Study data published in the Journal of Psychiatric Research indicate that keeping a sleep diary may be a useful means of tracking sleep disturbances in patients with bipolar disorder. In a cohort study of patients with bipolar disorder, subjective sleep measures were strongly correlated with objective sleep measures as captured by actigraphy.

Sleep disturbances are a key feature of bipolar disorder and can persist in both the euthymic and symptomatic periods. While objective measures of sleep provide more comprehensive data, their cost, availability, and invasiveness can limit patient acceptability.

To assess the correlation between subjective and objective measures of sleep quality, investigators conducted a cross-sectional cohort study of adult outpatients with bipolar disorder in either the euthymic or residual symptomatic phases. Patients were enrolled from clinics in Japan between 2017 and 2019. Patients with mania or current severe depressive symptoms were excluded.

Continue Reading

Subjective sleep was evaluated retrospectively using a patient-kept sleep diary and the self-administered Insomnia Sleep Index (ISI). Objective sleep was measured using a worn accelerometer for 7 consecutive days. The following metrics were compared between subjective and objective sleep measures: (1) total sleep time; and (2) sleep onset latency, or the time between bedtime and falling asleep. The Pearson correlation coefficient was used to assess the between-measure correlation. Bland-Altman plots were used to assess the mean differences between objective and subjective sleep scores.

The study cohort included 164 outpatients with bipolar disorder, among whom 98 (59%) were in the residual symptomatic phase and 66 (40%) were in the euthymic phase. Median IQR age was 43.0 (33.0-51.0) years and 53.7% were women.

Total sleep time as assessed by sleep diary was strongly correlated with total sleep time as assessed by actigraphy (r =.81; P <.001). Sleep onset latency as measured by diary was moderately correlated with latency measured by actigraphy (r =.47; P <.001). These correlations persisted across the euthymic and residual symptomatic groups.

The median (IQR) percentage difference in total sleep time between sleep diary and actigraphy parameters was small, at 6.2% (-0.2% to 13.6%). Median (IQR) percentage difference in sleep onset latency was larger: 44.1% (-21.1% to 143.6%), suggesting that time from bedtime to falling asleep is less accurately measured by diary entries compared with actigraphy.  

Results from this study support consistency between self-reported and objective measures of sleep in bipolar disorder. In particular, total sleep time as captured by a sleep diary was strongly correlated with time measured by actigraphy.

As study limitations, investigators noted that actigraphy alone may not be a sufficient objective measure of sleep quality in bipolar disorder. Further research using polysomnography as an objective metric may be necessary to confirm these findings. Additionally, patients in the symptomatic phase of bipolar disorder were excluded, meaning data are not generalizable to all patients with bipolar disorder.   

“In conclusion, subjective [total sleep time] assessed by sleep diary showed a high correlation with and a slight difference from objective TST [total sleep time] assessed by actigraphy,” investigators wrote.

“Thus…assessment using a sleep diary may be useful in clinical practice when actigraphy or polysomnography is not easily available.”

Disclosure: Several study authors declared affiliations with biotech, pharmaceutical, and/or device companies. Please see the original reference for a full list of authors’ disclosures. 


Fujita A, Esaki Y, Obayashi K, et al. Comparison of objective and subjective sleep parameters in patients with bipolar disorder in both euthymic and residual symptomatic periods. J Psychiatr Res. 2022;145:190-196. doi:10.1016/j.jpsychires.2021.12.015