Table 2.

Prioritised theoretical domains and mapping of intervention functions, policies and behavioural change techniques (BCTs) to support the three target behaviours

COM-B construct TDF domain Specific belief (linked to implementation factor)	Relevance to pharmacist-driven management of DTRs	Intervention functions considered	BCTs/policies for selected intervention functions/reasons for non-selection of intervention functions
TASK 1: applying clinical judgement
COM-B construct: psychological capability
TDF domain: skill	Clinical skills vary by pharmacist and type of drug therapy risk	Training	Instruction on how to perform a behaviour/modelling/ demonstration of behaviour: Provide brief and detailed written guidance on managing drug therapy risks targeted by the P-DQIP tool; demonstrate how to use the P-DQIP tool to identify and manage DTRs.
TDF domain: memory, attention and decision making	Support in prioritising patients and identifying DTRs valued	Enablement	Prompts/cues: P-DQIP tool identifies patients with drug therapy risks at practice level
COM-B construct: automatic motivation
TDF domain: reinforcement	Support in identifying DTRs at the point of review	Environmental restructuring	Prompts/cues: P-DQIP tool identifies drug therapy risks in individual patients
		Training	None: not feasible
		Incentivisation	None: not feasible or acceptable
		Coercion	None: not feasible or acceptable
TASK 2: collaboration with GPs
COM-B construct: psychological capability
TDF domain: skill	The quality of relationships between pharmacists and GPs varies	Training	Problem solving (to address interpersonal skills): Prompt pharmacists to analyse interpersonal barriers for collaboration with GPs and develop strategies to overcome them (eg, to build trust).
COM-B construct: social opportunity
TDF domain: social influences	Practices’ trust in pharmacists’ skills varies; practices‘ perceptions of pharmacist’s role as mainly cost-cutting limits collaboration in patient care; practices’ interest in P-DQIP work is crucial but expected to be variable.	Environmental restructuring	Communication/marketing: Promotion of the P-DQIP tool among GP clusters as a means to monitor and drive quality improvement in DTR management Restructuring the social environment: Financial incentives for GP practices to engage in P-DQIP work.
		Enablement	Self-monitoring of behaviour (GP practices): Provide tools to facilitate monitoring of review activity and trends in patients with DTRs.
		Modelling	None: not feasible because of the heterogeneity of social context.
		Restriction	None: not feasible (although pharmacists thought a policy that protects pharmacist time from other routine demands was deemed desirable).
TASK 3: fitting P-DQIP into work routines
COM-B construct: psychological capability
TDF domain: skill	Challenge of fitting P-DQIP work into work routines	Training	Action-planning: Encourage pharmacist to make detailed plans on how to introduce pharmacist-driven DTR management to practices.
TDF domain: behavioural regulation	Pharmacists struggle to fit medication reviews in with their routine work; pharmacists will avoid or procrastinate reviewing patients with more complex drug therapy risks.	Training	Goal setting (behaviour): Encourage pharmacist to set themselves achievable goals, for example, conduct at least one review per day Action-planning: see under skills.
		Enablement	Self-monitoring of behaviour (pharmacists): Provide tools to facilitate monitoring of review activity and trends in patients with DTRs.
COM-B construct: automatic motivation
TDF domain: reinforcement	Pharmacists may not perceive P-DQIP reviews as a health board priority	Environmental restructuring	Monitoring of behaviour by others: Pharmacists will include their DTR management activity in their monthly report to line managers.
		Training	None: not feasible
		Incentivisation	None: not feasible or acceptable
		Coercion	None: not feasible or acceptable
COM-B construct: physical opportunity
TDF domain: environmental context and resources	Belief that health board demands for cost-saving work will conflict with P-DQIP delivery; belief that practices’ demands on pharmacists’ work will conflict with P-DQIP delivery (especially when there is limited support from pharmacy technicians).	Training	Action-planning: see under skills
		Restriction	None: not feasible (although pharmacists thought a policy that protects pharmacist time from other routine demands was deemed desirable).
		Environmental restructuring	Guideline: health board specifies priorities for the P-DQIP work that initially contain the number of drug therapy risks to be targeted for review by pharmacists.
COM-B construct TDF domain specific belief (linked to implementation factor)		Relevance to pharmacist-driven management of DTRs.	Rationale for exclusion
TASK 1: applying clinical judgement
COM-B construct: psychological capability
TDF domain: knowledge	Knowledge of pharmacotherapy, task environment and patient preferences/ circumstances		Not prioritised for intervention, because reported limitations of pharmacists’ abilities appeared to relate more to managing DTRs (skill) rather than to gaps in pharmacotherapeutic knowledge. It was considered unfeasible to change pharmacists’ knowledge of the task environment (requires experience) and of patient preferences/circumstances (requires patient contact).
COM-B construct: reflective motivation
TDF domain: beliefs about capabilities	Belief in own capabilities is influenced by perceived knowledge, skill, other work demands and support from practices, which is variable.		Not prioritised for intervention since it would likely require an individually tailored intervention (infeasible). However, it was believed that this domain could be indirectly influenced by targeting skill, memory/attention/decision making, behavioural regulation, environmental context/resources and social influences.
COM-B construct: automatic motivation
TDF domain: emotion	Anxiety towards/professional satisfaction from autonomous decision making		Not prioritised for intervention since mitigating anxiety would require an individually tailored intervention (infeasible); Enhancing professional autonomy deemed infeasible as part of the intervention.
TASK 2: collaboration with GPs
COM-B construct: reflective motivation
TDF domain: professional/social role and identity		Pharmacist shares responsibility for therapeutic decisions with other clinicians.	Not prioritised for intervention. Although greater professional autonomy could facilitate P-DQIP implementation, it was deemed infeasible to enhance it as part of the intervention.
TDF domain: goals		Professional recognition	Not prioritised for intervention since aims of P-DQIP appeared to be aligned with personal goals.
TASK 3: fitting P-DQIP into work routines
COM-B construct: reflective motivation
TDF domain: beliefs about consequences		Impact on work processes and patient outcomes.	Not prioritised for intervention since pharmacists appeared to quickly understand the differences to current work processes as well as potential advantages and disadvantages; it was deemed infeasible to change pharmacists’ perception of increased workload.
TDF domain: optimism		Belief that P-DQIP work can be implemented despite increased workload.	Not prioritised for intervention since it would likely require an individually tailored intervention (infeasible).

COM-B, Capability-Opportunity-Motivation-Behaviour; GP, general practitioner; P-DQIP, Pharmacist and Data-Driven Quality Improvement in Primary Care; TDF, Theoretical Domains Framework.