Impact Of Medication Adherence On Health Outcomes Among Patients With Osteoporosis

Objectives:(1) To quantify the medication adherence among osteoporosis treatedwith bisphosphonate in Tianjin, China;(2) To assess the factors associated withmedication adherence;(3) To analysis the impact of medication adherence on fracturerate, health care utilization and costs.Methods: Data for the study were obtained from the Tianjin Urban EmployeeBasic Medical Insurance (UEBME) database (2008-2011) with30%random selection.Patients with≥1osteoporosis diagnosis, aged45years and older, and who had≥1pharmacy claim for bisphosphonate during2009-2010were selected.12-monthcontinuous enrollment before and after the first pharmacy claim for bisphosphonateswere required. No fracture for the first diagnosis, and no diagnosis of tumor or Paget’sdisease during2008-2011. Prevalence rate and medication treatment rate during2008-2011were counted. Medication possession ratio (MPR) was used to quantify themedication adherence and comparisons of fracture rate, health care utilization andcost were conducted between patients with different MPR. Multiple liner regression(MLR) was used to analysis factors associated with adherence. Cox proportionalhazards regression and logarithm MLR were conducted to analysis the impact ofmedication adherence on fracture rate and costs. Three subgroups including patientswith fracture during baseline, patients with osteoporosis diagnosis during baseline,and patients treated with other anti-osteoporosis drugs during baseline were used toanalysis the robustness of results.Results:(1) The number of osteoporosis patients increased during2008-2011and69,476osteoporosis patients were diagnosed in2011, prevalence rate was1.4%andmedication treatment rate was20.6%(N=14,431).(2)2,200osteoporosis patientsMPR reduced over time and the mean MPR was0.15(±0.19) in the12-monthanalyses. Only6.2%(N=136) and2.1%(N=46) patients were found with MPR above0.5and0.8respectively.(3) Patients with better compliance (MPR≥0.5) were foundwith higher fracture rate (18.4%vs.7.1%，p<0.001), more osteoporosis relatedoutpatient visits and higher costs.(4) Regression results showed that patients withprior medication treatment during the baseline period may have better compliance.And better compliance was associated with higher fracture rate and costs, for the patients with better compliance may have more serious disease.(5) More outpatientvisits and costs were found with patients with better compliance (MPR≥0.5) insubgroup analysis, while lower hospitalizations and inpatient costs were found withpatients with better compliance (MPR≥0.5) in patients with pre-osteoporosisdiagnosis and prior anti-osteoporosis drugs.Conclusions:(1) Low diagnosis rate and treatment rate were found amongpatients with osteoporosis in Tianjin, China.(2) Poor adherence to bisphosphonatewas found among osteoporosis patients in Tianjin, China.(3) The causalityrelationship between medication adherence and fracture rate, health care utilizationand costs was still uncertain.