LAUSR

Few studies have explored how patients sleep or what characteristics might be predictive of poor sleep during clinically-indicated polysomnography (PSG) in an in-laboratory setting. We reviewed clinically indicated diagnostic PSG studies completed over a 10-year period in a single academic sleep center. Total sleep time (TST) and sleep efficiency (SE) were used as proxies for sleep quality. Patients were categorized as normal or poor sleepers based on TST <4 hours or SE <50%. Multivariate linear and logistic regression analyses were performed to determine factors associated with sleep quality while controlling for demographics, medications, comorbidities and measures of sleep. We included 4957 patients, who were mostly female (58.9%), middle-aged (52.9 y), Caucasian (69.3%), and overweight or obese (91.3%). 3682 patients (74.2%) were diagnosed with sleep apnea (Apnea Hypopnea Index(AHI)>5/hr). Average TST was 5.75±1.43 hours (Interquartile range [IQR] = 4.94 - 6.73) and average SE was 75.1%±16.1% (IQR=66.9 - 87.2). TST and SE were lower for males compared to females (5.48 vs 5.93 hr, p<0.001; 73% vs 77%, p<0.001). In multivariable analysis, older age (TST: OR = 1.04, 95% CI:[1.03,1.05]; SE: OR = 1.04, 95% CI:[1.04,1.05]), male sex (TST: 1.38,[1.14,1.68]; SE: 1.34,[1.07,1.68]), normal body habitus (TST: 1.47,[1.02,2.08]; SE: 1.51,[1.01,2.27]) and a higher AHI (TST: 1.02,[1.02,1.03]; SE: 1.02,[1.003,1.03]) were significantly associated with being a poor sleeper for both TST and SE. Antidepressant use was associated with poor sleep for TST (0.77, [0.59,1]), but not for SE (0.98, [0.73,1.3]). Sleep quality during the in-laboratory PSG differed by sex, age and presence of sleep apnea. Sleep quality during in-lab PSG is thought to be compromised by obtrusive monitoring and an unfamiliar environment, but average sleep quality may be higher than expected for patients in the laboratory. Future studies should consider examining in-lab sleep quality in different patient populations. N/A