Factors Associated With Withdrawal From Dialysis Therapy in Incident Hemodialysis Patients Aged 80 Years or Older.

This was an observational cohort, so it shows associations with withdrawal, not proof that any factor caused withdrawal. The authors note there is no uniform definition of withdrawal from hemodialysis, so outcome misclassification is possible. The study did not collect symptom burden, mental or physical health status, pain, insomnia, functional status, or depression severity, which may influence withdrawal. It did not record who initiated withdrawal decisions, whether advance directives were available, or how patients interacted with providers. It could not describe end-of-life care or quality of life around withdrawal because those data were lacking. Patients who received hemodialysis for less than 60 days were excluded, so some very elderly or ill patients who withdrew or died early may have been missed. Single-pool Kt/V was missing for 16% of covariate data, and multiple imputation was used. The main very-elderly results come from 17,296 patients aged 80 years or older, not the full 208,820-patient cohort. The cohort was drawn from outpatient facilities of a large dialysis organization in the United States, so other settings may differ.

The article reports research support from the NIH extramural program and non-U.S. government sources. One author, KKZ, declared honoraria from Genzyme/Sanofi and Shire and being medical director of DaVita Harbor-UCLA/MFI during 2007 to 2012. Other authors declared no conflict of interest.

Do not conclude that dementia, low albumin, or clinic region caused dialysis withdrawal. This is an observational cohort, and the authors did not measure symptom burden, patient preferences, advance directives, or who initiated withdrawal, so these factors are associations only.