Population-level impact of information sources on HIV testing uptake in South Africa: gender and age disparities

Abstract

In South Africa, mass media and interpersonal communication sources have been used extensively to increase awareness of human immunodeficiency virus (HIV)/AIDS. A sub-population of 10,565 black Africans who participated in the 2011–2012 household survey was analysed. Overall compared to men, a higher proportion of women had ever been tested for HIV (76% versus 60%). In age-specific analysis, men and women aged 25–49 years were more likely to be ever been tested compared to the youngest (<25 years) and the oldest age groups (50+ years). More than 80% of the study population was exposed to mass media tools, while exposure to interpersonal information sources was less common (16–72%). Among adolescents, 37% (men) and 41% (women) of the HIV tests were collectively associated with exposure to interpersonal programmes, while mass media tools had substantial contributions on HIV testing rates among the older individuals. Information sources to promote HIV/AIDS awareness should be tailored to reduce age–gender disparities.

Keywords

HIV testing men women mass media South Africa

Introduction

South Africa is the worst human immunodeficiency virus (HIV)-affected country in the world.^1,2 The country has the highest HIV incidence and prevalence rates in the world.¹ The epidemic is currently described as hyper-endemic because of its generalized and clustered nature.^2,3 The number of infected individuals has increased significantly since the early 1990s. However, testing and prevention activities did not intensify until the 2000s³; antiretroviral therapy (ART) became available after 2003.⁴ Nationwide awareness efforts have been increased by expanding the Voluntary Counselling and Testing campaigns across the country.^3,4 Several non-governmental and faith organizations have also played significant roles in prevention and awareness efforts.⁵

South Africa has also adopted the World Health Organisation’s 90–90–90 goal (i.e. 90% of HIV-infected individuals tested, 90% of those on ART and 90% of those virally suppressed).⁶ Aligned with this strategy, HIV testing and treatment efforts have been significantly increased and expanded through primary health care and antenatal clinics.⁷ Although these efforts led to substantial increase in testing rates, they still remained below the target levels, particularly among men.⁷ Besides economic and structural barriers, other factors including stigma, lack of knowledge and perceived risk about HIV transmission have been frequently linked to low levels of testing and treatment.^8,9 Given the significant clinical and community-level benefits of early diagnosis and treatment, development and dissemination of linguistically and culturally appropriate information are a research priority.

Mass media and interpersonal communication sources have been used frequently around the world to increase awareness of HIV/AIDS-related information.^10–12 In South Africa, mass media tools particularly radio, television and print materials (e.g. leaflets, booklets and billboards) have been used extensively to increase awareness of HIV/AIDS.^12–14 Interpersonal information sources (i.e. face-to-face communication) have also been developed and successfully delivered by voluntary counselling, health care providers and peer educators.^15,16 These communication sources have been reported to be associated with reduction in high-risk behaviour and stigma surrounding the disease in South Africa and other sub-Saharan countries.^14,16,17 Despite substantial research to show their individual-level effectiveness, their population-level impacts have never been investigated.

The current study aims to evaluate population-level importance of mass media and interpersonal information sources on HIV testing rates using the Fourth South Africa National HIV, Behaviour and Health Survey (2011–2012).^18,19 We examined the associations between the uptake of HIV testing and media exposure. After more than three decades of research, we have a better understanding of how to address and improve HIV/AIDS awareness in populations. The current challenge is to evaluate their population-level relevance in order to determine the information sources having the highest influence on testing rates. As the epidemic continues its impact in the region without showing signs of slowing down, identifying and reducing the barriers for testing and treatment can potentially play a significant role in preventing new infections. To our knowledge, this is the first comprehensive study that assesses the population-level benefits of multimedia and interpersonal information sources on HIV testing rates among sexually active men and women in South Africa using nationally representative data.

Material and methods

Study design and population

We included 10,565 sexually active black African men and women aged 15 years or older who participated in the Fourth South Africa National HIV, Behaviour and Health Survey and agreed to be tested for HIV during the period of 2011–2012. Survey, study population, behavioural and biological data collection have been previously described.¹⁸ A multistage stratified cluster sampling design was used to sample 1000 census enumerator areas from the listing of 86,000 census areas where in each area 15 households were sampled using a systematic random sample.

Measurements

The primary outcome of interest was ever been tested for HIV. Current analysis considered a variety of socio-demographic characteristics and sexual behaviours including age, employment status, marital/cohabitation status, partner has another partner (no, yes, I don’t know), age at sexual debut (<20 versus 20 years or older), number of life time sexual partners (1, 2+ partners), condom use at first sex (yes/no), condom use at last sex (yes/no); for males only: circumcision status (yes/no).

Information sources

The mass media tools included radio, television and print media. Participants were asked to choose all possible answers that are relevant to them including: ‘how often listen radio?/watch television? read newspaper/magazines?’. The answers were grouped as never versus at least weekly. Exposures to leaflets/pamphlets/booklets and billboards were collected by asking ‘What has made you take the problem of HIV/AIDS seriously?’ Receiving useful information sources was ‘in the past 12 months from where/whom have you received HIV/AIDS information’: ‘AIDS/welfare organisation’ (yes/no), ‘clinic/hospital or doctors’ (yes/no), ‘family/friends’ (yes/no). Survey participants were also asked if they attended HIV/AIDS-related ‘training/workshop’, ‘community meeting’ and ‘play/educational event’(yes/no)

Statistical analysis

Descriptive data were presented as frequencies/percentages for categorical variables. The Chi square test was used to compare categorical factors. Weighted-logistic regression models were used to assess the associations between the demographic, socioeconomic and sexual behavioural factors, information sources and outcome variable, while accounting for the multistage sampling design. Information sources were grouped into four categories according to their dissemination types. The first group was exposure to the common mass media tools including television and radio on a weekly basis; these were combined as one variable due to the correlated nature of these measurements, while print media (newspaper/magazines) was considered as a separate variable. The third group was created by combining those receiving HIV/AIDS-related information from interpersonal information sources including AIDS organizations, clinics/hospitals or family/friends. The fourth group combined participants who attended a HIV/AIDS-related workshop, community events or plays/dramas in past 12 months. To avoid multicollinearity between the information sources, associations between HIV testing and each of the four information sources were investigated using separate logistic regression models after adjusting the results for the significant socio-demographic factors and sexual behaviours. The p-values from the Hosmer–Lemeshow ranged 0.2886–0.750 indicating reasonable fits for each information source-specific model. Adjusted odds ratios (aORs) and their 95% confidence intervals (CI) were presented. In an additional analysis, we assessed population-level impacts of the information sources on uptake in HIV testing²⁰ (Appendix 1). We also estimated the population attributable risk percentage (PAR%) in a multifactorial analysis setting. Analyses were conducted using Stata 14.0 (College Station, TX, USA) and SAS statistical software, version 9 (SAS Inc., Cary, NC, USA).

Results

In this sub-group analysis, we included data from 10,565 sexually active African adults aged 15 years or older who consented to be interviewed and tested for HIV in the Fourth South Africa National HIV, Behaviour and Health Survey during the period of 2011–2012.¹⁸ Of these 4193 (40%) were men and 6372 (60%) were women. The per cent distribution of population characteristics was presented and compared across the three age groups: 15–24 years, 25–49 years and 50+ years for men and women separately (Table 1). Overall compared to men, a higher proportion of women had ever been tested for HIV (76% versus 60%). In age-specific analysis, men and women aged 25–49 years were more likely to be ever been tested compared to the youngest (<25 years) and the oldest age groups (50+ years). Higher proportions of the study population were living in rural formal/tribal settings in both genders. Overall 75% of the men and 79% of the women were identified as not married and/or not living with their sexual partners; as expected younger participants were more likely to be unmarried compared to the older participants in both sexes (p < 0.001). The vast majority of the men (89%) and women (87%) aged younger than 25 years of age reported their age at sexual debut as <20 years compared to the older age groups (44 and 48% for men and women, respectively), while men and women aged 25–49 years old had the highest number of sexual partners compared to the youngest and the oldest groups (p < 0.001). There was a linear decreasing trend between the proportion of condom use (every time/almost every time) and older age groups in both gender (p < 0.001).

Table 1.

Characteristics of the sexually active men and women by age groups.

	Men (N = 4193, 40%)					Women (N = 6372, 60%)
Characteristics	Overall	15–24 years	25–49 years	50+ years	p-value	Overall	15–24 years	25–49 years	50+ years	p-value
Ever tested for HIV	60%	50%	68%	56%	<0.001	76%	83%	88%	52%	<0.001
Geographical location					<0.001					<0.001
Rural formal/Tribal	46%	40%	42%	54%		48%	48%	40%	52%
Urban formal	35%	36%	36%	32%		35%	34%	36%	33%
Urban informal	19%	17%	22%	14%		17%	18%	19%	12%
Education					<0.001					<0.001
None/Primary	27%	11%	23%	58%		28%	10%	21%	59%
Some secondary	44%	60%	40%	29%		42%	53%	43%	30%
Secondary complete/high school	29%	29%	37%	13%		30%	37%	36%	11%
Not married/not living	75%	99%	77%	47%	<0.001	79%	96%	75%	74%	<0.001
Unemployed/no income	40%	37%	38%	49%	<0.001	60%	55%	59%	66%	<0.001
Age at sexual debut <20	67%	89%	67%	44%	<0.001	66%	87%	68%	48%	<0.001
Lifetime sex partners					<0.001					<0.001
1	13%	19%	8%	13%		31%	42%	22%	37%
2+	79%	76%	84%	74%		63%	54%	73%	53%
No response	8%	5%	8%	13%		6%	4%	5%	10%
Used condom at first sex	16%	41%	100%	100%	<0.001	13%	59%	<1%	0%	<0.001
Partner has other partner					<0.001
No	56%	43%	57%	72%		43%	42%	42%	49%	0.054
Yes	18%	28%	17%	8%		24%	25%	24%	20%
Don’t know	26%	29%	26%	21%		34%	33%	35%	31%
Frequency of condom used^a					<0.001					<0.001
Every time/almost every time	38%	59%	36%	16%		33%	45%	32%	12%
Sometimes	15%	19%	16%	7%		19%	23%	19%	6%
Never	47%	22%	49%	78%		48%	32%	49%	82%
Condom used in last sex					0.016					0.571
No	18%	15%	21%	16%		25%	24%	26%	25%
Yes	82%	85%	79%	84%		75%	76%	74%	75%

^aw/recent sexual partner (past 12 months).

Exposure to information sources

Distribution of gender-specific exposure to mass media and interpersonal information sources by age groups was also presented in Table 2. Overall, 82 and 81% of the men were exposed to radio and television, respectively; these proportions were 75 and 80% among women, while 45% of the men and 43% of the women reported to have exposure to print media. Less than 10% of the population reported that they had seen HIV/AIDS-related leaflets/booklets and billboards. All these proportions declined with increased age for both genders. The most common information source for HIV/AIDS was reported as receiving information from health care institutes and health care providers (60 and 72% for men and women, respectively).

Table 2.

Distribution of gender-specific exposure to mass media and interpersonal information sources by age groups.

	Men (N = 4193, 40%)					Women (N = 6372, 60%)
Characteristics	Overall	15–24 years	25–49 years	50+ years	p-value	Overall	15–24 years	25–49 years	50+ years	p-value
Exposure to mass media
Radio	82%	83%	82%	81%	0.368	75%	73%	78%	72%	<0.001
Television	81%	85%	81%	76%	<0.001	80%	84%	82%	74%	<0.001
Print media	45%	55%	50%	26%	<0.001	43%	52%	52%	20%	<0.001
Exposure to other sources
Leaflets/booklets	6%	8%	6%	3%	<0.001	5%	7%	5%	2%	<0.001
Billboards	4%	5%	4%	3%	0.397	3%	4%	4%	2%	<0.001
Exposure to information sources
AIDS organization	17%	18%	18%	15%	0.044	21%	23%	23%	16%	<0.001
Clinic, hospital or doctors	60%	54%	61%	62%	<0.001	72%	71%	76%	67%	<0.001
Family/friends	61%	66%	62%	52%	<0.001	60%	66%	63%	52%	<0.001
Attending
Workshop	14%	11%	18%	11%	<0.001	14%	13%	18%	9%	<0.001
Community meeting	18%	14%	20%	18%	<0.001	18%	14%	21%	17%	<0.001
Play/educational event	15%	25%	13%	8%	<0.001	14%	21%	15%	8%	<0.001

The second most common information source was reported as ‘talking to family/friends’ (61 and 60% for men and women, respectively). This proportion was the highest among those younger than 25 years of age (66% both genders) and declined with increased age (p < 0.001). Less than 20% of the men and women reported attending a workshop, community meeting and play/drama. There were no apparent trends across the age groups.

Associations between testing for HIV and socio-demographic factors, sexual behaviours

Table 3 presents associations between exposure to information sources and ever been tested for men and women separately. Compared to the youngest age group (<25 years) men and women aged 25–49 years old were significantly more likely to get tested (aOR: 2.04, 95%CI: 1.65, 2.53 and aOR: 1.51, 95%CI: 1.19, 1.93 for men and women, respectively). Participants who reported more than two lifetime sexual partners were also significantly more likely to get tested for HIV compared to those who had only one sexual partner (aOR: 1.31, 95%CI: 1.02, 1.69 and aOR: 1.48, 95%CI: 1.22, 1.80 for men and women, respectively), while those reported no sex partners in the past 12 months were significantly less likely to report ever been tested for HIV in both genders (aOR: 0.55, 95%CI: 0.43, 0.70 for both men and female). Higher HIV/AIDS knowledge were also significantly associated with increased odds of testing, while we were not able to find associations with employment status, age at sexual debut, condom used in first and last sex.

Table 3.

Individual- and population-level impacts of information sources: ‘Ever been tested for HIV’.

	Men (N = 4193, 40%)			Women (N = 6372, 60%)
	15–24 years	25–49 years	50+ years	15–24 years	25–49 years	50+ years
Full PAR% (95% CI)	–	80% (71%, 86%)	80% (72%, 87%)	–	87% (84%, 90%)	68% (59%, 75%)
Mass media (radio + television)
aOR (95% CI)	1.14 (0.51, 2.58)	2.15 (1.28, 3.60)**	2.32 (1.36, 3.95)**	1.03 (0.49, 2.19)	1.72 (1.02, 3.03)*	1.73 (1.15, 2.60)**
PAR% (95% CI)	N/A	PAR%= 49% (95% CI: 41%, 58%)	$PAR % =$ 48% (95% CI: 38%, 59%)	N/A	PAR%= 43% (95%CI: 33%, 53%)	PAR%= 40% (95%CI: 32%, 47%)
Print media (newspaper/magazines)
aOR (95% CI)	1.02 (0.70, 1.51)	1.31 (1.02, 1.70)*	2.04 (1.40, 3.00)***	1.01 (0.73, 1.41)	1.47 (1.05, 2.05)*	2.19 (1.62, 2.19)***
PAR% (95% CI)	N/A	PAR%= 16% (95% CI: 13%, 20%)	PAR%= 22% (95% CI: 18%, 26%)	N/A	PAR%= 32% (95% CI: 28%, 37%)	PAR%= 23% (95% CI: 20%, 26%)
Interpersonal sources AIDS organization + clinic/hospital + family/friends
aOR (95% CI)	1.63 (1.10, 2.40)*	1.95 (1.38 , 2.75)***	3.09 (2.06, 4.62)***	1.47 (0.87, 2.48)	2.22 (1.50, 3.30)***	1.85 (1.38, 2.48)***
PAR% (95% CI)	PAR%= 37% (95%CI: 29%, 45%)	PAR%= 45% (95%CI: 39%, 51%)	PAR%= 47% (95%CI: 40%, 54%)	PAR%= 41% (95%CI: 32%, 50%)	PAR%= 54% (95%CI: 48%, 60%)	PAR%= 44% 95%CI: (38%, 49%)
Attending: Workshop + community + play/dramas
aOR (95% CI)	1.14 (0.81, 1.62)	1.96 (1.39, 2.77)***	2.62 (1.78, 3.85)***	0.91 (0.62, 1.35)	1.67 (1.12, 2.51)*	1.69 (1.24, 2.30)***
PAR% (95% CI)	N/A	PAR%= 19% (95%CI: 17%, 23%)	PAR%= 29% (95%CI: 24%, 33%)	N/A	PAR%= 25% (95%CI: 21%, 29%)	PAR%= 19% (95%CI: 17%, 22%)

PAR%: population attributable risk percentage.

Significant at level: *p < 0.05; **p < 0.01; ***p < 0.001.

N/A: PAR% was not estimated due to the no/weak associations.

Combined population-level impact of information sources on HIV testing uptake

Table 3 presents population-level impacts of information sources on HIV testing uptake. In age–gender-specific analysis, all the information sources were significantly associated with increased prevalence of HIV testing among those older than 25 years old. Among men, 49% (for 25–49 years old) and 48% (50+ years old) of the HIV tests were collectively associated with the exposure to the common mass media tools – radio or television. High levels of exposure (>80%) and relatively high aORs were responsible for these moderate impacts (2.15 and 2.32 for 25–49 and 50+ years old, respectively), while these mass media tools had slightly lower impact among women ( $PAR % =$ 43% and $PAR % =$ 40% for 25–49 years and 50+ years). Exposure to the print media was attributed to 16% (aged 25–49) and 22% (aged 50+) of the testers among men; these proportions were 32% (aged 25–49) and 23% (aged 50+) among women. There were no significant associations between any of the mass and print media and ever having been tested for HIV in youngest age group (i.e. it was not possible to calculate population attributable risks).

The prevalence of ever having been tested was significantly higher among men and women who have received HIV/AIDS-related information in the past 12 months through interpersonal information sources compared to those who did not. Receiving useful information from AIDS organizations, clinic/hospital or family/friends was collectively associated with 37, 45 and 47% of the testers among men aged <25, 25–49 and 50+ years old, respectively, while these proportions were 41, 54 and 44% of the testers among women. These relatively high impacts were again due to the significant age-specific odds ratios (ORs) which were estimated as 1.63, 1.95 and 3.09 (men) and 1.47, 2.22 and 1.85 (women).

There were no significant associations between attending any of these events and testing for HIV among the youngest age group (<25 years) in both genders. In an ancillary analysis, we also calculated age–gender-specific full $PAR %$ for men and women separately. For example, 80% of the testers among men aged 25–49 and 50+ years were collectively associated with an exposure to at least one of the information sources listed in Table 3.

Discussion

This study investigates the population-level effectiveness of various information sources on HIV testing uptake based on nationally representative data from South Africa. In our study population, the proportion of women reporting to have ever been tested for HIV exceeds that of men (76% versus 60%). Age–gender-specific disparities in HIV testing rates have been confirmed in a model with statistically significant interaction between age and gender (p < 0.001). Those aged between 25 and 49 years had the highest testing rates (88 and 68% for women and men, respectively). In addition, individuals with low levels of education and knowledge about HIV/AIDS were less likely to get tested for HIV. These gender–age-specific disparities have been previously reported in South Africa and other sub-Saharan countries.^21,22 Our results confirmed that a high proportion of South Africans are exposed to various forms of information sources concerning HIV/AIDS.¹⁴ However, their population-level impacts varied across the age groups.

In our study population, mass media, including radio and television, were identified as the most common information sources across the age groups in both genders. Overall, more than 85% of the study population was exposed to mass media on a weekly basis. This proportion declined with age (p < 0.001). After adjusting for underlying differences in socio-demographic characteristics and sexual behaviours, we found significant associations between mass media exposure and testing for HIV among those 25 years or older. Concurrent with these findings, their population-level impact was considerable with $PAR % :$ 40–49% (in both genders). Mass media had no significant impact on HIV testing among the younger adolescents; therefore we were not able to calculate $PAR %$ for this segment of the population.

Reception of HIV/AIDS-related information from interpersonal communication programmes such as AIDS organizations and clinics/hospitals was less common compared to information from mass media (range 16–72% versus 75–85%, respectively). However, their impact on HIV testing was substantial across age groups (in both genders). In fact, these information sources were identified as the only effective mechanisms on testing rates among younger adolescents. They were collectively associated with 41% of the tests on men and 37% of the tests on women. Other information sources, including print media, attending HIV/AIDS-related workshops, community events and plays/dramas in the past 12 months were significantly associated with increased prevalence of tests among those 25 years older. However, their $PAR %$ estimates were modest with a range of 19–32%, due to the low prevalence of these activities in the population and/or weak associations with the HIV testing rates.

These findings suggest that, despite their widespread use, HIV-related information delivered through mass media is not effective among younger adolescents. During the past two decades in South Africa, there have been numerous HIV/AIDS promotion strategies involving various mass media and interpersonal information sources.^5,14,22 Besides promoting condom use, these programmes particularly focussed on messages such as ‘abstinence’/‘delaying sexual debut’.²³ Reductions in HIV prevalence and incidence in Uganda were primarily attributed to ‘abstinence’/‘delaying sexual debut’.²⁴ However, effectiveness of these messages on changing sexual behaviours has not been fully established in other sub-Saharan African countries. Increasing HIV rates, particularly among young people, provides strong evidence for challenges in acceptability of abstinence and delaying sexual debut among South African men and women.²³ In our study population, the vast majority of adolescents (88–89%) reported their sexual debut as being younger than 20 years of age, compared to older adults (44–68%). Therefore, messages related to abstinence and delaying sexual debut may not be the most effective way to target those who are already sexually active. Integrating and disseminating simple non-confrontational information using the mass media to promote safe sex with emphasis on clinical benefits of early testing and treatments are likely to play a significant role among younger adolescents. Expanding the interpersonal sources (particularly through AIDS organizations and clinics/hospitals) has the potential to play a significant role in reducing the age–gender-specific gaps in HIV testing uptake. Overall, 70% of the study population did not complete secondary school. The prevalence of testing for HIV was significantly higher among those who had at least secondary school education (in both genders). In light of this evidence, the impact of interpersonal information sessions on HIV testing uptake is not surprising and concurrent with previous research.³ Other media sources, including leaflets/booklets and billboards, were also significantly associated with increased HIV testing. Although these visual presentations can be effective for those with low literacy, their prevalence in the study population was less than 10%. Nevertheless, expanding these print media to the larger population may not be a cost-effective strategy.

Although our findings have common elements with previous research, this is the first study concerned with the association between HIV testing and specific information sources using a novel statistical technique which can handle the multi-factorial nature of the research question.^25–28

Overall, our data indicated that improving HIV/AIDS-related messages delivered by the mass media and expanding interpersonal information sources have the potential to increase HIV testing practice. These results also provide evidence on how media sources can play a key role in shaping individuals’ knowledge about complex health issues by decoding such information into plain language.²⁹

Our study has limitations and therefore our results need to be interpreted with a degree of caution. The current study was designed as a cross-sectional survey. Therefore, it shares the same limitations of these kinds of designs including lack of temporal associations between HIV testing and the factors considered in this study. Our analyses used self-reported data. Therefore, they might be subject to recall bias. In addition, the response rates among Whites and Indians were very low; therefore, we have only included Black Africans in our analysis and the results cannot be generalized to Whites and Indians. Despite these limitations, this is one of the largest population-based surveys conducted in South Africa. Finally, investigating population-level effectiveness of the information sources on HIV testing uptake using a nationally representative sample has significant implications for current and future HIV prevention programmes.

Conclusion

These results showed that with increasing HIV prevalence and incidence in the region, messages delivered through the mass media tools need to be improved by including more realistic messages while those related to the interpersonal sources need to be expanded. If the messages delivered by the mass media tools were as effective as the interpersonal communication programmes, we would expect to see more than a 60% increase in HIV testing rates. Also if exposure to interpersonal information sources had the same levels as mass media tools, HIV testing rates would be increased by 50%. These results have significant public health implications by providing guidance for effective and cost-effective programmes to increase HIV/AIDS awareness.

Footnotes

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

References

Joint United Nations Program on HIV/AIDS (UNAIDS). Global AIDS update. Geneva, available at: http://www.unaids.org/ (2016, accessed August 2018).

South African National Department of Health (DOH). National antenatal sentinel HIV & syphilis survey report, available at: http://www.hsrc.ac.za/en/research-areas/Research_Areas_HAST/HAST_National_HIV_Survey (2015, accessed September 2018).

Koker

Pierre

Matthys

et al . Barriers to VCT despite 13 years of community-based awareness campaigns in a peri-urban township in northern Limpopo. South Afr Med J 2010; 100(6): 364–5.

Karim

Churchyard

Karim

et al . HIV infection and tuberculosis in South Africa: an urgent need to escalate the public health response. Lancet 2009; 374: 921–933.

Birdsall

Hajiyiannis

Nkosi

et al . Voluntary counselling and testing in South Africa: analysis of calls to the national AIDS helpline. Johannesburg: Center for AIDS Development, Research and Evaluation, 2004.

HIV/AIDS. 90-90-90: an ambitious treatment target to help end the AIDS epidemic. Geneva: UNAIDS, 2014.

Hansoti

Stead

Parrish

et al . HIV testing in a South African Emergency Department: a missed opportunity. PLoS One 2018; 13: e0193858.

Tshuma

Muloongo

Setswe

et al.

Potential barriers to rapid testing for human immunodeficiency virus among a commuter population in Johannesburg. HIV AIDS (Auckl), South Africa. 2015; 7: 11–19.

Tanser

Barnighausen

Cooke

et al . Localized spatial clustering of HIV infections in a widely disseminated rural South African epidemic. Int J Epidemiol 2009; 38: 1008–1016.

10.

Liskin

Using mass media for HIV/AIDS prevention. AIDS Care 1990; 2: 419–420.

11.

Agha

The impact of a mass media campaign on personal risk perception, perceived self-efficacy and on other behavioral predictors. AIDS Care 2003; 15: 749–762.

12.

Jesmin

Chaudhuri

Abdullah

Educating women for HIV prevention: does exposure to mass media make them more knowledgeable?

Health Care Women Int 2013; 34: 303–331.

13.

Bertrand

O’Reilly

Denison

et al . Systematic review of the effectiveness of mass communication programs to change HIV/AIDS-related behaviors in developing countries. Health Educ Res 2006; 21: 567–597.

14.

Katz

Explaining the increase in condom use among South African young females. J Health Commun 2006; 11: 737–753.

15.

Bankole

Biddlecom

Guiella

et al . Sexual behavior, knowledge and information sources of very young adolescents in four sub-Saharan African countries. Afr J Reprod Health 2007; 11: 28–43.

16.

Buseh

Glass

McElmurry

et al . Primary and preferred sources for HIV/AIDS and sexual risk behavior information among adolescents in Swaziland, Southern Africa. Int J Nurs Stud 2002; 39: 525–538.

17.

Romer

Sznitman

DiClemente

et al . Mass media as an HIV-prevention strategy: using culturally sensitive messages to reduce HIV-associated sexual behavior of at-risk African American Youth. Am J Public Health 2009; 99(12): 1–10.

18.

Shisana

Rehle

Simbayi

et al.

South African national HIV prevalence, incidence and behaviour survey, 2012. Cape Town: HSRC Press, 2014.

19.

Zuma

Shisana

Rehle

et al . New insights into HIV epidemic in South Africa: key findings from the National HIV prevalence, incidence and behaviour survey, 2012. Afr J AIDS Res 2016; 15: 67–75.

20.

Wand

Ramjee

Combined impact of sexual risk behaviors for HIV seroconversion among women in Durban, South Africa: implications for prevention policy and planning. AIDS Behav 2011; 15: 479–486.

21.

Takarinda

Madyira

Mhangara

et al . Factors associated with ever being HIV tested in Zimbabwe: an extended analysis of the Zimbabwe demographic and health survey (2010–2011). PLoS One 2016; 11: e0147828.

22.

Staveteig

Shanxiao

Head

et al. Demographic patterns of HIV testing uptake in sub-Saharan Africa. DHS Comparative Reports No. 30. Calverton, Maryland, https://dhsprogram.com/pubs/pdf/CR30/CR30.pdf (Accessed April 2019).

23.

Setswe

Abstinence and faithfulness programs for prevention of HIV/AIDS among young people: what are the current debates?

S Afr Fam Pract 2007; 49: 5–10.

24.

Asiimwe-Okiror

Opio

Musinguzi

et al . Change in sexual behaviour and decline in HIV infection among young pregnant women in urban Uganda. AIDS 1997; 11: 1757e63.

25.

Keating

Meekers

Adewuyi

Assessing effects of a media campaign on HIV/AIDS awareness and prevention in Nigeria: results from the VISION project. BMC Public Health 2006; 6: 123.

26.

Laga M, Schwartlander B, Pisani E, et al. To stem HIV in Africa, prevent transmission to young women. AIDS. 2001; 15(7): 931–934.

27.

Rotheram-Borus

et al . Mass media and HIV/AIDS in China. J Health Commun 2009; 14: 424–438.

28.

Sano

Sedziafa

Amoyaw

et al . Exploring the linkage between exposure to mass media and HIV testing among married women and men in Ghana. AIDS Care 2016; 28: 684–688.

29.

Kansanga

Asumah Braimah

Antabe

et al . Examining the association between exposure to mass media and health insurance enrolment in Ghana. Int J Health Plann Manage 2018; 33: e531–e540.