In this study, we designed and refined a multifaceted intervention to facilitate the de-implementation of low-value practices in injury care. The CFIR(23) and the ERIC(24) tool enabled us to design an intervention based on barriers and facilitators to the reduction of targeted practices identified by trauma interested parties and according to evidence. Barriers and facilitators mainly concerned healthcare professionals' beliefs and perceived benefits, as well as the self-efficacy and the complexity of practice change in organizations. As a result, an intervention comprising numerous strategies to de-implement low-value practices was designed. These strategies focus on individuals and system changes and include distributing annual A & F reports on performance through a recognized national authority for the evaluation of practice standards, involving trauma center leaders, and providing educational materials and de-implementation support tools and educational/facilitation visits.
Through usability testing among interested parties from trauma centers of different designation levels and representatives from a national body overseeing the performance evaluation of trauma centers, we identified critical, moderate and minor issues that led to multiple improvements to the intervention. Improvements included the addition of information on drivers and benefits of reducing targeted low-value practices, changes in the definition of low-value practices, the addition of proposed strategies to facilitate de-implementation, and the tailoring of the educational/facilitation visits to the needs of trauma teams.
Obtaining interested parties’ input on barriers and facilitators to the development of interventions aimed at improving clinical practices has been identified as a crucial step in increasing their penetration into the healthcare system, their perceived relevance, and their sustainability.(23, 39) This approach is also in line with the Choosing Wisely De-Implementation Framework (CWDIF), that was designed to promote the implementation of evidence-based strategies to effectively reduce low-value care.(40) Barriers and facilitators are considered as determinants of healthcare professionals’ behaviors.(41) Several of these determinants were identified as hindering or enabling the adoption of recommended care and the reduction of low-value care. As shown in our study, the barriers and facilitators most often identified as being involved in improving practices are related to healthcare professionals and the organizational context.(26, 42–44) Healthcare professionals’ beliefs, attitudes, knowledge, understanding, skills and confidence have been recognized as important determinants to consider in implementation and de-implementations processes.(26, 42–44) The same applies to organizational leadership and support for practice improvement, the efforts required to change practices, and the availability of data to monitor quality of care and progress.(26, 42, 43)
Despite similarities between the categories of barriers and facilitators identified for high- and low-value practices, the underlying drivers associated with the latter may make de-implementation different to implementation.(26) These drivers include healthcare professionals’ fear of the consequences of withholding a test or treatment, their desire to meet patients' expectations and the time needed to explain the non-necessity of a treatment, their willingness to adopt new treatments rather than abandon existing ones, and financial incentive.(26, 45) Hence, interventions that address barriers and facilitators need to be designed diligently to facilitate the reduction of low-value care.(45, 46)
A variety of strategies have been incorporated into interventions aimed at reducing low-value practices, such as those selected in our study.(45–47) A recent knowledge synthesis showed that the following ERIC strategies are the most frequently used in de-implementation interventions: training and education (i.e., educational materials, making training dynamic by promoting the active participation of interest parties, organizing educational meetings), the use of evaluative and iterative strategies (i.e., A & F, implementing a quality monitoring system) and support for clinicians (i.e., using a clinical decision support tool).(47) Similarly, strategies related to behavior substitution (i.e., replacing low-value practices by a desired practice, for example by using share-decision tools to better communicate with patients) and restructuring the social environment (e.g., requiring discussion of care with other colleagues or obtaining authority approval to use a potentially low-value practice) were found to be more frequently used in de-implementation interventions than in implementation interventions.(48)
Regardless of the number of strategies that make up an intervention, those that contain educational materials or involve the use of A&F were shown to be the most effective in reducing low-value practices.(45, 46) Also, the combination of educational materials and meetings, A&F, patient-directed strategies (e.g. educational material, share-decision making tools), and organizational strategies (e.g., tools for supporting decision-making, changing test ordering procedures and forms for de-listing practices) were shown to enhance the sustainability of the effect.(46) However, in order to overcome the powerful underlying drivers associated with low-value care, it was recognized that problem analysis and context assessment were fundamental to tailoring de-implementation strategies.(45, 46) This exercise was made possible in our study through usability testing, which enabled us to refine the intervention with interested parties to strengthen the ability of trauma teams to de-implement the targeted practices.
Previous systematic reviews have showed favorable outcomes associated with de-implementation interventions(45, 46, 49–51) including a recent review of RCTs that found a significant median relative reduction of 17% in low-value care.(46) However, a greater number of interventions tailored to contextual factors, as in our study, were recognized as a potential strategy for further increasing their impact on low-value care.(45, 46) In addition, most of the de-implementation interventions whose effectiveness has been evaluated to date have focused on medication in a primary care setting(46), and there is still limited evidence on emergency care, including trauma-specific de-implantation interventions. The field of trauma differs from many other healthcare areas due to the complexity of clinical practice resulting from the speed of decision-making, heterogeneous populations, the multiplicity of specialists involved, the wide range of diagnostic and therapeutic options, and the need to transfer many patients to specialized centers for definitive care.(12) The potential benefits of de-implantation interventions adapted to this context therefore need to be further investigated. Hence, the next step of this research project will be to evaluate the effectiveness of the developed multifaceted intervention on the reduction of targeted low-value practices.(12)
Study strengths and limitations
This study proposes an intervention aimed at reducing low-value practices in acute injury care, that is based on a rigorous process of design and refinement. The description of the process and the results obtained will help to advance the science of de-implementation in the field of emergency care and in other healthcare areas. Similarly, prior identification of low-value practices according to a thorough review of the evidence(52–55) and expert consensus(13), comprehensive analysis of barriers and facilitators to determine intervention strategies, consultation with interested parties to refine these strategies, and the embedment of the process within a system-based initiative, increase the likelihood of having developed an acceptable, effective and sustainable intervention.(23, 39, 40, 56)
This study also has limitations. First, the identification of barriers and facilitators that contributed to the selection of the intervention strategies was not carried out through direct clinical observations in each of the trauma centers involved in the provincial performance evaluation process. However, experts working in several centers across Canada and internationally were consulted and a systematic review of the most common barriers and facilitators in de-implementation approaches was used, increasing the probability that we targeted the most important ones. Second, intervention refinement was conducted only with trauma decision-makers, the majority of whom had less than 5 years’ experience in this role, and in a small number of rural trauma centers (Level III). Trauma medical directors and trauma program managers are responsible for ensuring the quality of patient care, and thus for analyzing performance data and working with the multidisciplinary team to develop improvement plans.(57) They were therefore the key interested parties to involve in refining the intervention to promote appropriate initiation of low-value practice de-implementation. The processes surrounding de-implementation will be evaluated with healthcare professionals concerned by the low-value practices as part of our clinical trial, enabling us to gather their feedback and continue to improve the intervention for future performance evaluation cycles. The same is true for the planned educational and facilitation sessions, which will not only inform us about further refinements required to better meet the needs of trauma centers according to their level of designation, but will also provide support for less experienced decision-makers.