Objective To evaluate and compare the measurement properties of the EQ-5D-5L and SF-6Dv2 among Chinese overweight and obesity populations.Methods A representative sample of Chinese overweight and obesity populations was recruited stratified by age, gender, body mass index (BMI), and area of residence. Social-demographic characteristics and self-reported EQ-5D-5L and SF-6Dv2 responses were collected through the online survey. The agreement was assessed using intraclass correlation coefficients (ICC). Convergent validity and known-group validity were examined using Spearman's rank correlation and effect sizes, respectively. The test-retest reliability was assessed using among a subgroup of the total sample. Sensitivity was compared using relative efficiency and receiver operating characteristic.Results A total of 1000 respondents (52.0% male, mean age 51.7 years, 67.7% overweight, 32.3% obesity) were included in this study. A higher ceiling effect was observed in EQ-5D-5L than in SF-6Dv2 (30.6% vs. 2.1%). The mean (SD) utility was 0.851 (0.195) for EQ-5D-5L and 0.734 (0.164) for SF-6Dv2, with the ICC of the total sample was 0.639 (p < 0.001). The Spearman's rank correlation (range: 0.186-0.739) indicated an acceptable convergent validity between the dimensions of EQ-5D-5L and SF-6Dv2. The EQ-5D-5L showed basically equivalent discriminative capacities with the SF-6Dv2 (ES: 0.517-1.885 vs. 0.383-2.329). The ICC between the two tests were 0.939 for EQ-5D-5L and 0.972 for SF-6Dv2 among the subgroup (N = 150). The SF-6Dv2 had 3.7-170.1% higher efficiency than the EQ-5D-5L at detecting differences in self-reported health status, while the EQ-5D-5L was found to be 16.4% more efficient at distinguishing between respondents with diabetes and non-diabetes.Conclusions Both the EQ-5D-5L and SF-6Dv2 showed comparable reliability, validity, and sensitivity when used in Chinese overweight and obesity populations. The two measures may not be interchangeable given the systematic difference in utility values between the EQ-5D-5L and SF-6Dv2. More research is needed to compare the responsiveness.

Introduction: The present study aimed to evaluate the effects of glucagon-like peptide 1 receptor agonists (GLP-1RAs) on clinical and safety outcomes including glycemic control and cardiometabolic indicators using network meta-analysis. Methods: MEDLINE, Embase, and Cochrane Library Central Register of Controlled Trials were searched from inception through June 30, 2019. Randomized clinical trials comparing one or more of six eligible GLP-1RAs with placebo or another eligible GLP-1RA were identified. We further screened studies that had 24-30 week follow-up periods and target endpoints. The primary outcome was change in hemoglobin A1c (HbA1c). Secondary outcomes included additional glycemic control indicators, cardiometabolic measures, and adverse events. Frequentist random-effect network meta-analyses were conducted for effect comparison. Results: The NMA synthesized evidence from 54 studies covering 23,209 patients and 18 GLP-1RA regimens. All included GLP-1RA regimens except liraglutide 0.3 mg once weekly (QW) significantly lowered HbA1c after 24-30 weeks compared with placebo. The pairwise comparison of HbA1c-lowering effect showed that dulaglutide 0.75 mg QW, dulaglutide 1.5 mg QW, exenatide 2 mg QW, liraglutide 0.9 mg QW, liraglutide 1.2 mg QW, liraglutide 1.8 mg QW, loxenatide 100 µg QW, and loxenatide 200 µg QW were not significantly outperformed by any of the other regimens. The effects on blood pressure, weight, and lipids were relatively mixed. The GLP-1RA regimens had comparable safety profiles with regard to hypoglycemia and adverse events. Conclusion: Regimens of GLP-1RAs had differential glycemic control and cardiometabolic effectiveness. Policymaking and patient-centric clinical decisions should take into consideration the comparative effectiveness profiles.