Background: Although hospitals should be a place of healing and restoration, multiple studies demonstrate that hospitalized patients face acute sleep deprivation due to potentially modifiable disruptions (vitals, medications, tests). Acute inpatient sleep deprivation is also associated with worse health outcomes both in-hospital and post-discharge. While staff-directed sleep interventions show benefits, no study has tested whether sleep education and empowerment of hospitalized patients can reduce modifiable inpatient sleep disruptions and increase inpatient sleep.
Methods: Participants were randomly assigned to either the I-SLEEP (intervention) arm or the standard care arm. I-SLEEP subjects received a short 5-minute video that educated patients on the importance of sleep hygiene while hospitalized and provided tips to empower patients to advocate with hospital staff for reduced sleep disruptions. Hospitalized patients in both the I-SLEEP and the standard arms received a sleep kit, which consisted of earplugs, eye masks, and a color-copy of the free NIH brochure, “Tips to get Healthy Sleep.” The previously validated Potential Sleep Disruptions Questionnaire was used to measure patient-reported sleep disruptions (vitals, tests, medications). Wrist actigraphy was used to measure nightly sleep duration in minutes. Mixed effects models were used to test the effectiveness of I-SLEEP on outcomes, controlling for subject random effects.
Results: A total of 175 patients were randomized (95 I-SLEEP patients for 265 nights; 80 standard for 174 nights). Both groups were similar with respect to baseline characteristics. In unadjusted analyses, there were fewer reported disruptions due to vitals (63% vs 74%), medications (49% vs 60%) and tests (57% vs 69%) on I-SLEEP patient nights compared to standard patient nights (p=0.02 for all). In mixed models controlling for subject random effects, patients in I-SLEEP reported significantly lower odds of being disrupted due to vitals (OR 0.61 [0.40, 0.93], p=0.02), medications (OR 0.62 [0.42,0.94], p=0.02), and tests (OR 0.59 [0.39, 0.88], p=0.01). These findings remained significant when controlling for study day, study day*I-SLEEP, age, gender, race, BMI, and apnea risk. Using actigraphy, while sleep in minutes was higher on I-SLEEP nights than standard nights (334 minutes vs 309 minutes, p=0.16), this difference was not statistically significant. In multivariable mixed effects models controlling for subject random effects, while the main intervention effect and study night were not significant, there was a significant interaction between I-SLEEP and study night such that patients in I-SLEEP received 15 minutes (SE 7.8 min) more sleep for each night in the intervention (p< 0.05).
Conclusions: In this first randomized controlled trial testing an intervention (I-SLEEP) which aimed to educate and empower hospitalized patients to improve sleep, we show that I-SLEEP was associated with reduced patient-reported sleep disruptions from vitals, medications, and tests and each day in the intervention was associated with greater objective inpatient sleep. Given the importance of improving sleep in hospitalized patients, I-SLEEP is a patient-centered intervention that has potential to empower patients to advocate for more patient-centered care and better sleep in the hospital.