Abstract
Osteoporosis, characterized by low bone mineral density (BMD) and increased fracture risk, is common yet underdiagnosed in older adults. Medications such as corticosteroids, levothyroxine, and antiepileptic drugs are well-documented contributors to BMD loss. This study examines how long-term use of these medications affects osteoporosis prevalence, specifically focusing on sex and age differences. This retrospective electronic health record study included adults aged 50–90 years. In the revised primary analysis, osteoporosis was defined using diagnosis codes only; osteoporosis screening codes and DXA procedure codes were excluded from the primary outcome. Associations between chronic exposure to corticosteroids, levothyroxine, or antiepileptics and osteoporosis ascertainment were estimated using modified Poisson regression with robust standard errors, adjusting for demographics, healthcare utilization, medication-specific indication diagnoses, and bone health covariates. Adjusted prevalence ratios and adjusted prevalence differences were reported. In the revised diagnosis-only analysis, associations were attenuated relative to the original broad outcome definition. Chronic corticosteroid exposure retained a positive but imprecise association with osteoporosis ascertainment (aPR 1.57, 95% CI 0.79–3.12; aPD +1.15 percentage points), whereas antiepileptic exposure (aPR 0.92, 95% CI 0.66–1.30; aPD −0.15 percentage points) and levothyroxine exposure (aPR 0.71, 95% CI 0.42–1.21; aPD −0.59 percentage points) were not associated with higher adjusted prevalence. Long-term use of corticosteroids, levothyroxine, or antiepileptics increases osteoporosis risk, with sex- and age-specific patterns. Men, especially younger ones, showed greater relative risk, while women exhibited higher absolute prevalence. These findings support targeted screening and early intervention strategies based on medication exposure, age, and sex.