Does long-term GLP-1 therapy weaken bone? It is one of the youngest and most contested questions in obesity pharmacology, and the published evidence is narrower than the volume of patient concern suggests. Rapid weight loss from any source — diet, bariatric surgery, or pharmacotherapy — reduces bone mineral density at the hip and lumbar spine, typically 1-3% over 12-18 months on DEXA, with larger reductions after bariatric procedures. Whether GLP-1 receptor agonists add a drug-specific bone effect on top of that weight-loss-mediated signal is the question this list traces. Iepsen's 2015 JCEM randomized trial of post-weight-loss women suggested liraglutide actively prevented bone loss compared with placebo plus weight maintenance, generating the hypothesis that GLP-1 signaling is bone-protective. Jensen's 2024 JAMA Network Open S-LITE secondary analysis later showed that combining liraglutide with exercise preserved BMD better than liraglutide alone, complicating the simple drug-protective story. RCT meta-analyses (Mabilleau 2014, Cheng 2019, Zhang 2025) consistently find no statistically significant fracture-risk increase across GLP-1 trials in type 2 diabetes, but those trials were not powered for fractures and run mostly under 2 years. Observational cohort work (Ko 2023 JAMA Netw Open) and FDA Adverse Event Reporting System pharmacovigilance (Xiao 2025) likewise show neutral-to-favorable signals versus other diabetes drugs. The honest summary in 2026 is that GLP-1s do not appear to raise fracture risk in the short-to-medium term, but dedicated fracture endpoint trials in older obesity populations do not yet exist.
Ranked papers
#1S-LITE (bone substudy)
Jensen SBK, Sørensen V, Sandsdal RM, et al. · JAMA Netw Open · 2024
Primary endpoint: Change in bone mineral density at hip and lumbar spine at 52 weeks
This pre-specified bone substudy of the S-LITE trial randomized 195 adults with obesity to liraglutide 3.0 mg, supervised exercise, both, or placebo for one year after an 8-week low-calorie diet run-in. Liraglutide monotherapy reduced hip BMD significantly more than placebo, while the combination of liraglutide plus exercise preserved BMD. The exercise-alone group also preserved BMD. This is the cleanest randomized evidence that GLP-1 monotherapy produces weight-loss-mediated bone loss that resistance and aerobic exercise can fully mitigate.
PMID 38916894 ↗DOI 10.1001/jamanetworkopen.2024.16775 ↗
#2
Iepsen EW, Lundgren JR, Hartmann B, et al. · J Clin Endocrinol Metab · 2015
Primary endpoint: Total hip BMD change at 52 weeks after weight loss
The foundational randomized trial of GLP-1 effects on bone after weight loss. Thirty-seven women with obesity who had completed an 8-week low-calorie diet were randomized to liraglutide 1.2-1.8 mg or placebo for 52 weeks. The placebo group lost roughly 2.8% of total hip BMD; the liraglutide group lost none. Bone formation marker P1NP increased on liraglutide. This trial generated the hypothesis that GLP-1 receptor signaling has a direct bone-anabolic effect independent of weight loss — a claim later complicated by Jensen 2024.
PMID 26043228 ↗DOI 10.1210/jc.2015-1176 ↗
#3
Zhang Y, Chen G, Wang W, et al. · Bone · 2025
Primary endpoint: Pooled fracture risk ratio across GLP-1 RCTs in T2D
The most recent randomized-trial meta-analysis of GLP-1 fracture risk in type 2 diabetes. Pooled data from 53 RCTs covering more than 60,000 participants found no statistically significant difference in fracture risk between GLP-1 receptor agonists and comparators (placebo or active comparator). Subgroup analyses by drug, dose, and duration likewise found no signal. The authors note that follow-up across these trials is mostly under 2 years and that obesity populations without diabetes were not represented, leaving long-term fracture risk in older Wegovy and Zepbound users unresolved.
PMID 39603373 ↗DOI 10.1016/j.bone.2024.117338 ↗
#4
Cheng L, Hu Y, Li YY, et al. · Diabetes Metab Res Rev · 2019
Primary endpoint: Pooled odds ratio for incident fracture across GLP-1 RCTs
An earlier RCT meta-analysis covering 38 trials and roughly 39,000 participants on liraglutide, exenatide, dulaglutide, semaglutide, lixisenatide, and albiglutide. The pooled odds ratio for fracture was 1.05 (95% CI 0.81-1.37), consistent with no effect. Trial-level heterogeneity was low. Like the later Zhang 2025 analysis, the limitation is that fractures were adverse-event captures rather than pre-specified endpoints, and trial duration averaged under 2 years.
PMID 30974033 ↗DOI 10.1002/dmrr.3168 ↗
#5
Mabilleau G, Mieczkowska A, Chappard D · J Diabetes · 2014
Primary endpoint: Pooled fracture odds ratio across early GLP-1 RCTs
The first RCT meta-analysis of GLP-1 fracture risk, covering 16 trials of liraglutide and exenatide. The pooled odds ratio was 0.75 (95% CI 0.28-2.02) — directionally favorable but not statistically significant, with wide confidence intervals reflecting the small absolute number of fracture events. The paper helped establish the working assumption that GLP-1 therapy does not raise fracture risk in type 2 diabetes, an assumption that subsequent larger meta-analyses have not overturned.
PMID 24164867 ↗DOI 10.1111/1753-0407.12102 ↗
#6
Ko HY, Bea S, Jeong HE, et al. · JAMA Netw Open · 2023
Primary endpoint: Composite fracture incidence in new SGLT2 vs incretin (GLP-1 or DPP-4) users
A nationwide South Korean cohort comparing roughly 87,000 new SGLT2 inhibitor users with matched new incretin-based-drug users (GLP-1 RAs and DPP-4 inhibitors) using active-comparator new-user design. Fracture rates were similar between groups — no increased risk with either class. This is among the largest real-world fracture comparisons including GLP-1 exposure and is informative for the practical clinical question of whether one class is bone-safer than another in T2D.
PMID 37751205 ↗DOI 10.1001/jamanetworkopen.2023.35797 ↗
#7
Karam L, Mabilleau G, Paccou J · Osteoporos Int · 2025
Primary endpoint: Narrative review of GLP-1 obesity-trial bone data
A 2025 narrative review focused specifically on bone outcomes in obesity populations rather than the larger diabetes literature. The authors catalog the limited DEXA substudy data from STEP and SURMOUNT programs, the Iepsen and Jensen RCTs, and emerging real-world fracture cohorts. Their summary: weight-loss-mediated BMD reduction is the dominant signal; a drug-specific bone-protective effect of GLP-1 is plausible but not established in obesity populations; resistance exercise and adequate protein intake are the most evidence-based mitigations.
PMID 40920189 ↗DOI 10.1007/s00198-025-07664-1 ↗
#8
Anastasilakis AD, Paccou J, Palermo A, et al. · Diabetes Obes Metab · 2025
Primary endpoint: Critical appraisal of bone effects across modern AOMs
A 2025 critical appraisal covering bone effects of all major anti-obesity medications — orlistat, phentermine-topiramate, naltrexone-bupropion, liraglutide, semaglutide, and tirzepatide. For the GLP-1 and GLP-1/GIP class, the authors conclude that fracture-event data from RCTs is reassuringly neutral but underpowered, that DEXA substudies consistently show some BMD loss proportional to weight loss, and that population-level long-term fracture data in non-diabetic obesity remains the key unmet evidence gap before millions of healthy adults take these drugs for decades.
PMID 40555693 ↗DOI 10.1111/dom.16541 ↗
#9
Xiao Y, Zhou M, Xiao W, et al. · Acta Diabetol · 2025
Primary endpoint: Disproportionality analysis of fracture reports in FAERS
A pharmacovigilance disproportionality analysis of the FDA Adverse Event Reporting System covering fracture reports for liraglutide, semaglutide, dulaglutide, exenatide, and tirzepatide through 2024. No drug in the class showed a positive disproportionality signal for fracture relative to the full FAERS background. The analysis is hypothesis-generating only — FAERS cannot establish incidence or causality — but it is the largest spontaneous-report dataset on GLP-1 fractures available and is consistent with the RCT meta-analyses showing no fracture excess.
PMID 39556224 ↗DOI 10.1007/s00592-024-02415-w ↗
#10
Al Refaie A, Baldassini L, Mondillo C, et al. · Calcif Tissue Int · 2024
Primary endpoint: Lumbar and femoral BMD plus trabecular bone score at 12 months
A prospective single-center observational study of 65 adults with type 2 diabetes starting semaglutide or dulaglutide, followed for 12 months with DEXA and trabecular bone score (TBS). The GLP-1-treated group maintained or modestly improved lumbar BMD and TBS, while a non-randomized control group on standard care declined. The signal is consistent with Iepsen 2015 but the observational design and absence of randomization mean the result is hypothesis-generating, not confirmatory.
PMID 38864922 ↗DOI 10.1007/s00223-024-01240-1 ↗
About this list
We curate ranked, citation-anchored PubMed paper lists for the most-searched questions in obesity medicine. Every citation on this page was checked against PubMed on 2026-05-28. Each paper card links directly to PubMed and to ClinicalTrials.gov where applicable.
Browse our full index of research lists or our long-form research articles.