Effective Intervention Is Possible in the Inner City

Transmission Rates Can Be Reduced, Even Among the Indigent, Homeless, and Socially Marginalized

The HIV epidemic in the United States is increasingly an epidemic affecting the poor and the young. HIV infection incidence rates among older white gay men -- the population segment that accounted for the greatest number of infections through the mid-1980s -- have largely leveled off in the cities that were the original epicenters of infection. More and more, new HIV infections are concentrated among persons younger, poorer, and more socially marginalized than in the past. This pattern is evident among heterosexually-transmitted infections, infections occurring among men who have sex with men, and infections associated with injection drug use.

Efforts to control this "second wave" of HIV infections require the development, evaluation, and eventual deployment of prevention programs tailored to meet the needs of -- and change high-risk behavior among -- the disadvantaged populations now vulnerable to HIV and AIDS. Fortunately, recent research has shown that HIV prevention approaches can be effective, even among those who are young, poor, or both. Many of these successful outcomes have been documented through randomized and controlled studies.

A number of studies published in the past two years show that women at high risk for HIV can be assisted in making sexual behavior changes through participation in small-group interventions offered in the settings of inner-city primary health care, reproductive health, and social service agencies (1-3). Each of these randomized intervention trials recruited women who were at high initial risk for AIDS, either because they had multiple male partners, had a relationship with a single but high-risk male partner, had a recent history of sexually-transmitted disease, or had experienced an unwanted pregnancy.

Each study engaged women to attend a series of four to five 90-minute small-group sessions that met in a clinic or social service agency. The interventions used in these projects focused on risk education and correcting participants misconceptions about risk; rehearsal of sexual assertiveness and negotiation skills; exercises to increase the women's self-confidence, improve their attitudes, and heighten their intent to make behavior changes that would reduce their risk of infection. Group problem-solving exercises focused on how to implement change in the context of current and future relationships, and on reinforcement and support of behavior change efforts.

Women do not confront the barriers to behavior change in the same way that men do. The differences begin with the fact that condom use is a male-controlled protective strategy, that men are often resistant to condom use, and that impoverished and minority women are often inhibited by traditional cultural and gender role stereotypes that discourage their initiation of safer sex discussions. For these reasons, most group HIV prevention interventions for women have been tailored to address gender-related risk barriers and have relied on problem-solving approaches to confront these barriers.

In spite of the obstacles to change faced by disadvantaged women, the interventions tested in these studies have produced evidence of substantial behavior change. In one study, Kelly et al. found that women who participated in an HIV prevention program increased levels of condom use from 26% of all intercourse occasions at baseline to 56% at follow-up, had less frequent unprotected sex, and used condoms with a greater proportion of their male partners than did control group women (1).

HIV risk-reduction interventions for adolescents

Similar HIV prevention group interventions have been undertaken in randomized outcome trials with adolescents at high risk for HIV, including homeless and runaway adolescents (4) and inner-city, predominantly African-American adolescent males and females (5-7). The HIV prevention interventions evaluated in these studies were like those used in programs aimed at inner-city women. All were based on social-cognitive or cognitive-behavioral principles, and all combined risk education with exercises to strengthen adolescents' risk-reduction attitudes, beliefs, and intentions. These programs also included training in skills needed to effect behavior change such as condom use, sexual assertiveness and communication, and risk-reduction problem-solving.

The group interventions evaluated in these studies have been intensive, involving programs of from five hours to as much as thirty 90-minute group sessions. However, they have also produced convincing evidence of sexual risk-behavior change: increased levels of condom use, increased rates of sexual abstinence, and more favorable attitudes, intentions, and skills for avoiding risk. Of practical relevance for public health education efforts, most of these successful HIV prevention programs for at-risk adolescents were conducted in community settings, including shelters and inner-city agencies that serve runaway youth (4), schools (7), and medical and social service programs serving disadvantaged young people (5).

Rates of new HIV infection remain high among minority men who have sex with men, especially in those who are young (8). Although many prior trials of HIV risk-reduction interventions have included some minority gay men, few interventions have been specifically directed at minority men who have sex with other men (MSMs, a group that includes men who do not define themselves as gay or bisexual). In a recent project, Peterson and colleagues (9) recruited 318 African-American gay and bisexual men, randomly assigning participants to a 9-hour, a 3-hour, or a no-intervention control cohort. Men in the 3-hour program participated in a single 3-hour workshop, while men in the 9-hour program attended three different workshops. The interventions included risk-reduction skills-building and elements designed to encourage safer sex -- but framed these exercises around discussions that reinforced positive messages related to pride in black gay identity.

When followed up 12 months after the interventions, men in the control cohort and those who attended a single 3-hour workshop showed little evidence of risk-behavior change. However, men who received the three-workshop sequence were found to be less likely to report engaging in unprotected anal intercourse. The proportion of men in this group who reported having high-risk sex declined from 46% at baseline to 20% (9). This suggests that brief, single-session workshop interventions have less effect and carry less behavior change benefit than more intensive multiple-session programs.

Community-level approaches hold promise for interventions focused on specific subpopulations

All of the interventions discussed to this point have been small-group or workshop programs that promoted behavior change among the individuals attending them. Another level of intervention focuses on communities at large. Community-level HIV prevention programs attempt to produce change within entire populations that are characterized by high levels of risk behavior and vulnerability to HIV and AIDS. Controlled-outcome trials of community-level interventions have been less frequently reported in the literature than small-group programs because their scale makes community interventions difficult to undertake. However, several recent controlled trials have demonstrated the effectiveness of particular HIV prevention interventions in particular urban populations.

The Centers for Disease Control and Prevention recently completed demonstration projects of community-level HIV prevention interventions for vulnerable population segments, among them injection drug users and their sexual partners, commercial sex workers, youth in high-risk situations, and nongay-identified MSMs in five U.S. cities (10). The community interventions used in this effort employed outreach workers and "focused media" -- including pamphlets, newsletters, and other materials -- to reach members of the target populations with messages that provided AIDS risk education, promoted norm changes, and encouraged taking steps toward risk reductions.

The C.D.C.'s interventions were evaluated by means of street interviews with members of the target populations, conducted in the communities that received intervention, as well as in several comparison neighborhoods. Although the effects were not uniform across all target populations or communities, the AIDS Community Demonstration Projects produced evidence of increased readiness to enact behavior change and increased action-taking to reduce risk among members of the targeted community populations.*

Finally, Sikkema, Kelly, and colleagues have presented the preliminary outcomes of a randomized, community-level HIV prevention trial undertaken among women living in 16 inner-city public housing developments (11). In half of the housing developments, women who were well-liked and considered opinion leaders among their peers attended risk-reduction workshops, recruited other women in the housing development to attend those workshops, and then organized community social events for women in the housing developments to strengthen peer norms, attitudes, and confidence in making behavioral changes.

Before and three months following the sustained 6-month community intervention, women in all 16 housing developments were surveyed to assess their HIV risk characteristics. Among women in the community intervention housing developments, levels of AIDS knowledge, risk-reduction attitudes, conversations between women about AIDS, and condom use all exhibited positive changes. For example, condom-use rates nearly doubled -- from 22% to 44% -- among the women in the intervention housing developments who reported having attending risk-reduction program activities.


There is now convincing evidence that HIV prevention interventions -- at the levels of group and community programs -- can help inner-city and indigent adolescents and adults make behavior changes to reduce their risk of contracting HIV infection. Although the programs reviewed here differ from one another in some respects, they also have much in common. All moved beyond simply providing AIDS education and addressed attitude, skill development, and motivational dimensions needed to promote behavior change. All of the programs described here were relatively intensive, whether they involved group or workshop programs or were sustained community-level interventions.

The evidence suggests that brief HIV prevention programs are likely to produce only brief effects. The studies cited here show that effective interventions take time and must be tailored to the risk issues, barriers, and cultural values of the target population. At the same time, it is clear that much remains to be done and, given the increasing incidence of HIV infection among the inner-city poor, these prevention needs are urgent. Some urban populations with high prevalence of HIV -- including the homeless, the severely mentally ill, and adolescents and adults with crack and cocaine use problems -- are still not adequately represented in HIV prevention trials, and their risk issues are not yet well understood. And, for all the attention given to HIV prevention among women, few projects have focused on heterosexual men at high risk of infection -- despite evidence that these men often decide whether condoms will, or will not, be used.

Finally, the impact of behavior change interventions for inner-city and underserved populations will be advanced when interventions are carefully tailored to address the social and cultural circumstances facing these groups, are integrated into inner-city social service and health care systems, and better recognize that risk for HIV occurs in the context of other social health problems confronting the disadvantaged.

*To obtain a copy of the C.D.C.'s recommendations and report on community-level prevention of HIV among high-risk populations, send a check or money-order for $3.00 to: Massachusetts Medical Society; C.S.P.O. Box 9120; Waltham, MA 02254-9120. Be certain to include your name and address.


  1. Kelly, JA, Murphy DA, Washington CD, Wilson TS, Koob JJ, Davis DR, Ledezma G, Davantes B. The effects of HIV/AIDS intervention groups for high-risk women in urban clinics. Am J Pub Health 1994; 84: 1918-22.
  2. Hobfoll SR, Jackson AP, Lavin J, Britton PJ, Shepherd JB. Reducing inner-city women's AIDS risk activities. Health Psychol 1994; 13: 397-403.
  3. DiClemente RJ, Wingood GM. A randomized controlled trial of an HIV sexual risk-reduction intervention for young African-American women. JAMA 1995; 274: 1271-6.
  4. Rotheram-Borus MJ, Koopman C, Haignere C, Davies M. Reducing HIV sexual risk behaviors among runaway adolescents. JAMA 1991; 266: 1237-41.
  5. Jemmott JB, Jemmott LS, Fong GT. Reductions on HIV risk-associated sexual behaviors among Black male adolescents: Effects of an AIDS prevention intervention. Am J Pub Health 1992; 83: 372-7.
  6. St. Lawrence JS, Brasfield TL, Jefferson KW, Alleyne E, O'Bannon RE, Shirley A. Cognitive-behavioral intervention to reduce African American adolescents' risk for HIV infection. J Consult Clin Psychol 1995; 63: 221-37.
  7. Walter HJ, Vaughn RD. AIDS risk reduction among a multiethnic sample of urban high school students. JAMA 1993; 270: 725-30.
  8. Update: Trends in AIDS among men who have sex with men -- United States, 1989-1994. MMWR 1995; 44: 401-4.
  9. Peterson JL, Coates TJ, Catania JA, Hauck WW, Acree M, Daigle D, Hillard B, Middleton L, Hearst N. Evaluation of an HIV risk reduction intervention among African American homosexual and bisexual men. AIDS 1996; 10: 319-25.
  10. Community-level prevention of human immunodeficiency virus infection among high-risk populations: The AIDS Community Demonstration Projects. MMWR 1996; 45: 1-24 (Supplement).
  11. Sikkema KJ, Kelly JA, Heckman TG, Roffman RA, Solomon LJ, Cargill V, Winett RA. Effects of community-level behavior change intervention for women in low-income housing developments. XI International Conference on AIDS, Vancouver, B.C., July 1996. Abstract Tu. C. 453.

Jeffrey A. Kelly, Ph.D., is Director of the Center for AIDS Intervention Research, Medical College of Wisconsin, Milwaukee, WI.