Objectives: The aim of the current study was to investigate longitudinal associations between patterns of sleep and activity, and daily mood variation amongst a small group (N = 11) of outpatients with Bipolar Disorder. It was expected that lower activity rhythm stability and reduced sleep quality would be associated with greater levels of mood variation the next day. Raw activity was expected to be positively associated with mood on the same day.

Methods: Participants provided an average of 133.8 (range: 14 - 231) consecutive days of subjective (mood, sleep) and objective (activity, sleep) data using ChronoRecord self-reports and actigraphy, respectively. Multilevel regression was used to investigate prospective associations between variables.

Results: Consistent with expectations, a positive linear fixed effect relationship was found between raw activity and mood on the same day. Contrary to expectations, there was no fixed effect relationship between activity rhythm stability and mood the following day. There was also no significant relationship between actigraph-derived sleep quality and mood, however a negative quadratic association was demonstrated between self-reported sleep length and mood the next day. There were no significant inter-individual differences (random effects) between participants in any of the investigated relationships.

Conclusions: The results suggest that changes in raw activity and self-reported sleep length may be beneficial for the prediction of mood variation amongst outpatients with Bipolar Disorder. The implications for clinical practice are discussed in the context of one participant who was hospitalized for mania during the course of the investigation. Limitations associated with the clinical applications of current monitoring technologies are also discussed.