, we did not have sufficiently complete data on aetiology of ESRF from our databases to analyse this issue further. Clinicians practising over the time span of the study did not feel that access to dialysis services had limited ESRF take on rates at any point. Black and Indo-Asian ethnicity is associated with higher incidence rates [2,9] and Birmingham has a large ethnic population (28% B or I in 21). We confirmed that the incidence rate for the combined ethnic minorities was high, with relative risks between 2.6 and 5.5 compared to Whites, depending on age. This is not the sole explanation for the high overall incidence rate for RRT in Birmingham as the rate of 129 pmp for the W population was still higher than the national rate. A variety of explanations for the excess risk of ESRF experienced by I and B populations have been proposed. It is well recognized that B and I have a higher prevalence of non-insulin-dependent diabetes mellitus and B in particular have a high prevalence of hypertension [11]. However, there is now good evidence that this, along with other environmental and social factors, does not completely explain the increased risk of ESRF. White diabetics are at greater risk of MI, congestive heart failure, stroke and lower extremity amputation than other diabetic populations but are at a significantly lower risk of ESRF [12]. Indo-Asian diabetics in Leicestershire were found to be 13 times more likely to develop ESRF than White diabetics [13]. Several large observational studies have shown that, even after correcting for factors such as diabetes, hypertension, access to care and social deprivation, there is a significant unexplained relative risk of ESRF [14 18] in ethnic minority populations. Using data from the third National Health and Nutrition Examination Survey (NHANES) and the United States Renal Data System, Hsu et al. demonstrated a greater rate of progression from CRF to ESRF [19] in African Americans, consistent with the finding of a similar prevalence of CRF in African Americans and American Whites in NHANES [2]. Ethnic minorities, particularly those with an estimated glomerular filtration rate of <6 ml/min, are also at higher risk of proteinuria, a known risk factor for CKD progression, irrespective of diabetes [21]. One hypothesis to explain this predisposition is the higher incidence of low birth weight among African Americans. Low birth weight has been shown to correlate with the number of glomeruli in the adult kidney (although ethnicity alone had no effect on nephron mass) [22] and also to the subsequent risk of ESRF, particularly during the first 14 years of life [23], as well as chronic kidney disease [24]. Genetic susceptibility is another possibility. A family history of ESRF is associated with a three- to nine-fold increase in the risk of developing nephropathy in African Americans [25,26]. Several different causes of ESRF, including types I and II diabetes mellitus, hypertension, HIV and lupus nephritis are associated with familial clustering [26,27]. Candidate loci for nephropathy susceptibility genes have been identified in different ethnic groups on chromosomes 3q, 1q and 18q [26,28 3] and linkage between African American ESRF and the locus for plasma kallikrein has also been demonstrated [31]. Less information is available on possible reasons for excess risk in Indo-Asians, although two studies did suggest idiopathic interstitial nephritis as well as focal, segmental glomerulosclerosis, chronic pyelonephritis and systemic lupus erythematosus were over-represented in this group [4,32]. UK studies have demonstrated a significant correlation between the incidence/prevalence of RRT and social deprivation [5,33]. Large studies in Australia, New Zealand and the USA have also demonstrated this link [17,34]. The effect of deprivation is complicated by the association this has with ethnicity. The UK Renal Registry published data in 23 showing a clear association between deprivation and ethnicity [3]. Multiple regression analysis found no correlation between social deprivation and incidence rates for RRT after correcting for ethnicity, although a further analysis (not published) excluding areas of high ethnicity did demonstrate a significant correlation. Geographical proximity to renal services has been shown to increase RRT incidence rates [5]. This may be relevant in Birmingham, a relatively small densely populated area with good internal transport links, although a recent study suggested that rural populations in the USA have a higher incidence of ESRF [35]. We have demonstrated much higher RRT incidence rates in all Birmingham populations compared to the rest of the

5 Ethnicity, age and incidence rates for RRT in Birmingham 3987 UK but particularly in older people from ethnic minority groups. As anticipated from trends in immigration, the Black population in Birmingham aged during our study and is predicted to age further. The Indo-Asian population is younger, larger and likely to follow this trend. This applies for ethnic minority populations in other areas of the UK. This combination has significant implications for the future supply of RRT in similar areas within the UK and other countries, and the need for chronic kidney disease prevention. Conflict of interest statement. Dr Nick Richards is now Medical Director of Fresenius Medical Renal Services but was Medical Director of the Renal Unit, Queen Elizabeth Hospital, for the duration of the study. There is no conflict of interest for the other authors. The results presented in this paper have not been published previously in whole or part, except in abstract format. References 1. Roderick PJ, Jones I, Raleigh VS et al. Population need for renal replacement therapy in Thames regions: ethnic dimension. BMJ 1994; 39: US Renal Data System. USRDS 23 Annual Data Report: Atlas of End Stage Renal Disease in the United States. National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, MD, Ansell D, Feest TG, Byrne C et al. UK Renal Registry Report 23. UK Renal Registry, Bristol, Lightstone L, Rees AJ, Tomson C et al. High incidence of end-stage renal disease in Indo-Asians in the UK. QJMed1995; 88: Roderick P, Clements S, Stone N et al. What determines geographical variation in rates of acceptance onto renal replacement therapy in England? J Health Serv Res Policy 1999; 4: The National Service Framework for Renal Services. Part I: Dialysis and Transplantation. London: Department of Health, Ansell D, Feest TG, Byrne C et al. National Renal Review 22: Summary Report on Adult and Paediatric Renal Services. UK Renal Registry, Bristol, Treatment of adults and children with renal failure: standards and audit measures. Royal College of Physicians of London and the Renal Association, London, Roderick PJ, Raleigh VS, Hallam L et al. The need and demand for renal replacement therapy in ethnic minorities in England. J Epidemiol Community Health 1996; 5: Higgins RM, Edmunds ME, Dukes DC. End-stage renal failure in Indo-Asians. QJMed1995; 88: Brancati FL, Kao WH, Folsom AR et al. Incident type 2 diabetes mellitus in African American and white adults: the Atherosclerosis Risk in Communities Study. JAMA 2; 283: Karter AJ, Ferrara A, Liu JY et al. Ethnic disparities in diabetic complications in an insured population. JAMA 22;287: Burden AC, McNally PG, Feehally J et al. Increased incidence of end-stage renal failure secondary to diabetes mellitus in Asian ethnic groups in the United Kingdom. Diabet Med 1992; 9: Klag MJ, Whelton PK, Randall BL et al. End-stage renal disease in African-American and white men. 16-year MRFIT findings. JAMA 1997; 277: Brancati FL, Whittle JC, Whelton PK et al. The excess incidence of diabetic end-stage renal disease among blacks. A population-based study of potential explanatory factors. JAMA 1992; 268: Tarver-Carr ME, Powe NR, Eberhardt MS et al. Excess risk of chronic kidney disease among African American versus white subjects in the United States: a population-based study of potential explanatory factors. J Am Soc Nephrol 22; 13: Byrne C, Nedelman J, Luke RG. Race, socioeconomic status, and the development of end-stage renal disease. AmJKidneyDis1994; 23: McClellan W, Tuttle E, Issa A. Racial differences in the incidence of hypertensive end-stage renal disease (ESRD) are not entirely explained by differences in the prevalence of hypertension. Am J Kidney Dis 1988; 12: Hsu CY, Lin F, Vittinghoff E et al. Racial differences in the progression from chronic renal insufficiency to end-stage renal disease in the United States. J Am Soc Nephrol 23; 14: Coresh J, Astor BC, Greene T et al. Prevalence of chronic kidney disease and decreased kidney function in the adult US population: Third National Health and Nutrition Examination Survey. Am J Kidney Dis. 23; 41: Bryson CL, Ross HJ, Boyko EJ et al. Racial and ethnic variations in albuminuria in the US Third National Health and Nutrition Examination Survey (NHANES III) population: associations with diabetes and level of CKD. Am J Kidney Dis 26; 48: Hughson M, Farris AB, 3rd, Douglas-Denton R et al. Glomerular number and size in autopsy kidneys: the relationship to birth weight. Kidney Int 23; 63: Vikse BE, Irgens LM, Leivestad T et al. Low birth weight increases risk for end-stage renal disease. J Am Soc Nephrol 28;19: Fan ZJ, Lackland DT, Lipsitz SR et al. The association of low birthweight and chronic renal failure among Medicaid young adults with diabetes and/or hypertension. Public Health Rep 26; 121: Freedman BI, Spray BJ, Tuttle AB et al. The familial risk of endstage renal disease in African Americans. AmJKidneyDis1993; 21: Satko SG, Freedman BI, Moossavi S. Genetic factors in end-stage renal disease. Kidney Int Suppl 25; 67: S46 S Freedman BI, Wilson CH, Spray BJ et al. Familial clustering of endstage renal disease in blacks with lupus nephritis. AmJKidneyDis 1997; 29: Bowden DW, Colicigno CJ, Langefeld CD et al. A genome scan for diabetic nephropathy in African Americans. Kidney Int 24; 66: Freedman BI, Rich SS, Yu H et al. Linkage heterogeneity of endstage renal disease on human chromosome 1. Kidney Int 22; 62: Moczulski DK, Rogus JJ, Antonellis A et al. Major susceptibility locus for nephropathy in type 1 diabetes on chromosome 3q: results of novel discordant sib-pair analysis. Diabetes 1998; 47: Yu H, Bowden DW, Spray BJ et al. Identification of human plasma kallikrein gene polymorphisms and evaluation of their role in endstage renal disease. Hypertension 1998; 31: Ball S, Lloyd J, Cairns T et al. Why is there so much end-stage renal failure of undetermined cause in UK Indo-Asians? QJM 21; 94: Maheswaran R, Payne N, Meechan D et al. Socioeconomic deprivation, travel distance, and renal replacement therapy in the Trent Region, United Kingdom 2: an ecological study. J Epidemiol Community Health 23; 57: Cass A, Cunningham J, Wang Z et al. Social disadvantage and variation in the incidence of end-stage renal disease in Australian capital cities. Aust N Z J Public Health 21; 25: Fan ZJ, Lackland DT, Lipsitz SR et al. Geographical patterns of endstage renal disease incidence and risk factors in rural and urban areas of South Carolina. Health Place 27; 13: Received for publication: Accepted in revised form: 9.6.8