Predictors of bone mineral density testing in patients at high risk of osteoporosis: secondary analyses from the OSTEOPHARM randomized trial

Abstract

In a randomized trial, we demonstrated that a community pharmacist osteoporosis screening intervention doubled the rates of bone mineral density (BMD) testing in high-risk patients. The purpose of this secondary analysis was to evaluate the potentially modifiable factors associated with BMD testing. From 2005 to 2007, 15 pharmacies randomized 262 patients to intervention (education, pamphlets, point-of-care quantitative heel ultrasound [QUS]) or usual care. The main outcome was BMD testing within 4mo. Multivariate regression was used to determine independent correlates of BMD testing. The median age of the cohort was 62yr, 65% were women, and 49% (n=129) were randomized to intervention. Compared with patients who were not tested, those with BMD were more likely to be women (p=0.007) and have excellent or very good health (p<0.001). Postrandomization correlates of BMD test were intervention (p=0.017), greater osteoporosis knowledge (p=0.004), and osteoporosis-specific physician visits (p<0.001). In adjusted analyses, only female sex (adjusted odds ratio [aOR]: 3.0; 95% confidence interval [CI]: 1.3-7.4) and osteoporosis-specific visits (aOR: 3.2; 95% CI: 1.4-7.8) were independently associated with BMD testing. In analyses restricted to intervention patients, abnormal QUS (aOR: 3.7, 95% CI: 1.4-9.1) was the only independent predictor of BMD test. Future interventions should incorporate the finding that osteoporosis-specific visits and abnormal QUS results were strongly associated with getting a BMD testing and should give greater attention to men.