Timing of blood draws among hospitalized patients: An evaluation of the electronic health records from a large health care system

Abstract

Importance: Early morning blood draws disrupt patients' sleep and adversely affect patient experience. Objective: To describe the trends in non-urgent early morning blood draws in a large health care system. Design: This cross-sectional study analyzed trends in non-urgent early morning (04:00-06:59) blood draws using the electronic medical records of patients hospitalized between November 2016 and October 2019. Setting: The study was done at Yale New Haven Health System, which is Connecticut's largest health care system consisting of 5 hospitals. Participants: We included all blood draw events for hospitalized adult patients (≥18 years at the time of admission). Main outcome and measures: We assessed the timing of non-urgent (excluding blood draws performed in the emergency department, critical care units, step-down units, and during the first 24 hours of admission) blood draws among hospitalized patients and determined if the trends of early morning (04:00-06:59) blood draws varied by patient characteristics or over the study duration. Results: Of the 5,676,802 blood draw events included, 38.9% occurred between 04:00-06:59 with only modest differences across age, race and ethnicity, and gender subgroups. The proportion of early morning blood draws increased over the study period from 36.9% to 41.4% (P<0.001). Of the early morning samples, those collected between 04:00-04:59 decreased from 25.6% to 18.5%, those collected between 05:00-05:59 increased from 39.6% to 45.5%, and those collected between 06:00-06:59 increased from 34.6% to 36.0% (P<0.001, each). Conclusions and relevance: The proportion of non-urgent early morning blood draws was persistently high over 3 years. There is a need to redesign hospital care to reduce sleep interruptions from non-urgent interventions.