Validity of Subjective Reports of Sleep Latency in Normal Subjects

Abstract The validity of subjective reports of sleep latency was tested in 28 normal subjects aged 20 to 55 yr who answered a general questionnaire about their sleep habits, described their sleep latency at home in a series of daily reports, and also slept in the laboratory for 3 to 12 nights each enabling objective measurements of their sleep latency to be made from the electro encephalogram. The geometric mean of the log-normally distributed sleep latencies for all 118 nights in the laboratory was 12·7 min, measured objectively, and was 13·3 min reported subjectively. After adapting to the laboratory, the mean subjective latency for each subject was highly correlated with (r = 0·6 to 0·8) and was not significantly different from either the mean objective latency in the laboratory or the subjective latency at home. The results indicate that subjective reports of habitual sleep latency are valid as well as reasonably reliable and accurate for groups of normal subjects, but are not very accurate for individual subjects.