This is Archived Content

This content is available for historical purposes only. It may not reflect the current state of science or language from the National Institute on Drug Abuse (NIDA).
View current meetings on


May 8, 2007 - 12:00am to May 9, 2007 - 12:00am
Natcher Conference Center, NIH, Bethesda, Maryland


Nora D. Volkow, M.D.
Director, National Institute on Drug Abuse

The 2007 Drug Abuse and Risky Behaviors: The Evolving Dynamics of HIV/AIDSmeeting responded to deep concern about the continuing spread of HIV/AIDS in the United States and abroad and invited the scientific community to engage in a dialog to further our understanding of the problem and how it is changing. It provided a timely forum to explore our growing knowledge of the behavioral and social factors that affect HIV infection rates among various populations, including the multiple influences of drug abuse and addiction on HIV risk. It underscored how research could inform public health policy and identified emerging opportunities to address this evolving pandemic.

Sponsored by NIDA in collaboration with multiple NIH Institutes and the Centers for Disease Control and Prevention, this meeting highlighted the drug abuse-HIV link. Speakers addressed a variety of related topics, including how drugs of abuse alter brain function and impair decisionmaking, how substance abuse affects HIV/AIDS risk in diverse demographics, how and to what extent substance abuse influences sexual risk behaviors, and how testing and counseling can be incorporated as key components of HIV prevention strategies for different drug-abusing populations.

Overview and Plenary

HIV/AIDS in 2007: A Brief Overview
Anthony S. Fauci, M.D.
Director, National Institute of Allergy and Infectious Diseases 

NIH AIDS Research Priorities
Jack Whitescarver, Ph.D.
Office of AIDS Research

Drug Abuse Treatment in HIV Prevention and Care: Past Successes and Future Challenges
David S. Metzger, Ph.D. 

Injection and noninjection drug use have been well documented as risk factors for HIV infection. They have also been shown to inhibit access to HIV care and to negatively affect adherence to HIV treatment regimens. Although most research has been focused on injection drug users, there is a growing awareness of the significant role of noninjection drug use in sexual transmission among the populations of both heterosexuals and men having sex with men. Substantial data exist on the potential for effective drug abuse treatments, particularly those with pharmacologic adjuncts, to have significant effects in preventing the transmission of HIV infections and in increasing the access to HIV care among those already infected. Although there are several limitations to past HIV research focusing on the impact of drug abuse treatments (with most studies having been focused on opiate-dependent injectors, the absence of randomized controlled trials, and short-term followup intervals), it is clear that drug treatments can play a major role in HIV prevention and care in domestic and international communities where drug use is propelling the epidemic. Yet to maximize their impact, these treatments need to be accessible, acceptable, and affordable to the populations at the highest risk of HIV infection. Research must continue to be focused on the development of more effective pharmacologic agents, counseling strategies, and delivery mechanisms with public health relevance.

The Role of Drug Abuse in the Evolving HIV Epidemic
Steffanie A. Strathdee, Ph.D.

According to the National Institute on Drug Abuse's International Program, there is "no greater public health goal for addiction researchers than to help stem the continued spread of the intertwined epidemics of HIV and drug abuse." More than 15 million people have been diagnosed with drug use disorders worldwide. There are an estimated 13.2 million injection drug users (IDUs) in 130 countries. Although 10 percent of the world's HIV infections are directly attributed to needle-sharing among IDUs, outside of Africa this percentage rises to 30 percent. In the United States, of the nearly 14,000 HIV infections associated with injection drug use diagnosed in 2005, 45 percent and 24 percent were heterosexual males and female IDUs, respectively; 16 percent were attributed to men who have sex with men (MSM)-IDUs; 15 percent were heterosexual partners of IDUs; and less than 1 percent were children of IDU parents.* Explosive HIV epidemics among IDUs have occurred in both developed and developing countries, including such recent examples as Ukraine, the Russian Federation, Vietnam, Iran, and China, with emerging epidemics in Pakistan, Tajikistan, Afghanistan, and northwestern Mexico. In mature HIV epidemics in developed and developing countries, diffusion from IDUs to non-IDU populations has been documented, with growing attention needed on bridging the populations (e.g., sex worker-IDUs, MSM-IDUs, paid blood donors). Examples of successful interventions to reduce HIV among drug users have been reported at the levels of the individual, social network, and community. In a recent meta-analysis of behavioral IDU interventions, the most successful strategies focused on both injection and sexual risks, but only 50 percent of the interventions were based on behavioral theory and only 6 percent had more than one followup visit. Only 12 percent were conducted outside the United States, even though 78 percent of IDUs reside in developing countries. Despite the heterogeneity of IDU-associated HIV epidemics between and even within countries, lessons from HIV prevention stories among drug users worldwide suggest that HIV risk environments are socially produced, and thus, prevention is a shared responsibility.

*The percentages listed are rounded and may not equal 100 percent.

Drug Abuse, HIV, and the Brain

Neuroeconomics: New Approaches to Risky Decisionmaking
Gregory S. Berns, M.D., Ph.D.

Within the past 5 years, neuroeconomics has matured as a new approach of using brain imaging to both test and expand economically-based theories of decisionmaking. In this presentation, the basic principles of expected utility and prospect theory are described. Recent brain imaging experiments based on these theories suggest common neural substrates in the brain that weigh the costs and benefits of potential decisions as well as the risk that an individual places on them. These experiments suggest promising approaches to using brain imaging to predict future decisions.

Neuroimaging To Predict High-Risk Behaviors in Methamphetamine Users
Martin P. Paulus, M.D.

Relapse is an important clinical aspect of substance dependence. Many factors contribute to relapse, but its neurobiological basis is poorly understood. We have used a simple decisionmaking task in the context of functional magnetic resonance imaging in 40 methamphetamine-dependent individuals who completed a 28-day inpatient program to predict whether the patterns of brain activity predict relapse. Brain activation levels in the right insula, right posterior cingulate, and right middle temporal gyrus correctly predicted the outcome in 19 of 22 individuals who did not relapse and in 17 of 18 who returned to methamphetamine use. This result demonstrates that functional neuroimaging may be useful for long-term clinical predictions in substance dependence.

NeuroAIDS in Drug Users 
Justin McArthur, M.D.

Methamphetamine and HIV: CNS Effects
Igor Grant, M.D.

Methamphetamine and HIV infection each can produce brain disease, and there is evidence that their co-occurrence may produce additive effects on neural injury. Approximately 40 percent of methamphetamine (METH+) abusers who are HIV-uninfected have some neurocognitive impairment, and a similar proportion of HIV-infected individuals (HIV+) who are not methamphetamine abusers are also impaired. More than 60 percent of persons who are HIV+ METH+ have cognitive disturbance, suggesting an additive effect. Although there may be some common pathways (e.g., the induction of neuroinflammatory cascades) that underlie this apparent synergy, there are also differences in the neuropathogenic mechanisms. For example, although atrophy of selected brain regions is a common correlate of HIV disease of the brain, one sees increases in the volumes of certain brain structures with methamphetamine. Furthermore, neuropathologic studies indicate that in addition to the generally accepted effect of methamphetamine on dopaminergic circuitries, there may be a selective loss of calbindin immunoreactive interneurons in the brains of those affected by both HIV and METH. Disturbances in neurocognitive functioning associated with methamphetamine may also be associated with declines in everyday functioning, and deficits in decisional abilities could influence the likelihood of engaging in risk behaviors or not adhering to treatment that could affect HIV disease progression and transmission.

Drug Abuse, HIV, and Sexual Risk Behaviors

Behavioral Risk and HIV-1 Molecular Diversity: Making the Connections

Chris C. Beyrer, M.D., M.P.H., Sodsai Tovanabutra, Ph.D., Gustavo Kijak, Ph.D., Teerada Sripaipan, M.H.S., Eric Sanders-Buell, Ph.D., Kittipong Rungruengthanakit, M.A., Jaroon Jittiwutikarn, M.D., David D. Celentano, Sc.D., M.H.S, and Francine E. McCutchan, Ph.D.


HIV-1 is a highly, genetically diverse pathogen, and its genetic diversity is increasing. Advances in epidemiologic understanding and in the full-length genetic sequencing of HIV-1 suggest that risk behaviors may play critical roles in the genetic diversity of the pandemic and may have important implications for HIV prevention, vaccine research, and eventual epidemic control. Individual-level risk behaviors and network-level interactions may play critical roles in the risk of dual-infection, super-infection, and HIV-1 intersubtype recombination. We investigated these interactions in the opiate users research (OUR) cohort in northern Thailand where CRF01_AE, subtype B, as well as unique and circulating recombinant forms (CRFs) have been detected. Recent work among sexual risk populations in East Africa and analyses linking Thai, Chinese, and Burmese epidemics suggest that risk dynamics may play similar roles in HIV-1 genetic complexity in epidemics driven by sexual, as well as injection drug use, risks.


The OUR study was conducted among 2,231 volunteers in Chiang Mai, Thailand, between 1999 and 2001. Serum samples from 347 participants who were HIV-1 seropositive at baseline were analyzed by the RT-PCR amplification of HIV-1 genome fragments and subtyping in 6-8 genome regions per strain with the Multiregion Hybridization Assay MHAbce v.2. The assay distinguishes subtypes from recombinant forms and detects dual infections. Social and demographic variables and risk behavior were assessed using a structured questionnaire administered in a confidential, face-to-face interview.


Among the 336 (96.8 percent) samples that were typed, 81.8 percent were CRF01_AE; 3.9 percent were subtype B; and 9.2 percent were recombinants, most combining CRF01_AE and subtype B. Dual infections were detected in 5.1 percent of subjects. Subtype B was more frequent among participants 30 years or older (OR=6.92, 95-percent confidence interval [CI]: 1.51-31.73). Dual infections were more frequent among those with a lower education level (OR=5, 95-percent CI: 1.39-17.51) and among those who had initiated injecting in the past 3 years (OR=3.41, 95-percent CI: 1.19-9.79). Recombinant strains and dual infections were more frequent among those who reported frequent needle-sharing in the past 3 months (OR=4.08, 95-percent CI: 1.42-11.73).


Dual infections were associated with those possessing longer injection histories and among injection drug users (IDUs) reporting more frequent and recent needle-sharing. Early intervention aimed at a reduction in needle-sharing, especially among new IDUs, might help limit the increasing complexity of HIV-1 strains. Data from the region, and from East Africa, suggest that these risk dynamic and genetic interactions may have regional implications.

Alcohol and HIV Risk Behaviors
Michael D. Stein, M.D.

Worldwide, alcohol is the most commonly used psychoactive substance. Since the early 1900s, alcohol use has been associated with sexual behavior that places people at risk for sexually transmitted infections. This implied causal association can now be rigorously investigated across a variety of populations. This presentation (1) discusses the pertinent methodological issues in alcohol use and sexual risk research, (2) describes the approaches to investigating the connection of alcohol and sexual risk, (3) presents event-level data from one Daily Diary study in South Africa, (4) provides a conceptual model of alcohol use in sexual situations, and (5) examines the intervention strategies that may offer future directions in research and clinical care.

Methamphetamine Use and HIV
Grant N. Colfax, M.D.

Methamphetamine use is a driving force in the HIV epidemic. Methamphetamine is associated with increased unprotected sex acts, increased numbers of partners, and increased risk for HIV infection. Among men who have sex with men (MSM), methamphetamine use is at least 10 times more prevalent than among the general population. Most methamphetamine-using MSM are infrequent users (< weekly), and the majority do not inject. Methamphetamine use doubles to triples the risk of HIV infection. Longitudinal cohort studies report that methamphetamine is independently associated with HIV seroconversion, even after controlling for other behavioral risk factors, including the numbers of sexual partners and unprotected sex acts. Although the direct effects of methamphetamine on HIV disease progression remain to be determined, methamphetamine users on antiretroviral therapy have higher viral loads compared to nonmethamphetamine users, after controlling for medication adherence. Among persons with recent HIV infection, methamphetamine has been associated with primary drug resistance. Entering drug treatment and participating in methamphetamine-specific treatment interventions are associated with decreases in sexual risk and methamphetamine use. However, most programs have been tested among treatment-seeking, frequent methamphetamine users. The majority of methamphetamine-using MSM have not been in drug treatment, indicating a continued need to develop new interventions and to expand on those already shown to be efficacious in reducing both methamphetamine use and HIV risk.

Drug Treatment as HIV Prevention

Does Treatment for Methamphetamine Dependence Reduce HIV Risk Behavior in non-MSM Treatment Samples?
Richard A. Rawson, Ph.D., Valerie Pearce, M.P.H., Rachel Gonzales, Ph.D., and Julie Brummer

Treatment for methamphetamine (MA) dependence has been shown to be an important HIV risk-reduction strategy for MA-dependent men who have sex with men (MSM). However, there are minimal data on the generalizability of this finding to broader treatment populations who do not exclusively include MSM. This study examines the extent to which treatment for MA dependence is associated with a reduction in HIV-related risk behaviors among three samples of heterogeneous MA users. Data from three clinical trials of MA users were analyzed: two MA pharmacotherapy trials (ondansetron and bupropion) conducted as part of the National Institute on Drug Abuse Methamphetamine Clinical Trials Group (MCTG) (N=320) and a Substance Abuse and Mental Health Services Administration-funded clinical behavioral trial called the Methamphetamine Treatment Project (MTP) (N=977). All participants were enrolled in an outpatient behavioral treatment program, either as a "platform" intervention for a pharmacotherapy trial (MCTG) or as a primary psychosocial treatment (MTP). Data specifically assessed HIV-related risk behavior, including injection drug use and sexual practices (defined as unprotected sexual intercourse with multiple sex partners and/or anal intercourse) using the HIV Risk Behavior Scale (MCTG) or AIDS Risk Assessment (Texas Christian University) (MTP), at the time of treatment entry and at termination from the treatment at 8-12 weeks for those in the MCTG and at 16 weeks for those in the MTP. Data from the MCTG study indicated that at baseline 68 individuals (21.2 percent) reported injection within the previous 30 days. At discharge there was a reduction to 24 (10.3 percent). Of the sample that engaged in unsafe sexual activity in the last 30 days at admission (n=199), there was a significant reduction in the unsafe sexual behavior mean composite score (5.28 to 4.4; t=2.544, p<.05). In the MTP trial, injection within the past 30 days decreased from 128 individuals (13.1 percent) at baseline to 53 individuals, or 5.4 percent, at discharge. Unsafe sexual behavior was reported by 612 individuals (62.6 percent) at baseline and decreased over the treatment period to 445 individuals (45.5 percent) at discharge. The sum average risk score (number of times they engage in risky behaviors) for the heterosexual population is 23.06 at baseline and 10.18 at discharge. Additional results are presented on the relationship between time in treatment and reduction in MA use with the magnitude of the reduction of HIV risk behavior. Furthermore, in the MTP study data at 3-year followup will be presented.

Combined Pharmacological and Behavioral Therapy and HIV Risk Reduction
Kenzie L. Preston, Ph.D., Jennifer R. Schroeder, Ph.D., David H. Epstein, Ph.D., and Annie Umbricht, M.D.

Cocaine use is associated with injecting and sexual HIV risk behaviors. This study was a randomized controlled trial testing the combination of two behavioral interventions for cocaine dependence and HIV risk behaviors among dually (cocaine and heroin) dependent outpatients. Methadone maintenance was augmented with contingency management (CM), cognitive-behavioral therapy (CBT), both (CM+CBT), or neither. CM rapidly reduces cocaine use, but its effects subside after treatment, while CBT produces reductions that emerge 6-12 months after treatment ends. Combined, the two treatments might be complementary. The study sample (n=81) was 52 percent female, 70 percent African-American, and 37.9-7.0 years old. The proportions reporting HIV risk behaviors at intake were 96.3 percent (78/81) injection drug use, 56.8 percent (46/81) sharing needles, 30.9 percent (25/81) unprotected sex, and 28.4 percent (23/81) trading sex for money or drugs. The proportions who no longer reported behaviors at the study exit were 48.7 percent (38/78) injection drug use, 91.3 percent (42/46) sharing needles, 88 percent (22/25) unprotected sex, and 91.3 percent (21/23) trading sex for money or drugs. Participants receiving CM+CBT were more likely to report the cessation of unprotected sex relative to controls (OR=5.44, 95-percent confidence interval 1.14-26.0, p=0.034), but this effect was reduced by adjusting for drug-negative urines. These results suggest the broad beneficial effects of methadone maintenance augmented with behavioral interventions for reducing HIV risk behaviors.

Behavioral Interventions for HIV Risk Reduction and HIV Prevention: An International Perspective
Marek C. Chawarski, Ph.D.

The presentation describes our initial work in Malaysia comparing a behavioral counseling platform only with counseling combined with naltrexone or buprenorphine maintenance and our current work developing and testing behavioral drug and HIV risk-reduction counseling that can be provided by available nursing and other health personnel in developing countries. In the initial study, HIV risks were reduced in all three groups, buprenorphine was associated with significantly better drug-related outcomes, and sexual risk behaviors were not reduced in any of the three groups. Based on the research outcomes and clinical feedback obtained from studies conducted in the United States and internationally, we have developed an improved integrated drug abuse and HIV-risk reduction intervention. This new intervention is founded on the principles of cognitive-behavioral treatments and utilizes evidence-based counseling approaches based on the research findings in cognitive and health psychology on effective behavior change. We demonstrate that medical personnel (nurses) available in developing countries could be trained to provide this type of counseling. Our data also show preliminary improved efficacy as compared with the limited psychosocial treatment typically provided. The treaDrtment is currently being implemented and tested in various research and clinical settings in Asia, the Middle East, and the United States.

Drug Abuse and HIV/AIDS: Women and Youth

Do Sex and Drug Behavior Patterns Account for HIV/STD Racial Disparities?
Denise D. Hallfors, Ph.D.

This paper examines the relationship between race and sexually transmitted disease (STD)/HIV prevalence, after stratifying by sex and drug behavior and controlling for gender, marriage, age at first sex, and socioeconomic status. Data are from Wave III of the National Longitudinal Study of Adolescent Health. Participants were 18 to 26 years old; the analyses were limited to non-Hispanic blacks and whites. Theory and cluster analysis yielded 16 unique behavior patterns. Bivariate analyses examined each behavior pattern by race, gender, and race by gender interaction and then compared biological STD/HIV prevalence by pattern and race. Logistical regressions examined race effects within patterns before and after controlling for the covariates. The unadjusted odds of STD/HIV infection were significantly greater in black young adults than in white young adults for 11 of the risk behavior patterns. Across the behavior patterns, covariates had little effect on reducing the odds ratios for race. White young adults in the United States are at elevated STD/HIV risk only when their behavior is high risk. Black young adults, however, are at high risk even when their behavior is normative. Factors other than individual risk behaviors and covariates appear to account for the racial disparities, arguing for population-level interventions.

Influences of Marijuana Use on HIV/STI Acquisition and Care
Jonathan M. Ellen, M.D.

Marijuana use is not rare among adolescents and young adults, including those most at risk for HIV or those infected with HIV. Evidence of the deleterious effects of marijuana use on adolescent brain functioning are just beginning to emerge. However, there are scant data about the impact of marijuana use on adolescents' and young adults' risks for HIV infection and the effect of marijuana on their ability to receive medical care once they have been infected. The objectives of this presentation are to highlight the recent findings from two completed studies funded by the National Institute of Allergy and Infectious Diseases (Bayview Networks Study), National Institute on Drug Abuse and the National Institute of Child Health and Human Development (Adolescent HIV Prevention Trials Network [ATN] Protocol 009) that demonstrate (1) that marijuana use is associated with HIV/STI risk-related sex network structures and (2) that marijuana use is associated with poor medical appointment-keeping among infected women. Additionally, the presentation describes on-going trials within ATN to address the HIV risks associated with marijuana use.

How Does Trauma Contribute to Substance Abuse and HIV Infection Among Ethnic Women?
Gail E. Wyatt, Ph.D.

There are seven factors that are overlooked with regard to African-descended women's risks for sexually transmitted diseases and HIV that are described. Historical and empirical documentation highlight the need for interventions that are beyond what is currently available.

Women's Interagency HIV Study: Association of Substance Use With HIV Clinical Outcomes, Metabolic Conditions, and Psychiatric Comorbidity
Kathryn Anastos, M.D.

The Women's Interagency HIV Study (WIHS) is a longitudinal observational cohort study of HIV infection in women, with 2,807 HIV-positive women and 959 HIV-negative women enrolled in two waves: 1994-1995 (69.6 percent of the cohort) and 2001-2002. Every 6 months participants are interviewed, receive a physical examination, and provide blood and gynecologic specimens for real-time testing and for the WIHS repository. WIHS resources available to investigators will be discussed, with a focus on those most relevant to the investigations of substance use and related illnesses (e.g., hepatitis B and C infections). There are now nearly 2 million aliquots in the WIHS repository, including serum, plasma, cervicovaginal lavage, urine, and cells, both dry pellets and viable PBMCs, which are available to both WIHS and non-WIHS investigators. WIHS investigations have included the association of substance use with HIV-related clinical and laboratory outcomes (e.g., participants with a history of injection drug use having significantly lower viral loads) both on and off highly active antiretroviral therapy, with risk-taking behaviors and adherence and with non-HIV clinical outcomes, for example, sequelae of HPV infection and the viral hepatitides. Recent WIHS investigations have demonstrated a stronger association of insulin resistance with opiate use than with hepatitis C infection. Unpublished translational and epidemiologic data are shared, and the mechanisms for increasing collaboration from outside researchers explored.

HIV Prevention in Criminal Justice Populations

HIV/AIDS and the Criminal Justice System
Theodore M. Hammett, Ph.D. This presentation summarizes the intersection of the epidemics of substance abuse, HIV/AIDS, and incarceration in the United States, showing how the "war on drugs" has led to an explosion in correctional inmate populations in the past 20 years. At least 80 percent of all correctional inmates have some form of a substance abuse problem, whether or not they are incarcerated for a drug-related crime. The large-scale incarceration of injection drug users and other drug users has also led to the disproportionate prevalence and burden of HIV/AIDS and hepatitis among prison and jail inmates. About one-quarter of all people living with HIV in the United States, and about one-third of all people living with hepatitis C virus (HCV), in a given year pass through a correctional facility that same year.

As a result of this epidemiologic situation, correctional facilities are critical settings for interventions to prevent, diagnose, and treat HIV, HCV, and other infectious diseases. Although there have been substantial improvements in these and other correctional health interventions in recent years, the important related opportunities to improve the health of inmates and their partners and families and to advance the public health have by no means been fully exploited.

A Multisite HIV Prevention Protocol for Drug-Involved Offenders Returning to the Community
James A. Inciardi, Ph.D.

Recent estimates suggest that HIV and hepatitis seropositivity rates in correctional populations are roughly 8 to 10 times higher than those in the general population. These high rates are related to risky drug-using and sexual behaviors prior to incarceration. Importantly, many offenders resume these risk behaviors after release from the institution, attempting to "make up for lost time." Thus, reentry is a pivotal period for prevention, yet effective programs for drug-involved offenders in transitional correctional settings are lacking.

Using focus groups and interviews, an HIV/hepatitis protocol was developed that addresses the risk-reduction issues and barriers of concern to community corrections populations. The targeted intervention consists of an interactive, 2-session, DVD-based risk-reduction module facilitated by a peer interventionist and adapted for race/ethnic and gender appropriateness. The targeted intervention was designed to speak to correctional clients in their own language through the use of both virtual (DVD-based) and real "peers" (interventionists) and to coincide with reentry. By integrating the relevant intervention messages into an engaging, interactive format appropriate for different learning styles, the program seeks to provide the maximum impact in a brief intervention. The effectiveness of this intervention is being tested in a multisite field trial with community corrections clients.

HIV Testing Strategies and Linkage To Care for Criminal Justice Populations
Timothy P. Flanigan, M.D.

It is estimated that 20 percent of all HIV-infected individuals pass through the correctional system. Correctional institutions present an ideal opportunity to implement routine testing for HIV both to improve diagnosis and to link to care and initiative-prevention counseling.

In Rhode Island over a 10-year period, the routine testing of HIV within the correctional facilities identified 28 percent of all positive tests within the State. HIV testing, when linked to medical care, can result in a decrease in the rate of AIDS within the correctional setting, which in the long run results in cost savings. Since the advent of effective antiretroviral therapy, the rate of AIDS within the correctional settings in New York and other States has dropped dramatically. Rapid testing is feasible; can be easily implemented in different jail settings with good acceptance; and can be done through oral testing, which avoids exposure to blood. HIV testing within jails and prisons offers an important public health opportunity for HIV prevention, diagnosis, and linkage to care and can have enormous impact, particularly among African-American men, who suffer from the highest rates of HIV.

Testing and Counseling Policy

HIV Screening and Care: Effect on Clinical Outcomes, Transmission, and Cost
A. David Paltiel, Ph.D., M.B.A.

Of the estimated 1 million Americans who are currently infected with HIV, as many as 250,000 remain unaware of their infection. These individuals receive neither life-prolonging care nor counseling that could prevent the further transmission of the virus. This sobering fact has prompted a reexamination of the public health approach to HIV screening in the United States, culminating in the September 2006 release of new guidelines from the Centers for Disease Control and Prevention (CDC). This presentation reviews the economic evidence base in support of CDC's recommendation of routine, voluntary HIV testing for all persons ages 13 to 64 years in health care settings. Dr. Paltiel discusses the promise and pitfalls of using model-based, cost-effective analysis as a tool to inform public health decisions. He presents recent results demonstrating the cost-effectiveness of routine, voluntary HIV testing for all adults, not just for those who are at extra high risk. Finally, Dr. Paltiel reports on our efforts to document both the dramatic survival gains that have been achieved in patients with HIV disease in the United States and the chronic disparities arising from inadequate detection, linkage, and access to the appropriate care.

CDC's Recommendations for HIV Screening in Health Care Settings
Bernard M. Branson, M.D.

The Centers for Disease Control and Prevention (CDC) estimate that about one-quarter of the 1.0-1.2 million persons living with HIV/AIDS in the United States are unaware of their HIV infection. They are unable to access effective treatment, and compared to those who know they are infected, they are more likely to transmit HIV to others. Pregnant women need to know if they are HIV-infected to help prevent HIV transmission to their children and to access care for themselves. Despite past CDC recommendations for HIV testing, many HIV-infected persons encounter the health care system but are not tested for HIV. CDC has revised the recommendations for HIV testing in health care settings to help increase the proportion of HIV-infected Americans who are aware that they are infected and who receive prevention, care, and treatment. The new recommendations advocate voluntary "opt-out" screening in health care settings; incorporate the consent for HIV testing as part of the consent for general care; promote annual rescreening for individuals with risk indications; and enhance the linkages to care and treatment. CDC issued the revised recommendations on September 21, 2006, and is now engaged with numerous professional organizations on the practical strategies for implementation.

Delay From Testing HIV-Positive Until First HIV Care for Drug Users: Adverse Consequences and Possible Solutions
Barbara J. Turner, M.D., M.S.Ed., and John Fleishman, Ph.D.

Our group and others have reported lengthy delays from testing HIV-positive until the first receipt of HIV care. In a national probability sample of HIV-infected persons, 29 percent of 1,540 persons diagnosed with HIV in the early 1990s delayed first care for more than 3 months, and in this group, the mean delay was 1.45 years. Minorities were significantly more likely to delay, while persons with a usual source of care were less likely to delay. Based on the earlier analysis, we hypothesize that illicit drug use is associated with the increased delay in HIV care. We also hypothesize that this delay will put others at risk through unprotected sex.

Among 1,330 persons in the sample that we previously studied and who responded to questions about drug use, 41 percent of hard-drug users (i.e., crack, cocaine, opioid drugs) delayed HIV care more than 3 months compared with 27 percent of non-drug users. After an adjustment for demographic factors, hard-drug users had 83-percent higher odds of a more than 3-month delay (95-percent confidence interval 1.24-2.70) versus non-drug users, whereas other types of drug abuse (e.g., sedatives, inhalants, amphetamines, marijuana) were not associated with an increased delay. Among those with a more than 3 month-delay, the duration of delay did not differ by drug use group. Having a usual source of medical care was not significantly associated with the severity of drug use, but it was associated with the lower adjusted odds (0.59 [0.48-0.75]) of a more than 3-month delay until first HIV care, regardless of drug use. A later survey of 1,421 patients in this national sample addressed sexual risk-taking. Unprotected sex in the past 6 months was reported by fewer persons who used no drugs (20.7 percent) or only marijuana (28.0 percent) than persons using hard drugs (38.5 percent) or other drugs (40.6 percent). After adjustment, both hard-drug users and other-drug users (except marijuana) had more than twofold greater adjusted odds of recent unprotected sex than non-drug users (p<0.001).

We conclude that after testing HIV-positive, drug users are more likely to delay initiating HIV care and are likely to continue to spread the infection to their partners through unprotected sex. Because drug users with a usual source of medical care appear to initiate HIV care more quickly, these data support increased efforts to link drug users to a longitudinal source of medical care.

Expanding HIV Testing And Counseling into Community Settings

Ethical Challenges for Research-Related HIV Testing and Counseling: Community Perspectives
Celia Fisher, Ph.D.

This presentation draws on the voices of African-American, Hispanic, and non-Hispanic white male and female street drug users with and without a diagnosis of HIV to illustrate the ethical issues that arise in nontreatment drug abuse research involving HIV testing. As part of a National Institute on Drug Abuse-funded 3-year multisite project, content analysis from 11 focus groups identified the community misconceptions about HIV, personal fears and social consequences of learning one's HIV status, and loss of privacy as deterrents to participation. Taking personal responsibility for one's health, protecting loved ones, receiving education and counseling, and establishing a trusting relationship with an investigator emerged as participation incentives. The implications for policies encouraging the provision of HIV-risk-reduction counseling and testing for research participants are discussed.

HIV Testing and Counseling in Drug Abuse Treatment
James L. Sorensen, Ph.D.

Recently, HIV testing and counseling have undergone significant changes in technology (simplifying the collection of biological samples and reducing the lag between sampling and providing results to only 20 minutes) and in policy (emphasizing the public health benefits of testing). These changes are bringing a new set of ethical and practical challenges to the field. Although considerable research has documented the desirable impact of HIV testing and counseling on sexual risk behavior, the impact on drug use risk behaviors is less well understood. Among the Nation's drug abuse treatment programs there is considerable variation in the practices related to HIV testing and counseling. In some settings these practices are routine; in others, they are not available and are seldom used as referral options. The National Institute on Drug Abuse Clinical Trials Network is developing a protocol to study the impact of providing rapid HIV testing and counseling in drug abuse treatment programs. The study will examine the impact of providing rapid HIV testing onsite and the impact of providing enhanced counseling to patients. There is an urgent need for a greater understanding of HIV testing and counseling in the context of substance abuse treatment settings.

Voluntary Counseling and Testing: An International, Randomized Community Trial
David D. Celentano, Sc.D., M.H.S.

National Institute of Mental Health (NIMH) Project Accept is an HIV prevention trial in which 34 communities in Africa (South Africa, Tanzania, and Zimbabwe) and 14 communities in Thailand are being randomized to receive either a community-based HIV voluntary counseling and testing (CBVCT) intervention plus standard clinic-based VCT (SVCT) or SVCT alone. The CBVCT intervention has three major strategies: (1) to make VCT more available in community settings, (2) to engage the community through outreach, and (3) to provide posttest support. These strategies are designed to change community norms and to reduce the risk for HIV infection among all community members, irrespective of whether they participated directly in the intervention.

A community-level intervention based on modifying community norms can change the environmental context in which people make decisions about HIV risk and has the potential to alter the course of the HIV epidemic in developing countries. This is the first international, randomized, controlled Phase III trial to determine the efficacy of a behavioral/social science intervention with an HIV incidence end point.

NIMH Project Accept is funded as an NIMH Cooperative Agreement involving NIMH and several U.S. and international institutions. The HIV Prevention Trials Network (HPTN) is also providing support to the project [HPTN 043]; support has also been provided by the Office of AIDS Research, National Institutes of Health.