Implementation science issue

It has been reported that it takes an average of 17 years for only 14% research evidence to become adapted into clinical practice, and this is often referred to as the “17-year odyssey”1 or, more depressingly, the “valley of death.”2, 3 Systemic deficiencies and gaps between evidence-based best care practices and healthcare delivery, the “waste that comes with poor execution or lack of widespread adoption of known best care practices,” have been identified as failures of care delivery, representing hundreds of billions of dollars in annual wasteful spending.4, 5 Creating new knowledge to improve patient care requires several “translations,” termed the “3T's” road map.6 The first step (translation 1 [T1]) involves translating bench or basic science knowledge into clinical research. The next step, T2, converts this clinical research into best-practice guidelines that may then be widely disseminated to clinicians and policy-makers. The final translation (T3) addresses the adoption and application of those evidence-based guidelines into practice. Although the first two translational steps have received much attention and funding, the final T3 step has, until recent decades, been relatively ignored. Implementation science (IS), comprising both implementation research and implementation practice, can be succinctly described as the theory-based study of how to integrate new interventions into clinical practice,7 and the goal is to create new knowledge concerning effective strategies to execute an intervention once it has already been proven to be effective (rather than test the effectiveness of the intervention). Although IS shares similar features with quality improvement (QI), it is important to distinguish the differences. While QI begins with a specific problem in a specific context and seeks to attain a very specific SMART (Specific, Measurable, Achievable, Relevant, and Time-bound) goal, IS begins with an evidence-based practice (EBP) that is underemployed and seeks to understand all the factors that impede (barriers) or facilitate (enablers) the successful adoption of that EBP with the goal to produce knowledge that is generalizable beyond the particular system under study.8 QI is focused on performance improvement, whereas IS seeks to develop and incorporate new processes.9 Minimizing the gap between what we know and what we do, the “know-do gap” is the focus of the IS field. Importantly, IS also attempts to understand how and why the adoption of best practices succeeds or fails. IS studies involve diverse targets, such as individual wards/clinics, organizations/systems, communities, and policies.8 Well-planned and well-executed IS studies involve a diverse array of collaborators who are uncommonly involved in medical science trials, such as social scientists, anthropologists, human-factors engineers, and economists. 7 In addition to introducing new effective interventions into clinical practice, an equally important (although often overlooked) focus of IS is the deimplementation or deadopting of wasteful or ineffective practices.10, 11 One pertinent example is the “NPO at midnight” fasting guidelines that have persisted for decades despite evidence of no benefit over modern shortened preoperative fasting guidelines.12 In the past few decades, the field of IS has experienced exponential growth, now comprising >100 different available theories13, 14 and behavior change techniques.15 With this bewildering array of available frameworks, allied health professionals are justifiably confused and overwhelmed, unsure where to start and how to choose an appropriate theory. It is thus unsurprising that many report low confidence16 in choosing a framework and selection is most often haphazard or based on familiarity rather than appropriateness.13 This special issue of Nutrition in Clinical Practice, “Implementation Science in Clinical Nutrition,” provides an overview of various IS frameworks to serve as a nutrition-focused practical guide for readers interested in understanding and performing intervention science. Selecting the appropriate implementation framework depends upon the context, project goals, stage, and type of data selected. Initially, one determines if they are trying to understand barriers/facilitators to implementation, to evaluate an intervention's implementation, to design an implementation strategy, or to measure outcomes. If understanding context is the goal, then Consolidated Framework for Implementation Research (CFIR)17, 18 or Theoretical Domains Framework (TDF)19-21 may be the appropriate framework, whereas Expert Recommendation for Implementing Change (ERIC),22 Behavior Change Wheel, or Implementation Mapping may be more appropriate for guiding a strategy design. If evaluating an implementation intervention is the purpose, then Reach, Effectiveness, Adoption, Implementation, and Maintenance (RE-AIM),23, 24 Practical Implementation Sustainability Model (PRISM), or Proctor's Outcome Framework are excellent options, whereas integrated Promoting Action on Research Implementation in Health Services (i-PARIHS)25 and Knowledge to Action (KTA)26 are potential frameworks to consider if the implementors plan to track the process over time. The Plan-Do-Study-Act (PDSA) model is an excellent choice for incremental and iterative QI.27 Early-stage frameworks include Exploration, Preparation, Implementation, Sustainment (EPIS), CFIR, and TDF and are focused on needs assessment and planning. Middle-stage frameworks focus on executing the implementation intervention and use RE-AIM, i-PARIHS, and Proctor's. Late-stage frameworks evaluating sustainability or scale-up use PRISM, RE-AIM, and Dynamic Sustainability Framework. Finally, PRISM, EPIS, and CFIR are useful for multilevel systems or complex organizational environments, whereas TDF or Capability, Opportunity, Motivation, and Behavior are better utilized to implement individual behavior change. In addition to invited review articles, this special issue also includes original research demonstrating IS in nutrition practice across a variety practice settings and interventions. These articles provide examples in designing and conducting a successful implementation project in the context of various framework, frequently using an interdisciplinary approach. Two examples of early-stage IS study are presented by Sullivan et al, who investigated the barriers and facilitators to implementing routine nutrition screening in a memory care clinic28 and Golder et al who investigated barriers and facilitators to providing vitamin C deficiency care in a general medicine ward.29 The findings from these studies will help the authors (and readers) plan and develop future effective strategies. Taakefala et al identified an evidence-practice gap, with only 25% of their post–liver transplant patients receiving best-practice early initiation of nutrition.30 By combining multiple IS frameworks (KTA, CFIR, ERIC), they identified important barriers and developed a tailored suite of theory-informed and stakeholder-informed intervention strategies to greatly improve compliance to 60%. In a similar example, Elcin et al sought to introduce a completely new gastrostomy care bundle into their adult neurosurgical intensive care unit (ICU).31 Using the KTA combined with the Institute of Healthcare Improvement development recommendations, these authors conducted interviews, involved key stakeholders, and successfully implemented an evidence-based bundle with >95% compliance. The Grol and Wensing implementation model underlies the study by Heijnen et al demonstrating successful implementation of a reduced preoperative fasting guideline preceding orthopedic surgery.32 In an interesting hybrid of IS and QI, Knippen et al describe using the PDSA QI methodology and CFIR IS framework supported by artificial intelligence (AI)-assisted software to analyze implementation of guidelines within the multiphase, multisite Gestational Diabetes Mellitus Registry Study.33 Long-term sustainment of an implementation intervention is investigated in this issue by Mayre et al using the KTA framework.34 Several examples of QI research are also included in this issue. MacDonald et al report their successful experience improving the utilization of a food insecurity screening protocol in a long-term acute care hospital, whereas Vendiola et al used the PDSA framework to greatly improve documentation of energy goals in a pediatric ICU.35 Finally, Munk et al also used the PDSA framework to greatly improve the percentage of elderly hip fracture patients meeting their energy and protein goals.36 When an initial implementation attempt is unsuccessful, evaluating what was supposed to happen, what actually happened, the gaps between the two, and any unintended consequences using structured reflection tools such as root cause analysis or Failure Model and Effects Analysis are helpful in evaluating these discrepancies. A framework shift may be necessary depending upon the identified issues such as a switch from a more comprehensive to context-sensitive framework (eg, change from RE-AIM to CFIR). If individual-level resistance is identified, then adding a behavior theory (eg, TDF) may help. Investigators may also identify shifts in climate, leadership, or competing priorities since the start of the project and determine that the original strategy is no longer a good fit. A continuous QI mindset like PDSA allows investigators to monitor and iterate rapidly along with transparency. Documentation of failures is an important contribution to IS and allows for faster advancement and sustainability in the IS space. For example, when a large-scale weight management program introduced nationally to the Veteran Affairs medical centers experienced wide variation in implementation, a “post-mortem” analysis using the CFIR framework identified 10 constructs distinguishing facilities with low implementation from those with high implementation effectiveness.37 Understanding and addressing these constructs can help stakeholders understand why an implementation intervention fails in a certain environment and can help inform an implementation strategy to address these barriers. IS in healthcare has moved beyond its infancy to young adulthood, recognized as a legitimate field of study with unique and important contributions to improving patient care. In fact, we are now entering the delightfully meta-implementation era (studying the implementation of IS)!38 Although practical guides to IS already exist,7-9, 39, 40 our sincere hope is that this special theme issue will aid readers of Nutrition in Clinical Practice to demystify IS and introduce them to the field from a nutrition-focused perspective by reviewing the most relevant conceptual frameworks and providing actual examples of theory-based real-world application in nutrition support. Dr. Yeh reports author royalties for review articles on small and large bowel obstruction from UpToDate; consulting fees from Takeda Pharmaceuticals, Baxter, and Wave Strategies; and is a member of the medical advisory board for Irrimax, LLC.