Determinants of knowledge of HIV status in South Africa: results from a population-based HIV survey

Background Over 30% of women and men in the South African national HIV household of 2005 indicated that they had previously been tested for HIV (of which 91% were aware of their test results). This paper seeks to describe the associations between socio-demographic, behavioural and social characteristics and knowledge of HIV status among a nationally representative population in South Africa. Methods A multistage probability sample involving 16395 male and female respondents, aged 15 years or older was selected. The sample was representative of the South African population by age, race, province and type of living area, e.g. urban formal, urban informal, etc. Respondents were interviewed on HIV knowledge, perceptions and behaviour and provided blood for research HIV testing. Bivariate and multivariate logistic regression was used to identify socio-demographic, social and behavioural factors associated with knowledge of HIV status. Results From the total sample 27.6% ever and 7.8% knew their HIV status in the past 12 months. In multivariate analyses being female, the age group 25 to 34 years old, other than African Black population group (White, Coloured, Asian), higher educational level, being employed, urban residence, awareness of a place nearby where one could be tested for HIV, impact of HIV on the household and having had two of more sexual partners in the past year were associated with knowledge of HIV status. Among HIV positive persons awareness of a place nearby where one could be tested for HIV and impact of HIV on the household were associated knowledge of HIV status, and among HIV negative persons HIV risk behaviour (multiple partners, no condom use), awareness of a place nearby where one could be tested for HIV, higher knowledge score on HIV and knowledge of serodiscordance were associated knowledge of HIV status. Conclusion Education about HIV/AIDS and access to HIV counselling and testing (HCT) in rural areas, in particular among the Black African population group needs to be improved, in order to enhance the uptake of HIV counselling and testing services, an essential step for the initiation of treatment.


Background
HIV Counselling and Testing (HCT) is an entry point to both prevention and treatment. People have to know their HIV sero-status to enable them to seek appropriate treatment as well as enable them to choose prevention strategies. It has become increasingly important in the global response to HIV/AIDS. HCT services need to be available and accessible to the entire public to enable their easy utilization. Although most South Africans are aware that HCT services are available, only one in five people in South Africa who know about HCT have been tested for HIV [1]. Bunnell [2] found in a cross-sectional and nationally representative study (2004)(2005) Uganda HIV/ AIDS Sero-Behavioral Survey) that 21% of adults knew their HIV status. Among countries in Southern Africa the percentage who took an HIV test in the last 12 months and who know their results ranges between 2% in Mozambique to 20% in Botswana [3], and in the US 10.4% [4].
Various studies in South Africa have also shown that the uptake of HCT also differs depending on the counselling and testing model used, the use of HCT testing kits, other than rapid testing kits across different populations of high HIV prevalence [20][21][22][23]. There is lack of information on determinants of knowledge of HIV status in South Africa. Therefore, this study investigates the determinants of knowledge of HIV status in South Africa using secondary data analysis from a national population-based HIV survey.

Sample and procedure
The survey targeted all persons over 2 years of age living in South Africa and residing in homes, i.e. excluding individuals living in educational institutions, old-age homes, hospitals and uniformed service barracks but including those living in hostels. The survey applied a multi-stage stratified sampling approach based on a master sample consisting of 1 000 enumerator areas (EAs) used by Statistics South Africa for the national census in 2001. Three persons in each household were potentially eligible to be selected for the survey; however only one was selected from each of the age groups 2-14 years, 15-24 years, and 25 years and older. The sample included in this analysis includes the age group 15 years and above (range 15 to 96) 16395 (59.1% Black Africans, 18.4% Coloureds, 11.7% Whites and 10.8% Indian or Asians) of which 12 032 were interviewed and tested for HIV and 4363 who were interviewed but not tested for HIV. Linked anonymous HIV testing was performed using dried blood spot (DBS) specimens. Socio-demographic and behavioural information was collected with questionnaires administered by trained fieldworkers [5].
Ethical approval for conducting the study was obtained from the Human Sciences Research Council's Ethics Committee (Application Number REC5/24/04). Informed consent was obtained separately for agreeing to participate in the interview and for providing a specimen for HIV testing. HIV testing history: The survey included questions concerning history of HIV antibody testing. These measures were used to classify participants into groups based on whether they had been tested for HIV and knew their results. Participants who reported having been tested for HIV indicated their HIV awareness status of their most recent test, or that they did not know the results.

Measures
HIV risk behaviour history: To assess HIV risk history, participants indicated the number of sex partners they had in the previous 12 months, had symptoms of a sexually transmitted infection (STI), and whether they had ever used a condom, a condom with their last sexual partner and their last sexual non-regular partner. All responses were dichotomous indicating the occurrence or nonoccurrence of each risk factor.
HIV knowledge: A 7-item HIV knowledge test was used, e.g. Is it possible to transmit HIV through unprotected sex? Response options were yes, no, does not know. Responses were scored for the number of correct responses; with don't know responses scored incorrect, range 0-6. Scores were coded into three levels low = 5 correct responses; medium = 6 correct responses, and high = 7 correct responses. Cronbach's alpha for the HIV knowledge index was .70 for this sample.
ARV knowledge was assessed with one item: "Have you ever heard about new drug treatments for people with AIDS called antiretrovirals or ARVs?" Response option was "Yes" or "No".
HIV impact: Participants responded to three HIV impact items, ever anyone in the household ever been diagnosed with HIV-AIDS, is there a person in the household who is bed-ridden with an AIDS related illness and past year occurrence of AIDS-related death of household member. Response options were "yes or no"; any yes was scored with "1". AIDS stigma attitudes: Five AIDS stigma items were adapted from previous research and developed for use in South Africa, e.g. "I would be willing to care for a family member with AIDS". Response options were, yes, no, do not know; "no and Do not know were coded one and yes coded "0". A total score was calculated, (range 0-5), and coded into three levels 0= scores low AIDS stigma, medium 1-2 and 3-5 scores high AIDS stigma. Cronbach's alpha for the AIDS stigma index was .62 for this sample.
Demographic variables included sex, age, marital status, population group (Black African, White, Coloured, Indian or Asian), formal education, employment status and place of residence.

Data analysis
Data analysis was performed using STATA software version 10.0 (Stata Corporation, College Station, Texas, USA). The analysis in STATA took into account the multilevel stratified cluster sample design of the study. We obtained frequencies as estimation of prevalence of knowledge of HIV status. We also conducted logistic regression analysis to estimate the association between relevant predictor variables and knowledge of HIV status. The predictor variables were identified from the literature as possible factors that may be associated with knowledge of HIV status [5][6][7][8][9][10][11][12][13][14][15][16][17][18][19]. We report unadjusted odds ratios for selected predictor variables (sex, age, marital status, formal education, population group, employment status, geolocality, awareness of HIV test site, HIV knowledge, knowledge of ARVs, knowledge of HIV serodiscordance, impact of HIV, HIV risk perception, HIV stigma attitudes, HIV status, history of STI symptoms, number of sexual partners in the past year, no condom use at last sex, no condom use with last non-regular partner) while considering knowledge of HIV status as a dependent variable, and knowledge of HIV status separately for HIV positive and HIV negative as dependent variables. We therefore report results of adjusted odds ratios for the factors, having controlled for factors as significant in the bivariate analysis. The dependent variable was knowledge of HIV status, and the independent variables were factors which significantly increased knowledge of HIV status in the bivariate analysis. In the analysis, weighted percentages are reported. The reported sample size refers to the sample that was asked the target question. The two-sided 95% confidence intervals are reported. The p-value less or equal to 5% is used to indicate statistical significance. Both the reported 95% confidence intervals and the pvalue are adjusted for the multi-stage stratified cluster sample design of the study.

Results
From the total sample of 16395 15 years and above, 27.6% (CI = 26.5-28.7) reported to have ever had an HIV test and had received their HIV test results (knowledge of HIV status). Of those who had been tested for HIV, 38.8% had been tested within the year preceding the survey, 33.1% 1 to 2 years previously, and 28.2% more than 2 years previously; 7.8% of the total sample had taken an HIV test in the past 12 months and knew their test result.
In bivariate analyses being female, the age group 25 to 34 years old, being married or cohabitating, Grade 12 and more formal education, other than African Black population group (White, Coloured, Asian), being employed, urban residence, awareness of a place nearby where one could be tested for HIV, higher knowledge score on HIV, knowledge of serodiscordance, impact of HIV on the household, high HIV risk perception, being HIV positive (from the survey), one, two or more sexual partners in the past year and non-condom use at last sex were associated with knowledge of HIV status. In multivariate analyses being female, the age group 25 to 34 years old, other than African Black population group (White, Coloured, Asian), higher educational level, being employed, urban residence, awareness of a place nearby where one could be tested for HIV, impact of HIV on the household and two or more sexual partners in the past year were associated with knowledge of HIV status, and marital status, HIV knowledge, knowledge of serodiscordance, HIV risk perception, HIV status and no condom at last sex were no longer associated with knowledge of HIV status. A large proportion (39.5%, 38.2, 40.7) indicated that they had not heard about antiretroviral treatment (see Table 1).
Bivariate analyses with demographic variables and knowledge of HIV status among HIV positive and negative persons separately found among HIV positive persons that higher levels of formal education, being White, Coloured or Asian, and urban residence were associated knowledge of HIV status, and among HIV negative persons being female, the age group 25 to 34 years old, being married or  Table 2).
Bivariate analyses with behavioural factors (adjusted for sex and age) and knowledge of HIV status found among HIV positive persons no associations between behavioural risk factors and knowledge of HIV status, and among HIV negative persons bivariate and multivariate analyses (adjusted for sex and age) found that one, two or more sexual partners in the past 12 month and no con-dom use at last sex were associated knowledge of HIV status (see Table 3).   negative persons awareness of a place nearby where one could be tested for HIV, higher knowledge score on HIV and knowledge of serodiscordance were associated with knowledge of HIV status (see Table 4).

Discussion
The study found from a large nationally representative population-based HIV survey that 27.6% had knowledge of their HIV status, and 7.8% of the total sample had taken an HIV test in the past 12 months and knew their test result. Similar to this finding, more than half of Southern African countries have less than 10% of their population who know their HIV status (in the past 12 months) [3]. Evaluations of traditional VCT systems suggest low uptake of VCT even in places where access to VCT is unlimited [1]. As access to ART increases, there is an urgent need for alternative VCT delivery systems to increase access to and the utilization of VCT. These alternatives include mobile VCT, routine offer of counselling and testing and home-based VCT. These models can increase access to and the uptake of VCT [24].
This study found in multivariate analysis that being female, the age group 25 to 34 years old, other than African Black population group (White, Coloured, Asian), higher educational level, being employed, urban residence, awareness of a place nearby where one could be tested for HIV, impact of HIV on the household and two or more sexual partners in the past year were associated with knowledge of HIV status. Other studies also found that socio-economic factors such as age and being female [5][6][7][8][9], educational level [6,10,11], occupation [7], area of residence [8], proximity and access to VCT site [6,9,12,16,17], knowledge of VCT sites [7,8] and HIV risk behaviour [7,8,10,11,18,19] were associated with knowledge of HIV status. Rural residence, African Black population group, lower educational level and being unemployed were significantly associated with not knowing one's HIV status; clearly efforts should be made to make HIV testing available and promote HIV testing among these people.
In this study HIV knowledge, knowledge about ARVs, HIV risk perception, HIV status, history of STI and AIDS stigma and discrimination were in multivariate analyses not found to be associated with knowledge of HIV status, unlike in findings from other studies where fear of unsolicited disclosure, fear of stigma and discrimination [1,6,7,9,12], marital status [8], HIV knowledge, HIV risk perception [7,8,10,11,18,19] and health status [5,7,8] were

Limitations
Caution should be taken when interpreting the results of this study due to certain limitations. Since this was a crosssectional study, causality between the compared variables cannot be concluded. A further limitation was that a number of factors known to be contributing to knowledge of HIV status were not assessed, which included attitudes towards and desire for HIV testing [1,25]. Some measures in this study were limited in length, e.g. HIV risk perception and ARV knowledge was only measured with one item. The HIV testing and risk history measures did not include assessments of time since the behaviours occurred, not allowing us to examine whether people who were recently tested or who recently engaged in risk activities differed from those practising these behaviours less recently.

Conclusion
In this setting, a disproportionate number of HIV-positive young, lower educated Black African rural men are failing to learn their status, which has implications for equitable access to onward referral for care and treatment services.