Seite - 151 - in Applied Interdisciplinary Theory in Health Informatics - Knowledge Base for Practitioners

1.5. Translating theories to interventions Although the above theories are among the best-known, there are many other behaviour change theories – 83 according to Davis et al [14]. Often more than one is used when designing an intervention. Michie et al. have worked extensively to aid the application of these theories by mapping elements of each theory (“constructs”) to behaviour change techniques. Their group has identified 93 distinct behaviour change techniques, such as goal-setting (setting a concrete and achievable behavioural goal), and mapped these to 14 theoretical domains [15]. These insights have been organized into the Behaviour Change Wheel and the Theoretical Domains Framework [16]. At the centre of the Behaviour Change Wheel is the COM-B model, a general model of behaviour which states that an individual will engage in a Behaviour if they have the Capability, Opportunity, and Motivation to do so. The next level of the wheel maps these to nine intervention functions (high-level techniques) and seven policy categories (reflecting policies that can facilitate the techniques). Thus, for example, "goal setting" would be classified under Motivation in the COM-B model. A practical example of applying this framework to design an intervention is given by Mangurian et al. [17]. Their goal was to increase the rate of screening for cardiovascular disease by primary care providers in patients with severe mental illness. They followed the eight steps outlined in the Behaviour Change Wheel framework. First, focus groups were formed with providers, patients, and managers to identify barriers and potential target behaviours, to identify potential targets and select the target behaviour: ordering of metabolic screening tests (e.g. HbA1c) by primary care providers at community mental health clinics. The behaviour was then described in detail, including its place in the providers' workflow, to identify what change was needed. The researchers identified barriers in 10 behavioural constructs, e.g. "Screening for diabetes is low on the priority list for these patients" (constructs: goal setting, motivation, attitudes [Reasoned Action]) and "I don't know exactly what the guidelines recommend" (constructs: knowledge, perceived behavioural control [Reasoned Action]). These problems were then mapped to 18 intervention functions. For example, lack of knowledge could be addressed through education, or through environmental restructuring so that the knowledge is available when it is needed, e.g. through a decision support system. They also identified 8 policy strategies to support the interventions, e.g. providing training on the content of the guidelines, and 7 behaviour change techniques applicable to this situation, e.g. self-regulation (Social Cognitive Theory) in the form of feedback on their individual screening rates. Finally, they defined the mode of delivery for each intervention. The authors reported that using the framework helped them in applying the underlying behaviour change theories to their intervention. As noted by Kok et al., designing an effective behaviour change intervention is not simple [18]. Success depends on ensuring that changing the selected determinants will result in the desired behaviour, choosing behaviour change methods that affect these determinants, and executing the methods correctly so that they can be effective. The authors propose an Intervention Mapping taxonomy and protocol. The taxonomy consists of 13 behavioural determinants with methods for changing each, the theories on which these methods are based, and evidence of their effectiveness by specifying the theoretical parameters under which the method is effective or not. The taxonomy is not limited to the individual level, but also identifies methods of change at higher ecological levels. For example, the determinant "environmental conditions" can be affected by Box 1: Stages and Processes of change in the transtheoretical model [10] S.MedlockandJ.C.Wyatt /HealthBehaviourTheory inHealth Informatics 151