A nonpharmacologic sleep protocol for hospitalized older patients

OBJECTIVES: To evaluate the feasibility of and adherence to a nonpharmacologic sleep protocol targeted to nurses for acutely ill older patients and to test the effectiveness of the protocol on enhancing sleep and reducing sedative-hypnotic drug (SHD) use. DESIGN: Prospective cohort study. SETTING: A 34-bed general medical unit in a university-affiliated teaching hospital. PARTICIPANTS: A total of 175 consecutive admissions aged 70 years or older. INTERVENTION: A nonpharmacologic sleep protocol con consisting of a back rub, warm drink, and relaxation tapes was administered by nursing personnel to patients who complained of difficulty initiating sleep or who requested a SHD. After 1 hour, if the patient still requested it, the nurse administered the SHD. MEASUREMENTS: The main outcomes of sleep quality and SHD use were measured by patient interview and chart abstraction. Feasibility and adherence to the protocol were tracked daily by patient and nurse interviews and chart abstraction. RESULTS: A cohort of 111 patients, mean age 79.3 (+/- 6.4), 68% women, received the sleep protocol. Patients required the protocol for a mean of 4.9 days per patient, totalling 539 patients-days. The overall adherence rate was 400/539 (74%) patient-days. The rate of complete nonadherence was 139/539 (26%), with reasons for nonadherence including nurse nonadherence in 30 (6%), patient refusal in 104 (19%), and medical contraindications in five (1%). The quality of sleep correlated strongly with the number of parts of the protocol received, suggesting a dose-response relationship, with the highest correlation for receiving two to three parts (rho = .64, P < 0.001). The sleep protocol was successful in reducing SHD use from the baseline preintervention rate of 51/94 (54%) to 34/111 (31%) (P < .002). The sleep protocol had a stronger association with quality of sleep (p = .75, P = .001) than did SHDs (rho = .07, P = .45). However, chronic SHD users were more likely to refuse the protocol than nonusers (64% vs 41%, P < .03) and received SHDs 4.5 times more often than nonusers (67% vs 15%, P = .001). CONCLUSION: The nonpharmacologic sleep protocol provides a feasible, effective, and nontoxic alternative to SHDs to promote sleep in older hospitalized patients. Use of the protocol can substantially decrease use of SHDs.