Sexual Behaviors and HIV Status: A Population-Based Study Among Older Adults in Rural South Africa : JAIDS Journal of Acquired Immune Deficiency Syndromes

Secondary Logo

Journal Logo


Sexual Behaviors and HIV Status: A Population-Based Study Among Older Adults in Rural South Africa

Rosenberg, Molly S. MPH, PhD*,†; Gómez-Olivé, Francesc X. MSc, MD, PhD†,‡,§; Rohr, Julia K. MPH, PhD; Houle, Brian C. MPH, PhD‡,‖; Kabudula, Chodziwadziwa W. Msc‡,§,¶; Wagner, Ryan G. MSc, PhD‡,§; Salomon, Joshua A. PhD#; Kahn, Kathleen MD, MPH, PhD‡,§,**; Berkman, Lisa F. MS, PhD†,‡,#,††,‡‡; Tollman, Stephen M. MD, MPH, PhD‡,§,**; Bärnighausen, Till MD, MSc, ScD†,#,§§,‖‖

Author Information
JAIDS Journal of Acquired Immune Deficiency Syndromes: January 1, 2017 - Volume 74 - Issue 1 - p e9-e17
doi: 10.1097/QAI.0000000000001173



Older adults have received relatively little attention in the context of HIV prevention research and interventions.1–4 Mathematical models and growing empirical evidence suggest that older adults make up a fast-increasing proportion of people living with HIV because of the impacts of large-scale HIV treatment on HIV mortality.5–9 It is likely that the future of the HIV response will be determined to a large extent by our ability to design HIV interventions that are targeted at and appropriate for older adults. However, the evidence on sexual HIV transmission and acquisition risks in older adults remains scarce, in particular in sub-Saharan Africa where most of the worldwide 37 million HIV-infected people live.10–15 In fact, HIV prevention policy and funding are largely focused on younger adults: most HIV prevention interventions have been developed and designed to fit the specific needs of this age group, and very few prevention interventions specifically targeted at older adults are available worldwide.16 With this study in rural South Africa, we aim to contribute to filling this persistent knowledge gap.

Risky sexual activity places an HIV-negative individual at higher risk of acquiring HIV infection from an infected partner, whereas risky sexual activity by an HIV-positive individual places them at higher risk of transmitting HIV to an uninfected partner. Evaluating both the potential for HIV acquisition and the potential for HIV transmission in HIV-vulnerable populations is important because the intervention toolbox to prevent HIV acquisition is only partially overlapping with the intervention toolbox to prevent HIV transmission. As a biomedical example, antiretroviral therapy is an important approach to prevent HIV transmission,17 whereas preexposure prophylaxis is used to prevent HIV acquisition.18

As observed in younger adults, sexual behavior patterns in older adults likely differ by HIV status, and documenting these differences will help provide insight into the potential drivers of the sexual transmission and acquisition of HIV in this age range. Certain sexual behaviors, like unprotected sex and multiple partnerships, clearly increase the risk of acquiring HIV,19–21 and are likely to be observed more frequently in HIV-positive compared to HIV-negative populations, but whether or not these behaviors confer the same risk in older adults as in younger adults is underexplored. HIV status and sexual behaviors may also be linked because knowledge of HIV status could promote protective behavior change to avoid transmitting HIV to partners or to maintain HIV-negative status. Again, this relationship is observed in younger adults,22,23 but is not yet explored in older adults.

In this article, we aim to add to the emerging body of literature on HIV risk in older adults. First, we report on the HIV prevalence in a population of men and women aged 40 years and older in rural South Africa. Second, we characterize this cohort with respect to their sexual behaviors. Finally, we compare sexual behavior profiles across HIV status categories, using both self-reported and laboratory-confirmed HIV status data.


Study Population

We analyzed data from the Health and Aging in Africa: Longitudinal Studies of INDEPTH Communities (HAALSI) study. The HAALSI study is a population-based survey that aims to characterize a population of older men and women with respect to health, physical, and cognitive function, aging, and well-being. In 2015, 5059 men and women aged 40 years and older were enrolled in the study, which was carried out in the Agincourt health and socio-demographic surveillance system (AHDSS) in the rural Agincourt subdistrict of Mpumalanga province, South Africa.24 Extensive survey and laboratory data were collected to assess (1) physical and cognitive functioning, (2) cardiometabolic health, (3) economic well-being, and (4) HIV status and HIV risk. In this study, we analyzed baseline data on HIV prevalence and sexual behaviors. Although most literature on the sexual health of older adults uses 50 years as the lower age cutoff, there is not a clearly established definition of “older age.”25 Our study population of older adults included men and women aged 40 years and older because life expectancy in southern African communities with high HIV prevalence, like our study community, is substantially lower than in most other world regions.24,26–28 To assist in making valid comparisons with previous studies using different age cutoffs, we provide, where possible, sexual behavior and HIV prevalence data stratified by decade of life. Ethical approval for HAALSI was obtained from the University of the Witwatersrand Human Research Ethics Committee (#M141159), the Harvard T.H. Chan School of Public Health Office of Human Research Administration (#13-1608), and the Mpumalanga Provincial Research and Ethics Committee.

Data Collection

All variables in this analysis were collected in HAALSI baseline surveys and biological testing. Trained, local fieldworkers collected survey data electronically using a Computer Assisted Personal Interview system. Surveys were conducted in the local Shangaan language, with instruments translated from English and back-translated to ensure reliability. In addition to the survey data, trained fieldworkers also collected blood through finger prick and prepared dried bloodspots from each participant who consented to blood collection.

Key Measures

Laboratory-confirmed HIV status was determined through screening and confirmatory HIV enzyme-linked immunosorbent assays using standard laboratory practices on the prepared dried bloodspots.29 Participants also self-reported their HIV status in the baseline survey. We defined self-reported HIV status based on participant responses to 2 questions: “Have you ever been tested for HIV? (yes/no)” and “Have you ever tested positive for HIV? (yes/no).” We categorized participants into 3 self-reported HIV status categories: (1) Those who reported testing positive for HIV; (2) Those who reported testing negative for HIV, and (3) Those who reported never testing for HIV, or testing with unknown status. Cross-referencing laboratory-confirmed with self-reported HIV status, we also calculated whether or not HIV-positive participants were aware of their positive status (under the assumption that self-reported data reliably reflected this information30), separating those who self-reported being HIV-positive from those who self-reported being HIV-negative or not knowing their status.

We also present data on sexual activity and key sexual behaviors self-reported by participants at baseline. Participants were asked to report the number of lifetime sex partners and number of recent sex partners (past 24 months). Each participant reporting at least 1 recent partner was asked to report on specific characteristics of up to 3 recent sex partners. We examined whether or not the most recent partner was categorized as casual or anonymous, as opposed to regular. The frequency of condom use with most recent partner was originally elicited with the response categories of “always”, “most of the time”, “sometimes”, and “never”, but dichotomized for this analysis into the two categories “at least some of the time” and “never”.


We calculated the prevalence of key sociodemographic characteristics, HIV status (laboratory-confirmed status, self-reported status, and laboratory-confirmed status broken down by self-reported status), and each sexual behavior. Log-binomial regression models were used to calculate the age-specific and sex-specific prevalence of HIV and recent sexual behaviors (past 24 months) with their associated 95% confidence intervals (CIs).

We also used log-binomial regression models to estimate the prevalence of each recent sexual behavior (past 24 months) within each HIV status category and to assess whether the relative prevalence of certain sexual behaviors differed across the status categories. These analyses were conducted with the intention to separately characterize the potential for HIV acquisition and the potential for HIV transmission, with HIV-negative adults at potential risk for HIV acquisition and HIV-positive adults at potential risk for HIV transmission. Further comparing sexual behavior profiles between HIV-positive adults who were and were not aware of their positive status allowed us to investigate whether the observed patterns were compatible with the hypothesis that sexual behavior may change as a result of status knowledge. To examine whether any observed differences in sexual behavior across HIV status categories were attributable to age and sex composition differences, we used marginal structural binomial regression models to age- and sex-standardize prevalence and prevalence ratio estimates to the age and sex distribution of the overall study population.31 We coded age in 10-year intervals. All analyses were conducted using SAS statistical software, version 9.4.32


Of the 5059 older adults enrolled in the study, 46% were men, 51% were currently married, and 46% had no formal education (Table 1). The median age was 61 years with an interquartile range (IQR) between the ages of 52 and 71 years. Compared with men, women were less likely to be married, live alone, report any formal education, and be employed.

Socio-Demographic Characteristics, Sexual Behavior, Sexual History Characteristics, and HIV Status, by Sex, Among HAALSI Participants in Rural South Africa, 2014–2015 (n = 5059)
Socio-Demographic Characteristics, Sexual Behavior, Sexual History Characteristics, and HIV Status, by Sex, Among HAALSI Participants in Rural South Africa, 2014–2015 (n = 5059)

HIV Prevalence and Trends by Age and Sex

Overall, laboratory-confirmed HIV prevalence was high (23%, 95% CI: 22 to 24) and did not differ between men and women. Among those aged 50 years and older, HIV prevalence was 20% (95% CI: 19 to 21). Fewer respondents self-reported being HIV-positive (12% in full sample). About one-third of respondents (36%) reported never having been tested for HIV, and among those with laboratory-confirmed infections, nearly half (49%) self-reported a negative or unknown HIV status. Laboratory-confirmed HIV prevalence was highest in the youngest age categories (as high as 39% among 40–45-year-olds), and generally decreased with age in both men and women (Fig. 1A). Put another way: median age among those who tested HIV-positive [median age (IQR): 54 (47–62)] was considerably younger than among those who tested HIV-negative [median age (IQR): 63 (53–73)]. The age distribution of prevalent HIV infections by sex is reported in Supplemental Digital Content, Table 1,

Prevalence of (A) HIV, (B) recent partnerships, (C) condom use, and (D) casual sex, by age and sex. Study population is HAALSI participants (men and women aged 40 years and above) in rural South Africa, 2014–2015 (n = 5059). Prevalence estimates and 95% confidence intervals were calculated using log-binomial regression models. HIV prevalence estimates are based on laboratory-confirmed testing. Condom use and casual sex outcomes were calculated among those reporting at least 1 sex partner in the past 24 months (n = 2765).

Although the true duration of each HIV infection is unknown, we calculated a crude estimate of the duration of HIV status knowledge by examining the timing of the most recent HIV test. Among those who tested HIV-positive and self-reported being positive, 38% reported their most recent test less than 6 months ago, 21% between 6 months and 1 year ago, and 41% more than 1 year ago.

Prevalence of Sexual Activity and Sexual Behaviors by Sex

Sexual activity and sexual risk behaviors were prevalent in this population (Table 1). Two-thirds of the population reported multiple lifetime sex partners (67%), and more than half reported at least 1 recent sex partner, defined as within the last 2 years (57%). Among those with at least 1 recent partner, three-quarters reported never using condoms with their most recent partner (75%). More than 1 in 10 reported their most recent sex partner was casual or anonymous (12%). Women reported fewer lifetime partners and fewer partners in the past 2 years, compared with men. However, condom use and casual sex were observed at similar rates in women and men with at least 1 recent partner.

Age Trends in Sexual Activity and Sexual Behaviors by Sex

Men maintained sexual partnerships at relatively high rates across older ages, only dropping to 52% at age 80 and older (Fig. 1B). However, the proportion of women with recent sex partners decreased more steeply with age, dropping steadily from 78% at age 40–44 years, to 30% at age 60–64 years, to 6% at age 80 years and older. Multiple recent partnerships were represented in small numbers across men of all ages, and only represented in the youngest age categories for women (Fig. 2). The proportion of those reporting any condom use with their most recent partner decreased with age in both men and women (Fig. 1C), as did the proportion of recent partnerships categorized as casual or anonymous (Fig. 1D). However, low levels of casual sex persisted even at high ages with 10% of men 80 years and older and 17% of women 80 years and older reporting this outcome, among those with at least 1 recent partner.

Number of sex partners in the past 24 months, by age and sex. Study population is HAALSI participants (men and women aged 40 years and above) in rural South Africa, 2014–2015 (n = 5059). Number of sex partners in the past 24 months is categorized into 0, 1, and more than 1 partner.

Sexual Behaviors Across HIV Status Categories

The distribution of some sexual behaviors differed between HIV-negative adults, HIV-positive adults aware of their status, and HIV-positive adults unaware of their status (Fig. 3). Recent sexual partnerships were reported at similar rates across all 3 categories; 57% (95% CI: 55 to 59) of HIV-negative adults, 58% (95% CI: 54 to 62) of HIV-positive adults aware of their status, and 55% (95% CI: 51 to 60) of HIV-positive adults unaware of their status reported recent sex partners. However, among those with at least 1 recent partner, condom use was very low among HIV-negative adults [15% (95% CI: 14 to 17)], higher among HIV-positive adults unaware of their status [27% (95% CI: 22 to 33)], and dramatically higher among HIV-positive adults aware of their status [75% (95% CI: 70 to 80)]. Among those with at least 1 recent partner, casual sex was reported at the lowest level in HIV-negative adults [9% (95% CI: 8 to 10)] with higher levels reported in both HIV-positive categories: among those with status knowledge, casual sex prevalence was 29% (95% CI: 24 to 34) and among those without status knowledge, casual sex prevalence was 18% (95% CI: 14 to 23). Multiple partnerships followed a similar pattern: this outcome was reported in 8% (95% CI: 6 to 9) of HIV-negative adults, 13% (95% CI: 10 to 17) of HIV-positive adults with status knowledge and 9% (95% CI: 6 to 13) of HIV-positive adults without status knowledge. The differences in sexual behaviors remained but tended to attenuate modestly after age and sex standardization. The unstandardized frequency of each sexual behavior, by HIV status, and stratified by sex, is presented in Supplemental Digital Content, Table 2, Unstandardized and age- and sex-standardized prevalence ratios comparing the relative frequency of each sexual behavior across HIV status categories are presented in Supplemental Digital Content, Table 3,

Sexual behavior profiles by HIV status based on self-report and laboratory testing. Study population is HAALSI participants (men and women aged 40 years and older) in rural South Africa, 2014–2015 (n = 5059). A, Prevalence estimates and 95% CIs were calculated using log-binomial regression models. B, Estimates standardized to the age and sex composition of the full study population were calculated using marginal structural log-binomial models. Condom use, casual sex, and multiple partners outcomes were calculated among those reporting at least 1 sex partner in the past 24 months (n = 2765). The HIV-negative category consists of all participants with HIV-negative laboratory test. The category “HIV-positive with positive self-report” consists of all participants with an HIV-positive laboratory test and HIV-positive self-report. The category “HIV-positive with negative or “DK” self-report” consists of all participants with HIV-positive laboratory test and HIV-negative or “don't know” self-report.


Overall, our results suggest that older adults in rural South Africa are at high risk for both HIV acquisition and HIV transmission. We found that HIV prevalence was high among both men and women, and that many of those who tested HIV-positive were unaware of their status. In contrast to stereotypes that older people are not sexually active, and in line with the few previous studies of sexual behavior in older adults in other settings,10–13,15 this population reported recent sexual partnerships even into very old ages and sexual risk behaviors were reported that are consistent with the sexual transmission of disease—low condom use, casual sex, and multiple recent partnerships. The high rates of risky sexual behavior across HIV status categories points to the potential for older HIV-positive adults to transmit HIV to their sex partners and for older HIV-negative adults to acquire HIV from their sex partners.

Our study is one of the first to report sexual risk-taking among older adults in sub-Saharan Africa. Sexual activity was prevalent in this study population, although women tended to report fewer partners compared with men, and sexual activity tended to decrease with age. Although casual sex and multiple partnerships tended to decrease with age, condom use also became less likely with age. Notably, even in the oldest age categories, sexual activity and sexual risk behaviors were reported at rates implying substantial risk among both men and women. The observed continued sexual activity into older ages, low levels of condom use, and associated trends by sex and age are consistent with the 3 previous studies of older adult sexual behavior in sub-Saharan Africa, from Malawi in 1998–2008,10 from the same Agincourt study site in 2010–2011,13 and from Uganda in 2013.15 Even globally, the evidence base on sex in older adults is scarce. However, the few previous studies on this topic (in the United States11,33 and Thailand34) have also shown behavioral trends similar to the ones we observed here.

Important comparisons can also be made between the sexual behavior profiles of older adults and adolescents, a population considered to be at critically high HIV risk.35–37 Compared with both a nationally representative sample of South African adolescents, and a population-based sample of adolescent young women in the same Agincourt study site, this cohort of older adults in rural South Africa was much less likely to use condoms, and among women, more likely to report casual sex.38,39 Thus, older adults in South Africa are likely to differ in their exposure or response to HIV prevention messages.

The HIV prevalence estimates we observed in this study were higher than the estimates from previous studies of HIV prevalence in older adults in South Africa. The overall national HIV prevalence among adults aged 50 years and older was 6.4% in 2007–200840 and 7.6% in 2012.41 The 2008 HIV prevalence among adults aged 50 years and older living in the Africa Centre surveillance site in rural KwaZulu-Natal was 9.5%.42 The 2010–2011 HIV prevalence among adults aged 50 years and older within the same Agincourt study site from which the HAALSI study population was drawn was 16.5%.43 Our HIV prevalence estimate for this same age range was 19.7%, more than 10 percentage points higher than the national and Africa Centre estimates. The prevalence differences over time suggests that HIV prevalence in older adults is currently increasing, a phenomenon likely largely explained by the HIV-positive population living longer because of antiretroviral therapy. Consistent with our results, the previous South African studies of HIV prevalence in older adults showed decreasing HIV prevalence with increasing age and similar trends in both men and women. It is important to note that these findings were primarily observed in older populations in rural South Africa. Extending the generalizability of these findings to other older populations in sub-Saharan Africa should be a priority for future work.

We also found that sexual behavior profiles differed by HIV status. Notably, more condom use was observed among those who tested HIV-positive compared with HIV-negative, but the magnitude of the protective effect was much stronger among those who reported knowledge of their positive status. This finding points to the important influence that HIV testing and counseling may have in changing the behavior of HIV-positive older people to avoid HIV transmission to uninfected partners. However, the very low condom use among sexually active HIV-negative older adults (15%) highlights the potential for incident HIV infections to occur in this high prevalence setting. At the same time, recent casual sex and recent multiple partnerships were also more common among those who tested HIV positive. These findings provide preliminary evidence that the same types of sexual behaviors established as risk factors for HIV among adolescents and younger adults are also risk factors for the sexual transmission of HIV among older adults.19–21 Further examination of whether the magnitudes and patterns of these risks may change in older age is warranted.

The associations we observe here generate hypotheses about causal relationships between HIV status and sexual behavior, but they cannot serve as a test of such hypotheses. For those who tested HIV-positive, the duration of infection is unknown—transmission may have occurred relatively recently for some, whereas infections may be much older in others. The sexual behaviors we evaluated were collected with reference to the most recent 2-year time period, although behaviors reported for this period could have been established much earlier and may be representative of behaviors that also occurred at younger ages. In theory, the association between HIV status and sexual behavior is bidirectional. Older adults may plausibly change their behavior in response to knowledge of their HIV status to avoid transmission to their partners or acquisition from their partners. Conversely, risky sexual behaviors among older people could lead to subsequent HIV infection. Data on the timing of the most recent HIV test among those who test positive suggest a range in the length of time participants have known about their positive status from relatively recent (less than 6 months) to relatively long (more than 1 year). The fact that we observe short lengths of positive status knowledge is compatible with the hypothesis that sexual behaviors influence HIV risk. The fact that we also observe longer lengths of positive status knowledge is compatible with the hypothesis that older people modify their behaviors in response to their HIV status. Future analyses of longitudinal data on sexual behavior and HIV status could provide insights into the relative contributions of each hypothesized association.

Many of the outcomes we present in this article were self-reported by the participants in face-to-face interviews and may thus have been subject to social desirability bias. Sensitive questions about personal sexual behaviors were likely underreported in this sample. Although we found that a large proportion of participants were willing to report on “undesirable” outcomes such as sexual activity, condom use, and casual sex, the prevalence values we report may underestimate the true values of these sexual behaviors.44 Likewise, HIV status is also a sensitive outcome to self-report and those who tested HIV-positive but reported a negative or unknown status may have, in fact, known they were positive, or avoided previous HIV testing because of suspicions about being HIV-positive. It is thus possible that the true prevalence of HIV infection by self-report among those with laboratory-confirmed HIV-positive status is higher than we observed. However, the laboratory-identified infections among those who tested are objective measures and not subject to the same biases as self-reported status.

Our findings add to the growing body of evidence suggesting older adults urgently require HIV interventions that are tailored to their particular prevention needs. Among HIV-positive older adults, “positive prevention” campaigns with intensified counseling and motivation about sexual transmission risks should be considered,45,46 with attention to HIV-related stigma that may be experienced by older adults.47,48 Specific programs ensuring universal HIV testing, rapid linkage of HIV-positive people to care, and supported HIV treatment retention and adherence should also be considered.17,49 Among HIV-negative older adults, it will be important to consider how established interventions, such as medical male circumcision50 and condom promotion, and novel interventions, such as preexposure prophylaxis,18 can be designed and delivered in ways that best meet the needs of older adults.46,49 In general, it is likely that to reach older adults with HIV prevention messages, interventions will need to explore disseminating messages through different channels than have been successfully used for younger age groups (ie, social media and social venues). Older adults may also face different physical and cognitive barriers to interventions access than found in younger populations. HIV intervention research is urgently required to intensify and improve HIV prevention in the important but neglected vulnerable group of older adults.


The authors thank all involved in ensuring productive field research in HAALSI and the Agincourt HDSS, including the field staff, analysts, and, most importantly, the study participants themselves.


1. Mills EJ, Rammohan A, Awofeso N. Ageing faster with AIDS in Africa. Lancet. 2011;377:1131–1133.
2. Negin J, Bärnighausen T, Lundgren JD, et al. Aging with HIV in Africa: the challenges of living longer. AIDS. 2012;26(suppl 1):S1–S5.
3. Rammohan A, Awofeso N. Addressing HIV prevention and disease burden among Africans aged over 50 years. Trop Doct. 2010;40:171–172.
4. Mills EJ, Bärnighausen T, Negin J. HIV and aging—preparing for the challenges ahead. N Engl J Med. 2012;366:1270–1273.
5. Negin J, Cumming RG. HIV infection in older adults in sub-Saharan Africa: extrapolating prevalence from existing data. Bull World Health Organ. 2010;88:847–853.
6. Mahy M, Autenrieth CS, Stanecki K, et al. Increasing trends in HIV prevalence among people aged 50 years and older: evidence from estimates and survey data. AIDS. 2014;28(suppl 4):S453–S459.
7. Nyigo V, Kilale A, Kilima S, et al. Magnitude of HIV infection among older people in Mufindi and Babati districts of the Tanzania mainland. HIV AIDS (Auckl). 2014;6:75–79.
8. Hontelez JA, Lurie MN, Newell M-L, et al. Ageing with HIV in south africa. AIDS (London, England). 2011;25:1665–1667.
9. Hontelez JA, de Vlas SJ, Baltussen R, et al. The impact of antiretroviral treatment on the age composition of the HIV epidemic in sub-Saharan Africa. AIDS (London, England). 2012;26(suppl 1):S19–30.
10. Freeman E, Anglewicz P. HIV prevalence and sexual behaviour at older ages in rural Malawi. Int J STD AIDS. 2012;23:490–496.
11. Pilowsky DJ, Wu LT. Sexual risk behaviors and HIV risk among Americans aged 50 years or older: a review. Subst Abuse Rehabil. 2015;6:51–60.
12. Lekalakala-Mokgele E. Understanding of the risk of HIV infection among the elderly in Ga-Rankuwa, South Africa. SAHARA J. 2014;11:67–75.
13. Mojola SA, Williams J, Angotti N, et al. HIV after 40 in rural South Africa: a life course approach to HIV vulnerability among middle aged and older adults. Social Sci Med. 2015;143:204–212.
14. UNAIDS. Fact Sheet 2015. Geneva, Switzerland: UNAIDS; 2015.
15. Negin J, Geddes L, Brennan-Ing M, et al. Sexual behavior of older adults living with HIV in Uganda. Arch Sex Behav. 2016;45:441–449.
16. Centers for Disease Control and Prevention. Effective interventions: HIV prevention that works—behavioral interventions. 2015. Available at: Accessed September 26, 2014.
17. Cohen MS, Chen YQ, McCauley M, et al. Prevention of HIV-1 infection with early antiretroviral therapy. N Engl J Med. 2011;365:493–505.
18. Baeten JM, Donnell D, Ndase P, et al. Antiretroviral prophylaxis for HIV prevention in heterosexual men and women. N Engl J Med. 2012;367:399–410.
19. Weller S, Davis-Beaty K. Condom effectiveness in reducing heterosexual HIV transmission (review). Cochrane Database Syst Rev. 2002;1:CD003255.
20. Mishra V, Hong R, Assche S, et al. The role of partner reduction and faithfulness in HIV prevention in Sub-Saharan Africa: evidence from Cameroon, Rwanda, Uganda, and Zimbabwe. DHS Working Paper No. 61. Calverton, Maryland: Macro International Inc.; 2009.
21. Chen L, Jha P, Stirling B, et al. Sexual risk factors for HIV infection in early and advanced HIV epidemics in sub-Saharan Africa: systematic overview of 68 epidemiological studies. PLoS One. 2007;2:e1001.
22. Denison JA, O'Reilly KR, Schmid GP, et al. HIV voluntary counseling and testing and behavioral risk reduction in developing countries: a meta-analysis, 1990–2005. AIDS Behav. 2008;12:363–373.
23. Fonner VA, Denison J, Kennedy CE, et al. Voluntary counseling and testing (VCT) for changing HIV-related risk behavior in developing countries. Cochrane Database Syst Rev. 2012;9:CD001224.
24. Kahn K, Collinson MA, Gomez-Olive FX, et al. Profile: Agincourt health and socio-demographic surveillance system. Int J Epidemiol. 2012;41:988–1001.
25. World Health Organization. Definition of an older or elderly person: proposed working definition of an older person in Africa for the MDS Project. 2016. Available at: Accessed September 26, 2014.
26. Bor J, Rosen S, Chimbindi N, et al. Mass HIV treatment and sex disparities in life expectancy: demographic surveillance in rural South Africa. PLoS Med. 2015;12:e1001905.
27. Bor J, Herbst AJ, Newell ML, et al. Increases in adult life expectancy in rural South Africa: valuing the scale-up of HIV treatment. Science. 2013;339:961–965.
28. Kahn K, Garenne ML, Collinson MA, et al. Mortality trends in a new South Africa: hard to make a fresh start. Scand J Public Health Suppl. 2007;35(69 suppl l):26–34.
29. World Health Organization. Guidelines for Using HIV Testing Technologies in Surveillance. Geneva, Switzerland: UNAIDS; 2009.
30. Rohr J, Rosenberg M, Gomez-Olive X, et al. Validation of self-reported HIV status among older adults in rural South Africa. Conference on retroviruses and opportunistic infections, Boston, MA, 23 February 2016.
31. Richardson DB, Kinlaw AC, MacLehose RF, et al. Standardized binomial models for risk or prevalence ratios and differences. International journal of epidemiology. 2015;44:1660–1672.
32. SAS [computer program]. Version 9.4. Cary, NC: SAS Institute Inc; 2014.
33. Lindau ST, Schumm LP, Laumann EO, et al. A study of sexuality and health among older adults in the United States. N Engl J Med. 2007;357:762–774.
34. Knodel J, Chayovan N. Sexual activity among older Thais: the influence of age, gender and health. J Cross Cult Gerontol. 2001;16:173–200.
35. UNICEF, UNAIDS, WHO. Young People and HIV/AIDS: Opportunity in Crisis. 2002. Available at: Accessed September 26, 2016.
36. UNAIDS. The Gap Report. Geneva, Switzerland: UNAIDS; 2014.
37. United Nation's Children's Fund (UNICEF). Opportunity in Crisis: Preventing HIV From Early Adolescence to Young Adulthood. New York City, NY: UNICEF; 2011.
38. Pettifor AE, Rees HV, Kleinschmidt I, et al. Young people's sexual health in South Africa: HIV prevalence and sexual behaviors from a nationally representative household survey. AIDS. 2005;19:1525–1534.
39. Pettifor A, MacPhail C, Selin A, et al. HPTN 068: A Randomized Control Trial of a Conditional Cash Transfer to Reduce HIV Infection in Young Women in South Africa—Study Design and Baseline Results. AIDS Behav. 2016;20:1863–1882.
40. Negin J, Nemser B, Cumming R, et al. HIV attitudes, awareness and testing among older adults in Africa. AIDS Behav. 2012;16:63–68.
41. Shisana O, Labadarios D, Simbayi L, et al. South African National HIV Prevalence, Incidence and Behaviour Survey, 2012. Cape Town, South Africa: 2014.
42. Wallrauch C, Bärnighausen T, Newell ML. HIV prevalence and incidence in people 50 years and older in rural South Africa. S Afr Med J. 2010;100:812–814.
43. Gómez-Olivé FX, Angotti N, Houle B, et al. Prevalence of HIV among those 15 and older in rural South Africa. AIDS Care. 2013;25:1122–1128.
44. Houle B, Angotti N, Clark SJ, et al. Let's Talk about Sex, Maybe Interviewers, Respondents, and Sexual Behavior Reporting in Rural South Africa. Field Methods. 2015;28:112–132.
45. International HIV/AIDS Alliance. Positive Prevention: HIV Prevention with People Living with HIV: A Guide for NGOs and Service Providers. 2007. Available at: Accessed September 26, 2016.
46. Davis T, Zanjani F. Prevention of HIV among older adults: A literature review and recommendations for future research. Journal of Aging and Health. 2012;24:1399–1420.
47. Kuteesa MO, Seeley J, Cumming RG, et al. Older people living with HIV in Uganda: understanding their experience and needs. Afr J AIDS Res. 2012;11:295–305.
48. Kuteesa MO, Wright S, Seeley J, et al. Experiences of HIV-related stigma among HIV-positive older persons in Uganda—a mixed methods analysis. SAHARA J. 2014;11:126–137.
49. Negin J, Rozea A, Martiniuk AL. HIV behavioural interventions targeted towards older adults: a systematic review. BMC Public Health. 2014;14:507.
50. Auvert B, Taljaard D, Lagarde E, et al. Randomized, controlled intervention trial of male circumcision for reduction of HIV infection risk: the ANRS 1265 Trial. PLoS Med. 2005;2:e298.

aging population; older adults; sexual behavior; South Africa; HIV acquisition risk; HIV transmission risk

Supplemental Digital Content

Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.