Date of Award

January 2025

Document Type

Open Access Thesis

Degree Name

Medical Doctor (MD)

Department

Medicine

First Advisor

Melissa Knauert

Abstract

The medical intensive care unit (ICU) can be a challenging place for patients to get adequate sleep. Multiple environmental and pathophysiological factors can cause sleep deprivation in the ICU, which have been associated with adverse short and long-term outcomes. Although various factors have been hypothesized to impact quality of sleep in the ICU, research on the nature of patients’ pre-ICU sleep is very limited. Understanding the role that pre-ICU sleep plays in sleep quality in the ICU has the potential to inform public health measures to improve sleep and risk stratify critically ill patients. The objective of this study is to characterize the sleep quality of ICU patients prior to admission using sleep questionnaire data obtained from patients and their surrogates at the time of admission. We hypothesized that ICU patients have lower sleep quality prior to ICU admission when compared to sleep quality in non-critically ill patients. The data for this study was collected as part of baseline data for a randomized controlled trial on the use of daytime bright light to promote circadian alignment and shorten or prevent delirium among patients in the ICU. Enrolled patients were interviewed regarding their sleep routines prior to admission via the Pittsburgh Sleep Quality Index (PSQI) and the Epworth Sleepiness Scale (ESS), and this data, along with baseline demographics and past medical history, were used for this study. Continuous and categorical variables were analyzed with descriptive statistical techniques using mean, standard deviation, and proportions as appropriate. Exploratory bivariate analysis was conducted to test for associations between demographic and clinical features and sleep characteristics using bivariate Pearson correlation, independent samples t-test, and analysis of variance (ANOVA). 68 participants were included in this study. The mean (+/- standard deviation) age was 70 +/- 10 years. 43% of the participants were female. 75% of the population was White and 21% Black. The most common reason for ICU admission was acute respiratory failure (43%). 29% of participants reported having a diagnosis of sleep apnea prior to admission. The average modified Rankin score (mRS) was 2 +/- 1.6, and the mean Acute Physiology and Chronic Health Evaluation (APACHE) score was 24 +/- 8.2. The mean global PSQI score was 6.6 +/- 4.2, and the mean ESS score was 10.0 +/- 5.9. Participants reported taking a mean of 34.4 +/- 36.8 minutes to fall asleep and getting a mean of 6.9 +/- 1.8 hours of actual sleep with a mean of 9.2 +/- 3.8 hours per night spent in bed. Reported mean sleep onset was 9:56 PM +/- 121.2 minutes and mean sleep offset (wake) time was 6:31 AM +/- 124.2 minutes. 44% of participants reported that they cannot get to sleep within 30 minutes at least one time per week. 19% required prescribed or over the counter medicine in the past month to help sleep, and 55% of participants characterized their sleep as very or fairly good. Exploratory bivariate analysis was conducted to assess possible associations between participants’ demographics and baseline medical history data and reported sleep habit data, global PSQI, and ESS scores. The following significant associations were found: non-White participants had a higher mean time to fall asleep than White participants (p=0.01); participants with diabetes had a higher mean ESS than those without diabetes (p=0.01); participants with hyperlipidemia had a lower global PSQI than those without hyperlipidemia (p=0.005); participants who reported sleep aid use had higher mean number of hours in bed than those who did not report sleep aid use (p=0.004). When compared to population norms, our findings of elevated global PSQI and ESS scores suggest that patients admitted to the ICU have worse habitual sleep and/or sleep-related symptoms. Often with the presence of many comorbidities, it is not unexpected that ICU patients might have worse baseline sleep than the general population, but the data in this domain is very limited. Further, the use of questionnaires like the PSQI and ESS to study sleep is an expanding area of research. Given the impact of poor sleep on health and the potential to prolong recovery time after the ICU, this and further research could inform interventions to improve sleep quality inside and out of the hospital. Prospective, large-scale research is needed to further investigate how baseline sleep characteristics impact critical illness onset and outcomes.

Comments

This is an Open Access Thesis.

Download
Open Access

This Article is Open Access

Share

COinS