Background: Hospital-acquired delirium induced via disruptions of the sleep cycle contributes to increased mortality and morbidity in hospitalized patients (1,2). Surveys of hospitalized general internal medicine patients at our institution, a tertiary academic medical center, have found that sleep duration is decreased in the hospital, with inpatients sleeping on average 4-6 hours, and reporting worse quality sleep than at home. These data supported implementation of a delirium prevention protocol. Implementation of a delirium prevention protocol including leaving IV pumps outside of rooms, later lab draws, and melatonin use.
Purpose: This project evaluates the effects of a delirium prevention protocol and unit-wide training on inpatient sleep.
Description: A multifaceted sleep improvement and delirium prevention protocol was developed and implemented in the form of an optional order set. Subsequently, a single adult medicine floor at UNMH was trained to adopt sleep improvement practices for all patients. A survey inquiring about sleep was administered to a convenience sample of 100 patients before and after initiation of the protocol and following unit-wide adoption of sleep improvement practices. Survey recorded subjective sleep hours and quality on a 0-5 scale. Data analysis in SAS software consisted of chi-squared tests for differences in proportions and t-tests for continuous variables.One year following initiation, 13% of patients surveyed received the delirium prevention protocol. Four years following initiation there was unit-wide adoption of sleep improvement practices. Prior to implementation, average hours of sleep reported was 4.1 (95% CI 3.7 – 4.4), with a reported average quality of 3.3 on a 0-5 scale. One year after implementation, average sleep time was 4.6 hours (95% CI 4.1-5.1) with average quality 3.15. Four years after implementation, average sleep time was 5.86 hours (95% CI 5.7-6.0) with average quality 3.81. The initial change in reported sleep time was an average increase of 0.5 hours (p=0.08). After unit-wide training there was an average 1.76 hour increase in sleep duration (p< 0.01). Sleep quality also improved at the four-year mark. Median length of stay, use of physical or chemical restraints, and sitter use was unchanged between pre-protocol and post-protocol patients. Benzodiazepine and Zolpidem use was unchanged, but the proportion of patients on melatonin increased from 8% to 42% at one year (p< 0.01).
Conclusions: This delirium prevention protocol was partially adopted into admission workflow with provider education, and was associated with improved quantity and quality of sleep. Unit-wide adoption of sleep improvement practices in combination with delirium prevention protocol further increased reported sleep duration among non-ICU adult inpatients.