CHALLENGES OF BEHAVIORAL INTERVENTION RESEARCH
There are numerous challenges in the conduct of behavioral intervention research. Foremost among them is that behavioral and health problems are complex and changing behaviors is tough; thus, this form of research can be as well. Advancing an intervention can be costly, recruitment is effortful and time consuming, the conduct of interventions (treatment and control groups) requires adequate staffing and standardization, follow-up assessments necessitate resources, and testing of protocols evolves over time. As grant dollars in most countries, including the National Institutes of Health in the United States, favor basic research and then moving findings to clinical applications (referred to as T2 research), behavioral intervention research does not currently command a significant proportion of the research dollars of its respective institutes and centers. Furthermore, budget limitations often prohibit researchers from addressing some of the most important issues concerning a behavioral intervention such as determining whether outcomes are maintained over time, whether booster sessions are required to enhance treatment receipt, its cost and cost benefit, or the relationship between subjective and objective outcome measurement points such as biomarkers.
A second challenge is related to the time scale for behavioral intervention research. Designing, evaluating, and then implementing an intervention in a practice setting can take a long time from inception of an initial intervention idea to the demonstration of its efficacy, effectiveness, and evaluation of its implementation potential (discussed in further detail in Chapter 2). Many doctoral and postdoctoral students are dissuaded from pursuing behavioral intervention research because of this complexity, the perception that it delays professional advancement, and that testing may need to evolve over a relatively long time frame, preventing productivity.
A third challenge is that owing to the complexity and multifaceted aspects of behavioral intervention research, developing effective intervention approaches typically requires multidisciplinary research teams in order to enable a complete understanding of the issues at hand. Such collaborations add another layer of complexity to this form of research as researchers from diverse disciplines typically have distinct languages, methodological approaches, and unique perspectives that may initially be challenging to understand and integrate. For example, the development of a technology-based intervention for family caregivers requires combining the expertise of scientists in behavioral sciences and family caregiving with the expertise in engineering and computer science. A team science approach still remains elusive to most researchers and is not fully celebrated and appropriately rewarded in academic institutions in the form of promotions, recognition, and time and space. This prevents moving forward with behavioral intervention work in novel and potentially more effective directions. This, combined with the need to involve end users and stakeholders, adds more complexities to the research endeavor and can also tax the expertise of the originator(s) of the intervention.
A related point is the need to bring individuals from diverse backgrounds together to derive a shared language and understanding of the issues and participate in joint problem solving in advancing a particular approach. Although this can be challenging, involvement of diverse disciplines and backgrounds is also exciting and can yield breakthroughs in approaches.
Another challenge is that the field is often stymied by the lack of adequate outlets for reporting the nuances of behavioral intervention studies. For example, the CONSORT guideline that is widely used in the reporting of trials does not address certain elements of high relevance to behavioral intervention research such as the theory base and fidelity plan used and how adherence affects outcomes (Schulz, Altman, & Moher, 2010). Many medical journals have significant word limitations and are typically uninterested in how theory drives the intervention and links to the outcomes. Few journals allow space for articles to fully detail an intervention and its delivery characteristics so that it can be adequately replicated. Similarly, access to treatment manuals may not be readily available or granted by investigators, and there are no agreed upon sets of criteria for developing such manuals. The focus on reporting positive outcomes in peer-reviewed journals is upheld to the exclusion of understanding why and how a particular outcome may or may not have been achieved. Knowing that an intervention is not effective for a targeted population can be as informative as understanding what does work, and can prevent the duplication of such unsuccessful intervention approaches.
One more challenge is that little is known about some of the fundamental practices of this form of research. There is limited empirical evidence, for example, as to: which blinding or masking techniques of research staff and study participants are most effective and for which types of interventions; what types of control groups are appropriate and when; what the best practices are for ethically consenting vulnerable populations; which recruitment/retention strategies and types of interventions work best for diverse populations; and which fidelity measures are most useful across different interventions. Documentation and evaluation of specific methodologies for use in each aspect of the conduct of behavioral intervention research is very much needed. Furthermore, there is significant conceptual confusion as to the steps or processes for advancing interventions. Funders, researchers, journals, editors, and reviewers all employ different terminology, definitions, and usages for concepts such as the pipeline, translation, implementation, diffusion, dissemination, fidelity, and so forth. Conceptual misuses and confusion cloud or muddy efforts and impede working toward general consensus as to key terms and methodologies for evaluating and advancing interventions.
Furthermore, the health care landscape and population demographics are changing dramatically. There is an unprecedented need for new research designs, methodologies, procedures, and intervention approaches. Treatments that work today may not be as effective in the future for aging cohorts. For example, the delivery of health care for many conditions is moving away from traditional clinical settings to nontraditional settings such as the home. Patients and caregivers are being asked to perform complex care tasks (e.g., attending to wound care or tube feeding) and use more medical technologies (e.g., infusion systems or blood pressure and heart rate monitoring devices) (Reinhard, Samis, & Levine, 2014). This, in turn, requires the development of intervention strategies to help ensure that patients and caregivers are able to deliver care protocols as intended and that are adhered to over time. Further, the increase in populations with special needs, such as the “oldest old,” individuals aging with developmental and other forms of disability, long-distance caregivers, and individuals without family support, to name only a few, requires understanding the types of interventions that may benefit distinct and highly diverse groups and the approaches that are optimal.