389 - Minimizing Inpatient Sleep Disruptions

Poster Number: 389
Publication Number: 389.243

Brianna Glover, Emory University School of Medicine, Atlanta, GA, United States; Leonid Bederman, Duke University School of Medicine, Durham, NC, United States; Anthony Cooley, Emory University School of Medicine, Atlanta, GA, United States; Christy Bryant, Children's Healthcare of Atlanta, Atlanta, GA, United States; Swaminathan Kandaswamy, Emory University School of Medicine, Atlanta, GA, United States; Evan W. Orenstein, Emory University School of Medicine, Atlanta, GA, United States; Nicole Hames, Emory University School of Medicine, Atlanta, GA, United States

Background: When hospitalized, children go to sleep later, wake up later, have more nighttime awakenings and get less sleep than when at home.1,2 

Objective: Our study aimed to reduce caregiver-reported sleep disruptions by 20% over a 6-month period for patients hospitalized on the pediatric hospital medicine (PHM) service.

Design/Methods: The outcome measure was caregiver-reported sleep disruptions. Beginning 10/15/2020, caregivers were asked “Was your child’s sleep interrupted last night?” with further categorization of what specifically interrupted sleep if the answer was affirmative. Parents could select multiple causes. During baseline data collection through 2/28/2021, 416 caregivers were surveyed with 96 (23%) caregivers reporting ≥1 sleep disruption. Vital signs (VS) measurements were the most frequently reported cause (Figure 1). Starting 4/12/2021, we added a new VS frequency of “every 4 hours (unless asleep between 2300 and 0500)” to the electronic health record. Our process measure was tracking how often this new VS frequency was executed. Residents and attendings were educated on patient eligibility with exclusion criteria as follows: age under 2 months, sepsis, respiratory distress, cardiac disease, or any condition deemed by the treating team as unsafe to forego overnight VS. To alert nighttime residents and nurses, patient lists had a column added to identify which patients had this new order. As a balancing measure, we tracked rapid responses called on patients with this VS frequency.

Results: A control chart was used to track the proportion of inpatient nights with ≥1 caregiver-reported sleep disruption, which decreased from 53% pre-intervention to 7% post-intervention (Figure 2). The new VS frequency was ordered for 11% (1061/9270) of patient nights on the PHM service. VS were recorded between 2300-0500 for 90% (955/1061) of patients with the new frequency ordered and 90% (7447/8209) of patients without the new frequency ordered (p=0.49). However, blood pressure was recorded between 2300-0500 for only 39% (421/1061) of patients with the new frequency and 87% (7195/8209) without the new frequency (p < 0.001). There were 2 rapid responses called on patients with the new frequency.Conclusion(s): A reduction in sleep disruptions was temporally associated with implementation of this new VS frequency with a significant reduction in blood pressures taken but not all vital signs. While we achieved our improvement aim, other unmeasured factors likely contributed as the reduction in sleep disruption was greater than the 11% of patient nights where the new frequency was used.
References
Figure 1