Translation of the Patient-reported Outcomes Measure of Pharmaceutical Therapy for Quality of Life and Its Validation in Elderly Patients with Polypharmacy

Background There is a high prevalence of polypharmacy among elderly patients with severe situation. There is an urgent need for more comprehensive and objective tools to assess the benefits and risks of medication use in patients to ensure maximum length of life and quality of life for patients. Objective To translate the Patient-reported Outcomes Measure of Pharmaceutical Therapy for Quality of Life (PROMPT-QoL) Scale into Chinese and evaluate its reliability and validity among elderly patients with polypharmacy. Methods With the authorization of the original author, Brislin's translation model of forward-back translation, cultural adaptation, and cognitive interviews were used to develop a test draft of the Chinese version of the PROMPT-QoL. Elderly patients with polypharmacy who visited outpatient clinics, received health checkups and took medicine at community health centers of Gongyuan street, Beishan street and Henan street, Yanji Hospital and Yanbian University Hospital from June to August in 2022 were selected as the survey respondents. Item analysis was conducted using the item-dimension correlation analysis and the critical ratio (CR). The content validity of the scale was evaluated using the item-level content validity index (I-CVI), universal agreement scale-level CVI (S-CVI/UA) and average scale-level CVI (S-CVI/AVE). Structural validity was evaluated using exploratory factor analysis (EFA) and confirmatory factor analysis (CFA). Internal consistency tests (Cronbach's alpha coefficient) and the split-half coefficient were used to assess reliability. Results A total of 590 patients were investigated and 564 valid data were recovered, with a recovery rate of 95.8%. Among them, 234 data were applied to the item analysis and EFA in the first stage, and 330 data were applied to the CFA in the second stage. The correlation coefficients between the scores of each item and the scores of each dimension ranged from 0.504~0.915 (P3.0 (P0.80, and the S-CVI/Ave was 0.99>0.90. A total of 8 common factors were extracted from EFA, which were basically consistent with the results of the source questionnaire, among which item G34 had a loading of