Aim:This study aimed to determine the comparative treatment effects of semaglutide 2.4 mg and placebo on health utility index scores [6‐dimension short‐form survey (SF‐6D)] with Australian weights in full analysis set (FAS) and in post‐hoc subgroups of the STEP 1 trial, defined according to different body mass index (BMI) cut‐off points and presence of comorbidities at baseline. The study also explored the correlation between baseline BMI and SF‐6D in the STEP 1 trial population.
Methods:The 36‐item SF survey (SF‐36) scores from STEP 1 were mapped to SF‐6D health states and converted to utility index scores using an Australian valuation algorithm. The change from baseline in SF‐6D utility score (95% confidence intervals) was compared between semaglutide 2.4 mg and placebo at week 68 using the mixed model for repeated measurements approach. The relationship between utility scores and BMI at baseline was assessed by multiple linear regression analyses, controlling for demographic and clinical parameters.
Results:The estimated mean treatment difference in SF‐6D utility score favoured semaglutide 2.4 mg, and, at week 68, it was 0.057 (0.038‐0.076) for the FAS. A greater treatment effect was noted in subgroups with presence of symptomatic comorbidities, i.e. 0.077 (0.027‐0.128) to 0.105 (0.030‐0.179) at week 68. A 1‐unit increase in BMI was associated with a utility loss of 0.0075 (−0.0089 to −0.0062) for the FAS population, while controlling for demographic and clinical parameters.
Conclusion:To our knowledge, this is the first study showing statistically significant and clinically meaningful improvements in SF‐6D utility scores with weight‐loss pharmacotherapy in Australia.