Introduction:
Osteoporosis pathogenesis involves a complex interplay of genetic predisposition and medication exposure. Although long-term medication use contributes to bone fragility, which is distinct from osteomalacia, genetic evidence linking common medication-use traits to osteoporosis remains limited.

Material and methods:
Two-sample Mendelian randomization (MR) analyses utilized GWAS data for 22 medication-use traits (UK Biobank) and osteoporosis outcomes (FinnGen R12). Multivariable MR (MVMR) was employed to adjust for confounding by underlying disease indications. Primary estimates were derived using the inverse variance–weighted (IVW) method, supported by comprehensive sensitivity analyses including MR-Egger, weighted median, and MR-PRESSO to evaluate heterogeneity and pleiotropy.

Results:
Univariable MR identified seven significant associations: anilide use with increased osteoporosis risk (OR=1.25, FDR=0.0201); immunosuppressant use with higher risks of osteoporosis with pathological fracture (OR=1.16, FDR=0.0074) and drug-induced osteoporosis (OR=1.36, FDR=0.0168); thyroid preparations with increased osteoporosis (OR=1.08, FDR=0.0016), drug-induced osteoporosis, and osteoporosis with pathological fracture; and an inverse association between antidiabetic drug use and osteoporosis (OR=0.88, FDR=0.0002). In MVMR, associations for anilide and thyroid preparation use remained significant after adjusting for back pain and hypothyroidism, respectively. However, the antidiabetic signal reversed after type 2 diabetes adjustment, indicating the observed benefit was likely confounded by the underlying metabolic condition rather than a direct pharmacological effect.

Conclusions:
Genetic evidence supports that anilide and thyroid hormone use may independently increase osteoporosis risk, while the protection from antidiabetic therapy is confounded by diabetes status. Medication-related osteoporosis should be considered in long-term pharmacologic management.