Health Behavior in Implementation Science
Why some interventions are hard to implement?
In population health science and health policy, a research-to-practice gap exists, wherein evidence-based innovations (e.g., interventions, practices, or programs) either fail to achieve broad-scale implementation in communities where they are needed or fail to achieve intended outcomes once implemented. For instance, there is plenty of evidence that the Human Papillomavirus Vaccine (HPV) is more than 95% effective in preventing cancer, helping prevent up to 400,000 cancer diagnoses annually worldwide, including 22,000 in the US (1). However, this vaccine is drastically underused. Based on data from the Centers for Disease Control and Prevention (CDC), National coverage is still at 60%, with high variation among States, with some under 40%. Another example is naloxone prescription to prevent opioid overdose in the U.S. The CDC recommended naloxone co-prescription in 2016, and standing orders became legal in nearly all states in 2017. However, opioid overdose deaths are still rising in the country (2). In 2018, only one naloxone prescription was dispensed for every 69 high-dose opioid prescriptions. These only two are among many examples of rigorous, sound, and good-quality research that fails to translate into successful implementation.
In response to this research-to-practice gap that the population health science field faces and to attain potential equitable public health impacts of innovations designed to improve care or access, implementation, and dissemination, sciences have emerged as systematic approaches to understanding how innovations can be integrated and sustained in diverse practice settings. Among the many aspects that the implementation science field covers, one of its primary focuses is understanding behavior change for effective interventions. Successful implementation at the system, district, or hospital level depends on behavior change from agents, such as hospital and insurance providers, patients, and nurses. There is mounting evidence that initiatives in public health and health promotion founded on social and behavioral science theories are more successful than those not (3)(4). Presseau et al. (2010) assert that much variation occurs within individuals at medical practices rather than between practices. This finding has direct implications for implementation science because it makes us realize that a change at the population level will hardly happen without affecting behavior change at the individual level. And likewise, interventions should not only be targeted at individuals. Still, they should also involve the various levels of influence and interpersonal, organizational, and environmental factors influencing health behavior and health status.
Behavior change: understanding the theories.
We have some individual-level theories of health behavior change to understand these behavioral patterns. I will focus on comparing and analyzing the following three: (i) the Theory of Reasoned Action (TRA), (ii) the COM-B Behavior Change Wheel, and (iii) the Consolidated Framework of Implementation Research (CFIR). All these theories are based on different frameworks but with the same goal of understanding the mechanisms focused on psychology and cognition that will lead to decision-making and behavior change.
The underlying hypothesis of the TRA is that changing attitudes and norms will change behavior (5). Under this theory, behavior is driven by behavioral intentions, determined by an individual's attitude toward the behavior and subjective norms surrounding the behavior. There are background factors:
(i) Individual level: personality, emotions, intelligence, values, attitudes, mood, stereotypes, and experience.
(ii) Social level: education, age, gender, income, religion, race/ethnicity, and culture.
(iii) Information level: knowledge, media, and intervention.
These are connected with the beliefs categorized as behavioral, normative, and controls, and all shape the subjective norms surrounding the behavior, the individual’s attitude toward the behavior, and behavioral control. Finally, these attitudes and norms determine the behavioral intentions which can influence population health interventions for health behavior change. Therefore, under this theory, implementation strategies should be targeted at changing attitudes and norms, such as education via social media or training at the hospital level. A criticism of this theory is that it leaves behavioral changes entirely up to the individual. I would say that, unfortunately, change may not be within the individuals’ control, regardless of the intention. For instance, if we want to implement a policy to change physical activity patterns in a low-income community, the individual's intention or “willingness” to exercise should not be the other variable considered. There is a significant component of social inequality, stratification, and mobility in access to physical activity. Therefore, an implementation strategy of this nature should also consider the mechanisms to increase overall physical activity, such as the promotion of financial support from national and local governments to increase public facilities, such as public parks with sports equipment, and to promote equality in sports.
The next mid-level theory of health-relevant behavior I want to analyze is the COM-B Behavior Change wheel, which is highly used in implementation science. COM-B behavior system refers to the center of this proposed framework involving three essential conditions: capability, opportunity, and motivation. This theory was developed to fill in the gap of having behavioral interventions that cover all possible levels for implementation, as well as a system for matching these features to the behavioral target, the target population, and the context in which the intervention will be delivered underpinned by a model of behavior and the factors that influence it. Further, the authors' (6) wanted a model covering the full range of possible influences to exclude potentially essential variables. For instance, the TRA model does not address the roles of impulsivity, habit, self-control, associative learning, and emotional processing for behavioral change. In this theory, the determinants of behavior become the center of the wheel, with factors related to physical and psychological capability, automatic and reflective motivation, and physical and social opportunities (6). After this, the wheel focuses on the intervention functions: education, training, enablement, modeling, restrictions, environmental structures, persuasion, incentivization, and coercion. Finally, in the outer circle, the wheel determines the policy categories that will shape our interventions. These categories could be changing the legislation, service provision, establishing a regulation, fiscal measures, guidelines, or creating plans for communication, marketing, or the environment. This model has been mostly well received since it captures a range of mechanisms that may be involved in change, including those that are internal (psychological and physical) and those that involve changes to the external environment (6). However, Ogden (2016) argues that the existing variability in persons and interactions is neglected and not captured in the COM-B behavioral change wheel. Nonetheless, they make this criticism an overall point in the systemization of behavioral change science; for instance, many gaps between various psychological determinants and behavior have been identified by health psychology, and they argue that as an additional gap, many interventions suffer from low fidelity. Other authors emphasize the ideas that the role of the behavior change frameworks and theories should be to have the ability to systematically apply behavior change science while allowing variability of the individual (7).
Finally, I want to explain the fundamental concepts of the Consolidated Framework of Implementation Research (CFIR) in behavioral implementation science. Rather than a theory of health behavior, the CFIR is technically a determinant framework mainly used for behavioral implementation research. And, to my knowledge of implementation science, the CFIR is the most well-known and used determinant framework. The CFIR provides a systematic, defined list of barriers and facilitators to consider when planning or evaluating an implementation effort. Further, it was designed to make these behavioral theories, such as COM-B, more accessible, providing a menu of constructs associated with effective implementation. The CFIR contains five critical domains:
(i) The inner setting: relates to the culture of an organization, implementation climate, structural characteristics, etc.
(ii) The outer setting: patients’ needs and resources, peer pressure, incentives, etc.
(iii) Intervention characteristics: evidence strength and quality, relative advantage, adaptability, complexity, trialability, etc.
(iv) Characteristics of individuals: knowledge & beliefs, self-efficacy, individual stage of change and identification with the organization, etc.
(v) Process: planning, engaging, executing, reflecting, and evaluating.
From a behavioral theory perspective and as a determinant framework, the CFIR highlights the behavior part of the individuals under the “characteristics of individuals” domain. I consider that it provides a unique perspective to behavior change theory, as it tries to balance internal and external validity for using an implementation effort in different settings. Thanks to this, the CFIR allows not only the assessment of barriers and facilitators of an implementation effort but also to develop context-specific logic models, help with selecting implementation strategies, and inform systematic assessment of causal pathways and mechanisms of implementation strategies.
Behavior theories: policy and intervention design
There is some mixed evidence in terms of the effectiveness of theory-based interventions. For instance, Prestwich et al. (2013) find little to no differences between interventions based on Social Cognitive Theory or the Transtheoretical Model, showing that they are no more effective than interventions not reporting a theory base (8). Nonetheless, there seems to be a higher increase in substantial evidence detailing public health programs and policy interventions based on social and behavioral science theories aiming to change health behavior. These have shown to be more successful than those without any theoretical behavioral framework (3)(9)(10).
Going back to Ogden's (2018) argument regarding the COM-B Behavior Wheel, it makes sense to think that we cannot categorize and systematize every individual in a framework's factor. Translating a coded protocol can limit the effectiveness of reducing patient variability, and theory variability is necessary for the health and well-being of a discipline, and practice variability is central to the professional status of our clinician (11). However, for usefulness in the process of health policy and intervention design, behavior change theories, frameworks and models could help us to have a sound intuition of behavioral patterns under certain conditions to create and implement effective public health policies without overshadowing individual variability and highlighting the multiple levels of health behavior. These behavioral theories are generally helpful for testing and adapting interventions, organizing and positioning knowledge, communication strategies, assessing barriers and facilitators in an organization, and informing the design of one or more implementation interventions or strategies.
I also argue that behavior change theories should benefit from an economic perspective for better intervention design. Behavioral economics studies the effects of psychological, cognitive, emotional, cultural, and social factors on the decisions of individuals and institutions and how those decisions vary from those implied by classical economic theory (12). Basically, it helps us examine the differences between what people “should” do and what they actually do and the consequences of those actions. Incorporating a behavioral economics approach into policy helps us explain, for instance, why we shouldn’t pay individuals to acquire better health behaviors, such as exercising or eating more nutritious food. Moreover, behavioral economics concepts such as cognitive heuristics or nudges can help design better programs and policy interventions in health behavior. Heuristics are those mental shortcuts, unconscious decisions that can facilitate problem-solving and probability judgments (13). Richard Thaler, a leader in behavioral economics, defines a nudge as any aspect of choice that alters people's behavior predictably. There is evidence that nudges that frame information in an easy-to-understand way, change default options or enable choice are the best to improve clinical decision-making (14). However, most evidence of heuristics and nudges comes from a clinical setting or organization. Future work should consider using these pieces of evidence to implement policies to create behavioral changes in a population intervention.
Furthermore, behavior change theories can facilitate communication between actors in an intervention because everyone would be trained in the same framework for assessing the best way to intervene in behavior change, which could guide policy and intervention design. In my opinion, the most helpful way to use health behavior theory is under the context that social, cultural, and economic factors and determinants contribute to the development, maintenance, and change of health behavior patterns, and we need models to map how interrelated those determinants are. With this information, we could more easily learn how to move innovations into practice, including how to assess multi-level barriers and facilitators of program implementation; the theories, models, and frameworks underpinning state-of-the-art implementation science, at the organization or system level, for effective public health and health promotion interventions that will translate into practical impact.
References
1. How Many Cancers Are Linked with HPV Each Year? | CDC [Internet]. 2022 [cited 2022 Oct 29]. Available from: https://www.cdc.gov/cancer/hpv/statistics/cases.htm
2. Multiple Cause of Death, 1999-2020 Request [Internet]. [cited 2022 Oct 29]. Available from: https://wonder.cdc.gov/mcd-icd10.html
3. Glanz K, Bishop DB. The Role of Behavioral Science Theory in Development and Implementation of Public Health Interventions. Annu Rev Public Health. 2010 Mar 1;31(1):399–418.
4. Presseau J, Johnston M, Francis JJ, Hrisos S, Stamp E, Steen N, et al. Theory-based predictors of multiple clinician behaviors in the management of diabetes. J Behav Med. 2014 Aug;37(4):607–20.
5. Fishbein M. Reasoned Action, Theory of. In: The International Encyclopedia of Communication [Internet]. John Wiley & Sons, Ltd; 2008 [cited 2022 Oct 29]. Available from: https://onlinelibrary.wiley.com/doi/abs/10.1002/9781405186407.wbiecr017
6. Michie S, van Stralen MM, West R. The behaviour change wheel: A new method for characterising and designing behaviour change interventions. Implement Sci. 2011 Dec;6(1):42.
7. Peters GJY, Kok G. All models are wrong, but some are useful: a comment on Ogden (2016). Health Psychol Rev. 2016 Jul 2;10(3):265–8.
8. Prestwich A, Sniehotta FF, Whittington C, Dombrowski SU, Rogers L, Michie S. Does theory influence the effectiveness of health behavior interventions? Meta-analysis. [cited 2022 Oct 29]; Available from: https://core.ac.uk/outputs/19778485
9. Health Behavior and Health Education : Theory, Research, and Practice [Internet]. [cited 2022 Oct 29]. Available from: https://web-s-ebscohost-com.ezp-prod1.hul.harvard.edu/ehost/ebookviewer/ebook/bmxlYmtfXzIzODQ1MF9fQU41?sid=febb9c10-8486-4ff2-b3dd-6c89cbc7fd7e@redis&vid=0&format=EB&rid=1
10. National Cancer Institute (U.S.). Theory at a glance: a guide for health promotion practice. [Internet]. Bethesda, Md.? U.S. Dept. of Health and Human Services, Public Health Service, National Institutes of Health, National Cancer Institute; 1995 [cited 2022 Oct 29]. 48 p. (Guide for health promotion practice). Available from: https://catalog.hathitrust.org/Record/002998067
11. Ogden J. Celebrating variability and a call to limit systematisation: the example of the Behaviour Change Technique Taxonomy and the Behaviour Change Wheel. Health Psychol Rev. 2016 Jul 2;10(3):245–50.
12. What is behavioral economics? | University of Chicago News [Internet]. [cited 2022 Oct 29]. Available from: https://news.uchicago.edu/explainer/what-is-behavioral-economics
13. Heuristics [Internet]. The Decision Lab. [cited 2022 Oct 29]. Available from: https://thedecisionlab.com/biases/heuristics
14. Last BS, Buttenheim AM, Timon CE, Mitra N, Beidas RS. Systematic review of clinician-directed nudges in healthcare contexts. BMJ Open. 2021 Jul;11(7):e048801.
