Translation, Adaptation, and Synthesis of Interventions For Persons Living With HIV

As HIV-infected persons live longer and healthier lives and continue to be at risk for transmitting HIV, prevention of transmission among persons living with HIV/AIDS—often called “prevention with positives” (PWP)—is becoming an increasingly important focus for study. A number of recent studies, some included in this issue, have shown how such prevention efforts can be implemented and sustained.1–3 In this commentary, we discuss 3 areas relevant to dissemination of PWP research: (1) the current process by which interventions are transferred, or as we emphasize, translated into practice; (2) the importance of efficacy and program effectiveness studies; and (3) the utility of synthesizing core elements from comparable interventions to facilitate adaptations in practice settings. Previously, these issues have been extensively discussed for prevention among high-risk HIV-negative persons. We believe that the lessons learned from these experiences apply equally to PWP interventions as they are being developed and disseminated, however. TECHNOLOGY TRANSFER AND TRANSLATION Technology transfer is the process by which interventions with evidence of efficacy are disseminated for implementation and continued use in program application, often with additional provision of training and technical assistance.2,3 In 1996, the Presidential Advisory Committee on HIV/AIDS recommended that federally funded research on HIV prevention should include specific mechanisms for rapid dissemination of findings.4 The Centers for Disease Control and Prevention (CDC) simultaneously initiated the HIV/AIDS Prevention Research Synthesis Project, which sought to inventory prevention research studies and analyze their efficacy.5 Using rigorous evaluation criteria, the CDC identified 24 interventions with evidence of efficacy, and these interventions are listed and disseminated via the compendium of effective HIV interventions with evidence of effectiveness (CDC compendium).6 Next, the CDC supported the translation and transfer of the interventions for use in public health and community settings.1 The rapid transfer of HIV prevention technology continues to be heavily promoted as an essential component of the national HIV prevention plan.7 The CDC also supports the widespread dissemination of model interventions through the network of sexually transmitted disease (STD)/HIV prevention training centers (PTCs). For example, the community-level intervention developed in the AIDS Community Demonstration Projects (ACDPs)8 was packaged as “Community Promise,” and training curricula are being piloted by the PTCs with health departments and community agencies. Likewise, an intervention package accompanied by training was developed based on the VOICES/VOCES video-based intervention for use in STD clinics.9 Several more HIV prevention interventions are being disseminated through training programs in the United States. The CDC has also attempted to diffuse HIV prevention interventions by encouraging community-based organizations (CBOs) to implement the CDC compendium interventions in its most recent HIV prevention request for applications.10 The technology transfer process using these mechanisms is still relatively new, however, and some critical questions remain unanswered. Although rigorous clinical trials can evaluate intervention efficacy, the true test of large-scale effectiveness in communities is their adoption and measurable outcomes in “real-world” settings.7 Until recently, little attention has been directed to helping AIDS prevention providers successfully adopt evidence-based interventions11 or increase the scientific understanding of the organizational factors that affect implementation in community settings.12 INTERVENTION EFFICACY DOES NOT GUARANTEE COMMUNITY EFFECTIVENESS There are implicit assumptions that interventions that work well in research settings can be “transferred” to program settings with relative ease, and it seems that most literature to date reflects the researchers’ perspectives of the benefits of this process.12 Efficacy studies have paid little attention to how interventions could be realistically carried out in program settings, however,13,14 and there is a paucity of real-world effectiveness studies. Glasgow et al14 have made strong arguments that the absence of such research stymies the adoption of behavioral interventions into public health practice. As Elford and Hart15 write: “Indeed, translating evidence into practice constitutes a research program in itself which warrants significant resource allocation.” Several factors that should be the focus of such research attention and resource allocation are described below. BARRIERS TO TRANSLATION: GAPS IN BASIC SKILLS AND ABILITIES AND LACK OF RESOURCES For more than 2 decades, public health and community-based agencies have received federal funding to conduct individual-, group-, and community-level HIV prevention interventions. For funding recipients to carry out these interventions successfully, qualifications include but are not limited to: knowledge of factors influencing risk behaviors, ability to conduct client-centered sexual and drug use risk reduction counseling, comfort in addressing serostatus disclosure with sex or needle-sharing partners, and group facilitation skills. In our experience, the capacity to implement even the most basic of these intervention skills is often missing, even though interest and motivation is high in many agencies. Prevention providers who lack these skills may alter interventions in ways that remove or change core elements and unintentionally jeopardize intervention effectiveness. Unfortunately, although technology transfer models have been developed to guide agencies toward effective adoption of science-based interventions,16 many sites lack the time or resources to use the guidelines effectively. Rigorous HIV prevention trials are expensive and labor-intensive, and resources available for primary research studies are significantly greater than what is typically budgeted for community agencies attempting to adopt an HIV intervention. Research budgets include resources for training and quality assurance to support the delivery of the intervention precisely as designed.17 How much does a lower funding level affect fidelity and implementation at the program level? HIV prevention providers are encouraged to implement interventions with high fidelity but may be provided little guidance on how to best use limited resources and how these allocation decisions may influence effectiveness. Should they apportion more of their allotment to recruitment of appropriate populations, training on use of study protocols, payment of any incentives to participate, and/or quality assurance strategies? Successful translation, if defined as precise implementation of the original research intervention, seems like a formidable challenge in this scenario. WHEN ARE ADAPTATIONS SO EXTENSIVE THAT THE ORIGINAL INTERVENTION IS “LOST IN TRANSLATION”? The CDC recognizes that programs adapt evidence-based interventions to meet their own needs. In response, some researchers who have received funding to package and disseminate their interventions directly to community agencies have developed guidelines such as agency readiness self-assessments and sample budgets that identify the capacities and resources needed to implement a particular intervention.2 In the introduction to the CDC compendium, the CDC acknowledges that once an intervention is adopted, its actual impact depends on how it is implemented and emphasizes the importance of achieving a balance between adaptation to suit local needs and maintaining the core elements and key characteristics that made the original intervention successful.6 Will prevention providers know how to achieve that balance or recognize at what point their adaptations have affected the core elements of a program? Our experience has been that in some settings, providers have chosen applicable pieces of a complicated research intervention and combined them with parts of an entirely different intervention that was developed for a similar high-risk group. It is likely that HIV prevention providers believe that this approach works better for them and their community, but the effectiveness is unknown. How will these new “hybrid” interventions be evaluated? Few programs have the resources to conduct outcome evaluation independently, especially because they have been told that these interventions (albeit in their original form) have already been rigorously evaluated. The fidelity versus adaptation problem increases exponentially as a well-intentioned provider seeks to implement an intervention further from its original development and efficacy studies, such as in different populations, cultures, or countries.15,18 RESEARCH FIDELITY VERSUS RESEARCH SYNTHESIS Although an imperfect and understudied science, the evolution of technology transfer or translation thus far seems to follow a logical sequence from the original research, packaging for replication, and training to support dissemination. From our perspective, however, a critical step in the process has been overlooked, namely, synthesis. For example, if one reviews the 3 community-level interventions cited in the CDC compendium, which include the ACDP model, the Popular Opinion Leader (POL) model,19 and the MPowerment model,16 one is struck by the similarity of the principles that underlie the interventions as well as the similarities of the interventions’ respective effects on the risk behaviors they targeted. Are there common intervention elements that make them “work?” If so, this knowledge might be useful, particularly to those who are attempting to adapt interventions to the local situation or to improve existing programs, especially if resources are limited. Unfortunately, studies to answer these questions have not been conducted. Consequently, the focus of technology transfer has been on the dissemination of individual models, with a strong emphasis on fidelity to the original research. Kelly et al20 state: “Service organizations that plan to implement a research-based HIV prevention intervention can do so with greatest confidence when they conduct the intervention as it was originally conducted, that is, with the same intensity, with the same activities, content, and procedure, and with settings and clients similar to those in the research trial.” RECOMMENDATIONS Increase Capacity of HIV Prevention Providers Continued investments are needed to increase capacity among HIV prevention providers that goes beyond prescribing what interventions can be used and for which target populations. Providers must understand the necessary steps required to implement an evidence-based research intervention in a program setting before implementation begins, including its complexity, core elements, and the extent to which adaptations can be made. Lack of guidance in this latter area is particularly noteworthy. Agencies must be willing to assess their organizational and programmatic capacity to implement interventions, including required resources, recruitment strategies, quality assurance guidelines, and staff qualifications, and recognize that increasingly complex interventions require staff with advanced skills. Even with highly skilled providers, many evidence-based interventions still necessitate modifications in program settings. For example, in medical care settings, where lack of available time presents an additional barrier to implementing PWP interventions, clinic managers or medical directors may consider training nonmedical staff to conduct behavioral interventions after providing brief risk reduction messages and then to conduct more comprehensive risk reduction in sessions separate from the primary care visit. Synthesize and Disseminate a “Common Elements” Guide Although programs are encouraged to implement evidence-based interventions as closely as possible, adaptations occur in varying degrees. Because many HIV prevention providers do not realize the extent to which their changes affect crucial elements of an intervention, common elements from studies that are similar in scope should be developed and disseminated to guide providers more efficiently. Brief program-friendly publications that summarize these common elements, such as the Characteristics of Reputationally Strong Programs (C-RSP) Project, should be widely disseminated.21 We would applaud efforts by academia, public health, and community stakeholders to collaborate in the development of such synthesis-based models. Improve Partnerships Among Providers, Researchers, Funders, and Affected Communities The widely used term technology transfer describes the unidirectional approach of how research is transferred down to programs but implicitly overlooks the role of providers and other partners in informing research. Contextual influences that are experienced in practice settings, which can be barriers or facilitators to program implementation, should be communicated to researchers and funders to improve intervention effectiveness and inform future research. We define this latter situation as technology exchange,22 because it emphasizes a multidirectional communication between all partners.17 Such partnerships are more apt to result in identification of health issues that are of greatest concern to communities, and therefore increase chances for success.23 We should especially continue to seek guidance from persons living with HIV in future intervention research. PWP research needs to address the heterogeneous needs of positive persons, including those influenced by race, gender, sexual orientation, age, and language.24 Studies have shown that most individuals, once aware of their HIV status, go to great efforts to prevent transmission to others.25 These persons have been and continue to be critical allies in preventing new infections. As efforts in this direction progress, the development of technology transfer/exchange can be further improved by greater participation of providers and clients at the earliest stages of research development.