'Around the edges': using behaviour change techniques to characterise a multilevel implementation strategy for a fall...

Implementation strategies are needed to ensure that evidence-based healthcare interventions are adopted successfully. However, strategies are generally poorly described and those used in everyday practice are seldom reported formally or fully understood. Characterising the active ingredients of existing strategies is necessary to test and refine implementation. We examined whether an implementation strategy, delivered across multiple settings targeting different stakeholders to support a fall prevention programme, could be characterised using the Behaviour Change Technique (BCT) Taxonomy.

Data sources included project plans, promotional material, interviews with a purposive sample of stakeholders involved in the strategy's design and delivery and observations of staff training and information meetings. Data were analysed using TIDieR to describe the strategy and determine the levels at which it operated (organisational, professional, patient). The BCT Taxonomy identified BCTs which were mapped to intervention functions. Data were coded by three researchers and finalised through consensus.

We analysed 22 documents, 6 interviews and 4 observation sessions. Overall, 21 out a possible 93 BCTs were identified across the three levels. At an organisational level, identifiable techniques tended to be broadly defined; the most common BCT was restructuring the social environment. While some activities were intended to encourage implementation, they did not have an immediate behavioural target and could not be coded using BCTs. The largest number and variety of BCTs were used at the professional level to target the multidisciplinary teams delivering the programme and professionals referring to the programme. The main BCTs targeting the multidisciplinary team were instruction on how to perform the (assessment) behaviour and demonstration of (assessment) behaviour; the main BCT targeting referrers was adding objects to the environment. At the patient level, few BCTs were used to target attendance.

In this study, several behaviour change techniques were evident at the individual professional level; however, fewer techniques were identifiable at an organisational level. The BCT Taxonomy was useful for describing components of a multilevel implementation strategy that specifically target behaviour change. To fully and completely describe an implementation strategy, including components that involve organisational or systems level change, other frameworks may be needed.