The prevalence of overtreatment of osteoporosis: results from the PAADRN trial.
Arch Osteoporos. 2018 Sep 28;13(1):103 Authors: Hall SF, Wright NC, Wolinsky FD, Lou Y, Edmonds S, Roblin D, Jones M, Saag K, Cram P
PURPOSE: There is a robust body of literature addressing undertreatment in osteoporosis, but limited data addressing overtreatment. Understanding overtreatment is important to minimize harm and decrease costs. METHODS: One of the pre-specified post hoc analyses of the PAADRN trial, a randomized, controlled, pragmatic clinical trial, was to quantify and identify risk factors associated with osteoporosis overtreatment. PAADRN included patients ≥ 50 years of age presenting for bone density testing between February, 2012, and August, 2014, at three US healthcare systems. We assessed 20,397 patients for eligibility and randomized 7749. Intervention patients received a tailored letter containing their dual-energy X-ray absorptiometry (DXA) results and an educational osteoporosis brochure. Control patients received usual care. Using the National Osteoporosis Foundation treatment guidelines, we defined overtreatment as the receipt of osteoporosis pharmacotherapy 12 weeks after DXA when treatment was not indicated. We evaluated the relationship between the following baseline variables-sex, race/ethnicity, educational attainment, and differences across health systems-and overtreatment using a series of multivariable logistic regression models. RESULTS: Among 3602 patients with no apparent indication for osteoporosis treatment, 292 (8.1%; 95% CI, 7.22 to 9.00%) received a new prescription for osteoporosis pharmacotherapy or were instructed to continue an existing medication (presumed overtreatment). Presumed overtreatment was more common among participants with prior DXA history, those who reported a history of osteoporosis or low bone mass, and those referred for testing by family medicine providers. CONCLUSION: In our sample of older adults, overuse of osteoporosis pharmacotherapy was only 8.1%. Nevertheless, overtreatment exposes patients to possible risk with negligible chance of benefit and should be minimized. TRIAL REGISTRATION: clinicaltrials.gov identifier: NCT01507662. PMID: 30267162 [PubMed - in process]