Firstly we need to analyse our data, including any collected from COM-B questionnaires. Imagine you worked with a team of midwives on offering the ‘flu vaccine to pregnant women in their antenatal booking appointment. This is a new part of their role and an audit of notes shows that this is only happening in around 10% of consultations. The 10 midwives in the team complete your COM-B questionnaire with statements related to each of the 6 domains, where more favourable views were higher (e.g. If I offer pregnant women the ‘flu vaccine, it will benefit their health: 1=strongly disagree, 7 = strongly agree). You have taken a mean of the scores for each domain and find this pattern of (fictional) data.  Which determinant or influence do you think is important?

The lowest average scores are for motivation. Looking at the individual scores, you see that midwives tended to disagree that 1) offering the ‘flu vaccine would be beneficial to women’s health (reflective motivation) and 2) that they were in the habit of doing this (automatic motivation). This tells us these may be key barriers for midwives in offering the ‘flu vaccine to pregnant women.

Similarly, when we analyse anonymised transcripts of interview or focus groups, through deductive coding we can match what the team said against the COM-B domains, to analyse what seem to be important barriers. Consider this (fictional) extract from a focus group with the team. What more does this tell us about the key determinants of behaviour?

This tells us more about the reflective motivation barriers, including some misunderstandings about the prevalence and severity of ‘flu in pregnancy, pointing towards tailor-made interventions that could help. You can also see here that ‘triangulation’ is useful – comparing data from several exploration methods to build up a full picture.

For more information about quantitative and qualitative data analysis, you may wish to consult a research methods book such as Walliman, N. (2017). Research Methods: The Basics: 2nd edition.

We then need to feed back findings to teams and decide together on ways forward. Interventions are tailor-made depending on the barrier(s) identified. The behaviour change wheel matches C,O and M barriers to 8 intervention types or ‘functions’: for example, if psychological capability barriers are common in a team, physical training may be needed; whereas physical opportunity barriers might mean that teams need to advocate for equipment or other resources, known as environmental restructuring.  To enhance reflective motivation, our midwife team may benefit from persuasion about the benefits of the ‘flu vaccine, such as through researching the evidence themselves, a talk by an expert in the topic, or a leaflet.

Their automatic motivation could be built over time through adding a reminder to patients’ file, another example of an environmental restructuring intervention. Some recommendations will need to be made to more senior managers or organisation leads: having evidence from this kind of work can help in advocating for structural changes which may require financial support, such as increased staffing aid with severe time pressures. The behaviour change wheel also highlights the role of organisational policies or policy categories to help support these change efforts to happen.

Every intervention contains one or more behaviour change techniques or active ingredients of change: there have been 93 identified in the BCT Taxonomy v1 (Michie et al. 2013). Depending on what is practical, affordable and acceptable to teams (known as the APEASE criteria in the Behaviour Change Wheel), many behaviour change techniques could be useful.

See if you can link behaviour change techniques to these COM-B barriers and interventions:

To help people learn more about BCTs, we have made the Cards for Change. Contact us for more information.