Arch Osteoporos. 2026 Jul 1;21(1):98. doi: 10.1007/s11657-026-01730-9.
ABSTRACT
We examined the association of COVID time periods and equity-related variables with pharmacotherapy in a large jurisdiction fracture liaison service (FLS). We did not observe a significant association between COVID time periods and medication prescription after adjusting for all covariates. This highlights another potential success of the FLS model.
PURPOSE: Our objective was to examine the impact of COVID on bone-active medication prescription in a fracture liaison service (FLS), after adjusting for fracture risk status and equity-related variables.
METHODS: We conducted a logistic regression analysis with medication prescription (prescription by a bone health specialist or primary care provider) as the outcome. The model included covariates COVID time periods (T1: “pre-COVID” (n = 2796); T2: “during COVID” (n = 1575); T3: “COVID recovery” (n = 2208)), fracture risk status (high risk/not high risk) and equity-related variables (sex, age, marital status, living arrangement, education status, geographic location, and presence of comorbidities). Goodness of fit was assessed with the area under the receiver operating characteristic curve (AUC) and the Hosmer and Lemeshow test.
RESULTS: Fracture risk status was the primary driver of treatment with high-risk patients 7.8 times more likely to receive a medication prescription compared to patients who were not high risk, after adjusting for all covariates (OR = 7.80 [95% CI 6.91, 8.79]). COVID time period was not statistically significant. Female patients, those married or in a common-law relationship, living alone, or residing in urban areas were more likely to be prescribed medication. The model had good prediction power and fit the data well (AUC: 0.77; Hosmer-Lemeshow test p-value: 0.83).
CONCLUSION: Among patients reached by the FLS, COVID time period was not significantly associated with medication prescription, although program reach decreased at T2 and T3. Fracture risk status, sex, marital status, living arrangement, and geographic location were significantly associated with medication prescription.
PMID:42384300 | DOI:10.1007/s11657-026-01730-9
