This supplement was initiated after a meeting sponsored by the Center for Mental Health Research on AIDS (National Institute of Mental Health [NIMH]) and the Centers for Disease Control and Prevention (CDC) held in conjunction with the 2003 National HIV Prevention Conference in Atlanta. The goal was to discuss state-of-the-science behavioral risk reduction with HIV-positive persons and to outline immediate research needs. We invited input from scientists, federal representatives, HIV treatment and prevention service providers, and community members. To stay ahead of a rapidly evolving epidemic, multiagency and multisectoral teamwork is essential in all research phases and stakeholders should be alerted in advance of interventions in the pipeline. Therefore, participants presented findings from completed trials, studies in the field that were nearly complete, and innovative projects just underway. We emphasized inclusion of randomized controlled trials (RCTs) conducted in the United States, because the meeting had a domestic focus. The articles contained in this supplement are a substantial representation of these presentations and other recent significant findings. The studies were funded by the National Institutes of Health (NIMH and the National Institute of Drug Abuse), the CDC, and the HIV/AIDS Bureau of the Health Resources and Services Administration (HRSA), federal agencies that cooperatively support the CDC’s Serostatus Approach to Fighting the HIV Epidemic (SAFE).1
Behavioral risk reduction for persons living with HIV/AIDS needs to be improved to stem the spread of HIV. It is heartening that treatment gains are improving the duration and quality of life for HIV-positive persons and that most persons who learn their HIV-positive serostatus take measures to reduce risk for themselves and others. Consistent safer sexual behaviors are difficult to maintain over long periods, however, and recent studies document ongoing behavior conferring risk for HIV transmission and incident sexually transmitted diseases (STDs) among some individuals living with HIV/AIDS.2,3 Efficacious interventions are needed by community-based organizations (CBOs) and clinics that seek to provide evidence-based services. Development and dissemination of such programs is only 1 aim of the domestic prevention initiative, which includes reduction of barriers to early HIV diagnosis; increased access to quality medical care and treatment; and access to, utilization of, and adherence to prevention services (see commentary by Janssen and Valdeserri in this issue). Thus, the need for targeted interventions for persons living with HIV is acute as a result of evidence of increased risk and the shifts in federal prevention strategy.
The current fiscal environment requires careful discussion to triage competing priorities; there are finite resources available for research and implementation but an expanding set of recommendations for allocation. Funders, researchers, and providers must grapple with how difficult it can be to sustain behavior change because of the complexities of sexual behavior, relationships, and HIV risk reduction and must balance this understanding with practical realities that call for feasible and cost-effective models for change, that is, for interventions to be less expensive and often briefer to implement. This prospect can be daunting when multiple factors and levels of influence may be associated with risk behavior (eg, drug use, poverty, stigma, unstable housing, disparate access to prevention and treatment systems). The body of research in this supplement represents a significant response to these challenges.
The structure and content of this supplement are intended to further the discussion started at the 2003 meeting and to facilitate rapid dissemination of promising interventions. This issue contains interventions with outcome data that demonstrate efficacy (articles by Wingood et al and Rotheram-Borus et al), detailed conceptual and methodologic overview reports of 2 RCTs that have completed data collection (articles by Fisher et al and Purcell et al) analyses of patient risk behavior (article by Richardson et al) and patient report of receipt of behavioral counseling in HIV primary care clinics (article by Myers et al) that have implications for the integration of risk reduction in treatment settings, as well as lessons learned regarding serostatus disclosure and sexual risk practices of racially diverse HIV-positive men who have sex with men (MSM, article by Wolitski et al). The invited commentaries that follow provide a forum for amplification of pivotal issues and dialogue from a range of perspectives regarding research, program, and policy directions. In this introduction to the supplement, we briefly describe each contribution to the supplement, use these studies and commentaries to illustrate challenges and progress, and conclude with recommendations for research priorities.
CHALLENGES AND PROGRESS
It has been challenging to respond to the unique prevention needs of women living with HIV/AIDS in the United States, particularly women of color living in the southeast region. As Wingood et al note in this issue, these women have significant barriers to accessing services, including social isolation, stigma, child care needs, and transportation problems. In response, Wingood and her colleagues in Georgia and Alabama developed and evaluated a small-group intervention tailored for African-American women with HIV, guided by social cognitive theory and the theory of gender and power. Sessions were facilitated by a trained health educator and an HIV-positive female peer and emphasized gender and cultural pride, communication and condom use skills, and development of supportive social network members, including sexual partners. Several strengths of this methodologically rigorous study include comparison to an attention-matched health promotion intervention and the ability to test intervention effects on STD outcomes to complement the evaluation of self-reported risk reduction. This is the first published intervention trial for HIV-positive African-American women to demonstrate behavioral risk reduction and fewer incident bacterial STDs over a 1-year follow-up period.
A similar challenge has been to determine how best to reach young people living with HIV/AIDS, another relatively underserved group with unique prevention needs and barriers to service access. It is also difficult to mount intervention trials with HIV-positive adolescents, because there are often not adequate numbers at a single site to have power to detect intervention differences. Moreover, conceptualization of interventions appropriate for youth requires attention to developmental issues, and recruitment and retention may hinge on contemporary marketing design and content. The field has not benefited enough from the principles of market research in the design and packaging of HIV prevention interventions. As reported in this issue, Rotheram-Borus and her colleagues sought to extend the marketability and potential uptake of an efficacious small-group intervention (Teens Linked with Care [TLC]4) by deviating from conventional replication strategy. In the current study, they made the delivery format more accessible (eg, including a condition with individual telephone sessions), reduced the length of the intervention, and focused on youth at particularly high risk (ie, with concomitant drug abuse). Their pattern of findings and discussion highlight some of the advantages and pitfalls of this adaptation approach.
The next 3 studies in the supplement address the challenges of prevention interventions with HIV-positive persons in the context of HIV treatment settings. As articulated by Mayer and Gordon in this issue, although there is tremendous opportunity and rationale for HIV prevention in primary care, there are numerous barriers to such intervention, including time and physician comfort with these topics. The report by Myers et al indicates what may occur if the decision to address HIV risk is left to providers’ individual preferences and philosophy of care. The authors analyzed patient reports of behavioral counseling at publicly funded HIV primary care clinics and found that formal clinic procedures to guide providers were associated with an increased likelihood that patients reported receipt of counseling messages. Richardson et al present patterns of risk behavior from patients in clinical care, suggesting that a brief assessment can help clinics to develop an algorithm for provision of preventive services that match individual patient needs. Fisher et al further report in this issue that brief, theory-based, clinician-initiated risk reduction messages are feasible in a high-volume urban clinic and can be delivered with fidelity. At least within the auspices of this controlled trial, they were able to overcome the clinic barriers cited previously; future studies will illuminate whether effectiveness can be maintained if such interventions are brought to scale.
Next, Wolitski et al provide insights from data collected in response to indications that risk behavior among MSM, particularly MSM of color, seem to be increasing. To better inform new interventions for HIV-positive MSM in the post–highly active antiretroviral therapy (HAART) era, 2 large samples of men in New York City and San Francisco (1997–1998 and 2000–2001) completed detailed assessments of sexual practices, relationship patterns, correlates of unprotected sex, and serostatus disclosure. The authors’ synthesis of these data suggests that the interrelations among these variables are complex but that untapped intervention targets remain. An intervention trial was subsequently developed. The preliminary results from the intervention trial (that the enhanced intervention did not lead to a sustained reduction in transmission risk) serve as an important reminder as to why intervention implementation continues to be empirically guided—because theoretically based and well-delivered programs sometimes do not lead to hypothesized outcomes. Although analyses of these outcome data are ongoing, commentary from Auerbach in this issue sheds some light on gaps in our understanding that may help to explain why interventions based on current models for change may not always be successful. In part, she argues that future prevention successes depend on a fuller appreciation that HIV is transmitted in inherently relationship-driven contexts and observes that little research has investigated relational dynamics, condom use decision making for HIV transmission and acquisition when one partner is HIV-positive, the ways that partnerships are affected by culture and social contexts, and the translation of these findings into effective interventions.
In the last empiric paper in this issue, Purcell et al provide an overview of the first efficacy trial of an integrated intervention to address the challenges of medical care, adherence, and prevention needs of HIV-positive injection drug users (IDUs). Their innovative peer-mentoring intervention is based on the concepts of empowerment, peer leadership, and advocacy. During the 10-session intervention, group members are encouraged to adopt a prosocial role to help peers adopt and maintain the same health behaviors that they learn and practice. They present baseline data that underscore the difficult life circumstances of these men and women (eg, low levels of education and income). The high rates of participant retention from baseline to randomization and follow-up indicate the level of participant interest and the acceptability of the intervention. Outcome reports are forthcoming. The investigators suggest that it will be feasible for CBOs to implement this intervention, should it prove efficacious, because it can be staffed by well-trained nonprofessional staff.
Finally, although this supplement was originally intended to focus exclusively on domestic research and priorities, the 2 commentaries by Beloqui and Valdeserri that conclude this supplement focus on international challenges and progress, because it is nearly impossible to discuss these issues without an eye toward the global HIV/AIDS crisis and the impeding rollout of antiretrovirals in many countries. It is imperative that lessons are learned from the unanticipated negative consequences that accompanied improved treatments in the United States and other countries. The same urgency that is being given to improved access to treatment should be given to develop and test proactive and systematic approaches to ensure that prevention and adherence messages are routinely provided to HIV-infected persons around the world. Funding agencies should encourage studies in developing countries, investigating the potential to address prevention with HIV-positive persons in these settings, with evaluation of cost-effectiveness in the research plan.
Overall, risk reduction interventions for HIV-positive persons have produced a number of successes, several of which are described in this supplement. In addition, there are ongoing trials with preliminary evidence for feasibility and efficacy, and other innovative interventions are underway that appropriately target high-risk subpopulations according to epidemiologic trends in the United States. Conceptual and methodologic advancements are still needed, however. The following priorities were identified during the preparation of this supplement and may be considered in addition to the authors’ recommendations in this issue.
Interventions for Multiple Levels of Influence
In general, theoretical models tested for prevention interventions with HIV-positive persons have been variants of the social-cognitive theory. Some trials incorporate contextual factors in the intervention conceptualization, and the importance of this work should not be underemphasized, but the intervention is nonetheless delivered at the individual or small-group level. For example, in this issue, Wingood et al addressed the social context of HIV-positive women by incorporating peer educators and social support networks as important ingredients in their small-group intervention. Likewise, Purcell et al paid heed to the multiple needs of HIV-positive IDUs, based on formative work with this population that called for an enhanced prosocial role in their communities and an intervention that integrated sexual behavior as just 1 of several important health priorities. Again, however, the focus of change was an individual’s behavior.
Future interventions for HIV-positive persons will focus more efforts on structural factors, community-level interventions, media, and multiple systems simultaneously. Studies are underway to explore modification of social settings to reduce risk behavior, couples-based approaches, Internet-based interventions, and mass media campaign evaluation, but further creativity is warranted. For example, the efforts based in HIV clinics that appear in this issue (see articles by Fisher et al, Myers et al, and Richardson et al5) and a previously published report by Richardson et al hint at the potential impact of structural-level interventions in these settings. It may be fruitful to examine skills building programs (to deliver risk reduction messages) at the provider level, modifications of clinic environments to maintain safer behaviors, the process of clinic uptake of policy recommendations for integration of behavioral interventions, quality assurance approaches to sustained behavior change of clinic personnel, and outcomes secondary to modification in state or local guidelines for service provision. In their commentary in this supplement, Rotheram-Borus and her colleagues also argue for a broader conceptualization of prevention with HIV-positive persons that encourages a stronger focus on family-level intervention so as to prevent long-term multigenerational negative consequences of HIV, including future mental health problems, substance abuse, and risk behavior.
We do not mean to suggest that there is no longer a need for individual and group-based interventions that are tailored to theoretically based mediators for behavior change. As Holtgrave6 has pointed out, perhaps only a minority of HIV-positive men and women may need more intensive intervention. It is likely that HIV-positive individuals who are struggling with problems such as substance use, severe and persistent mental illness, relationship abuse, poverty, or transient housing may still need more intensive interventions such as prevention case management. The effectiveness of such interventions designed for individuals with multiple problems continues to require careful study.7
Maintenance of Behavior Change
Despite the fact that the challenge of behavioral maintenance was recognized early in the HIV epidemic,8 the predominance of research on HIV prevention has focused on successful initiation and far less on long-term maintenance. This trend has also occurred in prevention research among people living with HIV. Of the published trials, the longest outcome assessment has been 12 months after the intervention. One solution is to extend the trial length and assessment intervals, and there may be questions where this response is appropriate. The evaluation of sustainability of behavior change is also linked with the conventional RCT design, however, which typically entails an intervention period of specific frequency, duration, and intensity, followed by an assessment period to evaluate efficacy. Alternative models may be developed that more comprehensively capture factors that affect real-world implementation. Such a design could allow for intervention duration to be titrated to the needs of the individual (or family, structure, or other target) and also permit the intervention to adapt and evolve based on changing needs.
Diffusion, translation, and operational research studies should be a high priority to bridge research and practice. Many understudied issues in this process are likely to affect whether an intervention works in community settings. What is the minimal and optimal assistance needed to help providers deliver interventions? In most cases, for interventions that have demonstrated efficacy, skilled interventionists are a key component. Resource intensiveness for intervention implementation seems to be a critical issue, yet it is not often discussed in outcome reports. In their commentary in this issue, Gandelman and Rietmeijer describe their ideas and recommendations from a provider perspective and outline some key questions for the next generation of effectiveness research. For example, what organizational factors are related to adoption of interventions and ongoing outcomes assessment? How much can interventions be tailored or adapted to suit local needs before intervention fidelity and/or effectiveness is lost?
Finally, although it is becoming more common for researchers to solicit input and conduct formative research with consumer/community members in study design and intervention development, the degree and quality of this process vary widely. Contextual influences that are experienced by individuals in their organizations, cities, and clinics should be routinely communicated to researchers and funders to improve intervention effectiveness and inform future research. Research in this area should be especially sensitive to these issues, because prevention with persons living with HIV/AIDS has the potential to be politically divisive and stigmatizing. We need to be mindful that there are possible adverse consequences of increasing positive prevention, especially related to criminalization of the sexual behavior of persons living with HIV/AIDS. Studies have shown that most individuals, once aware of their HIV status, attempt to prevent transmission to others. We are working together to extend evidence-based and cost-effective interventions to reach those who need additional assistance.
This supplement benefited from a meeting sponsored by the Center for Mental Health Research on AIDS (NIMH) and the CDC held in conjunction with the 2003 National HIV Prevention Conference. The authors thank the meeting attendees who shared their emerging data and recommendations for future directions, consumers who discussed their experiences and ideas for innovative interventions, Paul Volberding for his support of this supplemental issue, and especially the research participants who gave their time and effort for these studies.