Assessing physical activity and sedentary behavior in people with mental Illnesses: Do actigraphy and daily self-report measures agree?

Background and aims: Moderate to vigorous physical activity (MVPA) and sedentary behavior are important factors for somatic and mental health. However, less than half of the people with mental illness (MI) complies with the norms for physical activity. For improving their physical activity with interventions, accurate measurement methods are essential. Actigraphy and diaries are used to measure physical activity in people with a MI, but little is known about how they compare in this group. This study investigates the agreement between actigraphy and daily diaries for assessing MVPA and sedentary behavior in people with a MI. Methods: As part of a pilot-RCT on a lifestyle intervention, twenty transdiagnostic outpatients wore actigraphy and filled in evening diaries for 2 × 14 days (before/after intervention period) (t = 429 paired observations), measuring daily minutes of MVPA and sedentary time. A mixed-model limits of agreement (LoA) method was used to calculate the mean bias between the measurement methods, which was compared to the clinical accepted difference (MVPA: 10 min, sedentary time: 60 min). Bland-Altman plots were examined on patterns. Results: The mean bias between actigraphy and diaries was −29 min (95 % LoA −122 to 64) for MVPA and −165 min (95 % LoA −584 to 253) for sedentary time; diaries underreported more than clinically acceptable compared to actigraphy. Post-hoc analysis indicated that the bias differed between volume levels. Conclusions: Actigraphy and daily diaries appear incomparable in MI. Follow-up research is needed to uncover the nature of these differences and ways to overcome them. Until then, it is recommendable to use both.