Woods JD, Port FK, Stannard D, Blagg CR, Held PJ
Comparison of mortality with home hemodialysis and
center hemodialysis: A national study
(May) 49:1464-1470 1996
From 1983 to 1993, home hemodialysis has been decreasing in prevalence
despite a 9-13% increase in the prevalence of ESRD treatment. The
prevalence of continuous peritoneal therapies appears to have
plateaued [HCFA data - 1994]. Not much has been published about the
benefits of home hemodialysis, especially with regard to
rehabilitation or survival.
This paper presents survival data of patients starting dialysis
in 1986-1987 using the USRDS Case Mix Severity dataset. Using an
intent-to-treat design, survival was compared in patients initially
assigned to center dialysis vs. those initially assigned to home
hemodialysis. Previous reports have suggested decreased mortality in
home HD patients, but the results may have been due to differences in
patient selection and/or comorbidity. The selection bias is
impossible to compensate for in any retrospective study, but patient
comorbidity was controlled for in this study by adjusting survival
date for age, sex, race, presence of diabetes, and about 18
additional comorbid factors using a proportional hazards model.
The home HD patients were younger, had less comorbidity and were less
likely to be diabetic than patients assigned to center dialysis.
Crude mortality was 63% less in home HD patients. Even after
adjusting for age, sex, race, diabetes, and comorbid conditions, a 42%
lower mortality risk persisted in the home HD patients.
What are the other factors that might have resulted in the
better patient outcome for the home HD patients? The authors speculate
that patient selection may still play an important role. Also,
comorbidity was assessed using an ON/OFF design instead of a graded
scale. Comorbidity may have still been less in the home HD patients.
The authors also point out that center patients are now getting a
higher Kt/V than that of the patients analyzed in the present report,
due to the recent increase in Kt/V nationwide. Nevertheless, the
result remains that home HD patients still had a 42% lower mortality
after adjustment for a large number of confounding covariates.
Comment: Every nephrologist should read this brief,
simple, and very pertinent and cogent article. The data suggest that
we need to re-evaluate and reverse the recent decline in home HD
therapy in the United States. Parenthetically, readers of HDCN might
wish to peruse the April 1996 issue of Advances in Renal Replacemnt
Therapy which is devoted entirely to home hemodialysis. (Lionel U.
Mailloux, M.D., Manhasset, NY)