Abstract: 【Abstract】Objective Based on the data from new peritoneal dialysis patient treated in the Department of Nephrology, Xinqiao Hospital of Army Medical University, multidimensional analyses on patients with automated peritoneal dialysis (APD) or continuous ambulatory peritoneal dialysis (CAPD) were performed by using the SF-36 quality of life questionnaire. Methods Purpose sampling method was used to investigate the peritoneal dialysis patients treated in the Department of Nephrology, the Second Hospital of PLA Military Medical University in the period from January 2018 to June 2019 by using the questionnaire of SF-36 scale. A total of 241 valid questionnaires were received, including 135 questionnaires from APD group and 106 from CAPD group. Their reliability and the dimensions of collective validity, discriminant validity and structural validity were comprehensively analyzed to obtain the results of social regression in the two groups of patients. Results The overall Cronbach's α coefficients were 0.93 and 0.92 in APD group and CAPD group respectively, with better internal consistency reliability in both groups. The success rates of collective validity of the samples were 83% and 93%, and the success rates of discrimination validity tests were 96% and 98% in APD
group (n=135) and CAPD group (n=106) respectively. Both groups had better collective validity and discrimination validity (Cronbach’s α=0.93 and 0.92, P=0.034 and 0.025 in APD group and CAPD group respectively). For standard validity, the rank correlation coefficient calculated by the standard and scale scores of the samples in APD group was 0.593, greater than 0.403 in CAPD group (P=0.320 and 0.300 respectively), suggesting that SF-36 scale has better standard correlation validity used in APD group. For structural validity, the t-test of the unstandardized factor load coefficients demonstrated that the load coefficients were statistically significant at >0.1 level in both groups (P=0.017 and 0.030 for APD group and CAPD group respectively), indicating

that the designed entries in the scale are informative for both groups. Conclusion The SF-36 scale has good reliability and validity for statistical analysis of quality of life in peritoneal dialysis patients, and the internal reliability and standard correlation validity are better for APD group than for CAPD group, suggesting that social function and emotional state are better in APD patients than in CAPD patients.

Key words: Peritoneal dialysis, SF36, Social regression, Automated peritoneal dialysis, Statistics