Growth of the ESKD Population: Progress or Peril?
Year: 2018
Type: editorial
Abstract: Since 1973, when Congress created entitlement for treatment under the Medicare system, growth in the dialysis and kidney transplant population has been remarkable. In 2016, this population totaled >726,000 patients—511,000 on dialysis and 215,000 with a kidney transplant1; this is >18 times the initial expectation of 40,000 individuals. This number continues to rise. This remarkable growth in the ESRD population arises from a number of factors: improved survival of the general population secondary to pharmacologic advances; increased access to dialysis; advances in dialysis, such as bicarbonate dialysate and synthetic membranes; improvements in prescribing appropriate treatments; home-based dialysis; quality improvement programs addressing vascular access; and improved immunosuppressive therapy for patients with kidney transplants. Projecting the number of patients with ESRD using various assumptions about the determinants of growth in this population has been a perennial research effort intended to inform policy makers, insurers, and providers about the future patient population. In this issue of the Journal of the American Society of Nephrology, McCullough et al.2 developed projections to 2030—the longest projection to date—using a population-based transitional model and examining changes in incident rates that could unfold, taking into account obesity and diabetes rates and the aging of the general population in the United States. On the basis of these assumptions, their predicted range for the ESRD population is 971,000–1,259,000. Projections, of course, cannot simply look at new patients, because death rates have a substantial effect over time. Along with obvious new patients, the prevalent population comprises patients who have had an episode of AKI, recover kidney function, and return to ESRD treatment; patients who switch between dialysis modalities; and patients with kidney transplants—a modality with markedly higher survival compared with dialysis. Increase in the kidney transplant population is limited by access to available organs; this can change over time as seen with the recent increase in available organs resulting from opiate overdoses, leading to higher transplantation rates.3 Relatively healthy patients with ESRD on the waiting list for a transplant accumulate on dialysis, which leads to lower overall death rates. Growth in the ESRD population has not been accompanied by a comparable growth in the number of organs available for transplant. Between 2004 and 2014, the number of kidney transplants increased only 5% from 17,148 to 17,998.4 The waitlisted population, however, grew 61% from 53,419 to 86,064, adding 32,645 patients to the prevalent dialysis population in 2014—an increase of 7%.3 Accounting for such issues is a challenge when making projections. The authors hypothesized that obesity and diabetes would be major drivers of future growth in the ESRD population. However, the effect of death rates among existing patients on dialysis and patients with transplants on the overall growth of the prevalent ESRD population is four to five times the effect from changes in the incidence rate, insofar as obesity influences incidence. Although preventing the progression of CKD to ESRD is an important public health objective, it is often not apparent in the shorter term how such efforts might affect growth in the dialysis and transplant populations. For example, the competing effects on cardiovascular and renal outcomes of two classes of diabetes drugs, glucagon-like peptide-1 receptor agonists and sodium-glucose cotransporter 2 inhibitors, will take time to manifest in the general population. Access to kidney transplantation is a core issue, with transplant providing the best outcome at the most reasonable cost to the health care system. After transplant, drivers of mortality need greater consideration. Remarkable progress has been made in reducing the death rates; Wetmore et al.5 show a drop of almost 28% between 2000 and 2013, in parallel with a decline in mortality from cardiovascular disease and cancer in the general population.6 The recent flattening of the death rates in the ESRD population in the United States—particularly in the dialysis population—is a major concern.7 Assuming that rates remain flat, McCullough et al.2 project an ESRD population of 971,000 by 2030; this is a 34% increase from the 2016 total of 726,331 or 2.1% per year, an annual increase that would be the lowest in over two decades.8 The real test of any projection is how close it comes to actual data. This is where reality meets academic exercise. Two prior projections by Xue et al.9 and Gilbertson et al.10 were within 10% of the projected prevalent population. Xue et al.9 used a simplified autoregression technique with various smoothing techniques, an approach that simply assumes that historical patterns of growth and death rates will continue. Their projection of the incident population was markedly off, primarily because the approach did not consider a slowing of the historical rates; their projection of the prevalent population, however, was close to what transpired. Gilbertson et al.10 used a more detailed nonstationary Markov chain model, projecting 10-year trends in mortality and incidence on the basis of the most recent year’s data. This approach also overestimated the incident population but was within 10% for the prevalent population, erring on the high side by assuming falling death rates in the dialysis population. The lower end of the projection by McCullough et al.2 may be too conservative, because the incident 90-day death rates given in the US Renal Data System 2018 Annual Data Report show a decline.11 As noted above, the upper end projection of 1,259,000 does not consider the effect of the two basic modalities of kidney transplantation and dialysis. Because the authors did not use an integrated approach to project these populations, it is difficult to assess the inter-related issues of access to organs—which drives growth in the transplant population and removes healthy patients from the dialysis pool—and changing death rates in the prevalent dialysis population. It is important to acknowledge the inevitable uncertainties in 15-year projections. Ongoing efforts to prevent kidney disease and slow its progression may start to have an effect. Changes in organ availability and death rates will certainly have an important influence. Future models should build in a range of estimates of these possible changes. This would allow for greater clarity of the realities facing patients and providers today. Disclosures A.J.C. served as Director of the US Renal Data System from 1999 to 2014 when the prior two projections were published. A.J.C. is currently the Chief Medical Officer of NxStage Medical, Inc., a home hemodialysis and acute dialysis product company.
Author: Allan J. Collins
Institution: University of Minnesota
Language: English
Subfield: Nephrology
Field: Medicine
Domain: Health Sciences
Open Access status: green
OpenAlex ID: W2905445111