Dyadic interventions to promote physical activity and reduce sedentary behaviour: systematic review and meta-analysis. Issue 1 (2nd January 2019)

• Record Type: Journal Article
• Authors: Carr, R. M.; Prestwich, A.; Kwasnicka, D.; Thøgersen-Ntoumani, C.; Gucciardi, D. F.; Quested, E.; Hall, L. H.; Ntoumanis, N.
• Is Part Of: Health psychology review. Volume 13:Issue 1(2019)
• Page Start: 91
• Page End: 109
• Publication Date: 2019-01-02
• Subjects: Randomised control trials -- dyads -- transactive goal dynamics theory -- interventions -- systematic review -- meta-analyses; Clinical health psychology -- Periodicals; 613.019
• DOI: 10.1080/17437199.2018.1532312
• Languages: English
• ISSNs: 1743-7199

Abstract

ABSTRACT: Several interventions have targeted dyads to promote physical activity (PA) or reduce sedentary behaviour (SB), but the evidence has not been synthesised. Sixty-nine studies were identified from MEDLINE, PsycINFO, and Web of Science, and 59 were included in the main meta-analyses (providing 72 independent tests). Intervention details, type of dyadic goal, participant characteristics, and methodological quality were extracted and their impact on the overall effect size was examined. Sensitivity analyses tested effect robustness to (a) the effects of other statistically significant moderators; (b) outliers; (c) data included for participants who were not the main target of the intervention. Dyadic interventions had a small positive, highly heterogeneous, effect on PA g = .203, 95% CI [0.123–0.282], compared to comparison conditions including equivalent interventions targeting individuals. Shared target-oriented goals (where both dyad members hold the same PA goal for the main target of the intervention) and peer/friend dyads were associated with larger effect sizes across most analyses. Dyadic interventions produced a small homogeneous reduction in SB. Given dyadic interventions promote PA over-and-above equivalent interventions targeting individuals, these interventions should be more widespread. However, moderating factors such as the types of PA goal and dyad need to be considered to maximise effects.