How many hours do you sleep? A comparison of subjective and objective sleep duration measures in a sample of insomnia patients and good sleepers

Our objective was to assess the agreement and linear relationships amongst multiple measures of sleep duration in a sample of patients with insomnia disorder and good sleeper controls. We retrospectively analysed data from 123 patients with insomnia disorder and 123 age- and gender-matched good sleeper controls who completed a simple subjective habitual sleep duration question (Pittsburgh Sleep Quality Index), a sleep diary (5-14 days), 2 nights of polysomnography, and two corresponding morning subjective estimates of sleep duration. Descriptive statistics, linear regression analyses and Bland-Altman plots were used to describe the relationship and (dis)agreement between sleep duration measures. Relationships between polysomnography and the simple question as well as between polysomnography and sleep diary were weak to non-existent. Subjective measures and polysomnography did not agree. Sleep duration measured with the Pittsburgh Sleep Quality Index or sleep diary was about 2 hr above or up to 4 hr below polysomnography-measured sleep duration. Patients with insomnia disorder, on average, reported shorter sleep duration compared with polysomnography, while good sleeper controls, on average, reported longer sleep duration compared with polysomnography. The results suggest that subjective and objective measures apparently capture different aspects of sleep, even when nominally addressing the same value (sleep duration). They disagree in both patients with insomnia disorder and good sleeper controls, but in different directions. Studies assessing sleep duration should take into account both the investigated population and the assessment method when interpreting results. Future studies should continue to investigate possible psychological and physiological correlates of sleep (mis)perception.