To determine the point prevalence of medication errors at the time of transition of care from an ICU to non-ICU location and assess error types and risk factors for medication errors during transition of care.
This was a multicenter, retrospective, 7-day point prevalence study.
Fifty-eight ICUs within 34 institutions in the United States and two in the Netherlands.
Nine-hundred eighty-five patients transferred from an ICU to non-ICU location.
Measurements and Main Results:
Of 985 patients transferred, 450 (45.7%) had a medication error occur during transition of care. Among patients with a medication error, an average of 1.88 errors per patient (sd, 1.30; range, 1–9) occurred. The most common types of errors were continuation of medication with ICU-only indication (28.4%), untreated condition (19.4%), and pharmacotherapy without indication (11.9%). Seventy-five percent of errors reached the patient but did not cause harm. The occurrence of errors varied by type and size of institution and ICU. Renal replacement therapy during ICU stay and number of medications ordered following transfer were identified as factors associated with occurrence of error (odds ratio, 2.93; 95% CI, 1.42–6.05; odds ratio 1.08, 95% CI, 1.02–1.14, respectively). Orders for anti-infective (odds ratio, 1.66; 95% CI, 1.19–2.32), hematologic agents (1.75; 95% CI, 1.17–2.62), and IV fluids, electrolytes, or diuretics (odds ratio, 1.73; 95% CI, 1.21–2.48) at transition of care were associated with an increased odds of error. Factors associated with decreased odds of error included daily patient care rounds in the ICU (odds ratio, 0.15; 95% CI, 0.07–0.34) and orders discontinued and rewritten at the time of transfer from the ICU (odds ratio, 0.36; 95% CI, 0.17–0.73).
Nearly half of patients experienced medication errors at the time of transition of care from an ICU to non-ICU location. Most errors reached the patient but did not cause harm. This study identified risk factors upon which risk mitigation strategies should be focused.