A Systematic Review of Behaviour Change Techniques in Digital Health Interventions with Midlife Women (Preprint)

Abstract

Digital health interventions (DHIs) are efficacious in health-promoting behaviours (e.g., healthy eating and regular physical activity) that mitigate health risks and menopausal symptoms in midlife, however, integrated evidence-based knowledge about the mechanism of change in these interventions is unclear.

This systematic review aims to evaluate studies on behaviour change techniques (BCTs) and mechanisms of change in DHIs aimed to promote health-enhancing behaviours in midlife women (aged 40-65 years).

A systematic literature search of electronic databases MEDLINE/PubMed, Web of Science, PsycINFO and Cochrane Central Register of Controlled Trials (CENTRAL) in The Cochrane Library were conducted. Two independent reviewers selected 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 Behaviour Change Wheel (BCW) framework. Reporting of psychological theory use within these interventions were explored using the Theory Coding Scheme (TCS). Mode of delivery, psychological theory and BCTs were presented as descriptive statistics.

Thirteen interventions (including 1308 women) reviewed used 13 ± 4.30 (range 6–21) BCTs per intervention on average. ‘Shaping knowledge’ and ‘Repetition and substitution’ behaviour change categories (BCC) were used most frequently, with 12 interventions (92%) implementing at least one of these BCTs from these two categories. Only 14% (169/1209) of the 93 available BCTs were used, with ‘Instructions on behaviour’ most frequently utilised (12 of 13 studies). The BCW mapping suggests that half of the intervention content aimed to increase ‘Capability’ (50%; 49/98), 42% ‘Motivation’ (441/98), and 8% ‘Opportunity’ (8/98). ‘Behavioural Regulation’ was the most frequently used mechanism of action (15%; 15/98), followed by increasing ‘Knowledge’ (13%; 13/98) and ‘Cognitive and Interpersonal skills’ (10%; 10/98). Seven (78%; 7/9) intervention functions were used in the studies to change behaviour, primarily through ‘Enablement’ (36%; 60/169), while no study used ‘Restriction’ or ‘Modelling’ functions. Although 69% (9/13) of the studies mentioned a psychological theory or a model, the majority (77%; 10/13) stated or suggested rather than demonstrated 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 (68%; 9/47) modes of delivery, followed by phone/text (62%; 8/47) and wearables (54%; 7/47).

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 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 with women in midlife.

PROSPERO CRD42021259246 https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42021259246