Date of Award

January 2018

Document Type


Degree Name

Medical Doctor (MD)



First Advisor

Margaret A. Pisani

Second Advisor

Melissa P. Knauert


Sleep disturbance in the intensive care unit (ICU) is both common and severe. The extent to which sleep disturbance and other patient-centered outcomes persist after ICU discharge is not well studied. The aim of this thesis was to understand the impact of sleep disturbance and other functional outcomes both during and after hospitalization for critical illness.

First, a systematic review was performed to characterize studies of sleep disturbance in post-ICU patients. Three large databases (PubMed, MEDLINE, and EMBASE) were searched in accordance with the PRISMA guidelines. Second, using validated questionnaires, we conducted an observational cohort study of medical ICU (MICU) patients to assess the prevalence of and risk factors for sleep disturbance and functional disability after ICU discharge. Lastly, an observational study was conducted within our institution’s MICU, in which we surveyed the prevalence of simple environmental factors affecting room lighting and disturbance in patient rooms. The prevalence of sleep-disturbing environmental factors was compared before and after a sleep-promotion quality improvement intervention.

Our systematic review yielded 22 studies assessing sleep in post-ICU patients. Questionnaire studies found subjective sleep disturbance in 22-57% of patients from 3 to 6 months after discharge, and 5 of 5 polysomnography studies showed improved sleep parameters over time after discharge. In our own cohort of 112 post-ICU patients at median 4-month follow-up, 63% of patients reported sleep disturbance and 37% had functional decline after critical illness. Patient-reported sleep disturbance was significantly related to the occurrence of ICU delirium. Lastly, our observational study of the MICU environment found a baseline preponderance of sleep-disturbing factors at night. Notably, 59% of rooms had artificial lights turned on (21% using the brightest room lights), 46% had the television on, and 94% had the room door open. A sleep-promotion quality improvement intervention improved several nighttime factors, such as the number of rooms with lights turned off, door curtains closed, and window curtains closed.

Sleep disturbance and functional disability in post-ICU patients are highly prevalent and may be plausibly linked to factors such as ICU delirium. A sleep-promotion quality improvement intervention can modify environmental factors in the ICU related to light and other disturbance. Future prospective studies are needed to examine whether these environmental modifications can improve sleep quality and reduce delirium in the ICU, and improve patient-centered outcomes among ICU survivors after discharge.


This thesis is restricted to Yale network users only. It will be made publicly available on 06/25/2100