By Eli Friedman, MD        

AAKP Note: Though we agree with Dr. Friedman's opinions regarding why comparing dialysis outcomes in the United States to those in other countries may be misleading and may actually demonstrate that Americans receive better care, AAKP also strongly believes that dialysis treatment can be made better in the United States.

Criticism of hemodialysis in the United States causes uneasiness in patients and despair in dialysis facility directors.  Stated simply, it is alleged that in the United States avoidable deaths account for approximately 23 percent of first year dialysis mortality.  This is in stark contrast with Europe that has a rate of about 15 percent and Japan of nine percent.  Before American nephrologists are forced to wear a scarlet letter of shame, however, the impact of selection bias on reports of survival on dialysis should be examined.

Is There Excessive Mortality on Dialysis in the United States?

The National Kidney Foundation (NKF) in their press release of November 14, 1995, "called for a massive effort to lower the all-too-high mortality rate of dialysis patients in the U.S. and improve the care provided under Medicare's kidney dialysis program."  Pointing out that the "24 percent mortality rate for U.S. dialysis patients, the highest rate in the industrialized world, is simply unacceptable," NKF states "We can and should do better to lower the death rate."

Insufficient (inadequate) dialysis and suboptimal quality control of dialysis delivery were suggested reasons for the so-called excessive death rate in the United States.  Calling for enforcement by the Health Care Financing Administration (HCFA), NKF desired standards for defining minimal acceptable dialysis (time and frequency) and staffing in dialysis units.  Additionally, NKF linked the purportedly higher U.S. dialysis death rate, in part, to the decline in reimbursement by HCFA from $135 per dialysis in 1973 to $126 in 1995 despite inflation in labor costs.  NKF felt such a situation, "encourages corner cutting and results in poor quality service by dialysis units."

Echoing NKF's concern over a high U.S. death rate on dialysis, a Consensus Conference at the National Institutes of Health (CCNIH) concluded: "The dose of hemodialysis and peritoneal dialysis has been suboptimal for many patients in the United States."  The dose of dialysis may be estimated by measuring fractional urea clearance during the hemodialysis procedure and expressed as Kt/V.  A Kt/V minimal delivered hemodialysis dose of at least 1.2 was advocated by the Conference.  (Please see the aakpAdvisory: Inadequate Hemodialysis Increases the Risk of Premature Death).

 What is Meant by Selection Bias?

Oranges must be compared with oranges of the same size, freshness and growth history when reporting on samples in a particular grocery store. Illustrating this point, in the 1960's, transplant enthusiasts celebrated markedly.  A third of a century later, we understand how transplant surgeons practiced "cherry picking," meaning selection (and removal) from the general population of dialysis patients of relatively healthy individuals. These are the people placed on a waiting list for a cadaver donor kidney, thus leaving behind sicker dialysis patients at greater risk of death.  According to the U.S. Renal Data System (USRDS), "the standardized mortality ratio for patients on dialysis awaiting transplantation was 38 to 58 percent lower than that for all patients on dialysis. In other words, transplant teams start with patients less likely to die than those not picked for transplantation.

Bias Outside the United States

The contrast between death rates in the United States and the truly spectacular outcome of dialysis units in France is repeatedly cited as evidence of American deficiency.  Especially lauded is the unmatchable "landmark" longevity reported in Tassin, France.  In 1983, Drs. Laurent, Calemard and Charra analyzed 373 hemodialysis patients. They found a remarkable survival rate of 75 percent at 10 years and 65 percent at 15 years.  Although supposedly "an unselected population," there were actually only 15 patients (four percent) who had diabetes while another 15 (four percent) had systemic disease.  What was the fate of uremic people in Tassin who were not accepted for dialysis (diabetic, very old, extensive systemic disease)?  Obviously, they died, uncounted in any ESRD survival statistics. By providing dialysis to healthier patients, France and other European countries built a superior survival rate. American survival rates can never match this, as all American patients are accepted into our dialysis system.

 Europe Rations ESRD Treatment by Age, Gender and Co-Morbidity

Skewed selection and transplant rates are factors that make dialysis survival appear good on paper. International comparisons ignoring selection factors and attributing differences in survival to technique are meaningless.  An inspection of the 1995 registry report of the European Dialysis and Transplant Association (EDTA) affirms the persistent exclusion of older uremic patients by stating,  "The median age of ESRD patients in the UK in 1992 was 60 years, while in Sweden it was 66 years, in France 63 years and in Germany 62 years. Study of international registries reveals an economic influence on rationing limits for ESRD programs worldwide.  Rationing is a kind term for restricting admission to dialysis.

Clearly, treatment for ESRD patients is a product of governmental policies and economic pressures.  Treatment of renal failure in the United Kingdom under performs countries in Western Europe and the United States.  As recently as April 2000, Shaldon, a UK home hemodialysis pioneer, wrote, "Most distressing is the appalling state of dialysis availability in the United Kingdom.  Although, successive governments since 1964 have refused to budget adequately for the needs of the renal services, the conspiracy of silence entered into by the medical profession seems to be equally responsible.  As long as primary care doctors, acting as indoctrinated gatekeepers, refuse to refer elderly patients to renal units for dialysis, nephrologists can continue to claim that they accept all patients referred to them."

There is no doubt that outside the United States, many women, the elderly and racial minorities die untreated.  In Eastern Europe, after the Berlin Wall fell, it became evident that ESRD therapy, though equivalent in survival to that reported for Western Europe, was restricted to Communist Party members or the otherwise rich and powerful.  This means that a low ESRD treatment rate can hide deaths due to renal failure.

Low Kidney Transplant Rate in Japan Improves Dialysis Outcome

If the acceptance rate explains even partially the disparity between the United States and Europe, what can explain the superior survival of Japanese hemodialysis patients?  One need only look at international comparisons of the number of functioning kidney transplants by country to note that the United States leads, while Japan ranks at the bottom.  Japan's low transplant rate means that the youngest and healthiest patients remain within the hemodialysis pool. But in the United States, these same young healthy patients are eight times more likely to be removed from hemodialysis statistics after receiving a kidney.

U.S. ESRD Acceptance Rate is World's Highest

In fact, the United States has the highest rate of new treatment for ESRD among reporting registries worldwide.  The United States accepts twice as many patients than Europe (320 per million population in 1998), 40 percent more than Canada and 36 percent more than Japan.

Why this discrepancy?  There are three possible answers:

1. The United States has a higher incidence of ESRD than any other part of the world. (African Americans and Native Americans attribute to higher rates)

2. The United States treats ESRD in "marginal patients" who are otherwise quite ill. (i.e. extensive spread of cancer)

3. Other countries do not treat a substantial portion of ESRD patients who die from other listed causes.  The United States has over twice the treatment rate for ESRD than other industrialized European countries, while Canada treats approximately one-half the number of patients.

What Explains Lower Treatment Rates Outside the U.S.?

According to my analysis, both the higher treatment and higher death rates in the United States are indicators of relative health in the American system for ESRD care when compared with the rest of the world.  Think it through.  Could it really be the Canadian-U.S. border that causes two different worlds of ESRD?  Or is it the same system that drives Canadians to the United States for coronary bypass, hip replacement and other modern medical procedures?  We now know that this was the reality in East Germany before the Wall came down.  And, it certainly is what takes place in Mexico.  Put succinctly, rationing life-sustaining though expensive health care for budgetary reasons is accomplished at the expense of life.  So, if the frail, very sick and elderly are excluded from ESRD therapy to die from other causes, then ESRD survival rates are enhanced.

Skepticism over International Comparisons of ESRD Mortality

Over the past decade, I have personally inspected ambulatory hemodialysis facilities, including chart reviews and patient interviews, in more than a dozen overseas programs in Europe and Asia.  While subtle differences in management may have eluded my detection, the level of care delivered as judged by predialysis serum albumin, creatinine, phosphorous concentrations and hematocrit, appeared at best equivalent to that for a U.S. typical unit.  Lacking outside the United States, however, were late evening shifts geared toward fostering rehabilitation in patients employed during the day.  Also absent, except in the U.K., was the daily participation of our literature distribution by patient activist groups such as the American Association of Kidney Patients.  My impression of dialysis quality outside of the United States left me comfortable in the belief that our Medicare standard of dialysis was not routinely exceeded.

There is strong evidence of consistent under reporting of dialysis deaths throughout Europe leading Ritz, to state, "Recent problems in the EDTA Registry have led to considerable under reporting, so that the figures for international outcomes (deaths and complications) have become unreliable and are probably over optimistic." Furthermore, based on my examination of dialyzed patients in Kobe, Kyoto, Tokyo, Sapporo and Yokohama, I am reluctant to accept the premise that hemodialysis performance is uniformly better in Japan than in the United States.  Japanese patients complain about long waits, a lack of explanations or informed consent procedures and poor physical facilities.

Underscoring all of this, is the finding that survival in the general population after the age of 80 is statistically superior in the United States to that in England, France, Japan and Sweden.  Be cautious, therefore, in condemning the U.S. medicare. ESRD program, which, after proper study is conducted, may very well prove to be the best provider of universal uremia therapy in any nation for all of its citizens.

Eli Friedman, MD is the Chief of the Division of Renal Disease for State University of New York, Health Science Center at Brooklyn. Dr. Friedman is the chairperson of the AAKP Medical Advisory Board.

This article originally appeared in aakpRENALIFE, Vol. 16, No. 3, November 2000.

Back