Research projects

Automated longitudinal monitoring to predict and counter relapse in bipolar disorder: A pilot investigation of effectiveness

Principal Researchers

Associate Professor Greg Murraya, Professor Fiona Juddb


a Faculty of Life and Social Sciences, Swinburne University of Technology

b Centre for Rural Mental Health, Monash University



Award Type

beyondblue Victorian Centre of Excellence

Project completion year


Project brief

In Australia bipolar disorder affects approximately 1.3 per cent of people at some point in their lifetime.  Early identification of changes in behaviour is important in preventing relapse of the condition.

This project aimed to investigate the feasibility of long-term monitoring of activity patterns and daily mood in people diagnosed with bipolar disorder, and second, to test whether patterns of locomotor activity correlated with mood fluctuations.

Fifteen adults diagnosed with bipolar disorder took part in the study and recorded daily reports of their mood on an internet-based software package called ChronoRecord.  These self-reports were supplemented with wrist-worn actigraphs which detected locomotor activity over a 24-hour period.  Participants were enrolled in the study for between one and 12 months.

Key findings

Long-term monitoring of mood using ChronoRecord software was shown to be a viable method for collecting daily records of mood and sleep patterns, as was long-term monitoring of locomotor activity and sleep/wake cycles on the actigraph. 

Actigraphic monitoring of locomotor activity correlated well with changes in subjective recordings of mood.  For example, if participants slept for shorter or longer periods than normal, deviation from normal moods also occurred.  Multi-level modelling indicated that the relationship between sleep patterns and mood also varied between participants.  This suggested that individual differences produce different correlations between sleep and mood.

During this pilot study, one participant also experienced a manic relapse during monitoring.  Actigraphic recordings from this participant indicated decreases in circadian rhythm amplitude in the weeks before the episode.

Implications for policy, practice and further research

The results indicate further research into the use of these monitoring methods to prevent bipolar episodes is warranted.  There may be a clinical benefit in using these monitoring methods for some people with bipolar disorder.  Specifically, monitoring may be particularly useful for people with bipolar disorder who cannot be closely watched by others.

More definite evidence of the link between circadian function and mood change would require a more sophisticated method of measuring circadian function.  A collaboration with the Harvard University Division of Sleep Medicine has been formed for this purpose.

Stay in touch with us

Sign up below for regular emails filled with information, advice and support for you or your loved ones.

All done! You should’ve received a confirmation email, so please check when you’re finished here and click the link in the email. If you can’t see it, we might be in your junk mail.

Subscribe failed. Please try later or contact us.