Background
Digital health interventions are efficacious in health-promoting behaviors (eg, healthy eating and regular physical activity) that mitigate health risks and menopausal symptoms in midlife. However, integrated evidence-based knowledge about the mechanisms of change in these interventions is unclear.
Objective
This systematic review aimed to evaluate studies on behavior change techniques (BCTs) and mechanisms of change in digital health interventions aimed at promoting health-enhancing behaviors in midlife women (aged 40-65 years).
Methods
A systematic literature search of the electronic databases PubMed, Web of Science, PsycINFO, and Cochrane Central Register of Controlled Trials in the Cochrane Library was conducted. In total, 2 independent reviewers selected the studies for inclusion, extracted data, and completed BCT mapping of eligible studies. The mechanism of action and intervention functions of eligible studies were evaluated using the behavior change wheel framework. Reporting of psychological theory use within these interventions was explored using the Theory Coding Scheme. Mode of delivery, psychological theory, and BCTs were presented as descriptive statistics.
Results
In total, 13 interventions (including 1315 women) reviewed used 13 (SD 4.30, range 6-21) BCTs per intervention on average. The “Shaping knowledge” and “Repetition and substitution” behavior change categories were used most frequently, with 92% (12/13) of the interventions implementing at least one of the BCTs from these 2 categories. Only 13.98% (169/1209) of the 93 available BCTs were used, with “Instructions on behaviour” most frequently used (12/13, 92%). The behavior change wheel mapping suggests that half of the intervention content aimed to increase “Capability” (49/98, 50% of the intervention strategies), “Motivation” (41/98, 42%), and “Opportunity” (8/98, 8%). “Behavioural Regulation” was the most frequently used mechanism of action (15/98, 15%), followed by increasing “Knowledge” (13/98, 13%) and “Cognitive and Interpersonal skills” (10/98, 10%). A total of 78% (7/9) of the intervention functions were used in the studies to change behavior, primarily through “Enablement” (60/169, 35.5%), whereas no study used “Restriction” or “Modelling” functions. Although 69% (9/13) of the interventions mentioned a psychological theory or model, most (10/13, 77%) stated or suggested rather than demonstrated the use of a theoretical base, and none reported explicit links between all BCTs within the intervention and the targeted theoretical constructs. Technological components were primarily based on web-based (9/13, 69%) modes of delivery, followed by phone or SMS text message (8/13, 62%) and wearables (7/13, 54%).
Conclusions
The findings of this review indicate an overall weak use of theory, low levels of treatment fidelity, insignificant outcomes, and insufficient description of several interventions to support the assessment of how specific BCTs were activated. Thus, the identified limitations in the current literature provide an opportunity to improve the design of lifestyle health-enhancing interventions for women in midlife.
Trial Registration
PROSPERO CRD42021259246; https://tinyurl.com/4ph74a9u