Real-world clinical and economic outcomes for daily teriparatide patients in Japan.

A large medical and pharmacy claims database was used to evaluate outcomes in daily teriparatide (D-TPD) patients in Japan. 445 patients were identified (April 2008-July 2013) with 6+ months' pre- and 18+ months' post-index observation. D-TPD 20 µg subcutaneous injection is indicated for individuals at high risk of fracture. Descriptive analyses were conducted on clinical fractures, health care utilization, and costs. Adherence was measured by medication possession ratio (MPR) (High MPR > 0.8; 0.5 ≤ Medium MPR ≤ 0.8, Low MPR < 0.5). Adjusted analyses of Lower (Low + Medium) MPR vs. High MPR for fracture incidence and hospital admissions were performed using logistic and Poisson regression models; adjusted cost analyses used propensity bin bootstrapping methods. Baseline characteristics were: mean 74.7 years (standard deviation = 8.9); 90 % female; 20 % 1+ fracture. Post-index, 249 and 196 patients had High and Lower MPR, respectively. Mean incident fractures/1000 patient-years for Lower and High MPR patients were 77.4 and 57.7, respectively. Adjusted fracture risk was greater in Lower MPR patients [logistic odds ratio (OR) = 1.67, 95 % confidence interval 0.791-3.541; Poisson incidence rate ratio (IRR) = 1.505, 0.764-2.966]. Hospital admission risk was significantly greater in Lower than High MPR patients (OR = 1.85, 1.169-2.921; IRR = 1.47, 1.137-1.904). High MPR patients had numerically lower inpatient and total unadjusted costs than Lower MPR patients. Adjusted inpatient costs were significantly less in High MPR patients; outpatient and pharmacy costs were greater. Better adherence to D-TPD revealed a trend towards lower fracture risk, and significant reductions in hospital admissions and costs. The small sample size limited the robustness of these results.