Objective: The study's objective was to describe readily identifiable predictors of filling medication prescriptions after discharge from the pediatric emergency department (PED).
Methods: The study was a prospective cohort study of caregivers of children aged 0 to 17 years, discharged from the PED of an urban safety net hospital with a medication prescription. Prescription filling was confirmed by direct contact with pharmacies. Logistic regression was used to estimate the association between baseline characteristics and prescription filling.
Results: Overall, 36 (32%) of 111 families did not fill their children's prescriptions. We found no association between any predictors of interest and prescription filling. In the patient attributes domain, neither English as one's nondominant language (adjusted odds ratio [aOR], 0.72; 95% confidence interval [CI], 0.25-2.10) nor low health literacy (aOR, 0.78; 95% CI, 0.17-3.62) was associated with prescription filling. In the patient-provider interaction domain, poor physician-family communication (aOR, 1.52; 95% CI, 0.50-4.61), lack of trust in the medical provider (aOR, 0.68; 95% CI, 0.24-1.77), and caregiver disagreement with the treatment plan (aOR, 0.81; 95% CI, 0.14-4.92) had no association with prescription filling. In the patient-health system interaction domain, concern that the prescription would be unaffordable (aOR, 1.30; 95% CI, 0.48-3.53) and lack of an identified primary care physician for the child (aOR, 0.29; 95% CI, 0.08-1.04) were not associated with filling the prescriptions.
Conclusions: Among a low-income urban population, approximately one third of families do not fill prescriptions for their children after discharge from the PED. We were unable to predict which families would fill prescriptions and which would not.