Conducting Translational Research in the Context of Patient Care

Abstract

Although much has been written on the importance of translational research for bridging the continuum of basic science to clinical practice, few authors have described how such work can be carried out practically when working with patient populations in the context of ongoing clinical service delivery, where the priorities for patient care can sometimes conflict with the methods and goals of translational research. In this article, we explore some of the considerations for conducting this type of work while balancing clinical responsibilities that ensure high-quality patient care. We also discuss strategies we have found to jointly facilitate translational research and improve routine, clinical service delivery. A primary goal of this article is to encourage others working in applied settings to contribute to the increasingly important role that translational research plays in our science and practice by helping to better characterize and potentially lessen or remove barriers that may have impeded such investigations in the past.

A Gary Larson comic strip from his beloved The Far Side series depicts a herd of sheep grazing on a hillside with a single tree nestled on another small hill in the background. In the middle of the herd, one sheep stands upright with its front hooves outstretched and mouth open as though speaking to the others. A caption below the illustration reads: “Wait! Wait! Listen to me! . . . We don’t HAVE to be just sheep!”

The comic described above may resonate with fellow behavior analysts who conduct clinical research. We keep one foot firmly in the clinic, working hard to promote meaningful and durable behavior change, and the other in the laboratory, designed to further improve our understanding of the behavioral phenomena we treat clinically. For many of us, our clinical practice is our laboratory, and our laboratory is our clinical practice. We routinely shift between the roles of clinician and researcher, and what others may not readily see is that we take on this dual role (sometimes called scientist–practitioner) to improve the lives of the patients and families we serve, both in the immediate and more distant future through clinical practice and research, respectively.

When the work of clinical researchers skews more basic science than applied practice, and when the work of basic scientists is guided by the issues that confront practitioners, we call this translational research. Translational research refers to studies that bridge the basic–applied continuum and incorporate aspects of each to answer fundamental questions about behavior and its controlling variables as they relate to solving applied problems (see Critchfield, 2011; Sidman, 2011; Vollmer, 2011, for elaboration, proposed categorization, and discussion of the importance of translational work).

Translational research represents a fertile cross-pollination of literatures, perspectives, and perhaps most important, goals. Like the basic scientist, translational researchers desire a deeper understanding of behavior and its controlling variables. However, like the applied researcher, translational researchers remain fixated on solving applied problems. Straddling the basic–applied continuum often means that translational researchers, although having colleagues across the spectrum of behavior analysis, also have no place to call “home” either. Thus, much like the comic described earlier, translational researchers sometimes struggle to convey the importance of their work to basic and applied audiences alike (see Mace & Critchfield, 2010 for commentary on the longstanding divide between basic and applied research, and Mahoney et al., 2019 for updated data relevant to this concern), and relatively few behavior-analytic researchers have been successful in straddling this divide (cf. Poling & Edwards, 2011). A common hope is that basic scientists will lend their expertise to solving longstanding applied problems and that applied researchers and clinicians will be better informed by the results of basic science and, as a result, be more capable of infusing basic-science methods and findings into their work where appropriate. Perhaps a more apt caption to our comic would be, “Wait! Wait! Listen to me! . . . We don’t HAVE to be just basic scientists or applied researchers, clinicians!”

Translational Research as a Priority for Improving Patient Care

A primary goal of applied behavior analysis is to produce socially meaningful improvements in the behavior of individuals, and the integration of translational research within clinical practice can facilitate attainment of this goal. Translational research aims to bridge the gap between basic laboratory studies that are designed to acquire new knowledge and clinical studies that aim to improve clinical outcomes for patients. Although most researchers describe translational research as a unidirectional process in which the results of basic studies are transformed into novel clinical applications and then tested (Lerman, 2003), a thoroughgoing translational-research program involves bidirectional communication and translation. That is, not only can the results of basic research be used to inform clinical innovation and evaluation (so called bench-to-bedside translation; see Fisher, Greer, Mitteer, & Fuhrman, in press, for a recent example), but the results of clinical research can also lead to more meaningful and relevant basic research questions (sometimes referred to as reverse translation [see Browning & Shahan, 2021; Mitteer, Greer, Randall, Kimball, & Smith, 2021, for recent examples]).

In our clinic, we strive to produce the best possible outcomes we can with each patient and to do so in a systematic way using reliable, direct-observation measures to monitor patient progress and single-case experimental designs to introduce and withdraw intervention components strategically. We view the integration of such research-based strategies into our clinical practice as necessary for professional accountability, that is, to demonstrate empirically that the prescribed treatment caused the observed change in behavior and to rule out extraneous causes (Kazdin, 2011). This systematic approach allows us to construct effective, efficient, and individualized treatments by including treatment components that produce demonstrable improvements in the patient’s behavior and by systematically eliminating unnecessary components (see Fisher et al., 2016, and Greer, Mitteer, Briggs, Fisher, & Sodawasser, 2020, for examples of this latter work as applied to functional analysis).

This systematic approach to clinical care also facilitates applied and translational research. It facilitates applied research through the accumulation of systematic data from controlled case studies. We call this patient-oriented research, and this approach differs from what we call programmatic research (e.g., grant-funded investigations that may span multiple years). The aim of patient-oriented research is to provide the best possible outcomes one can for both the current patient and for future patients. As indicated above, systematic data collection and rigorous experimental designs facilitate the development of an effective treatment for the current patient, and it also provides benefit to future patients with the same or a highly similar clinical presentation (see Hagopian, 2020, for related discussion). Finally, when we accumulate systematic data on a series of similar patients, we often have data that are worth sharing with our peers through publications in journals like the Journal of Applied Behavior Analysis (e.g., Briggs et al., 2018; Fisher et al., 1997; Fisher et al., 2015; Greer, Fisher, Saini, Owen, & Jones, 2016; Haney, Greer, Mitteer, & Randall, 2022; Mitteer, Greer, Randall, & Haney, 2022; Owen et al., 2020; Tiger et al., 2009).

Translational research is facilitated using research-based methods in clinical practice because the accumulation of systematic data provides not only information about our successes, but also our failures, not just what we currently know about a clinical phenomenon but also what we need to learn to improve our clinical outcomes. Information on what we need to learn to improve our clinical services leads to more focused and relevant research questions for basic researchers to consider. For example, we have been working collaboratively on translational research on treatment relapse with a research laboratory at Utah State University headed by Timothy Shahan, who conducts basic and translational research with nonhuman animals and is a leading expert on relapse. This collaboration has been highly productive and has resulted in multiple grant-funded investigations and peer-reviewed articles aimed at improving clinical practice (Fisher et al., under review; Fuhrman, Fisher, Greer, Shahan, & Craig, 2021; Greer, Shahan, Fisher, Mitteer, & Fuhrman, in press; Greer & Shahan, 2019; Shahan & Greer, 2021). However, an equally important outcome of collaborating with Tim Shahan and his team is that it has challenged us to think differently about what we see in the clinic and to consider novel solutions to longstanding problems relevant to the assessment and treatment of problem behavior. The fruits of this collaborative work and that of other research teams are only beginning to be realized at the level of changed clinical standards of care. As one example, the Behavior Analyst Certification Board (BACB) recently announced that beginning in 2025, the Task List would include an item relevant to treatment relapse: H.5 Plan for and attempt to mitigate possible relapse of the target behavior.

This partnership has also pushed us to more closely examine the behavioral processes believed to account for problematic patterns of responding in the clinic and to develop new quantitative models of behavior that describe such phenomena. For example, we recently discussed similarities between our translational-research findings on resurgence of problem behavior and our clinical-research findings on and experience with extinction bursts. In our clinical data, we found evidence suggesting that similar behavioral processes may govern both phenomena, which led Shahan (under review) to refine and expand a quantitative theory of resurgence, called Resurgence as Choice (RaC; Shahan et al., 2020; Shahan & Craig, 2017), to also account for the extinction burst. Shahan relabeled this expanded theory the Temporally Weighted Matching Law (TWML), which is an expanded version of Herrnstein’s (1970) single-operant matching equation in which relative-response allocation is a function not only of relative reinforcement rates currently in effect but also of the individual’s reinforcement history as determined by the Temporal Weighting Rule (Devenport & Devenport, 1994). Thus, bidirectional translational research can improve clinical interventions and lead to the refinement of theories of behavior like the TWML.

In the sections below, we provide a primer on some of the aspects we believe to be important for conducting translational research successfully in the context of patient care. Our target audience is aspiring translational researchers; however, applied researchers and basic scientists alike may benefit from a working understanding of the variables that come to control the behavior of translational researchers working in clinical settings. Many of the aspects we believe to be important for conducting translational research in the context of patient care are not unique to this goal, but they are nevertheless vital to its success.

Staff Credentialing and Supervision

Individuals who provide direct patient care and case management require specialized training and expertise. We recommend that direct-care staff hold the BACB’s Registered Behavior Technician (RBT) credential because it ensures they meet basic eligibility requirements for working with patients (e.g., cleared background check, completed basic training) and adhere to ethical standards and ongoing supervision requirements. Individuals who manage and help oversee translational research projects should be masters- or doctoral-level Board Certified Behavior Analysts (BCBAs or BCBA-Ds) who have experience with the responsible conduct of research in a patient setting. Individuals who develop and lead translational research projects should be BCBA-Ds with clinical expertise and relevant research knowledge. We also recommend that all staff involved in developing or conducting research protocols complete annual training on the responsible conduct of research (e.g., protection of human subjects, conflicts of interest).

Before beginning a research protocol, behavior analysts leading the project will need to consider other types of training that direct-care staff may need to implement the protocol safely and accurately. For example, if working with patients with dangerous behavior such as severe aggression or self-injury, direct-care staff will require training on how to block and respond to instances of such behavior in the context of the research protocol, which may involve procedures that differ from how staff would typically respond. In most cases, practitioners will also need to conduct formal training on the specific research procedures in the protocol to ensure high-integrity implementation. To monitor procedural integrity and data accuracy on an ongoing basis, we recommend systematic collection of interobserver agreement and procedural-integrity data across phases and conditions. Supervisors should review these data (i.e., agreement amounts and coefficients) regularly—in an ideal situation, daily or weekly—so that they can address issues in data collection or protocol implementation in a timely fashion.

It is also important to maintain a robust case-supervision structure so that the behavior analysts leading the research project observe direct-care staff frequently while they implement the research protocols and meet with them often to address any questions or concerns that arise. We often find close supervision and frequent observation of staff to be an especially important component for maintaining effective and responsible research conduct, as well as clinical-service delivery. It is also beneficial to have group research meetings where colleagues discuss planned and ongoing research projects. Regular research meetings can provide practitioners with insight and informal peer review that can help ensure the completion of high-quality studies.

Physical Plant, Equipment, and Resources

Clinical researchers should also consider what resources and equipment they will need before beginning a translational investigation with patients. Although the needed resources and equipment will often overlap with those required for patient care, it is important to anticipate and plan for any additional resources needed to ensure the success of the project. For example, when conducting research with patients with problem behavior, practitioners should be prepared with various forms of protective equipment for staff and patients to minimize the risk of injury during the research procedures (e.g., probing sessions of extinction), which may result in temporarily higher or more dangerous levels of problem behavior than would otherwise be likely during routine, clinical care. Examples of protective equipment to consider include helmets, arm guards, shin guards, and large and small striking pads (see Fisher et al., 2013, and Fuhrman, Greer, & Fisher, 2021, for more information on such equipment).

We also recommend that clinicians ensure that the physical plant in which the translational investigation is to be conducted is both safe for those involved, given the procedures outlined in the research protocol, and amenable to answering the experimental question. That is, some applied settings (e.g., busy and distracting classrooms) may not lend themselves well to eliminating important extraneous variables (e.g., interactions with peers) that may affect responding, and thus the results of the translational study. Such threats to internal validity should be minimized to the extent possible, especially when early in the translation of basic-research findings. However, it is important to note that such environments may be an ideal setting for evaluating procedures that have shown considerable promise at earlier stages of translation (e.g., with clinical populations in more controlled settings). Ensuring that such findings extend to less-well-controlled settings and with nonclinical populations is a natural step in the process of translating promising procedures. Thus, clinicians should consider whether the setting in which they work can facilitate an unbiased answer to the experimental question posed by the translational-research project, and if not, whether an alternative setting would be more appropriate given the current state of the translation of the procedure or approach in question.

Staying Current with Relevant Research

Most behavior analysts will agree that it is important to be in tune with the latest research in one’s area of expertise, in particular as it relates to providing effective patient care. Thus, Section 1.06 of the Ethics Code for Behavior Analysts (BACB, 2020) states that behavior analysts should maintain their competence by reading relevant literature, attending conferences, and engaging in other professional-development activities so that recipients of behavior-analytic services receive best-practice care. Carr and Briggs (2010) and Briggs and Mitteer (2021) detail some of the potential skill deficits and financial barriers that pose challenges to staying current with the literature, as well as solutions for overcoming such barriers.

However, even after removing such barriers, behavior analysts may find it difficult to consume new research due to their many other competing responsibilities. Our clinical staff have an amalgamation of client responsibilities (e.g., managing direct-patient care, collaborating with caregivers and providers), research obligations (e.g., overseeing grant progress, submitting and updating protocols for the institutional review board [IRB] and the ClinicalTrials.gov database), and teaching loads (e.g., instructing graduate students, mentoring BCBA trainees, supervising theses and dissertations). Although we are fortunate to have minimal barriers for contacting and accessing research, finding the time to consume and integrate new research is certainly a challenge. We rely on three general strategies for maintaining competency in our clinic.

The first strategy is scheduling a weekly readings group in which we review both recent and seminal articles to spark new research ideas while ensuring that more novice members of the group establish the foundation from which to be able to interpret more advanced content. We discuss readings that span the basic–applied continuum to encourage broader perspectives about our subject matter and creative solutions to the problems we encounter in the clinic. With a diverse set of group members comprised of experienced doctoral-level researchers to new RBTs, a readings group provides a format to discuss theoretical and clinical implications of the assigned readings while offering opportunities for staff and trainee education. For example, leading up to our research on Behavioral Momentum Theory (BMT; Fisher et al., 2019; Fisher et al., 2020; Fisher, Greer, Craig, et al., 2018; Fisher, Greer, Fuhrman, Saini, & Simmons, et al., 2018; Fuhrman et al., 2016; Greer, Fisher, Retzlaff, & Fuhrman, 2020; Greer, Fisher, Romani, & Saini, 2016), we reviewed basic, translational, and applied studies on BMT across many weeks. This had the benefit of generating interesting study ideas, identifying potential treatment refinements based on the equations of BMT, and educating trainees and staff on basic-research methods and findings with which they were previously unfamiliar. Meeting monthly to discuss assigned readings can also promote learning (Parsons & Reid, 2011) if more frequent meetings are impractical for your program.

A second strategy we incorporate is consulting the literature when encountering novel or persistent challenges in the clinic, and when this exercise does not yield a satisfactory solution to that problem, designing our own study to address it. For example, we knew from our work in the clinic that treatment effects sometimes fail to generalize to new implementers and settings (see Muething et al., 2020, and Mitteer et al., 2022, for prevalence data on renewal in the clinic). Therefore, our group spent several weeks reviewing the basic research on operant renewal (i.e., recurred behavior following a change in context). However, from our reading of the literature, we identified that a gap existed between how basic researchers typically study renewal and the conditions under which renewal tends to occur in the clinic. Basic researchers studying renewal often program extinction as the sole treatment component, whereas clinicians rarely use extinction alone when treating problem behavior (cf. Lerman et al., 1999; Lerman & Iwata, 1995). Therefore, our group designed translational experiments that evaluated renewal when implementing interventions that combine DRA with extinction, akin to common treatment approaches for problem behavior, relative to extinction-only treatments. In one study (Kimball et al., 2020), we found that renewal tended to be more common and more robust with extinction-only treatments relative to our usual treatments that also arrange alternative reinforcement (see Brown et al., 2020, for a related investigation on resurgence following treatments with and without extinction arranged for destructive behavior).

A final strategy we use is encouraging team members to stay involved in professional-development opportunities that promote contact with current research. Our faculty members serve on editorial boards and as ad-hoc reviewers for behavior-analytic journals, which compels them to not only read recently conducted research but also to consult the literature relevant to each submission. Likewise, we incorporate both seasoned and novice staff into presentations, manuscript writing, and grant-submission activities to promote critical thinking about our own data and encourage staff to consult the extant literature during each of these activities. Taken together, our organization attempts to integrate several opportunities for remaining in touch with the literature, and our clinical care often creates contingencies for consulting and expanding upon published research.

Which Research Questions to Ask

A variety of factors should be considered when selecting a research question, including current gaps in the literature, the theoretical relevance of the question, the clinical applicability of the question, and a variety of practical considerations that affect the feasibility of completing a methodologically sound investigation to address the question in a clinical setting. It would be ideal if one looked for research questions that address several—or all—of these factors (i.e., research questions that are theoretically and clinically relevant, that address an important gap in the literature, and that can be feasibly explored in your facility and by your team).

We have found it useful to consider well-established quantitative models of behavior for their applicability to the assessment and treatment of problem behavior. According to Box (1976), all quantitative models are wrong, but some are useful. That is, they are less than perfect simplifications and representations of the phenomena they model, but they nevertheless may clarify important characteristics of the phenomena and lead to novel approaches to clinical problems related to those phenomena.

We have also tried to focus our translational research on areas where multiple theories of behavior make the same or highly similar predictions about the clinical phenomena we study. For example, basic scientists are often interested in looking for behavioral phenomena where two theories make divergent predictions to compare the accuracy of the two theories. By contrast, we believe that clinicians interested in improving their assessments and treatments may have more success when their translational studies are informed by a confluence of empirical findings or theoretical predictions. For example, we have used the predictions of BMT and the TWML to develop clinical procedures for mitigating resurgence of problem behavior, a type of relapse that occurs following successful treatment with differential reinforcement due to a worsening of reinforcement conditions. We have had the most success in mitigating resurgence by using the discriminative stimuli of a multiple schedule to signal periods during which neither problem behavior nor the alternative response produces reinforcement (e.g., Fisher et al., 2020; Fuhrman et al., 2016), and both BMT and the TWML rely on the concept of generalization decrement to predict that this clinical procedure should lessen resurgence. According to the generalization-decrement hypothesis, the similarity between the treatment phase and resurgence test (e.g., discriminability of reinforcer-rate and contextual-stimulus changes) predicts the degree to which responding will be affected by the termination of treatment (cf. Nevin et al., 2001). Therefore, we should observe greater generalization from a treatment phase to a resurgence test when the stimulus conditions of the two are highly similar (e.g., by continuing to present the S^Δ from the treatment phase during the resurgence test to signal the absence of reinforcement), and generalization should be lessened as the stimulus conditions become increasingly dissimilar (e.g., by terminating both the contingency maintaining alternative behavior and the S^Δ correlated with reinforcer unavailability). Thus, we posit that effective clinical translations of phenomena studied in basic research may be more probable when the clinician selects research questions founded on well-established behavioral principles common to multiple theoretical formulations, like generalization decrement.

One can identify gaps in the literature by reading recent empirical investigations or systematic reviews on a given topic (e.g., Fisher et al., under review; Kimball et al., under review), because such articles commonly discuss theoretical significance and potential directions for future research. In addition, we suggest that clinicians interested in initiating their first translational research study start by identifying a research question that is relevant to the population of patients they see the most and one for which sufficient prior research suggests it will be safe to evaluate within the context selected (e.g., human-operant arrangement, clinical evaluation). Clinicians know the needs of their patients. They know what assessment and treatment procedures are useful to their patients and which ones need improvement.

Strategies for Addressing Identified Research Questions

If you decide to recruit patients to participate in a translational investigation, it is particularly important to balance patient care with the goals of the research. That is, although research and good clinical practice often go hand in hand (e.g., accurate data collection, assessing behavioral function, systematically introducing treatment across settings; Fisher, Fuhrman, et al., in press; Hayes, 1981), there are times when the needs of the patient diverge from the priorities of the research investigation. That is, the clinician may confront situations in which the research protocol specifies one course of action (e.g., conduct a resurgence test by terminating all programmed reinforcers for at least five sessions) and their best clinical judgement indicates a different plan (e.g., discontinue the extinction challenge if the participant’s behavior intensifies to a point where the safety of the patient or others is in jeopardy). Whenever possible, clinical researchers should anticipate such potential conflicts and develop a-priori criteria that specify the conditions that should lead to continuation of the research protocol as planned and when to withdraw the participant from the study and proceed to an alternative intervention. In addition, gathering input from colleagues who are not providing care to the individual or collaborating on the research project can be helpful when the clinical researcher faces such decisions. In general, erring on the side of caution and participant safety is always best practice.

It also is important for the clinical researcher to weigh the potential risks and benefits associated with participating in the study before initiating a translational research project, especially when the project is exploratory or when prior research has been limited or has produced conflicting results. In general, translational studies may not provide much direct or immediate benefit to participants, and therefore minimizing the risks associated with participation in such studies is essential. If the study poses more than minimal risk, then government regulation and IRBs usually require that participation in the study has the potential to provide direct benefit to the individual participant. According to federal regulation §46.102(j), “Minimal risk means that the probability and magnitude of harm or discomfort anticipated in the research are not greater in and of themselves than those ordinarily encountered in daily life or during the performance of routine physical or psychological examinations or tests.” When we conduct research (clinical or translational) with individuals who display severe destructive behavior (e.g., aggression, self-injurious behavior), our IRB regards the research as constituting greater than minimal risk, even if our research procedures do not increase risks beyond those the same patient would experience during standard-of-care treatment. That is, the destructive behavior itself constitutes a risk greater than that typically experienced by similar individuals who do not display such dangerous behavior.

One approach that behavior analysts conducting translational research have used to ensure that the investigation does not pose more than minimal risk to the participant is to select participants who do not display significant problem behavior and to use innocuous target responses (e.g., hitting a strike pad, dropping a ball into an object-permanence box) as analogues of problem behavior (e.g., Kimball et al., 2020; Retzlaff et al., 2020). As discussed briefly above, Kimball et al. examined the effects of DRA on operant renewal with neurotypical children and ones diagnosed with autism, but who did not display severe destructive behavior, as participants. The investigators used the dropping of one ball in an object permanence box as the target response and the dropping of a second ball in another object-permanence box as the alternative response. The results from two experiments showed that adding differential reinforcement to extinction decreased the magnitude and persistence of target responding during a renewal test relative to treatment with extinction alone.

Researchers have also found success conducting human-operant experiments with participants who do not resemble those seen in clinic and with using arbitrary responses sometimes far removed in topography from the types of behavior treated clinically. Shifting the project to a computer-based task and/or enrolling nonclinical adult participants can be a useful approach when the research question does not lend itself well to being evaluated using a single-case experimental design (e.g., studies on history effects) or when the number of participants necessary to address the research question exceeds the number of participants available to the researcher (e.g., Fuhrman, Fisher, et al., 2021; Smith & Greer, 2022). For example, Smith and Greer (2022) recently enrolled 64 adult participants using Amazon’s Mechanical Turk (MTurk) recruitment platform for a computer-based study that examined how different durations of exposure to baseline and treatment contingencies affect resurgence. Participants were sorted across four groups that varied baseline and treatment duration (i.e., short baseline, short treatment; short baseline, long treatment; long baseline, short treatment; long baseline, long treatment). Using this approach, the researchers obtained orderly data across the groups of participants with the results being described well by the quantitative predictions of RaC. This same study conducted with patients referred for the treatment of problem behavior would be unethical and using a nonclinical population recruited via other means would have been resource intensive and likely infeasible.

The primary limitation of conducting such translational investigations with participants who do not display the clinical problem of interest or substituting innocuous target responses for clinically important ones is that the generality of the findings remains uncertain. For example, in the Kimball et al. (2020) study, it remains to be determined whether the effects of differential reinforcement observed with the object-permanence task under laboratory conditions with children who do not display severe destructive behavior would operate in a similar manner in a more typical environment when individuals with autism and other disabilities display severe destructive behavior. Nevertheless, when the empirical findings on a given phenomenon are scant, conducting translational investigations with nonclinical populations and with nonclinical target responses are often necessary to accumulate preliminary results. The procedures and results of initial translational investigations can then inform subsequent investigations conducted with clinical populations and socially significant responses so that these latter studies are conducted only when indicated by prior findings and only under the safest possible conditions.

Another issue that clinicians must consider when integrating translational research with clinical practice is how to determine what services should be billed to insurance companies or Medicaid. In general, if a given procedure is standard of care for a patient, then it is reasonable to bill for that procedure. For example, if the appropriate standard of care for a given patient and the translational-research protocol that patient is enrolled in both call for the completion of a functional analysis of aggression, then it is appropriate to bill the insurance company for that analysis. However, if the research protocol calls for a procedure that is not clinically indicated as standard of care (e.g., delivering reinforcement following problem behavior rather than appropriate behavior during treatment to evaluate the effects of procedural-integrity errors), then it would be inappropriate to bill the insurance company for that procedure. When the clinician is uncertain whether a given procedure qualifies as standard of care, it can be helpful to seek input from peers, a billing expert, or the IRB that provides oversight of the project.

Funding and Dissemination Considerations

Translational research conducted in the context of patient care is more likely to be viewed favorably by funding agencies that appreciate both the science-advancing and outcome-driven goals of translational research. The National Institute of Child Health and Human Development (NICHD) of the National Institutes of Health (NIH) has a history of supporting such translational research through R01, R03, and R21 mechanisms. The National Institute of Mental Health (NIMH), also of the NIH, has a similar history of funding such research. However, one difficulty behavior analysts have had with obtaining these sorts of grants in the past is that grant review panels that evaluate submitted proposals are infrequently filled with fellow behavior analysts. This presents a significant barrier for the funding of behavior-analytic research. One strategy for addressing this problem is to explicitly request reviewer expertise in the research area in which you are proposing to study. For example, in the cover letter for our grant applications, we explicitly request reviewers with expertise in operant relapse models of behavior and in severe problem behavior. This request is typically honored by the scientific review officer, and that individual then invites such experts onto the review panel in a nonmember capacity. Serving as a temporary member of an NIH study section is an excellent opportunity for learning more about grants and what reviewers seek in a competitive application. One would imagine that such an experience also makes one a better grant writer and more apt to obtain extramural funding.

Deciding which publication outlet to submit your completed translational research to is another difficult decision for those working in the context of patient care. We have had success publishing our translational research in both basic and applied outlets. However, we are often intentional with which outlet type to pursue prior to writing up our results, and thus, the submitted manuscript tends to be tailored to an intended audience. This likely improves the chances of a favorable decision regarding publication. There are no hard and fast rules with deciding on one publication outlet over another, but the perceived reception by the typical readership of the journal in question can be a good barometer. Considering the types of research that outlet has published in the past in relation to your manuscript can certainly help, too. Understanding the types of research valued by the editorial board, especially by the associate editors, can also help when making the decision. Again, writing in the language of the intended audience is a must.

Collaboration across the Basic–Applied Continuum

A question we have been asked on more than one occasion is how, despite our applied backgrounds and training experiences, our collaborative relationship developed with a prominent basic researcher (Tim Shahan). The answer is that we followed the advice laid out in the sections above for a few years on our own. Our then-current projects were inspired by what we read from the (mostly basic) literature on relapse and our own experiences with treating problem behavior. Dave Wacker, Bud Mace, and other pioneers in translational research on the applied side had already laid the groundwork for questioning the persistence of common treatments for problem behavior and for testing clinically viable solutions for improving treatment durability. We were particularly influenced by Nevin and Shahan’s (2011) article on BMT published in the Journal of Applied Behavior Analysis because it outlined a set of explicit predictions regarding what we might do differently in the clinic to mitigate resurgence. We read their article, built some of their models in Excel, consulted one another when we had questions, and simulated multiple experiments that we eventually conducted. However, as time progressed, and more data appeared in the basic literature on the shortcomings of BMT as a model of resurgence (cf. Nevin et al., 2017), we began to grow uneasy about what these findings meant for our growing translational research program. The publication of Shahan and Craig’s (2017) article on RaC was a watershed moment for us. We pored over their article, trying to understand as much as we could about this new theory of relapse. Fortunately for us, we failed in building their model in Excel on our own. Several of us felt that we could not fully understand RaC without building it ourselves.

It was at this point that we did what others might not have: we contacted Tim Shahan and asked for his help. We had some grant funds available and offered to host him for a couple days so that he could provide an in-person tutorial on RaC. Tim agreed and was genuinely surprised by our level of interest in understanding RaC. That brief visit led to the successful research collaboration between our teams that continues today. Andy Craig, a coauthor of RaC, even completed a postdoctoral fellowship with us as but one example of the integrated research program that came to be.

Contacting Tim Shahan was one of the best decisions our team has made to advance our translational research program. Although some of us were intimidated at the time to do so, it has paid dividends and enhanced our work in countless ways. Close contact with a researcher on the opposite end of the basic–applied continuum has the added benefit of overcoming language barriers that can make it difficult to enter a new area of research and it allows for the uncovering of nuanced, but potentially important, differences in experimental approach. Others would do well to follow a similar path. Identifying someone or a few individuals with common research interests and taking the initiative to contact them can be similarly rewarding.

The Promise of Translational Research

Translational research rarely leads to immediate improvements in practice or theory refinement, yet it enjoys a special role in tying together basic science and applied practice. We have found that clinicians who appreciate basic-research methods and findings tend to be more grounded conceptually in the principles of behavior and more flexible in their approach to practice. Clinicians with a working understanding of quantitative models of behavior have at their disposal the culmination of often decades of basic-research findings, presented as elegant expressions and capable of predicting the conditions under which common clinical procedures are more or less likely to be effective. Likewise, basic scientists who appreciate the challenges that clinicians routinely experience tend to be more apt to structure their basic-research methods in ways that resemble standards of care in the clinic, which allows for clearer detection of procedures ripe for translation and potentially capable of improving clinical practice in meaningful ways. Finally, straddling the applied and basic literatures can lead to new ways of thinking about and solving applied problems, and translational research is primed to extend the reach of basic research while occasionally nudging its direction by focusing the powerful lens of the experimental analysis of behavior onto those variables most amenable to manipulation in the clinic.

Declarations

Conflicts of Interest

The authors have no competing interests to disclose related to this work.