|Home | About | Journals | Submit | Contact Us | Français|
This study examined the impact of taking or not taking antiretroviral (ARV) medications on stigma, as reported by people living with HIV infection in five African countries.
A two group (taking or not taking ARVs) by three (time) repeated measures analysis of variance examined change in reported stigma in a cohort sample of 1,454 persons living with HIV infection in Lesotho, Malawi, South Africa, Swaziland, and Tanzania. Participants self-reported taking ARV medications and completed a standardized stigma scale validated in the African context. Data were collected at three points in time, from January 2006 to March 2007. Participants taking ARV medications self-reported a mean CD4 count of 273 and those not taking ARV self-reported a mean CD4 count of 418.
Both groups reported significant decreases in total HIV stigma over time; however, people taking ARVs reported significantly higher stigma at Time 3 compared to those not taking ARVs.
This study documents that this sample of 1,454 HIV infected persons in five countries in Africa reported significantly less HIV stigma over time. In addition, those participants taking ARV medications experienced significantly higher HIV stigma over time compared to those not taking ARVs. This finding contradicts some authors’ opinions that when clients enroll in ARV medication treatment it signifies that they are experiencing less stigma. This work provides caution to health care providers to alert clients new to ARV treatment that they may experience more stigma from their families and communities when they learn they are taking ARV medications.
AIDS stigma and discrimination have considerable influence on people living with HIV infection, particularly in sub-Saharan Africa, where the burden of AIDS is so significant (Phaladze et al., 2005). Stigma has developed into a major barrier to HIV prevention and care (Weiss & Ramakrishna, 2001). It is a deterrent to testing and treatment (Newman, 2002), which may result in delays to seek care and possibly increasing suffering and mortality. Many experts have concluded that programs must address stigma in order to be successful (L. R. Uys, 2000).
The development of antiretroviral (ARV) medications in the late 1990s altered the face of HIV/AIDS, transforming it from a fatal disease into a chronic illness. Many believed that as more people gained access to ARV treatment, and as the disease itself was transformed, stigma would decline. Such has been the case in Brazil, where a study of young people living with HIV (Abadia-Barrero & Castro, 2006) showed that access to ARV treatment reduced stigma. Thus, universal access has become an important intervention against stigma.
For some, however, the opposite has been true. Taking medications may indicate that someone is HIV-positive, so someone who does not wish to disclose his/her status may avoid taking medications. In a qualitative study with 25 adolescents and young adults (Rao, Kekwaletswe, Hosek, Martinez, & Rodriguez, 2007), half of the participants had missed medication doses for fear of disclosing their status and possible subsequent stigma. In a study of 204 HIV-positive people in the United States, Rintamaki and colleagues (Rintamaki, Davis, Skripkauskas, Bennett, & Wolf, 2006) similarly found that worry about disclosing one’s HIV status was a significant predictor of adherence to medications. However, a study in Botswana (Nam et al., 2008) found that acceptance of one’s HIV-status helped people avoid internalizing stigma and they were more adherent to their medications. These studies report the conflicting relationship between HIV stigma and outcome variables.
Various studies have also documented the negative impact of both treatment side effects and stigma on and individual’s ability to be adherent to medications (Hardon et al., 2007; Weiser et al., 2003). Not only can the side effects of ARVs be bothersome, but they can also cause visible side effects, or “stigma symbols,” which add to existing stigma that one may perceive or receive as a person living with HIV infection (Holzemer, Uys, Makoae et al., 2007). In a study of HIV-positive men and women on ARVs in Rwanda, participants noted visible body fat redistribution as a side effect of their treatment. The body fat redistribution negatively affected their quality of life, and could result in stigma and, ultimately, poorer medication adherence (Mutimura, Stewart, & Crowther, 2007).
Although the literature reports a great deal of research on adherence and stigma in various settings, there is little research on understanding the relationships between taking ARV medications or not, and HIV stigma in persons living with HIV infection. Program managers may assume that ARV therapy will improve people’s lives. However, if the side effects and potential resulting stigma are significant, people may decide to delay initiation of ARV therapy, or even forgo it altogether. If the second phase of The President’s Emergency Plan for AIDS Relief (PEPFAR) is to be successful in sub-Saharan Africa with further “roll out” of ARVs, program planners and managers need a greater understanding of the relationship between HIV stigma and taking ARV medications. The purpose of this study, therefore, was to examine whether there are differences in HIV stigma over time between persons who are taking ARV medications and those not taking ARV medications in five African countries.
A purposive convenient sample of 1,454 people living with HIV infection in Lesotho, Malawi, South Africa, Swaziland, and Tanzania was followed for one year in a longitudinal design that examined stigma change. Participants self-reported taking ARV medications, and completed a standardized stigma scale, which had been validated in the African context. The data were collected in three assessment periods, about six months apart, between January 2006 and March 2007. A two-group (taking or not taking ARVs) by three (time) repeated measures analysis of variance examined change in HIV stigma over time. This is part of a larger study on HIV/AIDS-related stigma in Africa (Dlamini et al., 2007; Greeff et al., 2008; Holzemer, Uys, Makoae et al., 2007; Holzemer, Uys, Chirwa et al., 2007; Kohi, 2006; Makoae et al., 2008; Naidoo et al., 2007; L. Uys et al., 2005).
The study protocol was reviewed and approved by all seven of the universities involved (see author list), providing protection of human subjects. As appropriate, permission for the study was also obtained from central and local government authorities. Participants received information about the background of the study, were informed that participation was completely voluntary, and were also told that they could withdraw at any time. They were also assured of the confidentiality of the information obtained. After receiving this explanation – in the local language of their country – each person who agreed to participate signed a written consent form. The participants received reimbursements for their transportation, and lunch was provided. The surveys were conducted either in English or in the local languages: English and Tswana in South Africa; Chichewa in Malawi; SeSotho in Lesotho; Swazi in Swaziland; and Kiswahili in Tanzania. Most participants completed the self-report surveys with some assistance from the research staff.
Each of the five sites had a goal of collecting data from 300 persons living with HIV infection, at three points in time over approximately 12 months (baseline, six months, and 12 months). Potential participants were recruited from HIV support groups or HIV clinics. If the person agreed to participate, the field worker made an appointment to meet at a convenient place and time to collect that data.
The participants completed a survey booklet with a series of instruments at each of the three points in time. Two instruments were used for the analysis reported here:
The multiple-item Demographic Questionnaire was used to obtain demographic and illness background data from the participants. These items included age, gender, education level, marital status, number of children, number of people living in the same home, work for pay, whether they were currently taking ARV medications, and the year of HIV diagnosis.
The HASI-P is a 33-item instrument that measures six dimensions of HIV-related stigma (verbal abuse, negative self-perception, health care neglect, social isolation, fear of contagion, workplace stigma) and total HIV stigma. The Cronbach alpha reliability coefficients for the scale scores range from 0.76 to 0.90. The instrument was developed based upon qualitative interviews with persons living with HIV infection in each of the five participating sites. Sample items from each of the six factors are presented in Table 4.
Responses to the questionnaires were entered into Statistical Package for the Social Sciences (SPSS) for Windows Version 15.0 software. Descriptive statistics (i.e., means, standard deviations, frequencies, and percents) were used to examine demographic characteristics of the sample and stigma. A two-group repeated measures analysis of variance was conducted to determine the relationship between taking and not taking ARVs and HIV stigma.
The total sample size was 1,454 and was approximately equally distributed by country (See Table 1). When comparing the people taking ARVs with those not taking ARVs with regard to demographic variables, the two groups differed significantly on age, marital status and setting. Younger participants, those not married, and those living in urban areas were less likely_to be taking ARV medications. CD4 counts were self-reported. Those not on ARVs reported a mean CD4 count of 418 (SD=225, n=314) and those taking ARVs reported a mean CD4 count of 273 (SD=196, n=440; t= 9.403, df=752, p=<0.000).
The overall participant attrition rate (see Table 1) or loss during the study was 31% (n=447) over the one year period. The attrition percentages varied by country with Swaziland losing the most and South Africa losing the least. Participant loss generally occurred between the second and third assessment period. A comparative analysis of participants who stayed and those who left the study over time demonstrated no differences in gender, age, marital status, taking and not taking ARVs, or setting.
The mean summed scores for total HIV stigma over time for people taking and not taking ARVs are reported in Table 3. At Time 1, those not taking ARVs reported more HIV stigma (M = 0.471) than those taking ART (M = 0.425). At Times 2 and 3, however, those not taking ARVs reported less HIV stigma (M = 0.410 and M = 0.250 respectively) than those taking ARVs (M =.0423 and M = 0.314, respectively) (see Table 3). Repeated measures ANOVA showed that there were significant differences in overall stigma levels over time (F=52.91; df = 2, 928; p<.0001), and there was a significant interaction between stigma and taking ARV medications (F=5.81; df = 2, 928; p<.006).
Six separate analyses of variance (ANOVAs) were conducted to explore where the change over time in HIV stigma occurred (See Table 4). Five of the six factors had a significant main effect of time, demonstrated by decreased total HIV stigma over time. Factor 6, Workplace Stigma, did not change over time. Three of the factor scores (Verbal Abuse, Negative Self-Perception, and Social Isolation) showed significant interactions, demonstrating that those taking ARVs reported higher scores on these three factors. There were no differences for the factors Health Care Neglect or Fear of Contagion. Since there was no significant main effect of change over time for Workplace Stigma, the interaction cannot be interpreted.
Total stigma scores were examined across time. HIV stigma was significantly different between Time 1 and 3 (p <.000) and Time 2 and 3 (p <.000) (see Table 3). The interaction between stigma and ARV use was significant between Time 1 and 3 (p <.001). The interaction of time and HIV stigma between the two groups is graphically presented in Figure 1. Participants not taking ARVs initially reported higher levels of stigma but, one year later, those participants taking ARVs reported greater stigma.
The results of this study indicate that the roll-out of ARV treatment is not a magic cure for HIV stigma, as many expected. However, based upon self-reported CD4 levels, this five country sample appears to have been appropriately treated with ARVs, given the significant difference in self-reported CD4 counts. HIV stigma significantly decreased over time for both groups, but those taking ARVs did report higher levels of total HIV stigma by Time 3. This might be related to the fact that it is difficult to hide one’s status when taking ARVs. Taking ARV medications requires regular visits to health care providers and daily intake of several pills. Frequent clinic visits, blood draws, and taking pills are aspects of antiretroviral therapy that are difficult to hide. The model of HIV stigma by Holzemer and colleagues (2007) outlines triggers that can identify a person as being HIV-positive and lead to HIV stigma. Given this theoretical work, it is not surprising that taking ARV medications increases HIV stigma for some individuals.
Another factor that might increase the stigma reported by those taking ARVs is the belief in communities that you only take ARVs when you are already very sick. Taking medications is seen as an indication of severity of illness that might, in itself, increase stigma, since it increases fear. This is a perception that needs to be addressed by health care providers doing health education. Taking ARV medication should imply that the individual is participating in appropriate health monitoring and seeking treatment to retard illness progression.
The finding that significantly more older people are taking ARVs can be explained by the fact that the medication is only taken in more advanced stages of the illness. There are a number of factors that might influence the finding that fewer people who were never married are using ARVs. One could speculate that single people are younger, and that this finding is linked to the previous one. Another factor is that many of the countries require a person who comes for ARV treatment to bring a “treatment partner.” Such a partner is probably easier to find for a person who is in a more permanent relationship. Being in a more permanent relationship might also motivate a person to be tested and get onto medication, while single people might find it easier to remain untested and not to confront their own status. It is more difficult, however, to explain why fewer people living in urban areas are on ARVs. This may be related to a sampling error, since more rural people than urban and peri-urban were lost from the sample.
Another interesting result of the study was the general decline in HIV stigma as reported by people living with HIV infection in the five countries. This is good news for all affected and infected and for the health services in general.
ARV treatment is currently the only medical hope for people infected with HIV who require therapy. To help people have a better quality of life with the disease in the long term, we must recognize that taking ARVs may increase HIV stigma and this increased stigma may create a barrier to continuing the treatment if it is not addressed by health care providers. Although community-level stigma may be decreasing, it may not be true at the individual level for persons taking ARVs who are identified because of their medications and thus become targets for stigma and discrimination. More information is needed about whether a fear of stigma may be keeping some HIV-positive people from seeking care and treatment.
This work was supported by NIH Research Grant #R01 TW06395 funded by the Fogarty International Center, the National Institute of Mental Health, and the Health Resources and Services Administration, U.S. Government.
Lucy N. Makoae, National University of Lesotho, P.O. Roma 180, Lesotho.
Carmen J. Portillo, University of California, San Francisco, 2 Koret Way, Box 0608, San Francisco, CA 94143-0608, USA.
Leana R. Uys, University of KwaZulu-Natal, School of Nursing, Durban 4041, South Africa.
Priscilla S. Dlamini, University of Swaziland, Faculty of Health Sciences, P.O. Box 369, Mbabane, Swaziland.
Minrie Greeff, North-West University, Potchefstroom Campus, Private Bag X6001, Hoffman Street, Potchefstroom, South Africa.
Maureen Chirwa, Kamuzu College of Nursing, University of Malawi, Private Bag 360, Blantyre, Malawi.
Thecla W. Kohi, Muhimbili University of Health and Allied Sciences, PO Box 65004, Dar es Salaam, Tanzania.
Joanne Naidoo, University of KwaZulu-Natal, School of Nursing, Durban 4041, South Africa.
Joseph Mullan, University of California, San Francisco, 3333 California Street, Box 0612, San Francisco, CA 94143-0612, USA.
Dean Wantland, Assistant Adjunct Professor, University of California, San Francisco.
Kevin Durrheim, University of KwaZulu-Natal, School of Psychology, Private Bag X01, Scottsville, 3209, South Africa.
William L. Holzemer, University of California, San Francisco, 2 Koret Way, Box 0608, San Francisco, CA 94143-0608, USA.