A comprehensive computerised six - task cognitive test battery, the profile of
mood states questionnaire and the symptom questionnaire (headache, concentration and task difficulty) were administered during each trial.
They also had to fill out
a mood state questionnaire at 0, 6, 12, and 24h of sleep deprivation with cycling exercise in between.
The initial
mood of each patient was assessed by a clinician, with changes in
mood monitored by self - reported
questionnaires; the
mood states were then used to determine the accuracy of the classifiers, which use HRV to predict
mood, with an accuracy of around 90 %.28 No data were collected on healthy controls (or another clinical group), making it difficult to establish the specificity of the findings to BD.