The Association of Adherence to Osteoporosis Therapies with Fracture, All-Cause Medical Costs, and All-Cause Hospitalizations: A Retrospective Claims Analysis of Female Health Plan Enrollees with Osteoporosis

BACKGROUND: Osteoporosis affects approximately 10 million people in the United States and is associated with increased fracture risk and fracture-related costs. Poor adherence to osteoporosis medications is associated with higher general burden of illness compared with optimal adherence. OBJECTIVE: To examine the associations of adherence to osteoporosis therapies with (a) occurrence of closed fracture, (b) all-cause medical costs, and (c) all-cause hospitalizations. METHODS: This retrospective analysis of administrative claims data examined women with osteoporosis initiating therapy with alendronate, risedronate, ibandronate, or raloxifene from July 1, 2002, to March 10, 2006. Data were from a large, geographically diverse U.S. health plan that covered about 12.6 million females during the identification period. Commercially insured and Medicare Advantage plan enrollees were observed for 1 year before (baseline period) and 540 days after therapy initiation (follow-up period). Outcomes included closed fractures, all-cause medical costs, and all-cause hospitalizations; all outcomes were measured starting 180 days after therapy initiation through follow-up. All subjects had at least 2 pharmacy claims for any of the targeted osteoporosis medications. Adherence was measured with a medication possession ratio (MPR) and accounted for all osteoporosis treatment. High adherence was MPR of at least 0.80; low adherence was MPR less than 0.50. Covariates included baseline fracture, "early" fracture (in the first 180 days of follow-up), baseline corticosteroid or thyroid hormone use, health status indicators, and demographic characteristics. Outcome fractures were modeled with Cox survival regression with time-dependent cumulative MPR. All-cause medical costs and all-cause hospitalizations were modeled, respectively, with generalized linear model regression (gamma distribution, log link) and negative binomial regression. RESULTS: The sample comprised 21,655 patients-16,295 (75.2%) commercial and 5,360 (24.8%) Medicare Advantage. During the entire follow-up period, 5,406 (33.2%) and 2,253 (42.0%) of commercial and Medicare Advantage patients, respectively, had low adherence. Adherence tended to decrease over the follow-up period. The Cox regression showed that commercial plan patients with low versus high adherence had 37% higher risk of fracture (hazard ratio = 1.37, 95% CI = 1.12-1.68). Adherence was not significantly associated with fracture in the Medicare Advantage cohort. Commercial and Medicare Advantage patients with low versus high adherence had 12% (exponentiated coefficient = 1.12, 95% CI = 1.02-1.24) and 18% (exponentiated coefficient = 1.18, 95% CI = 1.04-1.35) higher all-cause medical costs during months 7 through 18 of follow-up. Commercial and Medicare Advantage patients with low versus high adherence had 59% (incidence rate ratio [IRR] = 1.59, 95% CI = 1.38-1.83) and 34% (IRR = 1.34, 95% CI = 1.13-1.58) more all-cause hospitalizations during months 7 through 18 of follow-up, respectively. CONCLUSIONS: Low adherence to osteoporosis pharmacotherapy was associated with higher risk of fracture for commercially insured but not Medicare Advantage patients and with higher all-cause medical costs and more all-cause hospitalizations in both groups. These results are consistent with the literature and highlight the importance of promoting better adherence among patients with osteoporosis.