JMIR Publications

ABSTRACT

Background:

Psychological stress poses a risk to mental and physical health, and has become a major public health challenge. As physical behaviors (i.e., physical activity (PA), sedentary behavior (SB), and sleep) play a key role in mental well-being, their targeted modification could be an approach to cope with stress in everyday life. Previous studies have primarily either analyzed the associations between isolated physical behaviors and stress-related outcomes or employed cross-sectional designs. Accordingly, there is a need for deeper insights into the within- and between-person associations between physical behavior over a 24-hour cycle and psychological stress in naturalistic settings.

Objective:

This study aimed to investigate how 24-hour physical behavior compositions are associated with daily self-perceived stress and stress-related coping self-efficacy, and how replacing time in one behavior with another is linked to changes in both stress-related indicators.

Methods:

A total of 198 healthy university employees (mean age 35.87 years, SD 10.76; 54.8% (109/198) female) participated in a 15-day ambulatory assessment study. Participants reported their momentary stress and coping self-efficacy perceptions up to 6 times a day in electronic diaries. 24-hour physical behavior was measured using a thigh-worn Move 4 accelerometer. The movement data were then classified on a daily basis into the four behavior categories of sleep, SB, light physical activity (LPA), and moderate-to-vigorous physical activity (MVPA). In order to obtain physical behavior compositions an isometric log-ratio (ilr) transformation was applied, resulting in four different ilr-sets. The associations between physical behavior compositions, self-perceived stress, and coping self-efficacy were analyzed using two-level mixed multilevel models. Exploratory reallocation models were conducted to simulate the effects of time shifts from one behavior to another on both stress-related outcomes.

Results:

The geometric average day comprised 33.9% (8.1/24h) sleep, 45.2% (10.8/24h) SB, 15.8% (3.8/24h) LPA, and 5.1% (1.2/24h) in MVPA. More time spent sleeping compared to being sedentary was associated with lower self-perceived stress (standardized β=–.03; t1988=–2.045; P=.04) but not with coping self-efficacy (t1988=–1.333; P=.18) in the 24-hour cycle. The ratio of SB to the other physical behaviors and time spent in LPA or MVPA relative to SB showed no association with either stress-related outcome. Significant random effects indicate high individual variability between the analyzed associations. The exploratory substitution of SB by LPA, MVPA, or sleep showed no significant changes of self-perceived stress or coping self-efficacy within a 60-minute period.

Conclusions:

Adapting 24-hour physical behavior seems to be a promising approach to reduce stress in everyday life, for example by extending the sleep duration instead of being awake in SB. Further research should be conducted on contextual and personal influencing factors in order to develop tailored stress management interventions for the 24-hour cycle.