Back

Weight trajectories after last Tirzepatide or Semaglutide prescription across a federated health network

Murugadoss, K.; Varma, G.; Venkatakrishnan, A.; Gibson, M.; Soundararajan, V.

2026-01-28 endocrinology
10.64898/2026.01.26.26344839 medRxiv
Show abstract

GLP-1 receptor agonist (GLP-1RA) discontinuation has been associated with weight regain. However, the real-world association between discontinuation of GLP-1RA prescriptions and weight change has not been explored. We assessed weight trajectories of 4,182 patients in the six months following their last GLP-1RA prescription. Approximately two-thirds of patients showed stable weight or continued weight loss during this period post the last known GLP-1RA prescription. In a representative subset of patients with clinician-documented discontinuation near the last prescription (N=300), a similar distribution of weight regain in a minority of patients was observed vs no regain in the majority of patients during the six-month post-GLP-1RA prescription period. To mirror clinical trial-style discontinuation definitions, we also evaluated cohorts with no subsequent GLP-1RA prescription for 1 year after the last prescription (semaglutide N=1,755; tirzepatide N=1,312), observing weight regain in a minority of patients (39.3% semaglutide; 26.6% tirzepatide) and no weight regain in the majority (60.7% semaglutide; 73.4% tirzepatide) in the year following the last known GLP-1RA prescription. Exercise counseling was documented more frequently among patients with durable weight loss post-last GLP1 prescription compared with those with weight regain (26.2% vs. 14.7%; p=0.04). Further studies are warranted to infer the mechanisms underlying these real-world patterns.

Published in Biology Methods and Protocols (predicted rank #12) · training set

Matching journals

The top 11 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.