When they analyzed the results of a study conducted in Tanzania to assess how care and support for HIV-positive people can contribute to prevention efforts, Tanzanian and U.S. researchers were surprised.
"Our results reinforce the importance of voluntary HIV counseling and testing, at least in the short term, and that's not necessarily what we expected to find," said Dr. Joan MacNeil of Family Health International (FHI), which initiated the study in 1996 through the AIDS Control and Prevention (AIDSCAP) Project.
The randomized controlled study was designed to detect any differences in behavior change reported by HIV-positive people who received continuing support from counselors and by those who received only post-test counseling and regular medical services. But researchers found minimal differences between the two groups.
Instead, the study revealed significant levels of behavior change among all study participants. Reports of two important HIV prevention measures--condom use and having fewer sex partners--increased in both the experimental and control groups during the six-month study.
Like the more extensive trial of counseling and testing sponsored by FHI and the United Nations Programme on HIV/AIDS in Kenya, Tanzania and Trinidad, the study in Tanzania's Tanga district showed that learning one is HIV-positive and receiving professional post-test counseling can be a powerful motivator. However, the results also indicate that this effect on behavior may be short-lived.
"Even though condoms continued to be readily available throughout the district, condom use started to decline among members of the control group between three and six months after their enrollment in the study," said Dr. MacNeil, the associate director for care and research in FHI's HIV/AIDS Prevention and Care Department. "This relapse into unsafe behavior suggests that additional counseling and care and support at this later point may be critical in sustaining behavior change."
The Tanga Study
The Tanga AIDS Working Group (TAWG), one of dozens of local groups that implement the Tanzania AIDS Project with support from FHI, offers HIV counseling and testing at centers in the towns of Tanga, Muheza and Pangani. Positive results are not uncommon in the Tanga district, where an estimated 7 percent of the adult population is infected with HIV.
Even with such high infection rates, it was not easy to find enough people who met the study criteria of being HIV-positive, sexually active, 18 to 54 years old, and not seriously ill. Counselors report that most people are reluctant to get tested for HIV and many do not want to know their serostatus. As a result, most HIV tests are performed in the hospital after people have become ill.
By early 1997, however, 154 HIV-positive people had been recruited to participate--16 more than the 138 needed to allow researchers to make statistically meaningful conclusions.
All of the study participants had received post-test counseling no more than four weeks before their enrollment. They were randomly assigned to one of two groups: a control group with access to regular health services or an experimental group that received enhanced care and support.
For this study the researchers defined enhanced care and support as the services of a trained counselor at least once a month. The counselors, all public health workers from TAWG, provided counseling, condoms and--when necessary--referral for treatment. They also visited some participants' homes to counsel and educate family members.
Researchers and members of the TAWG agreed on this definition of care and support based on the needs of the community as well as budgetary considerations, Dr. MacNeil explained. TAWG staff members, who were spending most of their time managing AIDS-related illnesses and providing palliative care to the terminally ill, saw a need for a different kind of support for people in early stages of HIV infection.
"Since study participants were relatively healthy, we felt that counseling and education would be an appropriate level of support," Dr. MacNeil said.
All study participants were interviewed in Kiswahili at enrollment, after three months, and after six months to gather information for the study. A final round of interviews had been planned at 12 months, but the study ended after the six-month follow-up because of time and funding constraints.
Trained interviewers used a 57-item questionnaire that covered a wide range of issues, including discussing HIV with partners, HIV risk behavior and use of health services. Participants also talked about how they felt about their HIV status, its impact on their decisions about having more children, and the reactions of their families and communities.
Women at Risk
In many African countries, women are becoming infected with HIV at faster rates than men. This high risk of HIV is partly a result of women's greater biological vulnerability to the virus, but it is compounded by social and economic pressures that make it difficult for women to refuse unwanted sex or insist on condom use.
Such pressures are evident in the Tanga district. Women enrolled in the study were more likely than men to be widowed, divorced or separated. They also were more likely to be unemployed or to work in unskilled manual labor or small businesses. Incomes were low for all study participants, but the women earned significantly less: an average monthly income of U.S.$35, compared to $57 a month for men.
Earlier research conducted in Tanga to provide information for HIV/AIDS prevention planning found that women working in small businesses, such as street vending or market stalls, often supplement their income by having sex with customers.
Study data suggest a high risk of HIV infection among women in monogamous relationships. At the beginning of the study, only 16 percent of the women reported having more than one sex partner in the previous three months. Forty-five percent of the women--all of whom were HIV-positive--were married or widowed. Many of them said they had sought an HIV test after learning that a husband, partner or child was infected with the virus.
"In Tanzania, as in other parts of the world with a high prevalence of HIV, being in a marriage or a stable relationship is actually high risk for many women," Dr. MacNeil said.
Improving Health
Proponents of voluntary counseling and testing say that early detection of HIV infection can prompt people to take better care of their health. This seemed to be true in Tanga, where reports of illnesses by all participants decreased and researchers found significant increases in reported use of home care, folk healers, and medical, counseling and religious services.
"The significant decline in reported illness suggests that knowing one's serostatus can encourage those who are positive to seek care earlier for illnesses and symptoms that may be easily treated, such as diarrhea, rashes and headaches," Dr. MacNeil said.
The most striking decline was in reports of pain, burning or discharge on urination--common symptoms of sexually transmitted infection (STI). Although 35 percent of participants reported such symptoms at baseline, only 13 percent were experiencing them after six months.
The number of people seeking STI treatment after six months was also reduced, suggesting that symptoms declined because of effective treatment and preventive measures. This decline has important implications for HIV prevention because people are more likely to transmit HIV when they have an STI.
Knowledge of HIV serostatus had an impact on reproductive choice as well as reproductive health. Those reporting that they used any family planning method rose from one-fifth of all participants at the beginning of the study to almost one-half at the end as a result of increases in condom use. A substantial number of participants did not use any family planning method, however, because they wanted larger families.
Changing Behavior
Reports of condom use for family planning and disease prevention rose dramatically in both the experimental and control groups, with most of the increase occurring during the first three months. Fidelity to one partner also increased, as the number of people reporting more than one sex partner dropped throughout the six-month study.
Since these preventive measures were adopted by people in both groups at similar rates, the additional care and support given to one group did not appear to have played a role in reducing HIV risk behavior.
"Enhanced care and support in this early phase did not have a significant impact on behavior change, yet the study as a whole decreased risk behavior," Dr. MacNeil said.
The researchers attribute this behavior change to the quality of the post-test counseling all participants received and the easy access to condoms and information about condoms provided by an energetic social marketing project.
Condom use did begin to decline among the control group after three months, suggesting that care and support may be more important after the initial shock of a positive test result subsides.
"Although this decline was not statistically significant, there was really a drop-off," Dr. MacNeil said. "I would speculate that if we could have followed them at 12 and 18 months, it would have been significant."
Also troubling was the number of people in both groups who continued to put their partners at risk of HIV and themselves at risk of reinfection. After six months, 28 of these HIV-positive individuals reported having more than one sexual partner, and many others said they had not used condoms.
Such findings are not unusual and probably reflect the difficulty most people have in accepting a positive test result, Dr. MacNeil noted. "It just takes time," she said. "And this shows that even after the first six months, there is still a lot of work to be done."
Opening Up
Another important preventive measure--revealing serostatus--proved difficult for all participants. At the beginning of the study, four weeks after learning of a positive test result, only 29 of the 154 people had told anyone about it.
"Most of them wanted to get on with their lives and not even talk about HIV," Dr. MacNeil said.
The main reason cited for telling others about a positive result was the need for support. Only a handful of people--three during the first round of interviews and two during the third round--mentioned protecting partners as a reason for revealing their status.
"Counselors emphasize the preventive aspects of knowing one's serostatus, but sometimes we feel that our clients were infected when they had no knowledge of AIDS and HIV," said Dr. Firmina Mberesero, TAWG chairperson. "It takes them some time to realize their role in preventing the spread of HIV."
The number of people who had disclosed their serostatus increased by almost 64 percent during the study, with no significant difference between the experimental and control groups. After six months, however, 24 participants had not told anyone that they were HIV-positive, and 80 hadn't told their spouses or other sexual partners.
"Even after six months there were a lot of problems with revealing status," Dr. MacNeil said. "People were worried about abuse, discrimination and being abandoned."
In fact, most responses to the news of a positive HIV test were surprisingly supportive in a community where HIV/AIDS is considered a "self-inflicted" disease and people dying of AIDS are often abandoned by all but their mothers or grandparents.
Participants in both groups also reported increasing emotional and economic support from spouses and partners. And when asked about their interactions with family members, such as eating or sleeping together and being included in discussions or cared for when ill, 91 percent of participants reported active inclusion in family life.
Although the enhanced care and support had no impact on sexual behavior, they appear to have had a positive impact on the response to HIV/AIDS in families and communities.
After the study, TAWG staff noted greater interest in HIV counseling and testing and greater support from other health care workers. "The majority of medical personnel thought that counseling was useless, but they are now in the forefront of sending clients to us," Dr. Mberesero said.
Another important sign of change is the support group formed by some of the study participants--the first of its kind in the Tanga district. Members meet to talk about their experiences and problems, visit each other, and help counsel other TAWG clients who have trouble accepting that they are HIV-positive.
Counselors believe that the study helped make community members more willing to discuss HIV/AIDS and more accepting of people living with the virus. "Ongoing care and support seem to bring people together and also bring issues out in the open," Dr. MacNeil said.
-- Kathleen Henry
