Implementation science seeks to close the gap between what we do and what we know by identifying and addressing the barriers that prevent uptake of evidence-based practices. An intervention that is guided by evidence-based implementation strategies is more likely to produce desired results because proper implementation is a key component of an intervention’s success. These strategies include methods used to: improve adoption, implementation, sustainment, and scale-up of interventions.
Understanding the processes by which strategies (mechanisms) produce desired effects is important to determine why a strategy did or did not achieve its intended effect. It is also important for practice to ensure strategies are designed and selected to directly target determinants or barriers.
To better understand these mechanisms of implementation, researchers systematically reviewed implementation studies to characterize how mechanisms are conceptualized and measured, how they are studied and evaluated, and how much evidence exists for specific mechanisms.
In total, 46 implementation studies were reviewed for implementation mechanisms that included quantitative (72.3%), qualitative (10.9%), and mixed-methods (15.2%) data collection methods.
Data extraction focused on study information/background, methods (e.g. proposed mediation model), results, and criteria for establishing mechanisms (using Kazdin’s criteria).
Findings revealed:
- Substantial variation in the models that emerged from the studies including nine unique versions of models.
- Inconsistent use of relevant terms (e.g. mechanisms and determinants).
- Fifty-three percent of the studies met half of fewer of the quality indicators assessed by the Mixed Methods Appraisal Tool.
- The majority of studies (84.8%) only met three or fewer of the seven criteria stipulated for establishing mechanisms.
Results exposed substantive conceptual, methodological, and measurement issues that must be addressed in order to advance implementation mechanism research. The authors provide several recommendations for future implementation science research including:
- Improving hypothesis generation by laying out mechanistic research questions and generating theory-driven hypotheses.
- Focusing effort on development and testing of implementation theory.
- Applying moderated mediation models and mediated moderator models for quantitative analytic approaches.
- Developing hypothesis and generating interview guides to directly test mechanisms in qualitative approaches.
To read this article in full, click here.