NDT Advance Access originally published online on April 4, 2006
Nephrology Dialysis Transplantation 2006 21(8):2178-2183; doi:10.1093/ndt/gfl145
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© The Author [2006]. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org
Original Articles: Clinical Nephrology
Geographic, ethnic, age-related and temporal variation in the incidence of end-stage renal disease in Europe, Canada and the Asia-Pacific region, 1998–2002
Department of Preventive and Social Medicine, University of Otago, Dunedin, New Zealand, The ERA-EDTA Registry, Department of Medical Informatics, Academic Medical Center, Amsterdam, The Netherlands, CORR, Canadian Institute of Health Information, Toronto, Ontario, Canada and the ANZDATA Registry, Queen Elizabeth Hospital, Adelaide, SA, Australia
Correspondence and offprint requests to: Dr Margaret McCredie, Department of Preventive and Social Medicine, University of Otago, PO Box 913, Dunedin, New Zealand. Email: margaret.mccredie{at}stonebow.otago.ac.nz
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Background. Only unbiased estimates of end-stage renal disease (ESRD) incidence and trends are useful for disease control—identification of risk factors and measuring the effect of intervention.
Methods. Age- and sex-standardized incidences (with trends) were calculated for all-cause and diabetic/non-diabetic ESRD for persons aged 0–14, 15–29, 30–44 and 45–64 years in 13 populations identified geographically, and six populations identified by ethnicity.
Results. The incidence of ESRD varied most with age, ethnicity and prevalence of diabetes. All non-Europid populations had excess ESRD, chiefly due to rates of type 2 diabetic ESRD that were greater than accounted for by community prevalences of diabetes. Their rates of non-diabetic ESRD also were raised, with contributions from most common primary renal diseases except type 1 diabetic nephropathy and polycystic kidney disease. The ESRD rates generally were low, and more similar than different, in Europid populations, except for variable contributions from type 1 (high in Finland, Sweden, Denmark and Canada) and type 2 (high in Austria and Canada) diabetes. In Europid populations during 1998–2002, all-cause ESRD declined by 2% per year in persons aged 0–44 years, and all non-diabetic ESRD by a similar amount in persons aged 45–64 years, in whom diabetic ESRD had increased by 3% per year.
Conclusions. Increased susceptibility to type 2 diabetes and to kidney disease progression characterizes excess ESRD in non-Europid peoples. The decline in all-cause ESRD in young persons, and non-diabetic ESRD in the middle-aged, probably reflects improving management of progressive renal disease.
Keywords: Asia-Pacific; Canada; end-stage renal disease; Europe; incidence; trends
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The data describing incidence and trends of end-stage renal disease (ESRD) have two important purposes: planning renal replacement treatment (RRT) services, and devising and monitoring strategies for prevention. For the first of these, summary rates (that is, including persons of all ages) are acceptable if appropriate adjustments are made for demographic variations. However, summary rates, even if age-adjusted, are not suitable when the emphasis is on prevention, principally because they conceal, and do nothing to reduce, what has been a large and important source of error, namely differing (in time or between populations) access to RRT [1–3]. Moreover, they provide little information about younger patients as they are determined predominantly by risk factors for middle-aged and elderly patients, who comprise 80% of incident ESRD in economically developed societies. By and large, age-specific rates avoid these drawbacks, but unless several age groups are combined there will be insufficient statistical power (except in the largest populations or if data are accumulated for many years) to verify any, but sizeable, differences.
Change, with time, has been a notable feature of ESRD incidence in the past, chiefly reflecting steady improvement in access to treatment and ageing of the population [3]. At present, interest in trends is more likely to be disease-oriented, for example, in explaining the escalation in hypertensive or diabetic ESRD [3–6], or determining if recent methods of secondary prevention [7] have proved effective at the population level. Truncated, age- and sex-standardized rates are suitable for these purposes; the age bands should be selected so as to represent the different epidemiological characteristics of ESRD in children, and in young, middle-aged and elderly adults.
For this study, we have assembled ESRD registry data from three regions of the world for 13 populations determined geographically, and six populations identified by ethnicity. Incidence and trends were calculated for ESRD due to all causes, and to all diabetic and all non-diabetic renal diseases, in persons aged 0–14, 15–29, 30–44 and 45–64 years, for the period 1998–2002. Where significant differences have been identified, an explanation has been sought by scrutiny of disease-specific ESRD incidence.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Study populations
Data were sought from populations whose access to publicly funded RRT was restricted only by medical contra-indications and not by socio-economic or geographic circumstances, and for which recording of incident cases was believed to be complete for persons normally resident in the country or region. The exception was Malaysia, where RRT has been funded from a variety of sources, and is not provided as of right to all. Accordingly, we included data only from the economically developed region of Southern and Western Peninsula Malaysia, where treatment has been available to nearly all patients with ESRD for some years.
Each contributing renal registry provided (i) the numbers of new patients who first received treatment (dialysis or transplantation) for ESRD, by primary renal disease and (ii) populations, by sex, age (in 5-year age groups), and country, region or ethnic group where applicable, for the calendar years 1998–2002. The European, Malaysian and Northern Mariana Islands’ registries submitted data for all residents in a country or defined region (in the case of Malaysia, for the six contiguous states of Johor, Melaka, Negeri Sembilan, Selangor with Kuala Lumpur, Perak and Pinang). Canada, Australia and New Zealand each provided nation-wide data for two populations [Canada: Aboriginal comprising First Nations, Inuit and Métis, and non-Aboriginal which included the 11% of Canadian patients for whom ethnicity had not been recorded; Australia: Aboriginal and Torres Strait Islander (ATSI) and non-ATSI; New Zealand: Maori and Pacific Island people, and non-Polynesian]. In each of these three countries, the population not identified by ethnicity is chiefly of European descent (Europid).
For each patient, the treating nephrologist recorded the primary renal disease according to simple registry guidelines, as described previously [8]—in Canada, the codes used are similar to those used in Europe. ‘Unknown diagnosis’ was included in the category ‘all non-diabetic renal disease’.
Statistical methods
Mean annual incidence rates, with their 99% confidence intervals (CI), were calculated for persons aged 0–14, 15–29, 30–44 and 45–64 years, directly sex- and age-standardized to Segi's ‘world’ population. Trends with time were estimated by Poisson regression but, because of small numbers, statistical power was insufficient to verify separately determined trends in individual populations other than non-Aboriginal Canadians and non-ATSI Australians. Accordingly, trends are presented only for these two populations and for all European countries or regions considered together. Standardized incidence ratios (SIRs), in each case indirectly standardized to their own majority population, were calculated for rates in each age band for the non-Europid populations of Canada, Australia and New Zealand. Because multiple comparisons were made between directly standardized incidence rates, differences were regarded as statistically significant only if the 99% CIs did not overlap. However, for analyses of trends and SIRs, the more usual significance level of P<0.05 was accepted.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The age structure of the study populations varied greatly (Table 1), the median age being 35–39 years in the 14 Europid populations and 20–28 years in the five non-Europid peoples.
|
Preliminary estimation of incidence trends during the 1998–2002 period showed no significant (in this instance, P<0.05) increase of ESRD in any age band up to 75 years of age in any population, except in persons aged 65–74 years in Malaysia, Greece and Denmark. However, most populations did show a rising incidence of (treated) ESRD for persons aged
75 years, indicating that access to RRT was still improving for the elderly in this period. As we wished to compare rates that were not influenced by changing access to RRT, further analysis was restricted to persons aged <65 years.
ESRD incidence and trends in persons aged 0–44 years
The incidence of ESRD was low in children aged 0–14 years, who accounted for only 1–2% of all ESRD (Table 1). The high rates in Finland were due entirely to congenital nephrotic syndrome of the Finnish type, principally in infants aged 0–4 years, while the excess in Aboriginal Canadians involved a spectrum of nephritic, congenital and hereditary diseases and extended throughout the 0–14 year age band.
From the age of 15 years, ESRD rates rose with age, more steeply in non-Europid than in Europid populations, except in Canada (Tables 1 and 2). The significantly high rates in Aboriginal Canadians, Australian ATSI, and New Zealand Maori and Pacific Island people aged 15–29 and 30–44 years were due principally to type 2 diabetic nephropathy and glomerulonephritis—there was no excess of type 1 diabetic ESRD (Table 2). The disproportionately high SIRs for type 2 diabetic ESRD in non-Europid persons aged 30–44 years reflect the combined effect of excess incidence and earlier age of onset in these peoples.
|
Among the Europid populations, the significant differences in rates in persons aged 30–44 years were with respect to the frequency of (type 1) diabetic ESRD in Finland, Sweden, Denmark and Canada (Table 3), partially offset in the case of Finland by a low incidence of glomerulonephritic ESRD.
|
In persons aged 0–44 years, ESRD due to all-causes fell by 1.8% (95% CI 0 to –3.6) (P = 0.05) per year in eight European populations combined, 2.8% (95% CI –0.1 to –5.4) (P = 0.04) per year in non-Aboriginal Canadians and 1.7% (95% CI +0.2 to –5.1) (P = NS) in non-ATSI Australians. When all 10 populations were combined, the mean annual fall was 2.1% (95% CI –0.7 to –3.5) (P = 0.003) per year. Examination of trends for individual primary renal diseases indicated that the declining incidence was accounted for by falling rates of diabetic and glomerulonephritic ESRD. All ages up to 44 years were combined for this estimate to ensure sufficient statistical power, but perusal of the data showed similar trends in each of the three broad age groups.
ESRD incidence and trends in persons aged 45–64 years
In this age band, the descriptive epidemiology of ESRD changes perceptibly from that in younger persons, due to the considerable increase in rates and the change in causative renal disease, with type 2 diabetic, hypertensive and polycystic ESRD, which are uncommon at younger ages, contributing about half of the ESRD in Europid populations between 45 and 64 years of age. All-cause ESRD rates were greater in each non-Europid, than in any Europid, population. Significant differences were present within the group of Europid populations and between each of the four largest non-Europid populations (Table 1). Much of the observed variation was attributable to differences in diabetic ESRD (Tables 2 and 3). Among the Europid populations, Canada and Austria had the most, and Norway and the Basque region the least, diabetic ESRD in this age band, attributable to differences in rates of type 2 disease.
The incidence of all non-diabetic ESRD also varied, but to a smaller extent, both within the group of 14 Europid populations, and between Europid and non-Europid peoples (Tables 2 and 3). The low rates were attributable in Finland to less glomerulonephritic and hypertensive ESRD, and in Australia and New Zealand to less hypertensive ESRD, while hypertensive and polycystic renal disease contributed significantly to the high rates in Valencia. The large excess in ATSI Australians and the small excess in Aboriginal Canadians were evenly distributed between glomerulonephritis, hypertensive renal diseases and all other non-diabetic renal disease, while the excess in New Zealand Maori and Pacific Island people was greatest in hypertensive renal disease (Table 2). Polycystic kidney disease was uncommon in non-Europid peoples.
During the period 1998–2002, in persons aged 45–64 years from the combined populations of Europe (all 11 countries/regions), non-Aboriginal Canadians and non-ATSI Australians, the incidence of ESRD due to all-causes did not change significantly (–0.8%, 95% CI –2.0 to +0.4) per year. However, all non-diabetic ESRD fell by 2.5% (95% CI –1.3 to –3.7) per year (P<0.001), while diabetic ESRD rose by 3.3% (95% CI +0.9 to +5.8) per year (P = 0.007). These trends were similar in each of the three populations examined.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The purpose of this study was to describe similarities and differences in ESRD rates between a number of like and unlike populations in several regions of the world, taking into consideration the contributions made by common primary renal diseases, but avoiding errors due to variable access to treatment. Apart from Malaysia, which does not have open-access, publicly funded ESRD treatment, there was no evidence that variable access (between populations, or during the period of the study) affected any rates in persons aged <65 years. Malaysian cases may have been under-enumerated, but the similarity of their rates to those from Japan [6] in respect to variation with age, proportion of diabetic ESRD and overall incidence, indicates that they are likely to represent a good estimate of the true situation. Aboriginal Canadian cases also may have been under-enumerated because we counted all patients in whom ethnicity was not recorded as non-Aboriginal, having ascertained that most were treated in hospitals with few Aboriginal patients, and that the pattern of primary renal diseases and the age-structure of this patient group strongly resembled those of non-Aboriginal, rather than Aboriginal, Canadians. Non-Aboriginal Canadian rates could have been over-estimated, at the most by 5%, from this source of error. In non-Europid peoples, in whom life expectancy is reduced, ESRD rates also would have been under-estimated because of competing risks, due to the fact that some of the excessive premature mortality (for example, from cardiovascular disease or complications of diabetes) was likely to have been greater in persons destined to develop ESRD than in persons who would not have developed ESRD. However, much of the premature mortality results from unrelated causes, such as infection, trauma and substance abuse, which would have had no effect on ESRD rates. These three errors all contributed to under-estimation of non-Europid, relative to Europid, rates.
Age, ethnicity and frequency of diabetes were the chief sources of variation. The effect of age is well-known; it is likely to be both a risk factor in its own right, reflecting the effect of renal obsolescence on progression of renal disease, and a confounder due to its association with prevalence and duration of a variety of risk factors including primary renal disease, type 2 diabetes, hypertension and atheroma.
The greatest disease-specific variation was in respect to diabetic nephropathy. In Europid populations, differences in the population prevalence of type-specific diabetes [9] account for much of the observed variation. However, as has been demonstrated for Latinos, Asians and Blacks [10] in the US, Aboriginal Canadians [11] and Indo-Asians in the Netherlands [12], non-Europid persons with type 2 diabetes are more likely to progress to ESRD than are Europid type 2 diabetics. The protection against ESRD enjoyed by US whites, relative to Latinos, Asians and Blacks, did not extend to myocardial infarction, stroke, congestive heart failure or lower extremity amputation, indicating that non-white races experience, not a general, but a renal-specific, heightened susceptibility to diabetic vascular disease [10]. The vulnerability to type 2 diabetes, and to progressive diabetic nephropathy, both may have an origin in fetal life [13] (see subsequently).
In respect to non-diabetic ESRD, the largest between-population differences in rates were between Europid and non-Europid peoples, especially the three indigenous peoples of the South-West Pacific. Our only Asian population, Malaysia, like Japan [6], had rates for both non-diabetic and diabetic ESRD that were similar to, or lower than, those in Europid peoples up to the age of 30–40 years but rose more steeply at higher ages. The excess of non-diabetic ESRD in non-Europid peoples involved a broad spectrum of kidney pathology, comprising approximately equal contributions from glomerulonephritis, hypertensive renal disease and other non-diabetic renal disease, indicating that the increased susceptibility to progression of chronic renal failure was largely irrespective of the original primary renal disease. In our data, this vulnerability was manifest equally at all ages in Aboriginal Canadians, but was present only in adults, and rose steeply with age in the Asian-Pacific populations. This implies that there are at least some important between-race differences in pathogenesis.
Studies in two of the worst-affected communities, Aboriginal Australians living in remote locations [14] and American Indians from reservations [15,16], have shed some light on pathogenesis, and a unified hypothesis has been proposed, postulating that an increased susceptibility to ESRD, as with type 2 diabetes, originates in exposure of the fetus to an adverse intra-uterine environment due to, inter alia, maternal diabetes and/or malnutrition [13]. The preponderance of excess ESRD in middle-aged Asian and Pacific peoples is consistent with this theory if, as has been suggested [13,14], one major effect of the imperfect nephrogenesis is to hasten renal obsolescence. However, only cautious generalizations are allowable, for we have shown in this study that the propensity to ESRD differs, at least in degree and probably in nature, between our five non-Europid populations. Moreover, a hypothesis chiefly based on observations in American Indians and Aboriginal Australians may not hold for persons of African or Asian descent who also have excess risks for ESRD and type 2 diabetes [6,9,17].
On the whole, Europid peoples have rates of non-diabetic ESRD that are more similar than different. Nevertheless, it is apparent that Finland stands out as having less non-diabetic ESRD than Western or Southern Europe, and that rates are higher in US whites [17] and Valencia than in New Zealand, Australia, Sweden and Norway. Differences such as these warrant investigation because of their implications for prevention.
Cross-sectional comparisons of incidence rates are useful to identify possible risk factors for ESRD, but determination of temporal trends are of greater value because they testify to the success or to any other outcomes of recent preventive strategies, so long as sources of bias are avoided. In clinical trials, secondary prevention has been more effective against proteinuric nephropathies, which comprise the majority of incident ESRD in persons aged <45 years, than against the non-proteinuric nephropathies, such as polycystic, hypertensive and renovascular kidney diseases, that are common in older persons [18]. Thus, an unexpected finding of this study was that the favourable trend for all non-diabetic ESRD in persons aged 45–64 years was about the same as that for all-cause ESRD in persons aged 0–44 years, the latter reflecting falling rates of the two common proteinuric nephropathies: glomerulonephritis and (type 1) diabetic nephropathy. This is a heartening observation, indicating that secondary prevention may now be proving effective at the population level for a broad spectrum of non-diabetic renal disease. It is a matter of concern that diabetic ESRD has continued to increase by over 3% per year in persons aged 45–64 years in the Europid populations of this study.
Collaborating authors in the ESRD Incidence Study
Analysis and writing group: J. H. Stewart (nephrologist), M. R. E. McCredie (epidemiologist), S. M. Williams (statistician) (Department of Preventive and Social Medicine, University of Otago).
Contributing renal registries: Canadian Organ Replacement Register (CORR)—S. S. Fenton, L. Trpeski; Australia and New Zealand Dialysis and Transplant Registry (ANZDATA)—S. P. McDonald; European Renal Association-European Dialysis and Transplant Association (ERA-EDTA) Registry—K. J. Jager, P. C. W. van Dijk (Academic Medical Center, University of Amsterdam); Finnish Registry for Kidney Diseases—P. Finne; Swedish Registry for Active Treatment of Uremia—S. Schon; Norwegian Renal Registry—T. Leivestad; Danish National Registry on Regular Dialysis and Transplantation—H. Løkkegaard; Dutch-speaking Belgian Society of Nephrology (NBVN)—J.-M. Billiouw; Austrian Dialysis and Transplant Registry (OEDTR)—R. Kramar; Basque Renal Patients’ Registry (UNIPAR)—A. Magaz; Catalan Renal Registry (RMRC)—M. Cleries; Valencian Renal Registry (REMRENAL)—M. J. Garcia-Blasco; Hellenic Renal Registry—G. A. Ioannidis; Malaysian National Renal Registry—Y. N. Lim; Northern Mariana Islands’ (US) ESRD Program—I. Zahid.
|
Acknowledgments |
|---|
We thank Dr F. de Charro for data submitted from the Dutch End-Stage Renal Disease Registry (RENINE).
Conflict of interest statement. None declared.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
- Wolfe RA, Held PJ, Hulbert-Shearon TA et al. A critical examination of trends in outcome over the last decade. Am J Kidney Dis 1998; 32 [Suppl 4]: S9–S15
- Stewart JH, McCredie MRE, Williams SM, McDonald SP. Interpreting incidence trends for treated end-stage renal disease: implications for evaluating disease control in Australia. Nephrology 2004; 9: 238–246[CrossRef][Medline]
- Muntner P, Coresh J, Powe NR, Klag MJ. The contribution of increased diabetes prevalence and improved myocardial infarction and stroke survival to the increase in treated end-stage renal disease. J Am Soc Nephrol 2003; 14: 1568–1577
[Abstract/Free Full Text] - Perneger TV, Klag MJ, Feldman HI, Whelton PK. Projections of hypertension-related renal disease in middle-aged residents of the United States. JAMA 1993; 269: 1272–1277
[Abstract/Free Full Text] - Van Dijk PCW, Jager KJ, Stengel B, Grönhagen-Riska C, Feest TG, Briggs JD. Renal replacement therapy for diabetic end-stage renal disease: data from 10 registries in Europe (1991–2000). Kidney Int 2005; 67: 1489–1499[CrossRef][Web of Science][Medline]
- Wakai K, Nakai S, Kikuchi K et al. Trends in incidence of end-stage renal disease in Japan, 1983–2000: age-adjusted and age-specific rates by gender and cause. Nephrol Dial Transplant 2004; 19: 2044–2052
[Abstract/Free Full Text] - Taal MW, Brenner BM. Evolving strategies for reno-protection: non-diabetic chronic renal disease. Curr Opin Nephrol Hypertens 2001; 10: 523–531[CrossRef][Web of Science][Medline]
- Maisonneuve P, Agodoa L, Gellert R et al. Distribution of primary renal diseases leading to end-stage renal failure in the United States, Europe, and Australia/New Zealand: results from an international comparative study. Am J Kidney Dis 2000; 35: 157–165[Web of Science][Medline]
- Amos AF, McCarthy DJ, Zimmet P. The rising global burden of diabetes and its complications: estimates and projections to the year 2010. Diab Med 1997; 14: S7–S85[CrossRef][Web of Science]
- Karter AJ, Ferrara A, Liu JY et al. Ethnic disparities in diabetic complications in an insured population. JAMA 2002; 287: 2519–2527
[Abstract/Free Full Text] - Dyck RF. Mechanisms of renal disease in indigenous populations: influences at work in Canadian indigenous peoples. Nephrology 2001; 6: 3–7[Medline]
- Chandie-Shaw PK, Vandenbroucke JP, Tjandra YI et al. Increased end-stage diabetic nephropathy in Indo-Asian immigrants living in the Netherlands. Diabetologia 2002; 45: 337–341[CrossRef][Web of Science][Medline]
- Nelson RG. Intrauterine determinants of diabetic kidney disease in disadvantaged populations. Kidney Int 2003; 63 [suppl 83]: S13–S16
- Hoy WE, Rees M, Kile E et al. A new dimension to the Barker hypothesis: low birthweight and susceptibility to renal disease. Kidney Int 1999; 56: 1072–1077[CrossRef][Web of Science][Medline]
- Narva AS. The spectrum of kidney disease in American Indians. Kidney Int 2003; 63 [Suppl 83]: S3–S7
- Robbins DC, Knowler WC, Lee ET et al. Regional differences in albuminuria among American Indians: an epidemic of renal disease. Kidney Int 1996; 49: 557–563[Web of Science][Medline]
- US Renal Data System. USRDS 2004 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, 2004
- Modification of Diet in Renal Disease Study Group. Predictors of the progression of renal disease in the Modification of Diet in Renal Disease Study. Kidney Int 1997; 51: 1908–1919[Web of Science][Medline]
Accepted in revised form: 8. 3.06
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  D. Nitsch, R. Burden, R. Steenkamp, D. Ansell, C. Byrne, F. Caskey, P. Roderick, and T. Feest Patients with diabetic nephropathy on renal replacement therapy in England and Wales QJM, September 1, 2007; 100(9): 551 - 560. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Caskey, R. Steenkamp, and D. Ansell International comparison of UK registry data (Chapter 17) Nephrol. Dial. Transplant., August 1, 2007; 22(suppl_7): vii185 - vii193. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. J. Jager and P. C. W. van Dijk Has the rise in the incidence of renal replacement therapy in developed countries come to an end? Nephrol. Dial. Transplant., March 1, 2007; 22(3): 678 - 680. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||