Lesley Meng
Assistant Professor of Operations Management
Yale University, United States
Disclosure information not submitted.
Ann Huffenberger
Director of Operations for the Center for Connected Care
University of Pennsylvania, United States
Disclosure information not submitted.
Christian Terwiesch, n/a
Professor of Operations, Information and Decisions
University of Pennsylvania, United States
Disclosure information not submitted.
Krzysztof Laudanski, MD, PhD, FCCM
University of Pennsylvania
Philadelphia, Pennsylvania
Disclosure information not submitted.
Title: A quantification of the effect of shift-changes on the delivery of medications to ICU patients.
Introduction: The efficient delivery of care remains a high priority in all medical settings, yet the quantification of ineffective care has been difficult to construe. We hypothesize that medication delays in the intensive care unit (ICU) will increase due to the provider shift-change, resulting in patient care delays overall.
Methods: 41,370 patients admitted to the ICU of an academic hospital between January 2012 and December 2015 were enrolled in this retrospective observational cohort study. The potential effect of shift-changes on the delivery of vasodilators, antipyretics, and bronchodilators was quantified by the odds of a patient developing hypertension, a fever, or tachycardia, respectively. Two instrumental models were used to estimate the effects of medication delay on physiological parameters. A naïve approach was used where the effect of patient prioritization was not accounted for as well as an instrumental variable (IV) approach where this and patient clinical severity were accounted for.
Results: Medications ordered for immediate delivery to ICU patients are delayed by 88 minutes on average. The delay around a nurse shift change was 60.81 minutes (p < 0.001) on average for antipyretics, 39.51 minutes (p < 0.001) on average for bronchodilators, and 57.11 minutes (p < 0.001) on average for vasodilators. Only the model incorporating ICU patients' clinical severity demonstrated that delays in administering medication can result in increased odds of the patient developing a fever by 32.94% if antipyretics are delayed by 61 minutes, 79.5% for tachycardia in the case of a delay in bronchodilators, and 134% for hypertension in the case of vasodilators.
Conclusions: ICU provider shift-changes result in a quantifiable interruption of care delivery as seen through medication delays. We demonstrated a possible bias introduced when providers prioritize clinically severe patients when delivering medications.