Medication Errors and Adverse Drug Events (ADEs) in Ambulatory Pediatrics

The purpose of this study is to determine the rates of medication errors in pediatric outpatients in 6 office practices. Further, we wish to determine the effectiveness of a computerized physician order entry (CPOE) system in reducing errors.

This study has the following 2 goals: Aim 1: To determine the rates, types, and predictors of medication errors and ADEs. Aim 2: To perform a trial assessing the effectiveness of an intervention (CPOE) on reducing serious medication errors. We hypothesize that: Medication errors and ADEs are frequent in ambulatory pediatrics. Presence of the following predictors will be associated with higher error rates: cultural, racial, socioeconomic, educational, and linguistic barriers to communication, understanding, and successful completion of prescribed therapies; complex medical or chronic medical conditions; complex medication regimens; non-physician providers with limited clinical experience; high provider workloads; and complex prescription refill systems. Prevention strategies, including both technology-based and behavioral/ human factors-based interventions, will be effective and cost-effective in reducing rates of serious medication errors (defined as preventable ADEs and non-intercepted potential ADEs).

We performed a prospective cohort study of patients under age 21 from 6 office practices. Data collection methodologies included duplicate prescription review, 2 surveys, and chart review. All data were reviewed for medication errors, including those with the potential for harm (near misses) and those that actually caused harm (preventable ADEs). In addition, data were reviewed for harm from medications that were not associated with an error (nonpreventable ADEs). All ADEs were further characterized according to types and potential prevention strategies.

During the initial survey, we reviewed dispensed medications by having parents of patients with prescriptions under 21 read the medication label. We then asked questions about potential side effects, method of medication administration, communication regarding the medication with the pharmacist and the health care provider, and demographic information. Research assistants underwent several weeks of training to ensure a standardized approach to surveys.

If the parent completed the initial survey, we attempted a second survey 2 months after the index visit to capture persistent symptoms. Surveys were translated and back translated into Spanish and Cambodian and conducted by fluent bilingual interviewers. Surveys were pretested and revised based on 2 focus groups of parents representing a broad array of socioeconomic backgrounds.

Inclusion Criteria: Pediatric providers at the participating office practices during the study period as well as the parents of the patients of these providers. Exclusion Criteria: Pediatric providers or patients at other office practices.