Background: Many theoretical frameworks have been used in order to understand health behaviors such as physical activity, sufficient sleep, healthy eating habits, etc. In most research studies, determinants within these frameworks are assessed only once and thus are considered as stable over time, which leads to rather 'static' health behavior change interventions. However, in real-life, individual-level determinants probably vary over time (within days and from day to day), but currently, not much is known about these time-dependent fluctuations in determinants. In order to personalize health behavior change interventions in a more dynamic manner, such information is urgently needed.
Objective: The purpose of this study was to explore the time-dependent variability of emotions, physical complaints, intention, and self-efficacy in older adults (65+) using Ecological Momentary Assessment (EMA).
Methods: Observational data were collected in 64 healthy older adults (56.3% men; mean age 72.1 ± 5.6 years) using EMA. Participants answered questions regarding emotions (i.e., cheerfulness, relaxation, enthusiasm, satisfaction, insecurity, anxiousness, irritation, feeling down), physical complaints (i.e., fatigue, pain, dizziness, stiffness, shortness of breath), intention, and self-efficacy six times a day for seven consecutive days using a smartphone-based questionnaire. Generalized linear mixed models were used to assess the fluctuations of individual determinants within subjects and over days.
Results: A low variability is present for the negative emotions (i.e., insecurity, anxiousness, irritation, feeling down) and physical complaints of dizziness and shortness of breath. The majority of the variance for relaxation, satisfaction, insecurity, anxiousness, irritation, feeling down, fatigue, dizziness, intention, and self-efficacy is explained by the within subjects and within days variance (42.9% to 65.8%). Hence, these determinants mainly differed within the same subject and within the same day. The between subjects variance explained the majority of the variance for cheerfulness, enthusiasm, pain, stiffness, and shortness of breath (50.2% to 67.3%). Hence, these determinants mainly differed between different subjects.
Conclusions: This study reveals that multiple individual-level determinants are time-dependent, and are better considered as 'dynamic' or unstable behavior determinants. This study provides us with important insights concerning the development of dynamic health behavior change interventions, anticipating real-time dynamics of determinants instead of considering determinants as stable within individuals.
Keywords: COM-B framework; Determinants; EMA; Elderly; Time-dependent variations.
©2022 Maes et al.