Comparison of objective and subjective sleep parameters in patients with bipolar disorder in both euthymic and residual symptomatic periods

Sleep disturbance is a core feature of bipolar disorder; hence, sleep must be accurately assessed in patients with bipolar disorder. Subjective sleep assessment tools such as sleep diary and questionnaires are often used clinically for assessing sleep in these patients. However, the insight into whether these tools are as accurate as objective tools, such as actigraphy, remains controversial. This cross-sectional study included 164 outpatients with a diagnosis of bipolar disorder, including patients who had euthymic and residual symptomatic periods. Objective sleep assessment was conducted prospectively using actigraphy for 7 consecutive days, whereas subjective sleep assessment was conducted prospectively using a sleep diary. The correlations were high and moderate between sleep diary and actigraphy when assessing the total sleep time and sleep onset latency, respectively (r = 0.81 and 0.47). These correlations remained significant after correction for multiple testing (both p < 0.001) and in both euthymic and residual symptomatic states (total sleep time: r = 0.86 and 0.77; sleep onset latency: r = 0.51 and 0.40, respectively). The median (interquartile ranges) of the percentage difference (sleep diary parameters minus actigraphy parameters divided by actigraphy parameter) in the total sleep time was relatively small (6.2% [−0.2% to 13.6%]). Total sleep time assessment using a sleep diary could be clinically useful in the absence of actigraphy or polysomnography.