The development of delirium is a common source of morbidity and mortality in the hospitalized elderly, with an estimated incidence of 30-60%. Sleep deprivation is a common and modifiable risk factor for iatrogenic delirium in hospitalized patients. It is unclear from the literature how many hours of sleep hospitalized patients actually get on average during their stay. 


The purpose of this project is to determine average hours of sleep and barriers to sleeping for the inpatients at our institution, an academic medical center, and to use this information in the design and implementation of a delirium prevention protocol.


Anonymous surveys assessing sleep quantity, quality and barriers to proper sleep were collected and analyzed for 100 random hospitalized patients in a medical-surgical unit. The patients self reported an average sleep time of 4 hours (95% confidence interval, 3.7-4.4) and an average quality of sleep rated as 3.3 (95% CI, 3.0-3.5) on a scale of 0 to 5. Presence or absence of a roommate had no effect on average sleep duration (p=0.40) or quality of sleep (p=0.20) when the mean values between these groups were compared with a t-test. 41% of patients identified noise and/or beeping machines as barriers to sleep, 38% identified night vital signs checks as barriers and 29% identified nighttime or early morning blood tests as barriers. 


The results of this survey confirm the anecdotal impression that the quantity and quality of patients’ sleep is suboptimal. The multiple contributing factors identified above spurred design and implement of a delirium prevention protocol at our institution. The protocol is ordered by physicians for any patient over age 59. The protocol initiates nurse-driven interventions to prevent unnecessary interruptions to sleep from 10 pm to 6 am. These interventions include reducing non-urgent vital sign checks and moving the time morning lab are obtained to after 6 am. The protocol also uses clinical decision support to remind providers to discontinue unnecessary cardiac monitoring. A “quiet time” policy has also been implemented from the hours of 2:30-4:30 pm and 11 pm-2 am to help minimize environmental noise.  Post-protocol implementation sleep surveys are currently being collected, as well as aggregate quarterly outcomes data on total length of stay, use of pharmacologic and physical restraints, and transfers to psychiatric facilities. Our institution is also developing a daily nursing assessment protocol to perform CAM scoring on all patients daily to determine delirium risk and incidence.