Known knowns and known unknowns on behavior change interventions and mechanisms of action^*

* Karina W. Davidson served as Editor-in-Chief for this article.

ABSTRACT

Systematic reviews and meta-analyses play an important role in summarising current research on the efficacy of behaviour change interventions and their mechanisms of action. The reviews in this special issue represent a ‘step change’ in evaluating current evidence on behaviour change interventions and mechanisms. This article outlines the findings and emerging issues identified in the reviews (‘known knowns’), and summarises evidence gaps to be addressed in future research (‘known unknowns’). Findings indicate that tests of mechanisms of behaviour change interventions are not routinely conducted in primary studies and research syntheses; reviews and studies do not sufficiently account for study quality; substantive variability exists in descriptions of intervention content and putative mediators implicated in their mechanisms of action; limited data is available on the efficacy of many behaviour change techniques; and moderators of intervention effects and mechanisms are seldom taken into account. Possible solutions include testing effects of isolated behaviour change techniques and mechanisms of action; routine evaluation of study quality in behavioural intervention research; development of an evidence base linking behaviour change techniques with theory-based constructs that comprise mechanisms of action; adoption of fit-for-purpose methods for synthesising behavioural intervention mechanisms of action; and routine testing of moderators in intervention research.

KEYWORDS:

• Behaviour change technique
• mechanisms of action
• behavioural theory
• behaviour change intervention
• systematic reviews and meta-analyses
• putative mediators

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

* Karina W. Davidson served as Editor-in-Chief for this article.

1 Analytically, the mechanism of action is most effectively evaluated using mediation analyses (Hayes, 2018). To the extent that the proposed mechanism of action fully accounts for the effect of the technique in changing behaviour, the indirect effect represented by paths a and b Figure 1(a) should be approximately equivalent in size to the direct effect illustrated by path c in Figure 1(b). Interested readers are directed to Hayes’ (2018) lucid treatment of this subject and accompanying process diagrams for more details.

2 It is important to be note that the term mechanism of action encompasses the effect of the intervention (behaviour change technique) on behaviour change through the theory-based factor. Links between behaviour change techniques and modifiable factors (illustrated by path a in Figure 1(a)), and between factors and behaviour change (illustrated by path b in Figure 1(a)), represent constituent parts of the mechanism of action and not the mechanism of action itself.

Additional information

Funding

Martin Hagger’s contribution was supported by a Finnish Distinguished Professor (FiDiPro) award from Business Finland (1801/31/2015).