NDT Advance Access originally published online on September 19, 2007
Nephrology Dialysis Transplantation 2008 23(2):659-664; doi:10.1093/ndt/gfm597
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Dialysis in Israel, 1989–2005—time trends and international comparisons
1Braun School of Public Health and Community Medicine, Hebrew University-Hadassah, and 2Department of Health Information and Computer Services, Ministry of Health, Jerusalem, Israel
Correspondence to: Ronit Calderon-Margalit, MD, MPH, Braun School of Public Health and Community Medicine, Hebrew University-Hadassah, POB 12272, Jerusalem 91120, Israel. Email: ronitc{at}ekmd.huji.ac.il
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and methodsResultsDiscussionConclusionsReferences |
|---|
Background. A universal increase in the incidence of renal replacement therapy (RRT) was reported in developed countries during the 1990s, especially among the elderly and diabetic patients. We studied trends in RRT incidence and mortality in Israel between 1989 and 2001–2005.
Methods. The end-stage renal disease (ESRD) registry holds data on all RRT patients in Israel. Age-adjusted incidence rate ratios (RRs) were estimated comparing 2001–2005 with 1989. We compared incidence data between Israel and elsewhere using standardized incidence ratios (SIRs). Survival analysis was conducted by the Kaplan–Meier method and Cox's proportional hazards regression was used to compare survival of diabetic with non-diabetic ESRD patients.
Results. The mean incidence rates per million population increased from 99 in 1989–1991 to 179 in 2003–2005. In 2000, Israel was the second leading country for incidence of RRT.
Age-adjusted incidence rates increased by 67% [95% confidence interval (CI): 49–87%], from 1989 to 2001, but the trend was attenuated between 2002 and 2005. The increase in incidence was positively associated with age, the largest increase being among the elderly aged 
75 years (RR: 3.18, 95%CI: 2.72–3.70). Diabetes accounted for 41% of RRT in 2001 vs only 19% in 1989. There was no increase in 1-year survival between the beginning and the end of the study period. Patients with diabetes-associated RRT had 57% increased risk of 1-year mortality (adjusted HR: 1.57 95% CI: 1.51–1.63).
Conclusions. Despite a similar proportion of RRT attributed to diabetes in Israel and other countries, the age-adjusted incidence in Israel is considerably higher than most countries.
Keywords: end-stage renal disease; incidence; renal replacement therapy; secular trends; survival
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and methodsResultsDiscussionConclusionsReferences |
|---|
Reports from developed countries on the incidence of renal replacement therapy (RRT) indicate a universal increase during the 1990s. This increase has been manifested especially among the elderly and was evident especially for end-stage renal disease (ESRD) associated with diabetes mellitus (DM) [1–4].
In Israel, RRT is provided to all patients with ESRD regardless of insurance coverage or ability to pay. During the 1970s and 1980s, Israel had one of the highest reported rates of RRT, and an increasing trend during that period has been reported [5].
In 1995, Israel experienced major new health insurance legislation, according to which the four sick funds that provide medical care to the entire population are reimbursed for dialysis treatments.
We aimed to assess the trends in RRT incidence and mortality in Israel in view of the global increasing trends and the legislative changes.
|
Subjects and methods |
|---|
|
TopAbstractIntroductionSubjects and methodsResultsDiscussionConclusionsReferences |
|---|
The Israeli ESRD database is a national administrative registry maintained by the Ministry of Health. It contains information about patients receiving RRT, i.e. haemodialysis, peritoneal dialysis or kidney transplants. All nephrology, hypertension and dialysis units in Israel report to the Ministry of Health on new RRT patients and changes in treatment modality. The database includes demographic data, a primary diagnosis, initial type of RRT, and dates of starting dialysis, change of dialysis treatment modalities, renal transplants (either in Israel or abroad) and death.
Validation of the ESRD database includes periodic linkage with the Israeli population registry to update demographic and mortality data. Reports of cadaver-donor transplants in Israel are crosschecked with the National Laboratory for Tissue Matching, and reports on living donor kidney transplants are crosschecked with the National Transplant Center.
A single primary diagnosis is recorded for each new patient in the ESRD database. The proportion of patients with missing diagnoses declined from 20% in 1989–1993 to <10% in the following years.
Data analysis
Practically, all patients who undergo renal transplant receive dialysis treatments prior to transplantation; therefore, the incidence of dialysis is a fair estimate of the incidence of ESRD in Israel.
Patients undergoing temporary RRT were excluded from the analysis. Five percent of patients, presumed to be non-Israeli residents, were excluded from the analysis due to inability to link them to the population registry.
In the current study, data for the years 1989–2001 were linked to the Israeli population registry, re-examined and edited, yielding ‘confirmed cases’. As only aggregated data were available for the years 2002–2005, data obtained for these years were used only to describe trends in incidence and prevalence of RRT, and were corrected according to the ratio of confirmed to unconfirmed cases in 2000–2001. Data on underlying diagnosis and mortality were thus available for the years 1989–2001.
Crude, age- and disease-specific incidence rates were calculated as the number of patients starting dialysis treatment during a specified year, divided by the corresponding mid-year population of Israel recorded at the Israeli Central Bureau of Statistics. Prevalence rates were calculated as the number of dialysis patients at the end of the year divided by the corresponding population. We attempted fitting linear and exponential models to the incidence time trends. For all models, R2 was estimated. Despite the curved appearance, polynomial models of the third-order did not provide a significantly better fit. For time trend curves and rate ratios (RRs), we used the 3-year moving average incidence rates.
Rates were age-adjusted by applying the direct method, using the population of Israel in 1996 as the standard. For all rates and RRs, Fisher's exact 95% confidence intervals (CI) were calculated using the PEPI computer programs for epidemiologists software (version 4.04).
Survival analysis was conducted by the Kaplan–Meier method. Data were right censored upon patients' receipt of a renal transplant, permanently leaving the country or at the end of 2001. The 1-year survival of patients who started dialysis treatments at the beginning of the study period (1989–1991) was compared to the survival of patients who started dialysis treatment a decade later by age strata (0–17, 18–44, 45–64 and 65+ years). A comparison of survival between diabetic and non-diabetic patients was conducted, using Cox's proportional hazards regression with time-to-event as the dependent variable. The adjusted hazard ratio (HR) of death of diabetic patients vs those without diabetes was obtained, controlling for age, sex and date of onset of RRT.
To compare the Israeli incidence rates of ESRD with international data we used the indirect standardization method. Standardized incidence ratios (SIR) were calculated as the ratio of observed to expected incidence rates of RRT. Observed rates were obtained from published material [2,6–10]. Expected rates were calculated by applying the Israeli age- and sex-specific rates of dialysis treatment to the population distribution of every country as reported in the US census international database [http://www.census.gov/ipc/www/idbpyr.html]. Standard errors and 95%CI were calculated using the PEPI software following the methods of Armitage, Berry and Kahn.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and methodsResultsDiscussionConclusionsReferences |
|---|
Incidence and prevalence
During the years 1989–2005, a total of 14 326 patients received dialysis treatments in Israel due to ESRD. In 1989, patients receiving peritoneal dialysis comprised 30% of dialysis patients, while in 2005 only 10% of dialysis patients were treated with peritoneal dialysis. The proportion of new patients aged 65 years or more increased from 34% in 1989 to 63% in 2005. The largest increase was in the proportion of patients aged 75 years or more (from 10% to 30% in both incident and prevalent cases). Throughout the study period, three-fifths of all patients on dialysis were males. Males had higher incidence and prevalence rates than females with RRs of 1.93 (95%CI: 1.85–2.01) and 1.79 (95% CI: 1.75–1.83), respectively. There were no temporal trends in the male : female ratio, either in incidence or prevalence rates of dialysis.
In 1989–1991, the mean prevalence and incidence rates per million population were 303 and 99, respectively. In 1999–2001, these rates had increased to 524 and 170, and by 2003–2005 they were 606 and 179, respectively. Israel encountered an overall average annual increase in incidence of 4.2%. However, this was not consistent. The average annual increase between 1989 and 1993 was 3.9%, and between 1994 and 1999 was 8.1%, whereas in 2000–2005 the average annual increase was 0.4%. The corresponding annual increase rates in prevalence were 4.7, 6.3 and 3.4%. Neither linear nor exponential models seemed to explain well the increase in incidence, and although the best fit was for the polynomial regression of the 3rd order (R2 = 0.995), this model was not significantly better than the 2nd order regression.
The age-adjusted incidence rates increased by 67% (95%CI: 49–87%), comparing 2001 with 1989. This increasing trend was, however, attenuated between 2002 and 2005. Between 1989 and 2001, the age-adjusted prevalence increased by 71% (95% CI: 61–81%) and by 8.6% (95% CI: 4–14%) thereafter (Table 1).
|
The increase in incidence during the study period was positively associated with age. Comparing 2003–2005 with 1989–1991, there was no increase among those younger than 55 years of age, but the incidence rates were multiplied by 1.28 (95%CI: 1.12–1.46), 2.18 (95%CI: 1.93–2.47) and 3.18 (95%CI: 2.72–3.70) among the 55–64, 65–74 and
75 year groups, respectively. Age-specific rates are presented in Table 1. A test for interaction of age and time trend was significant (P < 0.001, Mantel–Haenszel test). The proportion of new patients receiving dialysis for whom diabetes was listed as the underlying cause increased from 19% in 1989 to 41% in 2001. The diabetes and non-diabetes-specific rates are presented in Table 1. The incidence of diabetes-associated ESRD increased by 3.36 (95% CI: 2.93–3.87), comparing the 3-year average incidence between 1999–2001 and 1989–1991. The corresponding non-diabetes-associated ESRD increased by 1.62 (95% CI: 1.49–1.75). The age-specific rate ratios of ESRD associated with DM increased significantly with age with RRs of 1.80, 2.07, 2.29, 4.37 and 7.18 for ages 18–44, 45–54, 55–64, 65–74 and 75+ years, respectively (P for interaction<0.001).
We studied a best-worst scenario to assess the effect of missing diagnoses on the ratios of diagnostic-specific incidence rates, assuming once that all missing diagnoses in 1999–2001 were diabetes and all missing diagnoses in 1989–1991 were not diabetes and then inverting both assumptions. The resulting RRs ranged between 1.66 and 3.91 for diabetes and between 1.25 and 1.77 for non-diabetes ESRD.
Survival
One year after starting dialysis treatment nearly one-fifth of all patients had died, nearly one-tenth had a functioning renal transplant and three-quarters of patients remained on dialysis treatments. These proportions were constant during the years 1989 and 2001. During this period, the 1-year mortality rates ranged between 0% and 8% for patients aged <45 years, 10–19% for patients aged 45–64 and 22–31% for patients aged 65 years and over. No increase in age-specific 1-year survival was noticed comparing patients who started dialysis in 1989–1991 with those who started treatments in 1998–2000 (Table 2).
|
The 1-, 3- and 5-year survival rates among diabetic patients were 78, 46 and 24% compared with 84, 60 and 43% among non-diabetics. The unadjusted HR for all-cause mortality comparing diabetic with non-diabetic patients was 1.60 (95% CI: 1.54–1.66). Adjustment for age, gender and dialysis starting date did not materially change the HR (adjusted HR 1.57, 95%CI: 1.51–1.63).
International comparisons of incidence data
Indirect age standardization of the incidence data in 2000 from the US, the Netherlands, Japan, Norway, Germany, Australia, New Zealand, Greece, Denmark, Sweden (2001), Belgium (1989–1999), the UK [2,6–10] and Israel, showed that the incidence in the US ranked the highest, with Israel as the second leading country in incidence and the UK ranked as the country with lowest incidence of RRT (Table 3). The proportion attributed to diabetes among new patients receiving RRT in Israel (37%) was similar to the proportions reported in Japan, New Zealand and Germany, lower than the proportion of patients in the US (44.5%) and higher than the proportions reported in all other countries studied.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and methodsResultsDiscussionConclusionsReferences |
|---|
The incidence of ESRD, as measured by RRT, increased between 1989 and 2001 in Israel, similar to other developed countries [2,8,10,11]. This increase was noted especially in diabetes-associated ESRD and ESRD among the elderly. Israel encountered an overall average annual increase of 4.2%, an increase that is similar to other European countries [1], and lower than the 6.9% mean annual increase noted in the US [12]; however, while in Europe the trend has been described as linear, and in the US, Japan, and the Netherlands, rates had stabilized towards the end of the 1990s [1,2,10], the increase in Israel during the study period seemed to accelerate in the 1990s and to level-off in the 2000s.
As of 2000, Israel had the second highest age-adjusted incidence rate of dialysis treatments after the US. Compared with the Israeli incidence data, European countries as well as Australia and New Zealand had 33–67% lower incidence. These international variations could be partially explained by different incidence and prevalence of chronic renal failure (CRF) or by different progression rates from CRF to ESRD [13]. The proportion of RRT attributed to diabetes in Israel is one of the highest in the countries studied and second only to the US. Germany and New-Zealand have similar proportions attributed to diabetes and yet these countries had considerably lower incidence of RRT than Israel. Moreover, according to international comparisons of the prevalence of diabetes among adults aged 20 years or more in the year 2000, the prevalence of self-reported diabetes in Israel (6.4%) was similar to that reported for Australia, New-Zealand and Japan and lower only to the rates reported in Greece and the US [14]. Therefore, despite a relatively high prevalence of diabetes and relatively high proportion of ESRD attributed to diabetes in 2000–2001, the frequency of diabetes itself could not entirely explain the high incidence of ESRD in Israel.
A decrease in competing risks for mortality, together with a change in administrative regulations as to eligibility for RRT could allow older and/or sicker individuals to benefit from dialysis treatments. Having no official age or comorbidity restrictions on the eligibility for dialysis treatments, we cannot assume that the increase is due to a change in such administrative regulations. During the 1990s, in Israel, life expectancy at age 65 years increased from 16.8 to 18.5 among females and from 15.2 to 16.4 years among males [http://www1.cbs.gov.il/shnaton57/st03_23.pdf], reflecting a decrease in mortality rates of the elderly population in Israel. The increase in incidence of dialysis treatments among the elderly in Israel, as well as other countries [4,12], could therefore represent a decrease in incidence of other serious illnesses or a change in doctors’ attitudes. One of this study's limitations is lack of data on comorbidities, other than the underlying disease, to further explore this hypothesis.
The increase in incidence might be attributed to changes in policy of enrollment of patients to dialysis treatments. A birth cohort analysis in the setting of the American NHANES II and III [12] indicated that incidence of RRT grew faster than the prevalence of chronic renal failure, suggesting policy as a major contributing factor. In Israel, since 1995, all sick funds are reimbursed for dialysis treatment. Such reimbursement could theoretically contribute to the increase in incidence of RRT. The accelerated rate of increase in incidence before 1995 would appear to contradict a legislative effect on decision making. The 1997 NKF-DOQI guidelines that indicated an increased GFR (glomerular filtration rate) threshold for initiating dialysis [15] have probably been responsible for at least part of the increase in incidence and prevalence of RRT evident by the end of the 1990s in other countries [3,4] as well as Israel. Although our data lack renal function, we assume that these guidelines cannot explain the entire increase, since this study demonstrates an increase almost a decade before the guidelines were published. Similarly, incidence rates were already increasing in Japan in the 1980s [2] prior to the publication of these guidelines, and trends in Europe in 1990–1999 have been described as linear [1], arguing against a major effect of these guidelines.
Another explanation for the increase in incidence and prevalence rates during the study period is demographic changes in Israel. During the 1990s, 1 million immigrants from the Former Soviet Union arrived in Israel. These immigrants were characterized by a high prevalence of chronic diseases [16]. Although there is no information on the prevalence of CRF or ESRD among these newcomers, it is reasonable to assume that having a higher proportion of hypertension than the Israeli population [16] and presumably also of diabetes, these immigrants should also have had a high prevalence of renal diseases. According to the 2003–2004 Israeli national health survey, former USSR immigrants reported about twice the rate of any renal disease compared to the Israeli adult population of (all: 13.3 vs 6.3%, 65 years of age: 21.6 vs 12.8%, respectively) (Central Bureau of Statistics and the Ministry of Health, personal communication).
The attenuation in the rate of increase in incidence, and to a lesser extent in prevalence, since 2000, could be partly explained by the dramatic decrease in in-migration of this population. Others have ascribed the decrease seen in the US to improvements in tertiary prevention (i.e. treatment) of DM and hypertension [17].
Survival rates among diabetic and non-diabetic patients as shown in this study are similar to those reported in Europe [11,18,19], Australia [20] and somewhat lower than those reported in Canada [21]. All reports demonstrated poorer survival among patients with diabetes compared with non-diabetic patients. Our data do not show improvement in survival during the 1990s and are in agreement with Jager et al. [18]. It is reasonable to assume that if patients are indeed enrolled to dialysis earlier in the course of their disease even without having any improvements in survival, an artificial longer survival (lead time bias) would occur. One must question why there was no improvement. A possible explanation for the stable 1-year survival rates is that while patients are starting RRT at an earlier stage of CRF, sicker patients, with more comorbidities are also starting treatment [22]. Alternatively, it could also be that starting dialysis earlier does not improve survival [23]. The 11% increase in age-adjusted prevalence rates after 2001 together with leveling-off of incidence rates might suggest improved survival after year 2000. Future studies are needed to confirm this hypothesis.
|
Conclusions |
|---|
|
TopAbstractIntroductionSubjects and methodsResultsDiscussionConclusionsReferences |
|---|
Incidence of ESRD in Israel is one of the highest in the world. The explanations for these high incidence rates are not clear. An increasing trend was noted in the 1990s especially, but not exclusively, in ESRD attributed to diabetes and among elderly patients. This increase seemed to stabilize in the early 2000s. Expansion of the database with additional follow-up and renal function data will enable further understanding of these trends. Lack of improvement in survival of dialysis patients warrants thorough investigation.
Conflict of interest statement. The results presented in this article have not been published previously in whole or in part, except in Hebrew language reports by the dialysis registry.
|
References |
|---|
|
TopAbstractIntroductionSubjects and methodsResultsDiscussionConclusionsReferences |
|---|
- Stengel B, Billon S, Van Dijk PC, et al. Trends in the incidence of renal replacement therapy for end-stage renal disease in Europe, 1990–1999. Nephrol Dial Transplant (2003) 18:1824–1833.
[Abstract/Free Full Text] - 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] - Termorshuizen F, Korevaar JC, Dekker FW, et al. Time trends in initiation and dose of dialysis in end-stage renal disease patients in The Netherlands. Nephrol Dial Transplant (2003) 18:552–558.
[Abstract/Free Full Text] - Stewart JH, McCredoe MR, 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]
- Lubasch S, Atzmoni S, Better OS, Epstein L. End-stage renal disease replacement therapy in Israel: demographic and regional differences. Isr J Med Sci (1989) 25:444–450.[Web of Science][Medline]
- UK Renal Registry, 2002. Available at: http://www.renalreg.com/Slides2002/Report%202002.pdf. Accessed 1 April 2007.
- ERA-EDTA annual report 2001. Available at: http://www.era-edta-reg.org/files/annualreports/pdf/AnnRep2001.pdf. Accessed: 1 April 2007.
- Australia and New Zealand Dialysis and Transplant Registry (ANZDATA). In: 24th Annual report (2001) Available at: http://www.anzdata.org.au/anzdata/AnzdataReport/24thReport/files/Ch.02%20new%20pts%202001.pdf. Accessed: 20 March 2005.
- Frei U, Schober-Halstenberg HJ. Annual Report about dialysis treatment and renal transplantation in Germany 2000. Available at: http://www.quasi-niere.de/berichte/online/en/00/report00_2.html. Accessed 1 April 2007.
- United States Renal Data System (USRDS). Reference tables. available at: http://www.usrds.org/reference.htm. Accessed 1 April 2007).
- van Dijk PC, Jager KJ, de Charro F, et al. ERA-EDTA registry. Renal replacement therapy in Europe: the results of a collaborative effort by the ERA-EDTA registry and six national or regional registries. Nephrol Dial Transplant (2001) 16:1120–1129.
[Abstract/Free Full Text] - Hsu CY, Vittinghoff E, Lin F, Shlipak MG. The incidence of end-stage renal disease is increasing faster than the prevalence of chronic renal insufficiency. Ann Intern Med (2004) 141:95–101.
[Abstract/Free Full Text] - Hallan SI, Coresh J, Astor BC, et al. International comparison of the relationship of chronic kidney disease prevalence and ESRD risk. J Am Soc Nephrol (2006) 17:2275–2284.
[Abstract/Free Full Text] - Wild S, Roglic G, Green A, Sicree R, King H. Global prevalence of diabetes: estimates for the year 2000 and projections for 2030. Diabetes Care (2004) 27:1047–1053. Appendix available at: http://care.diabetesjournals.org/content/vol27/issue5/images/data/1047/DC1/Wild_Appendix.xls. Accessed 1 April 2007.
[Abstract/Free Full Text] - NKF-DOQI clinical practice guidelines for hemodialysis adequacy: National Kidney Foundation. Am J Kidney Dis (1997) 30:S15–S66.[Web of Science][Medline]
- Baron-Epel O, Kaplan G. Self-reported health status of immigrants from the former Soviet Union in Israel. Isr Med Assoc J (2001) 3:940–946.[Web of Science][Medline]
- Centers for Disease Control, Prevention (CDC). Racial differences in trends of end-stage renal disease, by primary diagnosis–United States, 1994–2004. MMWR Morb Mortal Wkly Rep (2007) 56:253–256.[Medline]
- Jager KJ, van Dijk PC, Dekker FW, Stengel B, Simpson K, Briggs JD. The epidemic of aging in renal replacement therapy: an update on elderly patients and their outcomes. Clin Nephrol (2003) 60:352–360.[Web of Science][Medline]
- Van Dijk PC, Jager KJ, Stengel B, Gronhagen-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]
- McDonald SP, Russ GR, Kerr PG, Collins JF. ESRD in Australia and New Zealand at the end of the millennium: a report from the ANZDATA registry. Am J Kidney Dis (2002) 40:1122–1131.[CrossRef][Web of Science][Medline]
- Lok CE, Oliver MJ, Rothwell DM, Hux JE. The growing volume of diabetes-related dialysis: a population based study. Nephrol Dial Transplant (2004) 19:3098–3103.
[Abstract/Free Full Text] - Kazmi WH, Gilbertson DT, Obrador GT, et al. Effect of comorbidity on the increased mortality associated with early initiation of dialysis. Am J Kidney Dis (2005) 46:887–896.[CrossRef][Web of Science][Medline]
- Traynor JP, Simpson K, Geddes CC, Deighan CJ, Fox JG. Early initiation of dialysis fails to prolong survival in patients with end-stage renal failure. J Am Soc Nephrol (2002) 13:2125–2132.
[Abstract/Free Full Text]
Accepted in revised form: 1. 8.07
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||