Table 4 List of included/excluded BCTs with reasons for inclusion/exclusion

Information about health consequences (5.1)

Considered affordable, practical, potentially effective, potentially acceptable (for patients and HCPs), should have limited side effects

Self-monitoring of behaviour (2.3)

Considered affordable, practical, potentially effective, potentially acceptable (for patients and HCPs), should have limited side effects and shouldn’t create significant issues of equity

Biofeedback (2.6)

Considered affordable (as patients’ blood pressure and cholesterol, for example, are often monitored and discussed within routine care), practical, potentially effective and potentially acceptable (for patients and HCPs)

Information about antecedents (4.3)

Considered affordable, practical, potentially effective, potentially acceptable (for patients and HCPs), should have limited side effects and shouldn’t create significant issues of equity

Credible source (9:1)

Considered affordable, practical, potentially effective, potentially acceptable (for patients and HCPs), should have limited side effects and shouldn’t create significant issues of equity

Self-monitoring of outcome(s) of behaviour (2.4)

Considered affordable (as patients can access, for example, blood pressure monitors for free from local GP surgeries and pharmacists), practical, potentially effective and potentially acceptable (for patients and HCPs)

Pros and cons (9.2)

Considered affordable, practical, potentially effective and potentially acceptable (for patients and HCPs)

Prompts/cues (7:1)

Considered affordable, practical, potentially effective, potentially acceptable (for patients and HCPs), should have limited side effects and shouldn’t create significant issues of equity

Action planning (1:4)

Considered affordable, practical, potentially effective, potentially acceptable (for patients and HCPs), should have limited side effects and shouldn’t create significant issues of equity

Habit formation (8:3)

Considered affordable, practical, potentially effective, potentially acceptable (for patients and HCPs), should have limited side effects and shouldn’t create significant issues of equity

Social support (emotional) (3.3)

Considered affordable (as patients may be able to get this support from their own social networks or from community stroke support groups already running), practical, potentially effective, potentially acceptable (for patients and HCPs), should have limited side effects

Feedback on outcome(s) of the behaviour (2.7)

Not considered practical as most feedback on behavioural outcomes (for stroke medication adherence) routinely provided in NHS is a form of biofeedback and so would add additional workload if HCPs were providing feedback

Feedback on behaviour (2.2)

Not considered practical in this context. Although HCPs based in primary care/community pharmacy have access to prescription acquisition records, this is a proxy measure of adherence and so could be difficult to provide accurate estimates of adherence. Even if stroke survivors provided self-reported accounts of adherence to HCPs/carers and adherence rates were fed back this could be too onerous as an intervention strategy

Provide normative information about others behaviour/ experiences (BM5)

Not considered practical to deliver. BCT originally utilised for smoking cessation. The impact of stroke and varying medication regimens will make generalised comparisons challenging

Salience of consequences (5.2)

Not considered practical for negative consequences of non-adherence e.g. use images of the consequences of stroke - hard to demonstrate paralysis, aphasia and other stroke implications in an image. Potential to have unwanted side effects also, if BCT evokes upsetting emotional responses. Enhancing view about positive consequences of adherence may not be considered acceptable as patient could have suffered a stroke even when adherent to medication and may not find the salience of the consequences meaningful

Information about social and environmental consequences (5.3)

Not considered acceptable. Potential ethical issues. E.g. informing patients that it’s not socially acceptable to miss medication doses – patients can have valid reasons for not taking medications

Information about emotional consequences (5.6)

Not considered acceptable and could cause unwanted side effects if information about negative emotions is given. Provision of this information could be upsetting, and patients can have valid reasons for not taking medications so seems inappropriate in this context. Even if provide information about positive emotions (e.g. taking medicines give peace of mind) may not be considered acceptable to those who have suffered a stroke even when adherent to medication and so may not find the information meaningful

Anticipated regret (5.5)

Not considered acceptable and could cause unwanted side effects. Provision of this information could be upsetting, and patients can have valid reasons for not taking medications so seems inappropriate in this context. If a patient suffered a stroke following good adherence to medications BCT could be considered unacceptable by intervention facilitators delivering this BCT

Comparative imagining of future outcomes (9.3)

Not considered acceptable. Asking people to imagine different outcomes might not be something HCPs are confident doing or patients are familiar with

Reduce negative emotions (11.2)

Not considered practical and potentially not affordable. Not all patients would require this type of BCT and training in stress management, for example, would be costly and time/labour intensive

Provide reassurance (RC10)

Not considered practical. Patients experiences (e.g. of side effects) are likely not time limited and may vary person to person. Reassurance may not always be the appropriate response and so may not be considered acceptable