GLP-1 medications such as semaglutide have transformed type 2 diabetes treatment, but new research suggests their well-documented physical health benefits may not extend to measurable improvements in mental health, employment, or relationships. The findings were published as a National Bureau of Economic Research working paper.
Researchers at the University of Chicago analysed over a decade of data from the nationally representative Medical Expenditure Panel Survey, focusing on adults with diabetes between 2012 and 2023. Rather than simply comparing GLP-1 users with non-users, the team followed the same individuals over time to examine whether their circumstances changed after beginning treatment.
The study assessed several wellbeing measures, including symptoms of depression and psychological distress, self-rated health, employment status, and marital status. Initial comparisons suggested GLP-1 users differed from non-users on some measures. However, these differences largely disappeared when researchers tracked the same people over time, indicating the apparent gaps reflected pre-existing differences between people taking the medications rather than effects of the drugs themselves.
The researchers reached similar conclusions whether examining outcomes after approximately one year of GLP-1 use or over a longer two-year period. Across both analyses, they found little evidence that the medications produced meaningful changes in the broader outcomes studied.
Lead author Robert Kaestner, a research professor at University of Chicago Harris School of Public Policy, cautioned that this does not mean the drugs have no effects beyond physical health. Rather, such effects may be more subtle or harder to measure than commonly assumed. The study’s measures may not capture changes in self-esteem, relationship quality, or day-to-day experiences.
The findings do not diminish the medications’ established medical value, including better glycaemic control, weight loss, and cardiovascular risk reduction.
Source: Medical Xpress / University of Chicago (NBER Working Paper, 2026)
