Association Between Method of Prescribing and Primary Nonadherence to Dermatologic Medication in an Urban Hospital Population

IMPORTANCE Prescription underuse is associated with poorer clinical outcomes. A significant proportion of underuse is owing to primary nonadherence, defined as the rate at which patients fail to fill and pick up new prescriptions. Although electronic prescribing increases coordination of care and decreases errors, its effect on primary nonadherence is less certain. OBJECTIVES To analyze factors associated with primary nonadherence to dermatologic medications and study whether electronic prescribing affects rates of primary nonadherence. DESIGN, SETTING, AND PARTICIPANTS A retrospective review of medical records was conducted from January 1, 2011, to December 31, 2013, among a cohort of new patients prescribed dermatologic medications at a single, urban, safety-net hospital outpatient dermatology clinic. MAIN OUTCOMES AND MEASURES The primary outcome was the overall rate of primary nonadherence, defined as filling and picking up all prescribed medications within a 1-year period, and the difference in primary nonadherence between patients who received electronic prescriptions and those who received paper prescriptions. Secondary outcomes included the association of primary nonadherence with sex, age, relationship status, primary language, race/ethnicity, and number of prescriptions. RESULTS A total of 4318 prescriptions were written for 2496 patients (mean [SD] age, 47.7 [13.2] years; 849 men and 1647 women). The overall rate of primary nonadherence was 31.6%(n = 788). Based on multivariable analysis, the risk of primary nonadherence was 16 percentage points lower among patients given an electronic prescription (15.2%) than patients given a paper prescription (31.5%). Primary nonadherence decreased with age (< 30 y, 38.9%; 30-49 y, 35.3%; and 50-69 y, 26.3%), and then increased in elderly patients 70 years and older (31.9%). Of patients who were given 1, 2, 3, 4, or 5 prescriptions, rates of primary nonadherence were 33.1%, 28.8%, 26.4%, 39.8%, and 38.1%, respectively. Primary nonadherence decreased with age but then increased in elderly patients. Patients identifying English as their primary language had the highest rate of primary nonadherence (33.9%) compared with Spanish (29%) or other speakers (20.4%). CONCLUSIONS AND RELEVANCE Compared with paper prescriptions, electronic prescriptions were associated with less primary nonadherence. Number of prescriptions, language, race/ethnicity, and age were associated with increased rates of primary nonadherence. Efforts must be made to understand why primary nonadherence occurs, identify patients prone to primary nonadherence, and simplify medication regimens to maximize adherence and quality of care.