Genetic load for bipolar disorder is reflected in ecologically assessed mood

Alterations in mood and locomotor activity are thought to be among the core symptoms in bipolar disorder (BIP); if such alterations are also present in healthy but at risk individuals, they may be used as a markers to inform prevention. Using a longitudinal ambulatory assessment approach, the present study investigated how the presentation of these phenotypes in daily life differed depending on familial load and clinical diagnosis. Thirty individuals, comprising BIP patients (n=8) and healthy relatives (n=6) from families with a high density of BIP, and BIP patients (n=8) and healthy controls (n=8) from the general population, participated in the study. At study entry, affective symptoms of the participants was assessed by a clinician; BIP patients were in a euthymic state; and no difference in depressive symptoms were seen among the four groups (as measured by the CES-D). Over a period of 6+ weeks, participants completed 3 daily smartphone-based assessments of current mood while wearing actigraphs to monitor locomotor activity patterns. Mood was assessed using the multidimensional mood questionnaire (MDMQ), which measures valence, energetic arousal and calmness. Mood and locomotor activity levels were compared between the different groups, looking at effects of family and diagnosis. The longitudinal association between mood and activity levels was also examined. Significant effects of both family and diagnosis were found on valence and energetic arousal, while significant effects of diagnosis but not family were observed for calmness. Patterns of associations suggest that differences were pronounced between healthy population controls and all other groups; healthy relatives and familial patients did not significantly differ from each other on mood measures. Energetic arousal and calmness, but not valence, were significantly associated with locomotor activity. Overall, average locomotor activity was not found to significantly differ between groups. We observed both family and diagnosis based differences in terms of mood but not physical activity. Our results suggest that mobile-based assessments can differentiate not only patients, but also healthy individuals with a high familial load from healthy controls. Given that these individuals are at high risk, future research should explore whether such parameters can function as markers of risk for transitions into affective disorders.