By Drayton Hammond, PharmD, MBA, BCPS, BCCCP

Clinical Pharmacy Specialist, Adult Critical Care, Rush University Medical Center, Chicago

Dr. Hammond reports no financial relationships relevant to this field of study.

SYNOPSIS: Three factors associated with decreased odds of an error occurring were daily patient care rounds in the ICU, discontinuing and rewriting medication orders during the transition of care from the ICU to a non-ICU setting, and 16-20 ICU beds in the transferring ICU.

SOURCE: Tully AP, Hammond DA, Li C, et al. Evaluation of medication errors at the transition of care from an ICU to non-ICU location. Crit Care Med 2019;47:543-549.

Medication errors may occur at any point during a patient’s hospitalization, although transition points from different levels of care (e.g., ED to ICU, ICU to floor, and floor to outpatient setting) add an additional element of potentiating the error until the patient’s next formal interaction with a healthcare provider.1,2 The financial costs and mortality attributable to these errors in the United States represent at least $19.5 billion and 98,000 deaths, respectively.3 Guidelines and best practice statements for transition of care (TOC) focus predominately on hospital discharge, which has many similarities but also differences compared to an inpatient TOC.4,5 The extent to which medication errors occur during the TOC from the ICU to a lower acuity inpatient setting and risk factors associated with development or prevention of those errors has not been described.

Tully et al completed a multicenter, observational, seven-day study of patients’ first transfer from an ICU to a non-ICU setting within the same institution to describe the point prevalence and types of medication errors and patient-, medication-, and system-specific factors associated with their development.6 A pharmacist evaluated medication orders that were active within one hour pre- and post-ICU transfer for potential medication errors. These pharmacists were provided training and reference documents to facilitate valid and reliable identification of potential medication errors. Prevalence and characteristics of errors were determined using descriptive statistics. Characteristics between those TOCs with and without a medication error were compared. Characteristics with a P value < 0.05 were considered for inclusion in a multivariate logistic regression analysis to determine independent risk factors for medication errors at TOC.

Of the 985 TOCs evaluated, 450 had at least one medication error. Most patients experienced a single error (55.1%), although the mean number of errors was 1.88 (standard deviation, 1.30; range, 1-9). The most common error types were continuation of a medication with an ICU-specific indication (28.4%), untreated condition (19.4%), and medication without a clear indication (11.9%). The most common untreated conditions were cardiac (27.6%) or neurologic (12.9%) in nature. Three-quarters of errors reached patients, although 94.2% did not cause patient harm. For those errors that did cause patient harm, the most common types of errors were incorrect dose (22.6%) and untreated condition (18.9%), and medication classes were anti-infective (28.6%), cardiovascular (18.4%), and neurologic (12.2%). Patient-specific factors associated with increased odds of medication errors were renal replacement therapy during ICU stay (odds ratio [OR], 2.93; 95% confidence interval [CI], 1.42-6.05) and number of medications ordered following TOC (OR, 1.08; 95% CI, 1.02-1.14). Medication-specific factors associated with increased odds of medication errors were receipt in the ICU of an anti-infective agent (OR, 1.66; 95% CI, 1.19-2.32), hematologic agent (OR, 1.75; 95% CI, 1.17-2.62), and intravenous fluid, electrolyte, or diuretic agent (OR, 1.73; 95% CI, 1.21-2.48). System-specific factors associated with increased odds of medication errors were community teaching hospital (OR, 3.96; 95% CI, 1.79-8.79) and 500-999 total inpatient hospital beds (OR, 4.26; 95% CI, 1.05-17.32). System-specific factors associated with lower odds of medication errors were daily patient care rounds in the ICU (OR, 0.15; 95% CI, 0.007-0.34), discontinuing and rewriting medication orders during the TOC from the ICU to a non-ICU setting (OR, 0.36; 95% CI, 0.17-0.73), and 16-20 ICU beds in the transferring ICU (OR, 0.40; 95% CI, 0.21-0.74).


Medication errors occur in almost 50% of patients transitioning from the ICU to a non-ICU setting. While all errors placed patients at an increased risk for harm, approximately 5% resulted in patient harm during the hospitalization. The quantity and extent of harm from these errors likely are underestimated for at least three reasons. First, errors that were recognized and resolved during order verification likely were underreported because the data collection process was more complex for capturing and recording these types of errors. Additionally, the duration errors persisted, including presence of the error at hospital discharge, was not evaluated. Because of the retrospective nature of this research, interventions to resolve these errors were unable to be provided after they were identified. Finally, there was at least one dedicated pharmacist for each ICU from which patients were transferred. Pharmacists frequently recognize and resolve minor and major medication errors.7,8 However, approximately one-third of ICUs in the United States do not employ a partially or fully dedicated pharmacist.9 The quantity and harm from medication errors at institutions without dedicated ICU pharmacist services likely is greater than reported in this research.7

The three factors associated with lower odds of an error were system- and process-focused in nature and represent opportunities for improving patient safety while also likely improving other financial and patient care metrics.10 While implementing or improving the structure and formality of direct patient care rounds is a significant undertaking, the benefits can be substantial.10 Similarly, reducing the size of the critical care service to accommodate 16-20 patients at most may require additional resources but likely will increase the ability of all members of the healthcare team to adequately provide care for these patients. The final factor (discontinuing and rewriting medication orders during TOC) is straightforward to implement in most electronic health records and TOC workflow processes.11

The research by Tully et al may serve as both a call to action for institutions that are at an increased risk for medication errors and a trove of hypothesis-generating data for investigators interested in improving patient safety through process changes.


  1. Bell CM, et al. Association of ICU or hospital admission with unintentional discontinuation of medications for chronic diseases. JAMA 2011;306:840-847.
  2. Schnipper JL, et al. Role of pharmacist counseling in preventing adverse drug events after hospitalization. Arch Intern Med 2006;166:565-571.
  3. Andel C, et al. The economics of health care quality and medical errors. J Health Care Finance 2012;39:39-50.
  4. Agency for Healthcare Research and Quality. Strategy 4: Care Transitions From Hospital to Home: IDEAL Discharge Planning. Available at: Accessed June 10, 2019.
  5. World Health Organization. Transitions of Care: Technical Series on Safer Primary Care, 2016. Available at: Accessed June 10, 2019.
  6. Tully AP, et al. Evaluation of medication errors at the transition of care from an ICU to non-ICU location. Crit Care Med 2019;47:543-549.
  7. Leape LL, et al. Pharmacist participation on physician rounds and adverse drug events in the intensive care unit. JAMA 1999;282:267-270.
  8. Hammond DA, et al. Cost avoidance associated with clinical pharmacist presence in a medical intensive care unit. J Am Coll Clin Pharm 2019;1-6.
  9. MacLaren R, et al. Critical care pharmacy services in United States hospitals. Ann Pharmacother 2006;40:612-618.
  10. Bhamidipati VS, et al. Structure and outcomes of interdisciplinary rounds in hospitalized medicine patients: A systematic review and suggested taxonomy. J Hosp Med 2016;11:513-523.
  11. Barnsteiner JH. Medication Reconciliation. In: Hughes RG, editor. Patient Safety and Quality: An Evidence-Based Handbook for Nurses. Rockville (MD): Agency for Healthcare Research and Quality (US); 2008. Chapter 38. Available at: Accessed June 10, 2019.