Share this article on:

HIV/AIDS and Antiretroviral Treatment Knowledge, Attitudes, Beliefs, and Practices in HIV-Infected Adults in Soweto, South Africa

Nachega, Jean B*; Lehman, Dara A*; Hlatshwayo, Dorothy; Mothopeng, Rachel; Chaisson, Richard E*‡; Karstaedt, Alan S

JAIDS Journal of Acquired Immune Deficiency Syndromes: February 1st, 2005 - Volume 38 - Issue 2 - p 196-201
Epidemiology and Social Science

Summary: A cross-sectional study of knowledge, attitudes, beliefs, and practices (KABPs) toward HIV and antiretroviral therapy (ART) was conducted in Soweto, South Africa, using a standardized validated questionnaire. Of 105 HIV clinic patients evaluated, 70% of whom were not on ART, 89% had good knowledge about the cause of HIV infection and 83% knew about modes of transmission. Fifty-nine percent reported they were not worried about ART side effects. Sixty-five percent agreed that missing ART doses can lead to disease progression. Ninety percent had disclosed their HIV serostatus to 1 or more persons, but only 62% of those with a current sexual partner reported having told that partner. Approximately 80% reported that if they were taking ART, they would not be worried about family or friends finding out. Forty-nine percent believed that ART can cure HIV, a belief that was associated with a low level of education (P < 0.001). Overall, knowledge of the cause of HIV/AIDS, modes of transmission, and importance of ART adherence was good in our study population. Further research is warranted to assess the extent to which this knowledge and attendant attitudes predict ART adherence levels. The low rate of HIV serostatus disclosure to sexual partners calls for multidimensional interventions to reduce HIV-related stigma.

From the *Department of International Health, Johns Hopkins University, Bloomberg School of Public Health, Baltimore, MD; †Adult HIV Outpatient Clinic, Chris Hani Baragwanath Hospital, Department of Medicine, University of the Witwatersrand, Johannesburg, South Africa; and ‡Department of Medicine, Division of Infectious Diseases, Johns Hopkins University School of Medicine, Baltimore, MD.

Received for publication October 20, 2003; accepted May 24, 2004.

Support provided by the Johns Hopkins Center for AIDS Research (CFAR) under a National Institutes of Health (5P50AI42855) and Johns Hopkins Family Health and Child Survival (HRN-A-00-96-90006-00) cooperative agreement with the United States Agency for International Development.

Parts of this paper were presented as an oral presentation at the 10th Conference on Retrovirus and Opportunistic Infections, Boston, February 10-14, 2003 (ThOrAb#169) and at the Second International AIDS Society Conference on HIV Pathogenesis and Treatment, Paris, July 13-16, 2003 (Ab#1218).

Reprints: Jean B. Nachega, Department of International Health, Johns Hopkins University, Bloomberg School of Public Health, 615 North Wolfe Street, Room W5031, Baltimore, MD (e-mail:

South Africa is laboring under one of the worst HIV/AIDS epidemics in the world. It is estimated that more than 1700 people are newly infected every day and that a total of 4.2 million South Africans (13%) are infected, placing South Africa first in the world in the number of infected individuals (UNAIDS Surveillance Report, 2000). Projections suggest that 6 million South Africans will be infected with HIV by the year 2005 and that without treatment intervention, the mortality rate of HIV/AIDS will reach 800,000 deaths per year by 2010.1,2 Increasing access to highly active antiretroviral therapy (HAART), which has been shown to decrease morbidity and mortality, could alter this trend substantially.3-5

Although South Africa's ability to afford and supply HAART to treat HIV infection has been limited in the past, a recent combination of drug price reductions, private donations, and increased access to generic drugs as well as the establishment of the Global Fund to Fight AIDS, Tuberculosis, and Malaria, has decreased the financial barriers to HAART in Africa. The effective delivery of HIV therapies, however, requires an understanding of patient knowledge, attitudes, and behaviors to design and implement appropriate behavioral interventions and treatment strategies.

Previous studies in South Africa from the early 1990s assessed perceptions and knowledge regarding AIDS among patients attending clinics for sexually transmitted diseases (STDs) and family planning.6,7 These studies showed inadequate awareness among at-risk individuals of the mode of HIV transmission as well as of the lack of an AIDS cure. Few studies have investigated the knowledge, expectations, and assumptions about HIV and antiretroviral therapy (ART) among HIV-infected patients in a resource-limited country or whether there are likely to be barriers to treatment adherence.

The present study investigated the knowledge, attitudes, beliefs, and practices (KABPs) of HIV-infected individuals regarding HIV/AIDS and ART in an HIV outpatient clinic in Soweto, South Africa. In addition, we examined potential barriers to ART adherence.

Back to Top | Article Outline


Study Design, Sample, and Enrollment

Between August and October 2002, we conducted a cross-sectional study of KABPs toward HIV and ART at the Chris Hani Baragwanath Hospital HIV Outpatient Clinic in Soweto, South Africa. This government-owned health facility, one of the largest hospitals in the world, serves the 3 million residents in this urban township southwest of Johannesburg. The population served by the hospital is predominantly black African, with a middle to low education level and socioeconomic status (SES). At their initial visit, all patients at this clinic received counseling about HIV, its treatment, and the importance of adherence. This counseling was repeated every 3 months on return visits to the clinic. A sample of 105 HIV-seropositive adults aged 18 years or older was randomly selected by going to the HIV outpatient clinic on random days and asking for volunteer participants, without consideration of length of time since diagnosis, disease severity, being on ART, or length of treatment experience on ART. Patients who agreed to participate and had read and signed the consent form were interviewed in their preferred language by 1 of 3 trained interviewers. This study was approved by the Committee on Human Research of the Johns Hopkins Bloomberg School of Public Health and by the Committee for Research on Human Subjects of the University of Witwatersand in Johannesburg, South Africa.

Back to Top | Article Outline

The Study Questionnaire

A previously validated questionnaire about KABPs that had been used in developing countries was adapted to the local culture and context of this study.8 Validated questions from the 1998 South African Demographic and Health Survey9 were also included. Demographic information sought included age, sex, race, marital status, employment status, and education. To assess knowledge, beliefs, and attitudes toward HIV and ART, participants were asked questions regarding modes of HIV transmission, disease progression, effects of ART, and concerns about ART. Perceived importance of ART adherence was assessed by asking the patients if they agree or disagree with the following statement: “If someone on ART misses doses of their medication, they may become very sick with AIDS.'' Questions regarding sexual and ART use practices included information about disclosure of HIV status, ART use, and condom use.

Back to Top | Article Outline

Knowledge of HIV, Antiretroviral Therapy, and Socioeconomic Status Scoring System

Composite scores for HIV knowledge, ART knowledge, and SES were created. An unweighted HIV knowledge score was calculated by adding the number of correct answers to the questions regarding HIV transmission, the differences between HIV and AIDS, and disease progression. ART knowledge scores were calculated by adding the number of correct answers to questions regarding benefits of ART and the consequences of not adhering to ART. SES composite scores comprised 1 point each for being employed, for having indoor tap water at home, for having electricity at home, and for having less than 3 people per bedroom.

Back to Top | Article Outline

Statistical Methods

Data were entered twice in separate files and then validated using Epidata software. Data analyses were performed using STATA Release 6 (Stata Corporation, College Station, TX). Statistical analysis included t tests for continuous factors, whereas χ2, odds ratios (ORs), and 95% confidence intervals (CIs) were performed for categoric factors. Logistic regression was used to determine the OR for multivariate cofactors. For an independent variable, 105 study participants evenly divided into 2 categories, there is 80% power to detect a difference of 20%, given a 6% event rate in 1 category versus a 26% event rate in the other and a type I 2-sided error of 5%. Because ART use and the counseling that accompanies it could bias patients' knowledge of HIV and ART, current ART use was used as a covariate in all multivariate regression analyses performed. All P values reported are nominal and 2-tailed. A P value of <0.05 was considered statistically significant.

Back to Top | Article Outline


Demographic Characteristics, Socioeconomic Status, and Education

A total of 105 interviews were conducted between August 21 and September 12, 2002. Table 1 presents results from the demographic portion of the questionnaire. The majority of respondents were black African (99%), unmarried (81%), and female (72%). Eighty-nine percent had electricity, and 64% had running water in their home. Only 35% of patients were employed, and an average of 1.0 (SD = 0.87) household member of 5.2 adults (SD = 2.7) was employed at the time of the survey. Twenty-eight percent reported that no one in their household worked for money, with many stating that they relied on the pension payments of someone in their home. Although 70% of the participants had less than a high school education, 95% of those interviewed reported they could easily read a newspaper or magazine. The number of patients included in our study who were currently taking ART was 31 (29%).



Back to Top | Article Outline

Sexual Practices and Disclosure of HIV Status

Nineteen percent of the participants were married, and 50% reported having a steady partner. Thirty percent stated that they did not currently have a sexual partner. When asked how many sexual partners they had had over the past year, 14% reported having no sexual partners, 62% reported having 1, 14% reported having 2, and 9% reporting having 3 or more. Of the 69% who reported having been treated for an STD, 44% reported STD treatment within the past year. Any condom use was reported by 84% of those interviewed. Reports of condom use decreased with increasing age (adjusted odds ratio [AOR] = 0.93, 95% CI: 0.88, 0.99; P < 0.05) and increased with increasing SES composite scores (AOR = 2.2, 95% CI: 1.1, 4.4; P < 0.05), but such reports were not associated with gender, ART use, length of time a patient had known she or he was HIV-positive, education, or marital status.

All patients attending the clinic were HIV-positive, and the average length of time a participant had known she or he was HIV-positive was 3.5 (SD = 3.06) years. Ninety percent of those interviewed reported having disclosed their HIV status to at least 1 person (Table 2). Of the 73 participants who reported having a spouse or sexual partner, however, 38% had not disclosed their status to their sexual partner. In multivariate regression analysis, disclosure to a sexual partner was not significantly associated with the length of time known to be HIV-positive, length of time in the current relationship, ART use, age, gender, education, or SES. Among those participants who reported having disclosed their HIV status, they most frequently reported disclosing this information to siblings (Table 1).



Back to Top | Article Outline

HIV/AIDS Knowledge

As shown in Table 2, a high proportion of study patients demonstrated good knowledge of HIV/AIDS and its modes of transmission. Nevertheless, 59% of patients believed that HIV could be transmitted by mosquitoes, a finding that was significantly associated in bivariate χ2 analysis with a low level of education (P < 0.01), ART use (P < 0.05), and a low SES (P < 0.05). Eighty-four percent of patients agreed that being infected with HIV is different from having AIDS and that someone can have a positive blood test for HIV and not have AIDS symptoms. Similarly, 87% agreed that if someone tests positive but has no symptoms, he or she may develop AIDS later.

An unweighted composite score of HIV knowledge was created from the questions listed in Table 2. The mean knowledge score was 25.9 of 30 correct, or 86% (95% CI: 85%, 88%). In multivariate linear regression analysis, HIV knowledge score was not significantly associated with education, length of time known to be HIV-positive, gender, age, having a sexual partner, SES, or being on ART.

Back to Top | Article Outline

Antiretroviral Therapy Knowledge

Results from questions regarding the effects of ART showed that participants' impressions of ART were overly optimistic. Whereas most participants correctly agreed that ART can prevent disease progression (98%) and vertical transmission (90%) and can control HIV (88%), sizable minorities believed that ART could cure HIV (49%) and that ART would not cause side effects (36%) (Table 3). Bivariate logistic regression revealed that belief of HIV cure by ART was associated with lower education levels (P < 0.01), lower SES (P < 0.01), and increasing age (P < 0.01). After adjusting for gender, ART use, length of time known to be HIV-positive, and marital status, only increasing age remained significant as a predictor of this knowledge (AOR = 1.1, 95% CI: 1.03, 1.17, P < 0.01) (Table 4).





Patient perception of the importance of adherence was also assessed. A majority of patients agreed that missing ART doses can lead to disease progression (65%) and increased transmission rates (56%). Surprisingly, knowledge that missing one' s ART doses can lead to disease progression was significantly higher among those not taking ART compared with those on ART (AOR = 0.27, 95% CI: 0.08, 0.91; P < 0.05).

An unweighted composite score of ART knowledge was created. The mean ART knowledge score was 7.9 of 11 correct, or 72% (95% CI: 69%, 74%). In a multivariate linear regression model, ART knowledge score was not associated with age, gender, ART use, SES, length of time known to be HIV-positive, or education. Based on the hypothesis that ART adherence may be associated with concerns about ART side effects or about friends and family discovering the patient is taking medication, we asked questions regarding these concerns. The questions, listed in Table 5, are summarized here. Fifty-nine percent of patients reported they were not worried about potential side effects. Seventy-six percent said they were not concerned that the medication would be ineffective. A majority of patients reported that they would not be worried about family (83%) or friends (74%) finding out if they were to take ART. In multivariate logistic regression models, concerns regarding ART were not significantly associated with ART use, length of time known to be HIV-positive, age, gender, SES, or education.



Back to Top | Article Outline


Among our study patients, we found that knowledge scores about HIV, including the cause, mode of transmission, and progression of disease, were high, with an average score of 86%. These results are in concordance with the level of HIV knowledge found in the 1998 South Africa Demographic and Health Survey among women in both urban and rural areas,9 suggesting that knowledge in HIV-positive clinic patients in Soweto is comparable to that in the general population.

Given that HIV/AIDS is a common topic in the South African media, most of our study participants have been exposed to HIV education through radio, television, newspaper, pamphlets, billboards, churches, and women's or youth organizations. In addition, all clinic patients received pre- and post-HIV test counseling that included information about the disease, potential treatment, nutrition, and safe sex practices as well as prevention of mother-to-child transmission. Based on the high knowledge scores of our study participants, we can assume that the information they receive, both through the media and in counseling, is understood.

Knowledge and perceptions regarding ART among HIV-positive clinic patients have not been previously assessed in South Africa. Because cultural factors can influence attitudes toward medications and health care practices, it is important to evaluate the assumptions and beliefs toward ART and their potential impact on adherence. Among those who sought care at the Baragwanath Hospital HIV Outpatient Clinic, we found a high level of knowledge regarding ART and its potential for controlling HIV infection. The majority of patients acknowledged that taking ART can cause side effects and knew that missing doses can lead to disease progression. A majority of patients also stated they had little or no concern about potential side effects or that the medication would be ineffective. Although these results may suggest an excessively high confidence in ART, it is reassuring that the majority of our study population expects ART to work and are not afraid of it. It is interesting to note that ART knowledge was not associated with age, gender, ART use, SES, or education in the multivariate analysis.

Knowledge that missing ART doses can lead to disease progression was significantly higher among those not taking ART compared to those on ART. Although this result is only marginally significant and is based on a small sample of patients on ART, it could indicate that those on ART have missed doses in the past without clinical consequences-potentially, a finding of great concern. In any event, this result calls for appropriate and tailored education messages emphasizing that decreased ART adherence can lead to drug resistance and disease progression. Whether the overall high knowledge scores about HIV and ART will translate into good HAART adherence levels remains to be evaluated further.

We found that a relatively high proportion of individuals had not disclosed their HIV status to their sexual partners. This finding may be explained by the fact that persons with HIV infection are still highly stigmatized in South Africa; thus, a significant proportion of HIV-infected patients may be reluctant to disclose their HIV status because of fear of their partners' reactions. Given that most of the study respondents were women in a male-dominated society, culturally based gender differences in personal power could explain some of this non-disclosure. A study by Maman and colleagues10 at an HIV voluntary counseling and testing clinic in Dar es Salaam, Tanzania, found that the most salient barriers to HIV testing and serostatus disclosure included fear of partners' reactions, decision-making and communication patterns between partners, and partners' attitudes toward HIV testing. Reluctance to disclose HIV status to one's sexual partner could result in continued unprotected sexual activity with risk of transmission and superinfection11,12 that could continue to fuel the HIV epidemic in the community. Increasing the population' s access to HAART could help to decrease the stigma of HIV while also decreasing the risk of transmission from patients unwilling to disclose their HIV status to their partners.13 Indeed, providing treatment services in sub-Saharan Africa so that HIV infection is transformed from a deadly disease to a treatable chronic condition is likely to provide an incentive for individuals at risk for HIV to seek voluntary counseling and testing.

The low reporting rate of HIV status to sexual partners is in contrast to the high rate of disclosure to family, friends, or community members. Strikingly, 90% of the patients we interviewed reported having previously disclosed their status to at least 1 person. In addition, the majority of these same patients were not concerned that their friends, family, or sexual partners would find out about their HIV status if the patient was on ART. This discrepancy may be explained by the fact that ART may have a positive influence on serostatus disclosure or that respondents did not make the connection between disclosing ART use and disclosing HIV status. Finally, the high rate of serostatus disclosure to friends, family, and community members suggests that community-based interventions that improve adherence, such as community-based directly observed therapy, do not have a large impact on patient confidentiality. An example of such an intervention has been successfully piloted in Haiti, with family or friends in the community serving as treatment supporters, and it has shown promising results.14

Our study has several limitations. By including patients already registered in an HIV specialty clinic with ongoing research trials, this study has a potential for selection bias. In addition, our study data were generated by self-report and thus must be taken to reflect the knowledge, attitudes, and beliefs of a specific population at a certain point in time rather than objective fact or a population-based measure. Finally, the relatively small number of patients may have led to a failure to detect significant associations between potential risk factors and certain variables of interest.

In summary, the present study demonstrates good knowledge in the study population about HIV/AIDS, disease progression, and transmission. Nevertheless, there is a need to reinforce education messages that ART does not cure HIV/AIDS and that missing doses could breed drug resistance and lead to disease progression. Whether the overall good HIV and ART knowledge scores will translate into good levels of ART adherence needs to be evaluated further. Finally, the low rate of HIV status disclosure to sexual partners calls for urgent and culturally adapted, multidimensional interventions to decrease HIV/AIDS-related stigmatization and discrimination.

Back to Top | Article Outline


The authors thank the patients at Chris Hani Baragwanath Hospital's Adult HIV Outpatient Clinic for their participation in this study and Amy Knowlton and Carl Latkin, Johns Hopkins Bloomberg School of Public Health, Department of Health Policy and Management, Division of Behavioral Sciences, and David Dowdy for critical reading of the manuscript.

Back to Top | Article Outline


1. Dorrington R, Bradshaw D, Budlender D, et al. The current state and future projections of the HIV/AIDS epidemic in South Africa. SADJ. 2002;57:449-450.
2. Dorrington R, Bourne D, Bradshaw D, et al. HIV/AIDS data in South Africa. Lancet. 2002;360:1177.
3. Palella FJ, Jr, Delaney KM, Moorman AC, et al. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. HIV Outpatient Study Investigators. N Engl J Med. 1998;338:853-860.
4. Mocroft A, Vella S, Benfield TL, et al. Changing patterns of mortality across Europe in patients infected with HIV-1. EuroSIDA Study Group. Lancet. 1998;352:1725-1730.
5. Moore RD, Chaisson RE. Natural history of HIV infection in the era of combination antiretroviral therapy. AIDS. 1999;13:1933-1942.
6. Blecher MS, Steinberg M, Pick W, et al. AIDS-knowledge, attitudes and practices among STD clinic attenders in the Cape Peninsula. S Afr Med J. 1995;85:1281-1286.
7. Govender V, Bhana R, Pillay A, et al. Perceptions and knowledge about AIDS among family planning clinic attenders in Johannesburg. S Afr Med J. 1992;81:71-74.
8. Nachega J, Gounder C, Doherty M, et al. Knowledge, attitude, and perception to tuberculosis in HIV-infected Sowetan adults, South Africa [ThPeC7574]. Presented at the XIV International AIDS Conference, Barcelona, 2002.
9. Department of Health, Medical Research Council, and Macro International Incorporated. South Africa Demographic and Health Survey 1998, Full Report, 2002.
10. Maman S, Mbwambo J, Hogan NM, et al. Women's barriers to HIV-1 testing and disclosure: challenges for HIV-1 voluntary counselling and testing. AIDS Care. 2001;13:595-603.
11. Levy JA. Is HIV superinfection worrisome? Lancet. 2003;361:98-99.
12. Goulder PJ, Walker BD. HIV-1 superinfection-a word of caution. N Engl J Med. 2002;347:756-758.
13. Quinn TC, Wawer MJ, Sewankambo N, et al. Viral load and heterosexual transmission of human immunodeficiency virus type 1. Rakai Project Study Group. N Engl J Med. 2000;342:921-929.
14. Farmer P, Leandre F, Mukherjee JS, et al. Community-based approaches to HIV treatment in resource-poor settings. Lancet. 2001;358:404-409.

HIV/AIDS; highly active antiretroviral therapy; South Africa

© 2005 Lippincott Williams & Wilkins, Inc.