Frequency analysis of mood, sleep, and alcohol and nicotine use: investigating instability in bipolar I and II disorder

Background Individuals with bipolar disorder (BD) often experience unstable mood beyond mood episodes, which appears to be more prominent in BD II than in BD I. It is unclear whether the mood instability and associated sleep- and substance use patterns are different between BD subtypes and also present in the first treatment phase of the disorder. In this study, we compared the degree of instability of mood, sleep characteristics, and use of nicotine and alcohol across BD I and II. Methods We recruited 40 patients with BD in their first treatment phase (BD I n = 15, BD II n = 22 and BD Not Otherwise Specified n = 3) at Oslo University Hospital in Norway. The “MinDag” (MyDay) app was used to collect daily self-reports of mood (sadness, irritability, mood elevation and anxiety), sleep (length and quality) and weekly reports of alcohol and nicotine use for up to six months (mean no. of weeks = 18.5, a total of up to 740 weeks). To compare the degree of instability in the time-series across the two BD subtypes we used Root Mean Squared Successive Differences (rMSSD), representing the time domain, and power spectral density (PSD), representing the frequency domain. Cumulative difference periodogram (CDP) was used for group comparison. Results Significant differences were found between BD I and BD II in the mean levels of irritability and mean hours of sleep throughout the follow-up period. The trends for higher instability in anxiety (.029) and irritability (.058) in BD II compared to BD I did not survive Bonferroni correction. Similarly, the variability of sleep length and quality and the use of nicotine and alcohol was comparable across the two groups. Conclusion Applying longitudinal app-based self-monitoring and time-series analyses, our study did not replicate previous findings of more mood instability in BD II versus BD I. The pattern of similar instability across the groups also applied to sleep characteristics and nicotine and alcohol use. Larger samples are needed to investigate how BD subtypes can explain the instability of mood and other illness-related features, which could support the development of more personalized treatment.