Dr. Collins is affiliated with the New York State Psychiatric Institute and the Department of Psychiatry and the Department of Epidemiology, Columbia University, 722 West 168th Street, Room 1713, New York, NY 10032 (e-mail: firstname.lastname@example.org). Ms. Mestry and Dr. Lindegger are with the Department of Psychology, University of KwaZulu Natal, Pietermaritzburg, KwaZulu Natal, South Africa. Dr. Wainberg is with the New York State Psychiatric Institute, New York City. Mr. Nzama is with the Medical Research Council, Durban, KwaZulu Natal, South Africa. Preliminary results of this study were presented at the annual meeting of the American Psychiatric Association, New York City, May 1-6, 2004.
More than five million South Africans are HIV positive (1). Although the devastating results of the epidemic have prompted prevention and treatment initiatives for many populations, interventions for people with mental illness are lacking. Preliminary findings of HIV prevalence in two South African psychiatric facilities suggest that the epidemic also threatens this group. A prevalence of 9 percent in a state psychiatric facility and 29 percent in a psychiatric ward at a tertiary care hospital affirms the urgent need for education and intervention (2,3).
Although most South African health care providers know about AIDS, gaps exist between providers' AIDS awareness and HIV knowledge and the transfer of interventions to vulnerable groups. These disparities are particularly visible in marginalized sectors of the health care system such as the psychiatric institution. Although HIV risk has been well documented among people with serious mental illness in several settings, implementation of HIV prevention interventions often depends on the attitudes of mental health care providers toward sexual relations and HIV among people with mental illness and their skills for conducting interventions (4,5,6,7,8). Studies suggest that lack of knowledge about the disease, stigmatizing ideas, denial of sexual activity, and institutional barriers limit providers' readiness to respond to HIV prevention and care needs in psychiatric settings (4,5,9,10,11,12).
In the South African setting, political and social forces have also shaped providers' responses to the AIDS epidemic (4). The explosion of the AIDS epidemic coincided with South Africa's transition to democracy and development of mental health reforms. In 1996 a national consultative forum on mental health policy discussed issues related to the management of HIV infection in psychiatric hospitals (13). The consultants identified a human rights dilemma. Persons infected with HIV had the right to avoid being subjected to discrimination (by not being separated from noninfected patients) when they were in a psychiatric institution. Patients without HIV infection had the right to avoid it. This schism presented an important challenge to the management of HIV infection in psychiatric hospitals, but no specific guidelines were developed. To complicate matters further, AIDS prevention in the South African psychiatric setting requires enlisting the commitment of providers who may themselves be infected with HIV, in desperate need of services, or fearful of becoming infected, or they may be assisting in the care of ill family members or struggling with the stigma related to HIV.
This article describes the development and evaluation of a contextually relevant AIDS education program for mental health care providers at three institutions in South Africa. We used a pre- and posttest (no control group) design to evaluate whether teaching providers about HIV transmission and care, rehearsing skills for communicating with patients about sex, and addressing providers' anxieties about the AIDS epidemic in a 1.5-day workshop would change attitudes that could ultimately facilitate their ability to include HIV prevention among the services they provide to patients.
The study was approved by the institutional review boards of the University of KwaZulu Natal and the New York State Psychiatric Institute. All study participants gave informed consent.
The training activities took place at three public mental health facilities. Hospital 1 admits general psychiatric patients for acute care. Hospital 2 receives forensic patients who typically remain inpatients for extended periods. Hospital 3 cares for adults and children with developmental disorders.
We entered a setting in which growing numbers of staff members were dying of AIDS. Funerals had become so frequent that employees could no longer attend those that occurred during the workday and still adequately staff the hospitals. Yet few staff members openly acknowledged AIDS as the cause of these deaths.
Although staff and patients needed to talk about HIV and AIDS, doing so was not straightforward. Traditional isiZulu cultural norms predominated and discouraged talking about sex across gender and age groups. For example, many providers acknowledged that they did not discuss sexuality with their own children, as traditional culture deemed doing so inappropriate. A large portion of the white South African community strongly espoused the Calvinist tradition, which emphasized male authority, "`pure' New Testament principles, rigid austerity, and strictness in conduct and morals" (14). Administrators feared that talking about safer sexual activity might stimulate greater rates of sexual activity among inpatients. Finally, the South African Mental Health Act of 1973 stated that sex between psychiatric inpatients was illegal. Although a revised act was being introduced, the legacy of the old act (such as concerns that providing condoms condoned sexual activity) remained. At the same time, the growing human rights culture under the new South African constitution recognized the rights of all citizens, including psychiatric patients, to sexual expression. These variables led to considerable guardedness about recognizing, allowing, and managing sexuality in psychiatric patients.
We established a working group composed of four to nine providers at each facility. The group composition reflected the predominance of nurses and the smaller number of social workers, occupational therapists, and psychologists at the three institutions. Three focus groups with the core group members revealed that providers felt overwhelmed by the illness and deaths caused by AIDS at the hospital and outside. The participants identified a range of issues to address, including integrating HIV prevention into the daily work routine, developing policies on sexual behavior, managing violent patients and HIV risk, addressing hospital liability for HIV infection, managing sex in the institution, addressing the stigma of mental illness and HIV, and identifying staffing needs for HIV support and treatment. A major challenge for staff was to simultaneously recognize that patients have a right to sexual expression and that the staff are responsible for providing the necessary protection from HIV infection. The training curriculum arose from these needs.
We developed a nine-session curriculum that aimed to impart relevant information about HIV and to introduce skills for communicating with patients in order to increase confidence and efficacy of mental health care providers. We applied social-cognitive theory, which emphasizes provision of information, skills development, self-efficacy, and social support (15). We integrated risk assessment, role-playing, modeling, and problem-solving techniques throughout the workshop. Three members of the research team and four local medical and legal experts conducted the training.
Session 1, "What Is the Situation?" presented local HIV prevalence data and reviewed the basics of HIV infection and transmission, defined terms such as opportunistic infection and CD-4 cells, and explained concepts such as viral load.
Session 2, "Am I at Risk?" personalized the epidemic by inviting participants to complete a personal risk assessment. The exercise was followed by a guided discussion of stigma related to HIV. Providers were encouraged to brainstorm reasons for stigma as well as solutions for it.
Session 3, "What Can We Do?" began to offer solutions. We invited two physicians from the Communicable Disease Clinic at a local public hospital to share with providers the treatments available in the community for opportunistic infections and to describe what antiretroviral therapies were accessible. The physicians reviewed costs of medication, eligibility for antiretroviral therapy, medical insurance plans and their coverage for antiretroviral therapy, and nutrition and diet. They showed how to make referrals to the clinic.
Session 4, "What More Can We Do?" addressed the need to talk about sex with patients in the hospital in order to explain the institution's sexual activity policy and to teach sexual risk reduction. The session's exercises acclimated providers to using sexual terms and taught them to identify terms used by most of their patients.
In session 5, "Talking About Sex," providers demonstrated their approaches to discussing HIV with patients. Next, trainers modeled empathic interactions with patients. The group identified key messages to convey about sex and HIV and practiced delivering these messages in role plays.
In sessions 6 and 7, "Legalities 1 and 2," local human rights lawyers answered questions about confidentiality, disclosure of HIV status of psychiatric patients, and rights to privacy and protection for staff and patients. They explained the institution's liability with respect to HIV testing and prevention and provided guidelines for developing policies for sexual activity in the institution.
Session 8, "Our Daily Routine," focused on integrating HIV prevention messages into the ward routine. Providers identified which wards were amenable to integrating HIV prevention activities and the kinds of activities that would be appropriate.
Session 9, "Supporting Our Community," which was the final session, facilitated a discussion of staff support needs, strategies for sustaining the core group, and service provision to staff members affected by or infected with HIV.
We conducted the training workshops at all three institutions in April 2002. Each workshop lasted 1.5 days. Core group members recruited participants from their institutions. All participants attended the training workshop at their facility.
Participants completed a baseline assessment at the start of the workshop and a posttest immediately after the last workshop session.
We collected demographic information that included participants' ages, ethnicity, first language, level of education, job title, and reasons for attending the training.
We adapted the AIDS Stress Scale (16,17) for use in the psychiatric setting. The instrument assessed comfort with AIDS patients and friends of people with AIDS and perceived knowledge of the physical and emotional needs of AIDS patients. We expanded the instrument to include two additional questions derived from concerns expressed by the providers in the psychiatric setting: "I have sufficient knowledge to deal with the emotional needs of my coworkers who may have HIV/ AIDS" and "I have sufficient knowledge and skills to be able to convey messages about HIV/AIDS to mentally ill patients." Thus a total of five items assessed perceived knowledge (Cronbach's α=.84) on a 5-point Likert scale. Six items assessed comfort with AIDS patients (Cronbach's α=.66), such as "It will be hard for me to deal with a large number of HIV/AIDS patients in the future." Possible scores range from 1 to 5, with higher scores indicating greater perceived knowledge or greater comfort.
Factual knowledge was assessed with 12 items to which participants responded yes, no, or "don't know" (Cronbach's α=.67). Questions such as "All pregnant women infected with HIV will have babies born with AIDS" were developed on the basis of the topics of the focus group discussions.
Forty-four staff members (32 women and 12 men) from the three institutions participated in the training workshops. Forty-two participants returned baseline and posttest evaluations. The larger number of women than men in the study reflects the predominance of female staff members at the institutions. Participants ranged in age from 24 to 58 years (mean±SD=39±8.68). IsiZulu was the predominant first language of 17 participants (41 percent), and English was the predominant first language of 19 participants (45 percent). Fourteen participants came from hospital 1, 18 from hospital 2, and 12 from hospital 3.
Although all participants were staff members from the three institutions, seven employees who did not provide direct services to patients were asked by their supervisors to participate. Thirty-eight participants (88 percent) reported knowledge and skills building as their primary reason for attending the workshop. Twenty-five of 38 participants (66 percent) described themselves as comfortable or extremely comfortable with discussing sexual practices with patients. Thirty-three of 40 participants (83 percent) reported feeling comfortable or extremely comfortable discussing HIV prevention with patients; however most (41 participants, or 93 percent) reported that they did not provide HIV risk assessment or risk reduction interventions to patients. The participant group was largely a self-selected group with commitment to and perceived competence in addressing the issues discussed in the workshop.
Participants reported greater comfort with HIV and AIDS patients on the posttest than at baseline (baseline mean±SD=3.37±.56; posttest 3.69±.61; t=3.71, df=41, p=.001; effect size d=.54). The greatest change in attitude was observed in response to the question, "I am comfortable working in an environment where people may be infected with HIV/ AIDS" (baseline 3.29±.89, posttest 3.75±.99; t=2.55, df=41, p<.05; d=.48).
Participants reported an increase on posttest versus baseline in perceived knowledge about managing HIV and AIDS patients (baseline 3.11±.75; posttest 3.87±.54; t=5.86, df=41, p<.001; d=1.17). Providers showed the greatest change in attitude in response to the question, "I have sufficient knowledge to deal with the emotional needs of my coworkers who may have HIV/ AIDS" (baseline 2.90±.91; posttest 3.74±.77; t=4.90, df=41, p<.001; d=.99).
Baseline levels of knowledge were relatively high (73±19 percent correct) but showed significant improvement on the posttest (86±13 percent correct; t=5.53, df=41, p<.001; d=.74). Among the items on which participants showed greatest improvement was "All pregnant women infected with HIV will have babies born with AIDS" (baseline 57±50 percent correct, posttest mean=90±30 percent correct).
The workshop was well received by the participants. They engaged in conversations on culturally sensitive topics, including homosexuality and vaginal, anal, and oral intercourse. Exercises that concerned us before the workshop in fact facilitated dialogue. In particular, we wondered whether using the personal risk assessment tool might be painful in a setting where HIV affected so many people through the illness and death of friends, family members, and coworkers or as a result of their own infection. On the contrary, this exercise stimulated discussion, including coverage of AIDS-related stigma. Many providers took the personal risk assessments home and invited their partners to complete them.
Items in the assessment that showed the greatest change from pre- to posttest were those related to working in an environment where others may have HIV and the associated emotional issues. Given the visibility of HIV and AIDS among the staff, any intervention that addresses HIV prevention and care will need to address staff members' worries about their own well-being and the well-being of their colleagues. The results suggest that the intervention targeted this particular area of anxiety.
The evaluation of the training intervention has several limitations. The participants consisted of a small, self-selected group, we used no control group, and we did not follow providers past the immediate postintervention assessment. Alpha coefficients for two of the measures were in the .60 to .70 range, suggesting that further development and refinement of measures for use in this setting are needed. Finally, our study lacked a longitudinal follow-up to see how the attitude changes were reflected in clinical practice and integration of HIV interventions in these settings. Nevertheless, the gains in knowledge and increase in positive attitudes observed after the intervention were encouraging and form a platform from which focused intervention can begin.
This study was supported by a supplement to grant R01-MH-58917 and by grant K01-MH-16091, both from the National Institute of Mental Health. The authors thank Ms. Johan Byssainthe for her contributions to this work and the members of the International Core of the HIV Center for Clinical and Behavioral Studies (New York State Psychiatric Institute and Columbia University) for their helpful comments.