ABSTRACT: BACKGROUND:Little is known about differences in the clinical course between patients receiving maintenance dialysis who do and do not withdraw from dialysis therapy. STUDY DESIGN:Case-control analysis. SETTING & PARTICIPANTS:US patients with Medicare coverage who received maintenance hemodialysis for 1 year or longer in 2008 through 2011. PREDICTORS:Comorbid conditions, hospitalizations, skilled nursing facility stays, and a morbidity score based on durable medical equipment claims. OUTCOME:Withdrawal from dialysis therapy. MEASUREMENTS:Rates of medical events, hospitalizations, skilled nursing facility stays, and a morbidity score. RESULTS:The analysis included 18,367 (7.7%) patients who withdrew and 220,443 (92.3%) who did not. Patients who withdrew were older (mean age, 75.3±11.5 [SD] vs 66.2±14.1 years) and more likely to be women and of white race, and had higher comorbid condition burdens. The odds of withdrawal among women were 7% (95% CI, 4%-11%) higher than among men. Compared to age 65 to 74 years, age 85 years or older was associated with higher adjusted odds of withdrawal (adjusted OR, 1.61; 95% CI, 1.54-1.68), and age 18 to 44 years with lower adjusted odds (adjusted OR, 0.36; 95% CI, 0.32-0.40). Blacks, Asians, and Hispanics were less likely to withdraw than whites (adjusted ORs of 0.36 [95% CI, 0.35-0.38], 0.47 [95% CI, 0.42-0.53], and 0.46 [95% CI, 0.44-0.49], respectively). A higher durable medical equipment claims-based morbidity score was associated with withdrawal, even after adjustment for traditional comorbid conditions and hospitalization; compared to a score of 0 (lowest presumed morbidity), adjusted ORs of withdrawal were 3.48 (95% CI, 3.29-3.67) for a score of 3 to 4 and 12.10 (95% CI, 11.37-12.87) for a score ?7. Rates of medical events and institutionalization tended to increase in the months preceding withdrawal, as did morbidity score. LIMITATIONS:Results may not be generalizable beyond US Medicare patients; people who withdrew less than 1 year after dialysis therapy initiation were not studied. CONCLUSIONS:Women, older patients, and those of white race were more likely to withdraw from dialysis therapy. The period before withdrawal was characterized by higher rates of medical events and higher levels of morbidity.