 |
||||
| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
|
||||
| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
,
,¶,¶,#,¶,#
*Department of Epidemiology, School of Public Health and Tropical Medicine, Tulane University, New Orleans, LA; Department of Medicine, School of Medicine, Tulane University, New Orleans, LA; Departments of Epidemiology, Biostatistics, ¶Medicine, #Health Policy and Management, The Bloomberg School of Hygiene and Public Health and Department of Medicine, Johns Hopkins University, Baltimore, Maryland.
Correspondence to Dr. Paul Muntner, Department of Epidemiology, Tulane University SPHTM, 1430 Tulane Avenue, SL-18, New Orleans, LA 70112. Phone: 504-988-1047; Fax: 504-988-1568;
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionAppendixReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAppendixReferences |
|---|
The incidence of treated end-stage renal disease has increased at an annual rate of 8% over the past 20 yr (Figure 1) (11). In 1978, fewer than 15,000 people initiated treatment for end-stage renal disease, in contrast to more than 50,000 in 1991 and 96,000 in 2000 (11). According to a recent USRDS forecast (12), more than 172,000 persons are expected to begin treatment in 2010 (Figure 1), making chronic kidney disease an important area to focus prevention efforts. However, the contribution of factors that may underlie the unabated increase in treated end-stage renal disease incidence are poorly understood (13).
|
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAppendixReferences |
|---|
NHANES II and III
NHANES II and III were designed to allow the estimation of the prevalence of common chronic conditions, including diabetes mellitus and a history of MI and stroke, for the civilian, non-institutionalized population of the United States. NHANES II data were collected between 1976 and 1980 (midpoint: July 1978) and NHANES III from 1988 to 1994 (midpoint: April 1991). In addition to demographic information (i.e., age, race, sex), both NHANES surveys used in-home interviews to assess the presence of chronic health conditions including self-report of diabetes, MI, and stroke (Appendix B). Sampling weights to achieve US population estimates and variance estimators that account for the complex survey design employed in NHANES were applied in all calculations (18).
United States Renal Data System
Comprehensive data, including date of initiation of chronic outpatient therapy for end-stage renal disease, primary cause of end-stage renal disease and demographic characteristics are collected on every Medicare-eligible case of treated end-stage renal disease by the National Institutes of Diabetes and Digestive Kidney Diseases. Before 1996, data on comorbid conditions were not collected on all incident treated end-stage renal disease cases. However, in 1990, co-morbidity data were collected through medical chart review on a representative sample of treated end-stage renal disease cases from randomly selected dialysis centers in a special study (Case-Mix Adequacy Study). Therefore, for the current analysis, we assessed the prevalence of diabetes and a history of MI and stroke among the 1,903 Case-Mix Adequacy Study participants initiating end-stage renal disease treatment in 1990. The primary goal of the Case-Mix Adequacy Study were to evaluate the relationship of dialysis dose and dialyzer membrane on patient outcomes. Data collected as part of the Case-Mix Adequacy Study were abstracted from patient records kept at the dialysis facility where each patient received care. A copy of the abstraction form has been previously published.
Statistical Methods
The number of incident cases of treated ESRD was plotted by calendar year from 1978 through 1991. To investigate the increase in ESRD with the assigned cause of diabetes, incident counts were calculated by underlying cause of treated ESRD, diabetes or not diabetes, as assigned by the diagnosing physician. The difference in number of ESRD cases between 1978 and 1991 with an underlying cause of diabetes was divided by the overall increase in number of ESRD cases to estimate the impact of persons with an underlying diagnosis of diabetes on the incidence of ESRD.
Change in the number of persons in the US with diabetes, a history of MI or stroke, and without these conditions, between 1978 and 1991, was calculated as the difference in population estimates for each group using NHANES II and III (Appendix A; panel 2). Due to these conditions’ strong associations with age, calculations were performed using five age groupings (30 to 44, 45 to 51, 55 to 64, 65 to 74, and 
75 yr). NHANES II only included US citizens under age 75 yr; we therefore compared the prevalence of diabetes, MI, and stroke in 1978 and 1991 within each age category and imputed the average change to persons aged 75 yr and older for the 1978 estimate (Appendix A; panel 3).
The incidence rates of treated end-stage renal disease for persons with diabetes, with a history of MI or stroke, and without these conditions in 1991 were calculated as the number of incident cases of treated end-stage renal disease with each co-morbid condition divided by the number of people in the US population with the same condition (Appendix A; panel 4). The number of incident cases of treated end-stage renal disease with diabetes mellitus was estimated as the product of the total number of incident cases of treated end-stage renal disease in the US in 1991, obtained from the core United States Renal Data System data set, and the prevalence of diabetes among cases initiating end-stage renal disease therapy in 1990, as sampled in the Case-mix Adequacy Study (Appendix A; panel 5). Analogous methods were used to calculate the number of incident treated end-stage renal disease cases among persons with a history of MI or stroke and also for the population without diabetes, a history of MI, or stroke. Due to the low number of incident cases of treated end-stage renal disease between 18 and 29 yr old sampled in the Case-mix Adequacy Study (n = 82), we limited our data analysis to persons 30 yr and older.
Two sets of sensitivity analyses were performed. First, sensitivity analyses were performed by repeating all analyses stratified by race (White and Black), using four mutually exclusive categories (a history of MI/stroke only, diabetes mellitus only, both a history of diabetes mellitus and MI/stroke, and none of these conditions), and by both race stratification and use of mutually exclusive categories. Finally, analyses for diabetes were repeated capturing undiagnosed diabetes by defining diabetes as either a self-report of diabetes or a fasting plasma glucose 126 mg/dl.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAppendixReferences |
|---|
30 yr of age, respectively (total increase n = 36,881). In 1978, 1281 (10%) of the incident cases of ESRD 30 yr of age were diagnosed with the assigned cause of diabetes compared with 18,218 (37%) in 1991. Overall, the increase in incident treated ESRD cases among the US population 30 yr of age with an assigned cause of diabetes (n = 16,937) is equivalent to 46% (16,937/36,881) of the overall increase in ESRD between 1978 and 1991.
US Population Characteristics
Characteristics of the overall US population, persons with diabetes, persons with a history of MI or stroke, and incident cases of treated end-stage renal disease older than 30 yr of age in 1978 and 1991 are listed in Table 1. Compared with the general US population, in both years examined, persons with diabetes were older and more likely to be African-American, have a history of MI and stroke, and have hypertension. Additionally, those with a history of MI or stroke and incident cases of treated end-stage renal disease were older, and more likely to be male and have diabetes and hypertension. Overall, the US population grew by 26.7 million persons. As stated previously, the incident treated ESRD population increased by 36,881 cases from 1978 and 1991.
|
75 yr of age was used to estimate a diabetes prevalence of 8.3% for this population in 1978. Overall, between 1978 and 1991, the number of persons with diabetes mellitus in the US increased by 4.282 million.
|
75 yr of age in 1978 of 23.8% was imputed from 22.6% in 1991. In 1991, there were 1.224 million more persons with a history of MI or stroke compared to 1978. Additionally, there were 22.261 million more persons in the US without diabetes or a history of MI or stroke.
Incidence of Treated End-Stage Renal Disease
Overall, 50% and 27% of incident cases of treated end-stage renal disease in 1990 in the Case-mix Adequacy Study had a history of diabetes and history of MI or stroke, respectively (Table 3, column c). In contrast, 39% of this population had none of these conditions. The prevalence of these co-morbidities varied by age. The overall and age-specific incidence rates of treated end-stage renal disease was higher among persons with diabetes mellitus and persons with a history of MI or stroke when compared with the general US population (Figure 2 and columns f and g of Table 3). In contrast, the incidence was lower among persons without diabetes mellitus or a history of MI or stroke.
|
|
|
|
Additional sensitivity analyses were performed defining diabetes as either self-report of a previous diagnosis or fasting plasma glucose 126 mg/dl. Using this definition of diabetes, there were 5.6 million more persons in the US population with diabetes in 1991 compared with 1978 (compared with 4.3 million using self-report). However, compared with the incidence of treated ESRD among persons with self-reported diabetes, the incidence of treated ESRD within each age group was lower among those with either self-reported or undiagnosed diabetes. The larger increase in number of persons with diabetes but lower incidence rate of treated ESRD associated with this broader definition of diabetes offset one another, and the overall increase in incident treated ESRD cases was similar (n = 10,673) to the analyses that relied on self-report to define diabetes (Table 4, n = 10,183).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAppendixReferences |
|---|
On the basis of these reports, we expected a large percentage of the increase in treated end-stage renal disease incidence to be due to an increased number of MI and stroke survivors. In the current study, the age-specific risk of treated end-stage renal disease was much higher for persons with a history of MI or stroke compared with the overall population. However, in each age group, the change in MI and stroke prevalence was small, and the associated standard error for this change was large, indicating that the prevalence of MI or stroke may not have changed substantially between 1978 and 1991. This may explain our finding that this population is responsible for only a small percentage of the increase in treated end-stage renal disease incidence.
Although the effect of a lower MI and stroke case-fatality rate on treated ESRD incidence was directly analyzed in the current study, the impact of primary prevention of MI and stroke was not. It is possible that the successful efforts aimed at preventing MI and stroke has resulted in a longer life expectancy and, as such, more time at risk for developing ESRD requiring treatment. Unfortunately, the current analysis did not have data available to assess such secondary effects. In contrast, this reasoning seems incongruous with the purported renoprotective effects of statins (19) and certain anti-hypertensive medications (20). However, it can be argued that the methodology used in the current study indirectly accounts for increased longevity in the "Neither MI/CVA or diabetes" group.
The results shown in Table 4 provide details of the impact of each of the risk factor investigated stratified by age. The increased prevalence of diabetes accounts for incident ESRD cases within each age group, while the higher number of MI/stroke survivors in 1991 primarily impacts the older age groups, and the increased population size without diabetes or a history of MI/stroke had a substantial impact on both the 30 to 44 yr olds and the 75 yr old group. Overall, these results appear to make intuitive sense. There was a large increase in diabetes prevalence among all age groups. In contrast, most persons with a history of MI/stroke were older than 65. Finally, the large increase in the US population 30 to 44 yr of age (22.3 million) may explain the increase in cases of ESRD in this age group and the moderate increased size of the US population 75 yr of age, a population with a high incidence of treated ESRD, explained a substantial proportion of the increase in treated ESRD among persons 75 yr of age.
The results of the current study should not be interpreted as suggesting that diabetes, MI, and stroke are not important risk factors for treated end-stage renal disease. In contrast, compared with the general population of similar age, as shown in Figure 2, the risk of treated end-stage renal disease is substantially higher among persons with diabetes and a history of MI or stroke. While the majority of the increase in treated end-stage renal disease incidence during the study period examined may not have resulted from an increased prevalence of these conditions, the high end-stage renal disease risk and large number of persons with diabetes, MI, and stroke in the US indicate that the overall impact of these conditions on end-stage renal disease incidence are substantial.
Reasons for a majority of the increase in treated end-stage renal disease incidence remain unresolved. However, broader access to dialysis therapy among persons with renal failure in the US has been reported and may explain a substantial proportion of the higher incidence of treated end-stage renal disease. For example, after initial Medicare entitlement for dialysis therapy, older adults may have had limited access to treatment. In 1978, 30% and 7% of incident treated end-stage renal disease patients were greater than 65 and 75 yr of age, respectively, compared with 50% and 19% in 1991. This secular trend may reflect more liberal access to dialysis therapy for older persons in the US. Furthermore, although 28% of the increase in ESRD was attributable to an increased prevalence of diabetes, the percentage of incident treated ESRD cases with an assigned cause of diabetes increased from 10% to 37%, an absolute increase of n = 16,937, between 1978 and 1991. Also, approximately 50% of incident ESRD cases in 1991 had diabetes. While the current study concluded that 10,183 of incident ESRD cases in 1991 were due to an increased prevalence of diabetes, less restrictive access to dialysis therapy or more severe diabetes may explain a substantial proportion of the increase in ESRD among persons with diabetes. However, the absence of barriers to end-stage renal disease therapy was noted in 1992, and the incidence of end-stage renal disease is still increasing (13).
Other potential reasons for the increase in ESRD may be higher usage rates of nephrotoxic agents such as analgesic medications. Although environmental exposure to lead at the population level has decreased dramatically, other environmental exposures may be related to the increase in ESRD. Unfortunately, data were not collected to permit the systematic analysis of these risk factors in the investigation of the increase in ESRD. Because the forecasted Medicare expenditures for ESRD are projected to increase to $28.3 billion by 2010, population-level research including the analysis of the USRDS may help better understand the increase in ESRD and avenues for prevention
This study has several strengths. First, it is a systematic study of the increase in treated end-stage renal disease incidence that occurred between 1978 and 1991. Understanding the secular changes that have impacted treated end-stage renal disease incidence is important in both defining the end-stage renal disease epidemic and may also be imperative in end-stage renal disease prevention. Second, the data used in the current study were derived from nationally representative samples of the US population. The second and third National Health and Nutrition Examination Surveys each include interview response from more than 14,000 participants 30 yr of age and older. Data were also used from the United States Renal Data System; a registry containing records for more than 90% of incident treated end-stage renal disease cases. Finally, because many MI and stroke survivors also have diabetes, we repeated all of the analyses using four mutually exclusive categories (a history of MI/stroke only, a history of diabetes mellitus only, a history of both diabetes mellitus and MI/stroke, and none of these factors) for the overall population and by race. Although the standard errors were larger, the results were similar using this alternate approach. Such consistency provides confidence in the robustness of our study results.
Despite the strengths of using national population-level data, there are some limitations to our methodology. First, the data components necessary for the current analysis were only available in 1978 and 1991. As new data become available, analysis of treated end-stage renal disease over a longer time period will be useful. Second, the assessment of co-morbidity differed across the data sets that we used. The United States Renal Data System relied on chart abstraction, and NHANES II and III relied on self-report. Patient recall can be inaccurate even for major events such as MI and stroke (21,22). While the methodology used provides confidence that the NHANES II and III estimates are comparable, using two separate data sources (NHANES and USRDS) may have produced a bias in the calculated incidence rate. Nonetheless, the National Center for Health Statistics surveys provides the best available estimates of morbidity prevalence for the US population. In addition, estimates of the relative risks of treated end-stage renal disease associated with diabetes and MI obtained in the general population using this methodology are similar to previously published relative risks from observational studies (6,8).
Our estimates of the proportion of treated end-stage renal disease attributable to a higher prevalence of diabetes and MI and stroke survival may be conservative because the lack of data on the presence of a history of MI or stroke at the time end-stage renal disease treatment was initiated in 1978. It is quite likely that persons with diabetes or a MI and stroke were less likely to receive end-stage renal disease treatment in 1978 compared with 1991. Therefore, these populations may explain a larger proportion of the increase in the incidence of end-stage renal disease treatment. For example, under the scenario that no MI or stroke survivors initiated end-stage renal disease therapy in 1978, the increase in treated end-stage renal disease attributed to this population would account for 37% (13,786/36,881) of the total increase in treated end-stage renal disease incidence between 1978 and 1991. Applying this scenario to persons with diabetes mellitus, 67% (24,676/36,881) of the increase in treated end-stage renal disease incidence could be attributed to this population. However, such attribution is misleading in that the increase in treated end-stage renal disease incidence could arguably be attributed to broader access to dialysis therapy among persons with co-morbid conditions rather than increased prevalence of these co-morbidities.
Although persons with diabetes, a history of MI, and a history of stroke are at very high risk for treated end-stage renal disease, the current analysis indicates that reasons other than a higher prevalence of diabetes, improved MI and stroke survival, and an increased US population size explain the majority of the increase in treated end-stage renal disease incidence between 1978 and 1991. Given the results of the current study, it remains possible that more liberal access to therapy or greater severity of kidney disease in these populations may explain a majority of the increase in ESRD therapy. Costs for ESRD treatment have increased to over $15 billion and abatement of these costs and the incidence of treatment for ESRD are not expected anytime soon. To stem the tide of rising ESRD incidence and costs, it is first necessary to determine reasons for the increase in treated ESRD. Therefore, the continued study of explanations for the increase in ESRD incidence in the US is necessary.
|
Appendix |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAppendixReferences |
|---|
|
|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAppendixReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  S. V. Jassal MB BCh MD, L. Trpeski MPH MD, N. Zhu MD, S. Fenton MD, and B. Hemmelgarn PhD MD Changes in survival among elderly patients initiating dialysis from 1990 to 1999 Can. Med. Assoc. J., October 23, 2007; 177(9): 1033 - 1038. [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]
|
|
|
|
 |
|
 M. Kurella, K. E. Covinsky, A. J. Collins, and G. M. Chertow Octogenarians and Nonagenarians Starting Dialysis in the United States Ann Intern Med, February 6, 2007; 146(3): 177 - 183. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C.-y. Hsu, A. S. Go, C. E. McCulloch, J. Darbinian, and C. Iribarren Exploring Secular Trends in the Likelihood of Receiving Treatment for End-Stage Renal Disease Clin. J. Am. Soc. Nephrol., January 1, 2007; 2(1): 81 - 88. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Kiberd The Chronic Kidney Disease Epidemic: Stepping Back and Looking Forward J. Am. Soc. Nephrol., November 1, 2006; 17(11): 2967 - 2973. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
The ESRD Incidence Study Group 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 Nephrol. Dial. Transplant., August 1, 2006; 21(8): 2178 - 2183. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Stengel and C. Couchoud Chronic Kidney Disease Prevalence and Treated End-Stage Renal Disease Incidence: A Complex Relationship J. Am. Soc. Nephrol., August 1, 2006; 17(8): 2094 - 2096. [Full Text] [PDF]
|
|
|
|
|
|
W. M. McClellan The epidemic of renal disease--what drives it and what can be done? Nephrol. Dial. Transplant., June 1, 2006; 21(6): 1461 - 1464. [Full Text] [PDF]
|
|
|
|
 |
|
 P. Muntner, K. B. DeSalvo, R. P. Wildman, P. Raggi, J. He, and P. K. Whelton Trends in the Prevalence, Awareness, Treatment, and Control of Cardiovascular Disease Risk Factors among Noninstitutionalized Patients with a History of Myocardial Infarction and Stroke Am. J. Epidemiol., May 15, 2006; 163(10): 913 - 920. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Hegarty, R.J. Middleton, M. Krebs, H. Hussain, C. Cheung, T. Ledson, A.J. Hutchison, P.A. Kalra, H.C. Rayner, P.E. Stevens, et al. Severe acute renal failure in adults: place of care, incidence and outcomes QJM, September 1, 2005; 98(9): 661 - 666. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. M. Stuveling, S. J. L. Bakker, H. L. Hillege, P. E. de Jong, R. O. B. Gans, and D. de Zeeuw Biochemical risk markers: a novel area for better prediction of renal risk? Nephrol. Dial. Transplant., March 1, 2005; 20(3): 497 - 508. [Full Text] [PDF]
|
|
|
|
|
|
J. Coresh, D. Byrd-Holt, B. C. Astor, J. P. Briggs, P. W. Eggers, D. A. Lacher, and T. H. Hostetter Chronic Kidney Disease Awareness, Prevalence, and Trends among U.S. Adults, 1999 to 2000 J. Am. Soc. Nephrol., January 1, 2005; 16(1): 180 - 188. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C.-y. Hsu, E. Vittinghoff, F. Lin, and M. G. Shlipak The Incidence of End-Stage Renal Disease Is Increasing Faster than the Prevalence of Chronic Renal Insufficiency Ann Intern Med, July 20, 2004; 141(2): 95 - 101. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. M. McClellan, R. D. Langston, and R. Presley Medicare Patients with Cardiovascular Disease Have a High Prevalence of Chronic Kidney Disease and a High Rate of Progression to End-Stage Renal Disease J. Am. Soc. Nephrol., July 1, 2004; 15(7): 1912 - 1919. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Ritz and W. M. McClellan Overview: Increased Cardiovascular Risk in Patients with Minor Renal Dysfunction: An Emerging Issue with Far-Reaching Consequences J. Am. Soc. Nephrol., March 1, 2004; 15(3): 513 - 516. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
HOME
CURRENT ISSUE
ARCHIVES
JASN Express
ONLINE SUBMISSION
AUTHOR INFO
EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP |
