Acceptability, reliability, and validity of a brief measure of capabilities, opportunities, and motivations ("COM-B").

Abstract

Objectives The Capabilities, Opportunities, Motivations, Behaviour (COM-B) model is being used extensively to inform intervention design, but there is no standard measure with which to test the predictive validity of COM or to assess the impact of interventions on COM. We describe the development, reliability, validity, and acceptability of a generic 6-item self-evaluation COM questionnaire. Design and methods The questionnaire was formulated by behaviour change experts. Acceptability was tested in two independent samples of health care professionals (N=13 and N=85, respectively) and a sample of people with low socio-economic status (N=214). Acceptability (missing data analyses and user feedback), reliability (test-retest reliability and Bland-Altman plots) and validity (floor and ceiling effects, Pearson's correlation coefficient [r], exploratory factor analysis [EFA], and confirmatory factor analysis [CFA] were tested using a national survey of 1,387 health care professionals. Results The questionnaire demonstrated acceptability (missing data for individual items: 5.9-7.7% at baseline and 18.1-32.5% at follow-up), reliability (ICCs .554-.833), and validity (floor effects 0.6-5.5% and ceiling effects 4.1-22.9%; pairwise correlations rs significantly <1.0). The regression models accounted for between 21 and 47% of the variance in behaviour. CFA (three-factor model) demonstrated a good model fit, (χ2 [6]=7.34, p=.29, RMSEA=.02, CFI=.99, TLI=.99, BIC=13,510.420, AIC=13,428.067). Conclusions The novel six-item questionnaire shows evidence of acceptability, validity, and reliability for self-evaluating capabilities, opportunities, and motivations. Future research should aim to use this tool in different populations to obtain further support for its reliability and validity. Statement of contribution What is already known on the subject? The Capability, Opportunity, Motivation (COM), Behaviour (-B) model is being used extensively to inform intervention design. The lack of an accepted universal measure hinders progress in behaviour change. What does this study add? There is evidence of acceptability, validity, and reliability for self-evaluating COM. Our measure may be sufficiently generic for any behaviour or population, although this requires further testing.