Medication Discrepencies

I have wanted to do EMS research for many years. At long last, my first official research project with official IRB approval was completed, and our poster below was presented at the New England Regional Society of Academic Emergency Medicine Conference last March. Mark Andreae, a third year medical student and EMT, was the key person on the project. He had the patience and good nature to get us through the complicated IRB process, and sat with me as we went through the records. Dr. Richard Kamin provided initial guidance, and Dr. João Delgado, a research maven, brought the project home for us, helping turn all we had done into a polished product. I will post commentary on the project next week. For now, here is the poster. We hope to write our findings into a full research article for journal submission.


Medication Administration Discrepancies Between Prehospital and Emergency Department Records
Mark Andreae MS III, Peter Canning Paramedic, RN, Richard Kamin MD, João Delgado MD

Communication errors between providers are ubiquitous in the healthcare system and can compromise patient safety. Patient handoff’s between the prehospital setting and the emergency department (ED) are particularly challenging because of the need to communicate quickly while actively treating the patient, among other factors. Accurate communication of medications and dosages administered are especially important because they can influence subsequent care dramatically.

Our primary objective was to determine the discrepancy rate between medications and dosages noted on the prehospital patient care report (PCR) and the ED chart.

We also sought to determine whether completion of a brief form would influence discrepancy rate.

Study Design
Prospective, observational before/after study of a single paramedic intercept service with an annual call volume of 2432, transporting primarily to a community ED with an annual census of 30,000. Only intervention was the introduction of a brief medication handoff form.

Data Analysis
Descriptive calculations were performed. Means and proportions between phases were compared using t-tests or chi-square analysis as appropriate.

Medication Discrepancies
Discrepancies were categorized according to the following:
Medication listed in EMS PCR and not listed in the ED record.
Medication not listed in EMS PCR and listed in the ED record.
Medication listed correctly but the dose given did not match or was not recorded in either record.

Form Implementation
The form was introduced at the regular staff meeting. Paramedics were instructed to complete the form and hand it to the patient’s assigned nurse prior to exiting the ED.

Data were collected before and after form implementation with a 1 week interval between phases.

Sample size determination
Assuming a baseline discrepancy rate of 30%, a sample size of 118 patients per group would provide 80% power to detect a 50% decrease in discrepancy rate (two-tailed alpha = 0.5). We added 10% to account for missing data or a different effect size to reach the final sample size of 129 per group.


We analyzed a total of 258 total record pairs.

Mean patient age was 64.3 +/- 22.3; 54% were women.

Overall discrepancy rate was 36.4%.

No significant difference between the two periods: 33.3% before vs. 39.5% after form implementation (p = 0.1329)

In both time periods, nebulized medications (22.3%), ondansetron (20.2%), and aspirin (16%) accounted for the majority of discrepancies.

Controlled medications were listed in 33 cases. Four of these (12.1%) were not recorded in the ED record.

Distribution of Medication Discrepancy Types
Medication listed in EMS PCR and not listed in the ED record (69%).
Medication not listed in EMS PCR and listed in the ED record (3.2%).
Medication listed correctly but the dose given did not match or was not recorded in either record (29.8%).


Observational study design of a single EMS/ED system.

We were unable to determine whether the lack of effect of the brief medication form was a result of inconsistent use of the form or another factor.


In this pilot study of a single paramedic intercept service and a single community ED, the discrepancy rate between medications recorded in the EMS PCR and the ED chart was substantial.

Medications listed in the EMS PCRs but not in the ED record were the most common discrepancy noted.