Institutional factors influencing regional differences in the 1-year survival of dialysis patients.

Abstract

There are regional differences in the survival of incident dialysis patients, but few studies have investigated the reasons. We assessed the effect of institutional factors on factors on survival (by both cross-sectional assessment and after a 1-year investigation) in a cohort of the Japanese Society for Dialysis Therapy (JSDT). We investigated 20 institutional factors from 3958 dialysis institution data of the 47 prefectures in Japan in 2005 and the JSDT database of 102,011 patients who commenced dialysis during 2004-2006. Univariate regression analysis between 20 factors and 1-year survival rate, Kaplan-Meier method, log-rank test, and Cox's proportional hazard model between the upper 24 and the lower 23 prefectures of the significant factors were carried out. The age-adjusted 1-year survival rate was 0.832 ± 0.027. Deaths occurred in 15.0% in 24 upper survival prefectures and 18.7% in 23 lower survival prefectures (P < 0.0001, unadjusted hazard ratio [HR] of death in lower survival prefectures: 1.26, 95% confidence interval [CI]: 1.17-1.40). A total of five factors among males were significantly correlated with 1-year survival according to the univariate regression analysis. Among them, three factors (night-time center/total dialysis center ratio [males: P < 0.0001, age, genders, and presence of diabetes adjusted HR: 0.88, 95% CI: 0.81-0.93], number of full-time dialysis nurses [males: P = 0.0427, 0.94, 0.87-1.00], number of full-time dialysis dietitians [males: P = 0.0084, 0.92, 0.85-0.98], respectively) were significant in Kaplan-Meier analysis, log-rank test, and the Cox's model. Institutional factors were closely related to the survival of incident dialysis patients, and regional differences in the survival may be explained, at least partly, by these factors.