by Rhonda Mullen

ome 15 years ago, when Ralph DiClemente began his study of adolescent risks of HIV infection, he had access to a copier but little else. At night after his full-time job and on weekends, he copied the seven-p age questionnaire he developed for high school students in San Francisco. With no collating features on the copier, he produced fewer than 100 questionnaires a night. Even after he completed the copies and distributed them to more than 1,500 students, the going was slow as he entered the data by hand on a computer.

Despite these obstacles, DiClemente, then a cancer prevention and control research fellow with the American Cancer Society, doggedly pursued this study. Early on, he recognized its importance. "Given the mode of transmission of HIV and the high rate of sexually transmitted diseases among adolescents in San Francisco, it logically followed that this disease posed a risk to this population," DiClemente says.

In fact, when DiClemente analyzed the data, he found that the teens who filled out the questionnaires in a dozen San Francisco schools did indeed have high risk factors for HIV. First of all, as a group, they knew little about the AIDS epidemic. Their use of condoms was low, and they perceived themselves to be invulnerable to disease. Finally, AIDS, they reported, was a disease that affected gay men, certainly not them.

While scientific evidence since has proved these conclusions wrong, at that time, the beliefs of these high school students weren't all that different from those of the research establishment. There was little interest in studying adolescents as a pote ntial risk group for AIDS, says DiClemente. There was certainly little funding, and HIV education for teens was sparse and controversial.

The results of DiClemente's first, bare bones study made a dramatic impact on the direction of AIDS research when it appeared in the American Journal of Public Health in 1986. "It rang a bell," he says. "It was the first article in a major journal alerting the field to the risk of adolescents to AIDS. It helped us realize that our health education had to reach this group if we were to have an impact on slowing the epidemic." Soon after, behavioral scientists initiated early AIDS education efforts for adolescents.

For DiClemente, the study also became the germ of subsequent investigations focusing on adolescent risk-taking. He became so interested in adolescent decision-making about health that upon completing his cancer prevention fellowship, he abandoned that line of research to instead focus on a career in HIV prevention.

That career has been marked by a series of prestigious appointments at the University of California/San Francisco medical center, Bayview-Hunter's Point Foundation Multicultural Research Institute, the University of Alabama's Center for AIDS Research, and as of last summer, the Rollins School of Public Health. It has been distinguished by the publication of one landmark study after another. Just this year, he was awarded the Society for Adolescent Medicine's Visiting Professorship in Adolescent Health Research, allowing him to carry his expertise to campuses across the country.

More important than identifying at-risk groups was developing interventions to motivate individuals to adopt health-promoting behaviors, DiClemente says. In working with adolescents, he knew that peer norms influence behavior. "To get teens to adopt he althy habits, we have to change their perceptions. Madison Avenue has known this for a long time. Health educators, too, need to bring marketing skills to bear on changing behaviors."

There he was to find not only a fertile ground for research but also a collaborator who shared a similar passion for HIV prevention. Gina Wingood, a behavioral scientist with an MPH from Berkeley and a Doctor of Science from Harvard, brought research i nterests in women's health to DiClemente's efforts to design HIV interventions for adolescents. They were to become a close team.

In an early collaboration, they developed an HIV prevention intervention called SISTA (Sisters Informing Sisters about Topics on AIDS). The program's goal was to enhance consistent condom use for sexually active, heterosexual, African-American women ag ed 18 through 29 years. They chose this group because surveillance data at the time indicated that African-American women were substantially more likely to be diagnosed with AIDS than white women. Also, African-American women had markedly higher HIV preva lence rates.

They theorized that HIV interventions designed for gay men in the early 1990s would be inappropriate for these women. Instead they turned to social cognitive theory and the theory of gender and power to design an intervention tailored to this understud ied and underserved population. The intervention involved five sessions focusing on gender and ethnic pride, HIV risk-reduction information, sexual assertiveness and communication training, condom use skills, and cognitive coping skills.

"We started with trying to increase the women's sense of self-worth," says Wingood, "because we realized we were not going to change behavior unless we made an impact on self-esteem." The sessions also used role playing to allow young women to learn to talk with partners about sexual preferences, including condom use.

In the randomized, single-blind trial, published in the Journal of the American Medical Association (JAMA) in 1995, the researchers assigned women to either the social skills intervention, a single-session HIV education program, or a delay ed HIV education session. The study demonstrated significant changes in important areas of HIV prevention for those women who participated in the skills intervention. These positive changes occurred in several areas, including interpersonal skills, cognit ive coping skills, and most important, in their consistent condom use.

The findings suggested that culturally appropriate, gender-specific interventions that emphasize social skills training may be effective HIV prevention strategies. The results also spearheaded the next course of study for the researchers.

One of these interventions builds on the SISTA program DiClemente and Wingood developed in San Francisco. The new intervention, SiHLE (a Swahili word for beauty that stands for sistering, informing, healing, living, and empowering) seeks to lower the n umber of STDs and pregnancies among Alabama teens aged 14 to 18. Delivered by peer educators on Saturday mornings in a Birmingham clinic, it begins the educational program by enhancing ethnic and gender pride. Like the earlier intervention, SiHLE presents information on HIV especially tailored to this age group, and it ends with a fourth session that discusses coping strategies for problems such as abusive partners or alcohol and drug use. SiHLE recruits the peer facilitators from local schools to serve a s role models and teachers. Unlike its predecessor, this program includes a long-term follow-up of 18 months as well as multiple assessments of STDs using newly developed and noninvasive nucleic acid amplification technology.

The broader Emory community and colleagues at the School of Public Health give them an opportunity to continue their theory-based interventions with Atlanta adolescents. "The schools of public health and medicine and the Grady health system have a rich history of collaborative research," says DiClemente. "We hope to continue and expand this commitment with deep investments in HIV and STD prevention research."

For example, he and Wingood are currently collaborating with colleagues at Emory School of Medicine in a multisite grant that includes adolescent males and females as well as pregnant teens. The new interventions seek to reduce HIV risk behavior and th e incidence of STDs.

At Emory, they've also embarked on a new trajectory of studies that involve media. In a first foray, they found that teenage girls who were pregnant and teenage girls who engaged in HIV-related sexual risk-taking viewed more television programs that il lustrated violence, depicted abusive acts against women, or were sexually demeaning toward women.

Sponsored by a grant from the the Office of AIDS, National Institute of Mental Health, the study recruited 240 African-American women from 14 to 18 years of age from adolescent health clinics. Participants completed surveys about their television viewi ng habits and sexual history. They answered questions on the number of hours they watched television, the percentage of programs they viewed that illustrated violence, the frequency with which they used condoms, the occurrence of multiple sex partners, an d whether they had ever had an STD. Participants also received a pregnancy test.

In analyzing the data, the researchers found that the content of television shows was associated with either being pregnant or HIV sexual risk-taking. "It is not the amount of television viewing per se that is associated with behavior and adverse child hood experiences but the sexual and violent content of the programming that influences the teenage girls' risk for pregnancy, STDs, and HIV," says Wingood. "Adolescents are particularly influenced by the norms created by television. Media reinforce their reality, creating situations where life imitates art."

For many teenagers in underserved communities, "television is their portal to the world," says DiClemente. "Watching TV often reinforces violence toward women and promotes dangerous and unhealthy sexual norms and attitudes."

JAMA recently printed an abstract of these findings, which Wingood also presented at the most recent American Public Health Association conference. She will build on the work with DiClemente through observational studies on media, includin g movies, radio, magazines, and music videos. They want to develop an intervention that considers how media can be used to promote the adoption and maintenance of healthy lifestyles. They plan to deliver such a broad-based intervention using peers to begi n to change community norms.

Changing norms is no small undertaking, they realize. The work is slow going. Their studies are labor intensive and time-consuming. "Changing behaviors is not like turning on a light," DiClemente says. "You need to be receptive to partnering with commu nities to see what ideas and skills they bring to an intervention. As a field, we have to collaborate more closely with communities and community-based organizations as well as a broad array of social, medical, and behavioral scientists to create truly in terdisciplinary research efforts."

Neither DiClemente nor Wingood are likely to give up the risky business of changing the high-risk behaviors of adolescents to healthy habits. One girl at a time, these behavioral experts have seen the difference that an intervention can make on individ ual lives.

There are collateral benefits of these projects. "The success of these interventions is not measured only in disease averted," says DiClemente, "but in changing the trajectory of adolescents' lives. The growing self-esteem and self-confidence of the gi rls in these interventions often translates into healthier lifestyles and more fulfilled and productive futures."

Wingood smiles when she talks about the success of the peer educators. Many of these girls, most of whom had no aspirations for an education beyond high school, are now bound for college. One has even applied to an Ivy League school. That's a risk wort h taking.