BACKGROUND: Identification of early predictors of medication nonpersistence may allow timely adherence-promoting interventions and potentially reduce the risk of negative health outcomes. OBJECTIVE: This study was conducted to determine whether delay in filling an initial statin prescription predicts subsequent nonpersistence with medication. METHODS: This observational study of a cohort of adult patients (>18 years) who newly initiated statin therapy between December 1997 and June 2000 employed data from the administrative claims database of a large US managed care organization. Patients initiating statin therapy had to have at least 18 months of continuous eligibility and no statin use in the 6-month period before the index prescription. A new measure, dispensation delay, was measured as the gap between the most recent physician or hospital visit and the fill date of the index prescription. Five categories of dispensation delay were created--no delay, 1 to 7 days, 8 to 30 days, 31 to 183 days, and >183 days. Nonpersistence was defined as a gap of >or=30 days in the statin prescription supply during the follow-up period. Cox proportional hazards regression was used to model the risk of the initial dispensation delay on the time to discontinuation, controlling for such variables as demographic characteristics, comorbidities, physician specialty, and previous health care utilization. RESULTS: The final sample included 19,038 patients. Among all variables studied, the dispensation-delay variables were the most significant predictors of non-persistence, with a longer delay predicting a higher risk of early discontinuation. Patients with delays in filling the initial prescription of >30 days but <183 days were 30% more likely to discontinue therapy than those without delays (hazard ratio=1.30; 95% CI=1.20-1.40). CONCLUSIONS: The delay in filling the first statin prescription significantly predicted future non-persistence. Use of this measure may allow early identification of patients at high risk for early discontinuation.
BACKGROUND: Identification of early predictors of medication nonpersistence may allow timely adherence-promoting interventions and potentially reduce the risk of negative health outcomes. OBJECTIVE: This study was conducted to determine whether delay in filling an initial statin prescription predicts subsequent nonpersistence with medication. METHODS: This observational study of a cohort of adult patients (>18 years) who newly initiated statin therapy between December 1997 and June 2000 employed data from the administrative claims database of a large US managed care organization. Patients initiating statin therapy had to have at least 18 months of continuous eligibility and no statin use in the 6-month period before the index prescription. A new measure, dispensation delay, was measured as the gap between the most recent physician or hospital visit and the fill date of the index prescription. Five categories of dispensation delay were created--no delay, 1 to 7 days, 8 to 30 days, 31 to 183 days, and >183 days. Nonpersistence was defined as a gap of >or=30 days in the statin prescription supply during the follow-up period. Cox proportional hazards regression was used to model the risk of the initial dispensation delay on the time to discontinuation, controlling for such variables as demographic characteristics, comorbidities, physician specialty, and previous health care utilization. RESULTS: The final sample included 19,038 patients. Among all variables studied, the dispensation-delay variables were the most significant predictors of non-persistence, with a longer delay predicting a higher risk of early discontinuation. Patients with delays in filling the initial prescription of >30 days but <183 days were 30% more likely to discontinue therapy than those without delays (hazard ratio=1.30; 95% CI=1.20-1.40). CONCLUSIONS: The delay in filling the first statin prescription significantly predicted future non-persistence. Use of this measure may allow early identification of patients at high risk for early discontinuation.
Authors: Marsha A Raebel; Julie Schmittdiel; Andrew J Karter; Jennifer L Konieczny; John F Steiner Journal: Med Care Date: 2013-08 Impact factor: 2.983