Rationale (2) PD census among facilities with active PD programs.Analytical Approach: We applied mixed-effects logistic and negative binomial models with random facility effects to estimate changes in PD program presence and mean census over time.Models adjusted for region, organizational size, profit status, and urbanicity, and included interaction terms to test for heterogeneity in temporal trends. Results:The number of U.S. dialysis facilities with active PD program increased from 38.4% to 44.6% between 2010 and 2021 and amongst those programs, mean PD census rose from 14.7 to 18.2.Relative to 2010, facilities in 2021 had higher odds of providing PD (OR 1.61, 95% CI 1.35-1.92),and active programs treated substantially more patients (IRR 1.50, 95% CI 1.45-1.56).Non-profit and small dialysis organizations showed the greatest likelihood of initiating PD programs, whereas for-profit and large organizations demonstrated stronger census growth.Regional differences were observed, with the West showing the largest increases in program size.
Shah et al. (2026) studied this question.