Article Title

Minimizing Medication Histories Errors for Patients Admitted to the Hospital Through the Emergency Department: A Three-Arm Pragmatic Randomized Controlled Trial of Adding Admission Medication History Interviews by Pharmacists or Pharmacist-Supervised Pharmacy Technicians to Usual Care

Publication Date



adverse drug events, medication reconciliation


Background/Aims: Meta-analysis shows hospitalized U.S. patients suffer 106,000 fatal adverse drug events (ADE) annually. Errors in admission medication histories (AMH) frequently cause preventable ADEs. There is also concern that electronic health records (EHR) propagate AMH errors into ADEs by making AMHs easily orderable. We hypothesized that for patients with complex AMHs, pharmacist-supervised pharmacy technician (PSPT) and pharmacist AMH interviews would reduce AMH errors.

Methods: At a large university-affiliated hospital, we conducted an IRB-approved, pragmatic randomized trial with informed consent waived. Inclusion criteria, accessed via EHR, were: ≥ 10 chronic prescription medications, history of acute myocardial infarction or congestive heart failure, admission from skilled nursing facility, history of transplant, or active anticoagulant, insulin or narrow therapeutic index medications; patients were excluded if admitted to pediatric or trauma services or transplant services with pharmacists. All arms received usual care for patients admitted from the emergency department. This included registered nurses obtaining AMHs, and natural variation in checking, correcting and ordering from these AMHs by admitting providers. In the two intervention arms, PSPTs and pharmacists obtained initial AMHs. They contacted family, pharmacies and/or providers to resolve questions. As per prior studies, we obtained reference standard AMHs from all patients (usually one day after admission), and initial AMH errors were independently rated by ≥ 2 pharmacists as significant, serious or life-threatening. Three error severities were assigned –– 1, 4 and 9 points, respectively. We calculated weighted error scores for each patient, and mean scores/patient for each arm. We had 80% power to identify arms with higher/lower scores of ≥ 11.2/patient.

Results: Of 311 patients enrolled, 28 patients were later excluded (most not ultimately admitted, or discharged before reference standard AMH obtained). Median age and number of medications were 76 and 14 (interquartile ranges: 63–84 and 10–19), respectively. Patients in the usual care, pharmacist and PSPT arms had 7.4, 1.4 and 1.5 AMH mean errors/patient, respectively (P<0.0001); mean error scores were 21.2, 3.9 and 3.9 per patient, respectively (P<0.0001). Physician scoring of admission medication order errors due to AMH errors is ongoing.

Discussion: Pharmacists and PSPTs substantially reduced AMH errors and weighted error scores, and thus show promise for reducing ADEs. Limitations include our single site, and using intermediate endpoints.




March 30th, 2015


April 28th, 2015