Background:
Use of theories, models and/or frameworks (TMFs) in implementation practice is recommended for developing implementation strategies. However, TMF selection is made difficult by the high number of TMFs to choose from. Our first objective was to systematically choose TMFs to guide an ongoing project to facilitate implementation of hip protectors in long-term care (LTC). Hip protectors, consisting of shields or pads held in pockets covering lateral aspects of the proximal femur, can prevent hip fractures due to falls among older adults living in LTC. Our project goals are to explore determinants of hip protector use and to investigate organizational readiness for change in LTC homes with low use of hip protectors. Our second objective was to propose a model for TMF selection based on the process we developed in our hip protector project.
Methods:
We developed and used a five-step, systematic, consensus-based and integrated knowledge translation (iKT) process, involving: (1) a structured search strategy to identify TMFs; (2) screening of TMFs against pre-established eligibility criteria based on the purpose of TMFs; (3) appraisal of the relevance of TMFs using a structured questionnaire adapted from the Theory Comparison and Selection Tool (T-CaST); (4) identifying the three TMFs most relevant for each project goal; and (5) the final selection based on the top three TMFs through an on-line consensus-meeting using an adapted Nominal Group Technique with knowledge users and final approval by our institutional partners.
Results:
We identified 66 TMFs in step 1, of which 23 met our eligibility criteria in step 2. Five investigators appraised the 23 TMFs in step 3, over two months. The top three TMFs relevant to each project goal identified in step 4 were: Project goal 1=Normalization Process Theory, the Exploration, Preparation, Implementation, Sustainment Model, and the Practical, Robust Implementation and Sustainability Model (PRISM); Project goal 2= Organizational Readiness Theory, the Consolidated Framework for Implementation Research (CFIR), and the Precede-Proceed Model. Nine investigators, one patient partner, and two institutional partners participated in the consensus meeting in step 5. Two combinations, each comprised of two TMFs, tied with the most votes after three rounds of voting. The tie was broken through collaboration with three institutional partners, who selected the PRISM and CFIR. We propose that the process we developed may be referred to as the Implementation Theory Selection (ITS) Model.
Conclusions:
The ITS Model we developed offers a systematic, consensus-based, iKT process to select TMFs, which we used to inform investigations into determinants of hip protector use and organizational readiness for change in LTC. The value of this approach is best illustrated by our selection of the PRISM, whom no investigators or partners were previously aware of. Our approach also provided an opportunity for investigators to engage with knowledge users early in the project’s lifespan, which may facilitate meaningful partnerships in later phases of the project. Although continued research is needed to evaluate and refine our model, it has laid the foundation for a first iteration a process model for identifying and selecting TMFs for implementation practice.
Alexandra KORALL