Changing behaviour ‘more or less’—do theories of behaviour inform strategies for implementation and de-implementation? A critical interpretive synthesis

Patey et al. Implementation Science (2018) 13:134 https://doi.org/10.1186/s13012-018-0826-6 RESEARCH Open Access Changing behaviour ‘more or less’—do theories of behaviour inform strategies for implementation and de-implementation? A critical interpretive synthesis Andrea M. Patey1,2*, Catherine S. Hurt1, Jeremy M. Grimshaw2,3 and Jill J. Francis1,2 Abstract Background: Implementing evidence-based care requires healthcare practitioners to do less of some things (deimplementation) and more of others (implementation). Variations in effectiveness of behaviour change interventions may result from failure to consider a distinction between approaches by which behaviour increases and decreases in frequency. The distinction is not well represented in methods for designing interventions. This review aimed to identify whether there is a theoretical rationale to support this distinction. Methods: Using Critical Interpretative Synthesis, this conceptual review included papers from a broad range of fields (biology, psychology, education, business) likely to report approaches for increasing or decreasing behaviour. Articles were identified from databases using search terms related to theory and behaviour change. Articles reporting changes in frequency of behaviour and explicit use of theory were included. Data extracted were direction of behaviour change, how theory was operationalised, and theory-based recommendations for behaviour change. Analyses of extracted data were conducted iteratively and involved inductive coding and critical exploration of ideas and purposive sampling of additional papers to explore theoretical concepts in greater detail. Results: Critical analysis of 66 papers and their theoretical sources identified three key findings: (1) 9 of the 15 behavioural theories identified do not distinguish between implementation and de-implementation (5 theories were applied to only implementation or de-implementation, not both); (2) a common strategy for decreasing frequency was substituting one behaviour with another. No theoretical basis for this strategy was articulated, nor were methods proposed for selecting appropriate substitute behaviours; (3) Operant Learning Theory makes an explicit distinction between techniques for increasing and decreasing frequency. Discussion: Behavioural theories provide little insight into the distinction between implementation and deimplementation. Operant Learning Theory identified different strategies for implementation and de-implementation, but these strategies may not be acceptable in health systems. Additionally, if behaviour substitution is an approach for de-implementation, further investigation may inform methods or rationale for selecting the substitute behaviour. Keywords: Implementation, De-implementation, Behavioural theory and model, Behaviour change, Health professional, Intervention, Implementation research, Critical interpretive synthesis * Correspondence: 1 School of Health Sciences, City, University of London, 10 Northampton Square, London EC1V 0HB, UK 2 Centre for Implementation Research, Ottawa Hospital Research Institute – General Campus, 501 Smyth Road, Ottawa, Ontario K1H 8L6, Canada Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Patey et al. Implementation Science (2018) 13:134 Background Developing theory and evidence about interventions to support de-implementation is an important priority in implementation research. In 2014, Implementation Science issued an editorial arguing the need for more research to identify strategies to de-implement low-value or harmful care [1]. Since then, Implementation Science has published six research articles and protocols [2–7] investigating de-implementation strategies in a number of clinical contexts. Despite increasing policy interest in de-implementation, with international programmes such as the Choosing Wisely campaign [8–10] and Preventing Overdiagnosis initiative [11–13], relatively little work has been reported to understand and address systematic methods for designing de-implementation interventions. Researchers have noted de-implementation will likely involve different approaches than those used to promote people to do more of some things, but there is little evidence to support this notion [14, 15]. This raises the question of whether approaches for implementation versus de-implementation are similar or distinct. It is unknown whether or not this is the case, suggesting an investigation into whether implementation and de-implementation approaches should differ is imperative. Currently, the literature appears to lack clear guidance about what those approaches should be [16, 17]. Implementation or de-implementation as behaviour change is an important and productive thread within implementation research. A focus on reducing the frequency of overused clinical behaviours may offer a perspective that is currently lacking in the discourse on de-implementation. Behavioural theories can aid in developing a better understanding of the main effects, mediators (mechanisms), and moderators (effect modifiers) between behavioural influences and interventions in the environments (policy, system, organisation, team) [18] in which healthcare professionals work. Evidence and theory from behavioural science have informed methods for identifying factors that explain and influence behaviour and for selecting techniques to support behaviour change of healthcare professionals [19–21]. There have been major methodological and theoretical developments in the field of health psychology in designing and evaluating multi-level interventions. Advances in intervention mapping using behavioural theories have improved the design and implementation of health promotion interventions (community-level) and school-based programmes (system-level) [22, 23]. In addition, the Behaviour Change Wheel (BCW), a guide for designing interventions with its foundation in the behavioural sciences, illustrates that interventions can be delivered at any level by including policy-, system-, and individual-level components [24]. However, it is unclear to what extent theories from behavioural science propose different mechanisms for implementation and Page 2 of 13 de-implementation. This study reviewed published literature to investigate whether theories of behaviour differentiate between t (...truncated)