The role of patient inexperience in medication discontinuation: A retrospective analysis of medication nonpersistence in seven chronic illnesses

Background: Previous studies have reported that patients who are medication naive in some medication classes have a higher risk of medication discontinuation during the first 30 days of treatment and shorter median times to discontinuation than do medication-experienced patients.Objectives: This study compared the risk of discontinuation during the first 30 days after the index fill and the median time to discontinuation for medication-naive and medication-experienced patients who were prescribed drugs for asthma, diabetes mellitus, high cholesterol, cardiovascular disease, breast cancer, glaucoma, or osteoporosis.Methods: Deidentified outpatient pharmacy records from 4 large US retail chains were searched for patients who had obtained a prescription for one of the index medications between January 1, 2007, and January 31, 2007. Included medication classes were asthma inhalers, asthma pills, oral breast-cancer agents, cardiovascular medications, oral antidiabetic medications, insulin, glaucoma drops, osteoporosis medications, and statins. Patients were divided into 2 groups: those for whom an in-class medication was not dispensed in the previous 180 days were identified as medication naive, and those for whom an in-class medication was dispensed during this period were identified as medication experienced. Discontinuation was defined as being >30 days late for a scheduled refill. Time to discontinuation was measured using Kaplan-Meier analysis during a 360-day follow-up period.Results: Data for 2.17 million patients who received prescriptions from 3821 pharmacies were analyzed. During the first 30 days of therapy, rates of discontinuation for medication-naive patients were 17.4% to 42.6% higher than for medication-experienced patients, and their median times to discontinuation were 14.2% to 28.9% as long. The difference in persistence between medication-naive and medication-experienced patients emerged during the first 30 days of medication use, after which rates of discontinuation were relatively consistent. Patients prescribed nonoral medications for diabetes mellitus (insulin), asthma (inhalers), and glaucoma (drops) had an especially high risk of discontinuation during the first month of treatment.Conclusion: This study found that medicationnaive patients had a higher risk of discontinuation during the first 30 days of therapy than did medicationexperienced patients, regardless of the medication class prescribed, leading to shorter median times to discontinuation.