Title :
Covariation of Depressive Mood and Spontaneous Physical Activity in Major Depressive Disorder: Toward Continuous Monitoring of Depressive Mood
Author :
Kim, Jinhyuk ; Nakamura, Toru ; Kikuchi, Hiroe ; Yoshiuchi, Kazuhiro ; Sasaki, Tsukasa ; Yamamoto, Yoshiharu
Author_Institution :
Dept. of Psychosomatic Res., Nat. Inst. of Mental Health, Tokyo, Japan
Abstract :
The objective evaluation of depressive mood is considered to be useful for the diagnosis and treatment of depressive disorders. Thus, we investigated psychobehavioral correlates, particularly the statistical associations between momentary depressive mood and behavioral dynamics measured objectively, in patients with major depressive disorder (MDD) and healthy subjects. Patients with MDD (n = 14) and healthy subjects (n = 43) wore a watch-type computer device and rated their momentary symptoms using ecological momentary assessment. Spontaneous physical activity in daily life, referred to as locomotor activity, was also continuously measured by an activity monitor built into the device. A multilevel modeling approach was used to model the associations between changes in depressive mood scores and the local statistics of locomotor activity simultaneously measured. We further examined the cross validity of such associations across groups. The statistical model established indicated that worsening of the depressive mood was associated with the increased intermittency of locomotor activity, as characterized by a lower mean and higher skewness. The model was cross validated across groups, suggesting that the same psychobehavioral correlates are shared by both healthy subjects and patients, although the latter had significantly higher mean levels of depressive mood scores. Our findings suggest the presence of robust as well as common associations between momentary depressive mood and behavioral dynamics in healthy individuals and patients with depression, which may lead to the continuous monitoring of the pathogenic processes (from healthy states) and pathological states of MDD.
Keywords :
biomechanics; medical disorders; medical signal processing; patient diagnosis; patient monitoring; patient treatment; psychology; statistical analysis; MDD; activity monitor; behavioral dynamics; continuous monitoring; depressive disorder diagnosis; depressive disorder treatment; depressive mood; ecological momentary assessment; local statistics; locomotor activity; major depressive disorder; momentary depressive mood; multilevel modeling approach; pathogenic processes; pathological states; psychobehavioral correlates; spontaneous physical activity; watch-type computer device; Biomedical measurement; Data models; Informatics; Mathematical model; Monitoring; Mood; Predictive models; Depressive mood; ecological momentary assessment (EMA); locomotor activity; major depressive disorder (MDD); multilevel modeling;
Journal_Title :
Biomedical and Health Informatics, IEEE Journal of
DOI :
10.1109/JBHI.2015.2440764