Sleep in Older Hospitalized Patients: Effects of Disruptions and Perceived Control

Background:

Better sleep is associated with improved cognition, mood, and physical health in older adults. Despite the potential health benefits, obtaining quality sleep when hospitalized has been described as difficult. Additionally, perceived control may be a key determinant of sleep, as patients' control is dramatically reduced during hospitalization because of dependency on nursing staff and medical therapies. We aimed to characterize sleep among hospitalized seniors and to understand how potential sleep disruptions and perceived control are related to inpatient sleep loss.

Methods:

From June to August 2009, we conducted a prospective cohort study of community‐dwelling ambulatory seniors admitted to a general medicine service at a single teaching hospital. Patients with a known sleep disorder, cognitive impairment, ICU stay, or in isolation were excluded. Baseline sleep habits and perceived control over sleep were obtained via the Pittsburgh Sleep Quality Index and the Sleep Locus of Control, respectively. Wristwatch actigraphy was used to determine sleep duration and efficiency. Each morning, patients reported in‐hospilal sleep quality (Karolinska diary) and rated potential sleep disruptions (lab draws, alarms, etc.) on a 10‐point scale. To account for correlation among disruption items, factor analysis was used. Multivariate linear regression models, clustered by subject and adjusted for study day, demographics, and baseline sleep, were used to test the association between sleep disruptions and perceived control with sleep outcomes (duration and efficiency).

Results:

Twenty patients completed at least 1 night of actigraphy and surveys. Although baseline sleep duration prior to admission was relatively normal (427 ± 113 minutes), in‐hospital sleep duration was markedly reduced (277 ± 107 minutes). Mean sleep efficiency (66%) was similar to that of insomniacs (95% Cl 59.0–73.0), with 67.5% of nights below the normal range (79%–84%). Factor analysis identified 3 independent factors: “medical care,” “patient symptoms,” and “environmental.” All patients reported sleep disruptions, but “medical care” items were rated the worst (blood draws rated as severely disruptive 38% of the time, vital signs 33%, medications 32%). Medical care disruption was negatively associated with subjective sleep quality (P = 0.02), but not with objective sleep duration or sleep efficiency. Interestingly, those patients who had higher levels of baseline perceived control over sleep were more likely to have decreased sleep time (P = 0.03) and worse sleep efficiency (P = 002).

Conclusions:

Hospitalization is a time of severe sleep loss for seniors. Hospitalized patients, particularly those with more perceived control over their sleep, may endure numerous sleep disturbances, largely because of nocturnal disruptions related to ongoing medical care. Given the severity of sleep loss and its known detrimental effects on health, future work to understand and improve sleep loss for hospitalized seniors is warranted.

Author Disclosure:

K. Chang, none; E. Nunez, none; K. Knutson, none; D. Meltzer, none; E. Van Cauter, none; V. Arora, none.