Self-reported sleep efficiency may not precisely reflect objective sleep patterns. We assessed whether psychosocial factors and affective responses are associated with discrepancies between subjective reports and objective measures of sleep efficiency.
Participants were 199 working women aged 20 to 61 years. Standardized questionnaires were used to assess psychosocial characteristics and affect that included work stress, social support, happiness, and depressive symptoms. Objective measures of sleep were assessed on one week and one leisure night with an Actiheart monitor. Self-reported sleep efficiency was derived from the Jenkins Sleep Problems Scale. Discrepancies between self-reported and objective measures of sleep efficiency were computed by contrasting standardized measures of sleep problems with objectively measured sleep efficiency.
Participants varied markedly in the discrepancies between self-reported and objective sleep measures. After adjustment for personal income, age, having children, marital status, body mass index, and negative affect, overcommitment (p = .002), low level of social support (p = .049), and poor self-rated heath (p = .02) were associated with overreporting of sleep difficulties and underestimation of sleep efficiency. Self-reported poor sleep efficiency was more prevalent among those more overcommitted at work (p = .009) and less happy (p = .02), as well as among those with lower level of social support (p = .03) and more depressive symptoms (p = .048), independently of covariates. Objective sleep efficiency was unrelated to psychosocial characteristics or affect.
The extent to which self-reported evaluations of sleep efficiency reflect objective experience may be influenced by psychosocial characteristics and affect. Unless potential moderators of self-reported sleep efficiency are taken into account, associations between sleep and psychosocial factors relevant to health may be overestimated.
BMI = body mass index;
CVD = cardiovascular disease;
PSG = polysomnography