A Practical Guide to Analyzing Time-Varying Associations between Physical Activity and Affect Using Multilevel Modeling

Kim, Jinhyuk; Marcusson-Clavertz, David; Togo, Fumiharu; Park, Hye-Jin

doi:10.1155/2018/8652034

Cited by 11 publications

(6 citation statements)

References 112 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We used linear mixed modeling (LMM; e.g., [25]) to answer our research questions. This type of modeling is suitable for analyzing hierarchical EMA data in which multiple observations are nested within subjects, with the number and timing of observations varying between subjects, and in which some observations are usually missing [26]. In both hypotheses, the outcome was the experience of stress symptoms, and the predictor was the experience of stressful events.…”

Section: Statistical Analysesmentioning

confidence: 99%

Does laughing have a stress-buffering effect in daily life? An intensive longitudinal study

et al. 2020

View full text Add to dashboard Cite

Positive affect is associated with alleviating mental and physiological stress responses. As laughter is a common physiological operationalization of positive affect, we investigated whether the effects of experiencing a stressful event on stress symptoms is lessened by frequency and intensity of daily laughter. Using an intensive longitudinal design, we ambulatory assessed the self-reported experience of stressful events, stress symptoms and the frequency as well as the intensity of laughter in university students' daily lives. Our hierarchical ecological momentary assessment data were analyzed with multilevel models. The results support the stress-buffering model of positive affect: We found that the frequency of laughter attenuated the association between stressful events and subsequent stress symptoms. The level of intensity of laughter, however, was found to have no significant effect. Future studies should use additional psychophysiological indicators of stress and straighten out the differential contributions of frequency and intensity of daily laughter.

show abstract

Section: Statistical Analysesmentioning

confidence: 99%

Does laughing have a stress-buffering effect in daily life? An intensive longitudinal study

et al. 2020

View full text Add to dashboard Cite

show abstract

“…One method used in literature to analyze the time-varying relations between longitudinal variables is multilevel modeling of time-varying relations (Kim et al, 2018).…”

Section: Established Modeling Strategies To Inform Decision Rules For a Jitaimentioning

confidence: 99%

“…Therefore, the average amount of changes in an outcome associated with momentary deviation from a personal mean is represented by the effect of a time-varying covariate. While multilevel modeling provides a basis and is extendable to analyze time-varying effects (see Kim et al, 2018 for time-varying association between physical activity and affect), it is limited to simple change patterns that can be modeled with a relatively few parameters; e.g., linear or quadratic trends, or it becomes complicated in terms of implementation and interpretation.…”

Section: Established Modeling Strategies To Inform Decision Rules For a Jitaimentioning

confidence: 99%

Dynamical Outcome Modeling of Fragmented Intensive Longitudinal Data

Lagoa¹,

Hojjatinia²,

Conroy³

2021

Preprint

View full text Add to dashboard Cite

The advent of new sensing technologies has enabled the collection of high frequency individual data. Not only can this allow for the identification of personalized behavior models, it also opens up the possibility of developing just-in-time interventions that leverage the information collected to determine when and which micro intervention should be provided. However, there are significant challenges in the analysis of this type of data. First, due to the high rate of data collection, one can no longer assume that the stimulus (independent excitation or micro intervention) only has an instantaneous effect on the outcome, one has to also allow for delayed effects. Moreover, one is also frequently faced with fragmented data; i.e., poor placement of sensors, non-wear of the data collecting device and/or external disturbances can lead to intervals of time where the data collected is not reliable; i.e., missing or corrupted. To deal with these challenges, we leverage concepts from the areas of dynamical systems and signal processing to develop tools that i) can identify models that take into account the delayed stimuli effects and ii) are able to handle fragmented data. In this paper, we provide both the mathematical foundation of the tools proposed and the description of a package that implements them. We also discuss ways to interpret the results obtained.

show abstract

“…Furthermore, it is often assumed that both fatigue and PA are constant or trait constructs that do not fluctuate over time with fatigue expressed in a single score based on retrospective questions or PA represented as an average level derived from questionnaires or accelerometer data [ 12 – 18 ]. However, levels of fatigue and PA may not be stable between and within days and they may interrelate dynamically in daily life [ 19 – 22 ]. By simultaneously assessing fatigue and PA multiple times per day, using for example Ecological Momentary Assessment (EMA) and accelerometry, direct relationships between fatigue and PA in daily life might be revealed that would have been hidden when measured otherwise [ 20 , 22 ].…”

Section: Introductionmentioning

confidence: 99%

“…However, levels of fatigue and PA may not be stable between and within days and they may interrelate dynamically in daily life [ 19 – 22 ]. By simultaneously assessing fatigue and PA multiple times per day, using for example Ecological Momentary Assessment (EMA) and accelerometry, direct relationships between fatigue and PA in daily life might be revealed that would have been hidden when measured otherwise [ 20 , 22 ].…”

Section: Introductionmentioning

confidence: 99%

Unraveling the interplay between daily life fatigue and physical activity after subarachnoid hemorrhage: an ecological momentary assessment and accelerometry study

de Vries,

Heijenbrok-Kal,

van Kooten

et al. 2023

J NeuroEngineering Rehabil

View full text Add to dashboard Cite

Background Fatigue is one of the most commonly reported symptoms after subarachnoid hemorrhage (SAH) and is indirectly associated with physical activity (PA). Associations between fatigue and PA are primarily examined based on conventional measures (i.e. a single fatigue score or average PA levels), thereby assuming that fatigue and PA do not fluctuate over time. However, levels of fatigue and PA may not be stable and may interrelate dynamically in daily life. Insight in direct relationships between fatigue and PA in daily life, could add to the development of personalized rehabilitation strategies. Therefore we aimed to examine bidirectional relationships between momentary fatigue and PA in people with SAH. Methods People (n = 38) with SAH who suffer from chronic fatigue were included in an observational study using Ecological Momentary Assessment (EMA) and accelerometry. Momentary fatigue was assessed on a scale from 1 to 7 (no to extreme fatigue), assessed with 10–11 prompts per day for 7 consecutive days using EMA with a mobile phone. PA was continuously measured during this 7-day period with a thigh-worn Activ8 accelerometer and expressed as total minutes of standing, walking, running and cycling in a period of 45 min before and after a momentary fatigue prompt. Multilevel mixed model analyses including random effects were conducted. Results Mean age was 53.2 years (SD = 13.4), 58% female, and mean time post SAH onset was 9.5 months (SD = 2.1). Multilevel analyses with only time effects to predict fatigue and PA revealed that fatigue significantly (p < 0.001) increased over the day and PA significantly (p < 0.001) decreased. In addition, more PA was significantly associated with higher subsequent fatigue (β = 0.004, p < 0.05) and higher fatigue was significantly associated with less subsequent PA (β=-0.736, p < 0.05). Moreover, these associations significantly differed between participants (p < 0.001). Conclusions By combining EMA measures of fatigue with accelerometer-based PA we found that fatigue and PA are bidirectionally associated. In addition, these associations differ among participants. Given these different bidirectional associations, rehabilitation aimed at reducing fatigue should comprise personalized strategies to improve both fatigue and PA simultaneously, for example by combining exercise therapy with cognitive behavioral and/or energy management therapy.

show abstract

A Practical Guide to Analyzing Time-Varying Associations between Physical Activity and Affect Using Multilevel Modeling

Cited by 11 publications

References 112 publications

Does laughing have a stress-buffering effect in daily life? An intensive longitudinal study

Does laughing have a stress-buffering effect in daily life? An intensive longitudinal study

Dynamical Outcome Modeling of Fragmented Intensive Longitudinal Data

Unraveling the interplay between daily life fatigue and physical activity after subarachnoid hemorrhage: an ecological momentary assessment and accelerometry study

Contact Info

Product

Resources

About