Antiretroviral therapy to prevent HIV acquisition in ...

Clinical Infectious Diseases Advance Access published May 20, 2016

1 Antiretroviral therapy to prevent HIV acquisition in serodiscordant couples in a

cr ipt

hyperendemic community in rural South Africa

Catherine E. Oldenburg1, Till Bärnighausen2,3,*, Frank Tanser2,4, Collins C. Iwuji2, Victor De Gruttola5, George R. Seage III1, Matthew J. Mimiaga1,6,7, Kenneth H. Mayer3,6,8,

Department of Epidemiology, Harvard T.H. Chan School of Public Health, Boston, MA, USA

2

Africa Centre for Population Health, Mtubatuba, South Africa

3

Department of Global Health and Population, Harvard T.H. Chan School of Public Health,

M

Boston, MA, USA

an

1

Faculty of Health Sciences, University of KwaZulu-Natal, Durban, South Africa

5

Department of Biostatistics, Harvard T.H. Chan School of Public Health, Boston, MA, USA

6

The Fenway Institute, Fenway Community Health, Boston, MA, USA

7

Departments of Behavioral and Social Sciences & Epidemiology; The Institute for Community

pt ed

4

Health Promotion, Brown University School of Public Health, Providence, RI, USA Department of Medicine, Beth Israel Deaconess Medical Center, Boston, MA, USA

Ac

ce

8

© The Author 2016. Published by Oxford University Press for the Infectious Diseases Society of America. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.

Downloaded from http://cid.oxfordjournals.org/ by guest on May 24, 2016

us

Deenan Pillay3, Guy Harling2,3

2 *

Corresponding Author: Till Bärnighausen, Africa Centre for Population Health, PO Box 198,

3935 Mtubatuba, South Africa, Harvard T.H. Chan School of Public Health, 665 Huntington

cr ipt

Avenue, Boston, US, Email: [email protected], [email protected], Telephone: +1 617 379 0372

an

Background. Antiretroviral therapy (ART) was highly efficacious in preventing HIV transmission in stable serodiscordant couples in the HPTN-052 study, a resource-rich

M

randomized controlled trial. However, minimal evidence exists of the effectiveness of ART in preventing HIV acquisition in stable serodiscordant couples in real-life population-based settings in hyperendemic communities of Sub-Saharan Africa, where health systems are typically

pt ed

resource-poor and overburdened, adherence to ART is subpotimal, and HIV status disclosure to sexual partners is inconsistent.

Methods. Data arose from a population-based open cohort in KwaZulu-Natal, South Africa. HIV-uninfected individuals present between January 2005 and December 2013 (n=17,016) were

ce

included. Interval-censored time-updated proportional hazards regression was used to assess how the ART status affected HIV transmission risk in stable serodiscordant relationships.

Ac

Results. Of 17,016 individuals, 1,846 had an HIV-uninfected and 196 had an HIV-infected stable partner over the follow-up period. HIV incidence was 3.8 per 100 person-years (100PY) among individuals with an HIV-infected partner (95% confidence interval [CI] 2.3-5.6), corresponding to 1.4 per 100PY (95% CI 0.4-3.5) among those with HIV-infected partners on


us

ABSTRACT

3 ART and 5.6 per 100PY (95% CI 3.5-8.4) among those with partners not on ART. Use of ART

(aHR=0.23, 95% CI 0.07-0.80).

cr ipt

was associated with a 77% decrease in HIV acquisition risk amongst serodiscordant couples

Conclusions. ART initiation was associated with a very large reduction in HIV aquisition in serodiscordant couples in rural KwaZulu-Natal. However, real-life effectiveness was

substantially lower than in the HPTN-052 trial. To eliminate HIV transmission in sero-discordant

an M pt ed ce Ac


us

couples, additional prevention interventions are likely needed.

4 INTRODUCTION HIV transmission within stable heterosexual partnerships is thought to be a major

cr ipt

contributor to new HIV infection in Sub-Saharan Africa (SSA).1-3 In the pre-antiretroviral

therapy (ART) era, HIV incidence has been estimated at approximately 5 cases per 100 personyears among men and 10 cases per 100 person-years among women in serodiscordant

relationships in Tanzania4 and Uganda.2 Higher HIV incidence among women in serodiscordant

than men.5 In some studies from SSA, marriage has been implicated as a risk factor for HIV

an

acquisition among women,6,7 perhaps due to greater vulnerability to HIV within marriage (e.g., difficulty with condom use negotiation), or to women being more likely to acquire HIV prior to

M

marriage.8-11

The introduction of ART has resulted in dramatic decreases in HIV transmission globally.12-16 The landmark HPTN-052 study demonstrated near elimination in HIV incidence in

pt ed

serodiscordant couples with immediate versus delayed initiation of ART.12 A recent metaanalysis showed greatly reduced HIV transmission in serodiscordant couples where the HIV positive partner is virally suppressed.13 Increasing ART coverage at the community14 and household16 level is associated with decreased HIV transmission, and modeling studies have

ce

suggested that HIV transmission could be eliminated at 90% ART coverage.17 However, evidence of the efficacy of ART has primarily arisen from randomized trials or cohorts of

Ac

serodiscordant couples who volunteered to participate in a clinical study by enrolling in serodiscordant couple clinics or partner testing programs. These studies were highly controlled and resource-rich and thus do not represent the real-life settings of resource-poor public-sector health systems in the hyperendemic communities of SSA. In HPTN-052, viral suppression was


us

relationships may be related to higher per-act probability of HIV transmission among women

5 greater than 89% among individuals immediately initiating ART for the course of the study,18 compared 77% in public-sector ART clinics in KwaZulu-Natal19, which could lead to reduced

cr ipt

effect sizes. Importantly, all individuals in the prior studies were aware of their HIV status and

had disclosed their status to their partners.20-25 In real-life settings in SSA, individuals commonly do not disclose their HIV status to their partners for fear of negative consequences to their lives.26

serodiscordant couples exists from real-life population-based settings with resource-poor health

an

systems. In this study, we aim to establish such evidence in an HIV hyperendemic rural

M

community in KwaZulu-Natal.

METHODS

Study setting. Data arose from the population-based longitudinal surveillance program conducted

pt ed

by the Africa Centre for Population Health (“Africa Centre”).27 This surveillance program is located in a predominantly rural community of uMkhanyakude district, KwaZulu-Natal. The Africa Centre has conducted an HIV surveillance program since 2003, including confidential HIV testing, collection of sexual history and behaviors, relationship status, and household

ce

demographic data. The surveillance covers a 438-km2 demographic surveillance area (DSA) and includes all members of all households.

Ac

KwaZulu-Natal is the province that bears the largest HIV burden in South Africa.28-30

Age-specific incidence peaks at more than 6 cases per 100 person-years (100PY) among women

in KwaZulu-Natal and 4 cases per 100 person-years among men.14,31,32 Adult HIV prevalence in this region is 29%,33 and only 34% of working-age adults are employed.34 ART coverage has


us

To date, little empirical evidence on the effectiveness of HIV treatment-as-prevention in

6 increased rapidly since 2004 primarily via nurse-led public sector ART programs,14,33 but viral suppression and immunologic recovery among ART patients is comparatively low.19 ART is

cr ipt

provided primarily via primary care clinics that are run by nurses.19 Although ART coverage has expanded rapidly14, the program suffers from drug stock-outs, difficulty with retention in care, and lower viral suppression compared to other settings and studies.18,19

Participants were included in this analysis if they had a first negative HIV test followed by at

an

least one subsequent HIV test (regardless of result) and, for those in stable relationships, reported only a single stable partner. Baseline was considered the earliest date at which an individual

M

tested HIV negative.

Ethical approval for data collection, linkage, and analysis was obtained from the University of KwaZulu-Natal Biomedical Research Ethics Committee, and written informed

pt ed

consent was obtained from all participants. The analyses presented here were exempted from additional ethical review by the Harvard School of Public Health Institutional Review Board due to their use of anonymized secondary data.

ce

Outcome ascertainment. The primary outcome for analyses was time-to-HIV seroconversion, where seroconversion required at least one negative HIV test followed by at least one positive

Ac

test.

Exposure ascertainment. The primary exposures of interest were the HIV and ART status of each respondent’s partner. In this region, a large number of individuals are in primary, co-


us

Participants and procedures. Data were available from January 2005 through December 2013.

7 habiting, ‘conjugal relationships’ (henceforth, ‘stable partnerships’), who are members of the same household, may have had children together, and are socially recognized as being partners.35

cr ipt

These partnerships are more common among older individuals than among younger individuals, who more frequently live in households headed by their parents or kin. Stable partnerships were identified via demographic household surveillance data. Unique Africa Centre identification

codes were used to link partners to one-another. Stable partnerships remain under observation

The HIV status of each stable partner was ascertained from HIV surveillance data.

an

Respondents were considered to be in a serodiscordant relationship if their stable partner had tested positive before or at the same time as the respondent’s current negative test, or if they had

M

initiated ART prior to the respondent’s current test. Respondents were considered to be in a seroconcordant negative relationship if their partner tested negative on or after the day the respondent tested negative. Respondents were censored at the end of the relationship, on the date

pt ed

of their first positive HIV test, or on the date of their last negative HIV test, whichever was first. To determine ART status of HIV-infected partners, we linked demographic surveillance data to information from the Hlabisa HIV Treatment and Care Programme, which provides ART through 17 public-sector primary care clinics locally. HIV-infected partners were defined as

ce

being on ART at the time of a demographic surveillance visit if they had a record of having initiated ART prior to this date.

Ac

Respondents not in stable partnerships were categorized as being in a relationship or not

based on self-reported relationship status. Participants who reported being married or engaged, but not co-habiting with their partner and thus not identified by the demographic surveillance,


us

until they are reported to have ended by either partner during a household visit.

8 were coded as being in a non-co-habiting relationship. Those who did not report a current relationship were coded as such.

cr ipt

Partnership status was time-varying, as individuals could begin or end relationships, and

partners’ HIV or ART status could change. Individuals could contribute person-time to multiple exposure categories.

education (none or primary: 0-7 years; secondary: 8-12 years; tertiary: >12 years), household

an

wealth (quintiles of the first component identified by principal components analysis of 32 household assets and characteristics), and in the past 12 months having had more than one

baseline and sex.

M

partner or inconsistent condom use with any partner. Additional covariates included age at

pt ed

Statistical methods. Exact HIV seroconversion dates are not typically observed in cohort study research. Instead, the event time is ‘interval censored’, in that it is known that it occurs within an interval of time (between the date of the last negative HIV test and the date of the first positive HIV test). The standard approach to interval censored data is impute an exact seroconversion

ce

date, for example by assuming that the HIV seroconversion occurred at the midpoint between last negative and first positive dates.36 However, this approach has been demonstrated to lead to

Ac

underestimated standard errors, and can lead to misclassification if the date of the exposure is known with precision but not the event date.36 Here, the date of exposure is known with

precision as ART initiation dates are captured in ART clinic data, but HIV seroconversion dates are not known with precision due to the annual nature of the HIV surveillance. To avoid such


us

Covariates. Time-varying potential confounders included respondent-reported current completed

9 problems, we fitted interval-censored parametric proportional hazards models, which

varying nature of the exposure and potential confounders.37,38

cr ipt

appropriately account for both uncertainty regarding HIV seroconversion timing, and the time-

A series of three models were included for each contrast of interest (HIV-infected versus HIV-uninfected partner; HIV-infected partner on ART versus not on ART). Model 1 contained

an indicator term for each partnership status category, the individual’s age, sex, and an indicator

further added sexual behavior variables. For models considering the effect of partner HIV status

an

on HIV incidence, the referent category was having an HIV-uninfected partner. For models considering the effect of partner ART status, HIV-infected partner’s ART status was modeled as

range from 0 to 1.

M

the proportion of the interval that was covered by ART, and thus the indicator variable could

Several sensitivity analyses were run. First, given that ART use does not result in

pt ed

immediate viral suppression, we modeled partner’s ART use with a ‘wash-in’ period of 30 days, assuming that the first 30 days following ART initiation were untreated in an interval censored model. Second, a Cox proportional hazards model using a midpoint imputed HIV seroconversion date was run, to compare the interval censored approach to the more standard midpoint

ce

imputation approach. Analyses were performed in SAS 9.2 and Stata 13.1. Role of the funding source. The funding source did not have a role in the design, analysis,

Ac

interpretation, or decision to publish this analysis.


us

for year of observation. Model 2 added educational attainment and household wealth. Model 3

10 RESULTS Between January 2005 and December 2013, 17,016 individuals contributing 60,349

cr ipt

person-years of follow met the inclusion criteria. Of 2,029 individuals with a cohabiting partner during the follow-up period, 196 individuals had an HIV-infected cohabiting partner, and 1,846 individuals had an HIV-uninfected stable partner. Of individuals with an HIV-infected

cohabiting partner, 20 were on ART at baseline, and a further 56 initiated ART over the follow-

21.4% three times, 32.1% four to five times, 17.5% six to seven times, and 4.6% eight or nine

an

times). The median time between tests was 374 days (interquartile range 352 to 700 days). Table 1 lists baseline characteristics for the study sample. Approximately 63% of the study sample was

M

female, and women tended to be older than men (mean age of 36 versus 29). At baseline, 29% of participants reported being aware of their HIV status.

Table 2 shows HIV incidence by partnership status. Overall HIV incidence was 2.7 new

pt ed

infections per 100PY (95% CI 2.6 to 2.8 per 100PY), with an incidence of 5.6 per 100PY among individuals with an HIV-infected partner who was not on ART (95% CI 3.5 to 8.4 per 100PY), 1.4 per 100PY among individuals with an HIV-infected partner who was on ART (95% CI 0.4 to 3.5 per 100PY), and 0.3 cases per 100PY among individuals with an HIV-uninfected partner

ce

(95% CI 0.2 to 0.5). Men with HIV-uninfected partners had higher HIV incidence than women with HIV-uninfected partners, whereas women with HIV-infected partners had higher HIV

Ac

incidence than men with HIV-infected partners. All partnership categories had significantly elevated HIV incidence compared to

individual in stable partnerships with HIV-uninfected partners (Table 3). Individuals with an HIV-infected stable partner had the greatest elevation in HIV incidence, with more than 10 times


us

up period. Individuals were tested for HIV between two and nine times (24.4% tested twice,

11 the HIV incidence compared to individuals with HIV-uninfected stable partners (aHR 10.04, 95% CI 5.51 to 18.32; Table 3). This result was robust to adjustment for socioeconomic and

cr ipt

sexual behavior-related factors (Table 3). The use of ART by an HIV-infected partner was

associated with a 77% reduction in HIV incidence compared to no ART use by an HIV-infected partner (aHR 0.23, 95% CI 0.07 to 0.80; Table 4). As expected, this effect was slightly stronger

than that estimated by the midpoint imputed Cox proportional hazards model (aHR 0.28, 95% CI

DISCUSSION

an

assuming no effect of ART during the first 30 days of ART initiation (Table S5).

M

In controlled and resource-intensive randomized trials and clinical cohort studies, early initiation of ART dramatically reduces HIV transmission to uninfected partners in serodiscordant relationships.12,15,39 In the landmark HPTN-052 study, immediate ART initiation resulted in a

pt ed

96% reduction in linked HIV transmission and 89% reduction in overall HIV transmission among serodiscordant couples.12 In the present study we tested the effectiveness of ART in reducing HIV acquisition in serodiscordant couples in a real-life population-based setting. We find that ART is highly effective in reducing HIV acquisition in serodiscordant couples in a poor

ce

HIV hyperendemic community, despite severe resource constraints in the public-sector health system in this setting. In this community, ART is delivered through primary care clinics that are

Ac

staffed and led exclusively by nurses. Supply chain failures in this setting have frequently led to drug stock-outs over the observation period, and patients commonly miss scheduled visits to the ART program and adherence to ART is lower than is seen in more tightly controlled study settings. Despite these imperfections of real-life ART delivery, our results demonstrate that


us

0.10 to 0.80; Table S4), with a slightly wider confidence interval. This result was robust to

12 ART is highly effective in preventing HIV transmission in serodiscordant couples. Because this study was population-based, including all stable discordant couples in the community, our

cr ipt

findings are generalizable beyond those of the previous controlled clinical studies, which

enrolled only individuals who utilized health care targeting couples and who had disclosed their HIV status to their partner. 12,20-24,40

Our results must be considered in the context of several limitations. First, as with any

could be introduced if those who tested systematically differed from those who declined testing.

an

Second, we cannot rule out bias due to unmeasured confounding. For example, individuals with greater health-seeking behavior may both be more likely to link to care and initiate ART and

M

more likely to use condoms. However, the longitudinal nature of this analysis allowed for adjustment of time-varying confounding by a range of sociodemographic and behavioral factors, that likely at least partially control for such effects. Importantly, the large effect sizes in this

results.

pt ed

study mean that any bias due to confounding would have to be considerable to nullify these

Third, while this study represents a true population cohort, our findings may not be generalizable to all contexts. In rural South Africa, co-habitation, which would lead to

ce

documentation as being in a “conjugal relationship”, tends to occur at older ages,35 while HIV incidence is highest in younger men and women.14,29 Similar, sexual partnership structures are

Ac

found in other poor rural settings in Southern Africa. However, our results may not be generalizable to communities with different partnerships structures, e.g., where serodiscordant partnerships are prevalent among younger individuals. Finally, observed HIV incidence among individuals with HIV-uninfected stable partners was relatively low, and while this may partially


us

long-term population-based study, there was some attrition and nonresponse. Selection bias

13 reflect overall lower HIV risk behaviors among this older population, it may also reflect a selection effect: in this very high HIV prevalence setting, those who have not already acquired

cr ipt

HIV by the time they enter a seroconcordant-negative stable partnership may have lifelong risk behaviors that are qualitatively different from the rest of the population.

Conclusion

even when it is provided in a resource-poor, public-sector health care system in a hyperendemic

an

community. While the HIV prevention effect in this ‘real life’ study is large, it falls short of the near-complete elimination of HIV transmission observed in the HPTN-052 study, suggesting that

Ac

ce

pt ed

serodiscordant couples in SSA.

M

additional prevention interventions will likely be required to eliminate HIV transmission in


us

ART is a highly effective strategy to prevent HIV transmission in serodiscordant couples,

14 ACKNOWLEDGMENTS The Africa Centre for Population Health, University of KwaZulu-Natal, South Africa receives

cr ipt

core funding from the Wellcome Trust (www.wellcome.ac.uk) grant 082384/Z/07/Z. FUNDING

The Hlabisa HIV Treatment and Care Programme was funded by the generous support of the

American people through the United States Agency for International Development (USAID) and

This work was partially supported by the National Institutes of Health awards T32DA013911

DISCLAIMER:

an

(CO), R25MH083620 (CO) and R01HD084233.

M

The contents are the responsibility of the authors and do not necessarily reflect the views of any of the funders or the U.S. government. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

pt ed

CONFLICTS OF INTEREST None to declare.

ETHICAL APPROVAL

Ethical approval for data collection, linkage, and analysis was obtained from the University of

ce

KwaZulu-Natal Biomedical Research Ethics Committee, and written informed consent was obtained from all participants. The analyses presented here were exempted from additional

Ac

ethical review by the Harvard School of Public Health Institutional Review Board due to their use of anonymized secondary data.


us

the President’s Emergency Plan (PEPFAR) under the terms of Award No. 674-A-00-08-0001-00.

15 REFERENCES 1

Biraro S, Ruzagira E, Kamali A, Whitworth J, Grosskurth H, Weiss HA. HIV-1

cr ipt

Transmission within Marriage in Rural Uganda: A Longitudinal Study. PLoS ONE 2013; 8: e55060. 2

Carpenter LM, Kamali A, Ruberantwari A, Malamba SS, Whitworth JA. Rates of HIV-1

partners. AIDS 1999; 13: 1083–9.

Dunkle KL, Stephenson R, Karita E, et al. New heterosexually transmitted HIV infections

an

3

in married or cohabiting couples in urban Zambia and Rwanda: An analysis of survey and

4

M

clinical data. The Lancet 2008; 371: 2183–91.

Hugonnet S, Mosha F, Todd J, et al. Incidence of HIV infection in stable sexual

pt ed

partnerships: A retrospective cohort study of 1802 couples in Mwanza Region, Tanzania. JAIDS Journal of Acquired Immune Deficiency Syndromes 2002; 30: 73–80. 5

Hughes JP, Baeten JM, Lingappa JR, et al. Determinants of Per-Coital-Act HIV-1 Infectivity Among African HIV-1-Serodiscordant Couples. J INFECT DIS 2012; 205: 358–

6

ce

65.

Tenkorang EY. Marriage, widowhood, divorce and HIV risks among women in sub-

Ac

Saharan Africa. International Health 2014; 6: 46–53.

7

Clark S. Early Marriage and HIV Risks in Sub-Saharan Africa. Studies in Family Planning 2004; 35: 149–60.


us

transmission within marriage in rural Uganda in relation to the HIV sero-status of the

16 8

Nalugoda F, Guwatudde D, Bwaninka JB, et al. Marriage and the Risk of Incident HIV Infection in Rakai, Uganda. JAIDS Journal of Acquired Immune Deficiency Syndromes

9

cr ipt

2014; 65: 91–8.

Glynn JR, Carael M, Auvert B, et al. Why do young women have a much higher prevalence of HIV than young men? A study in Kisumu, Kenya and Ndola, Zambia. AIDS 2001; 15 :

10

us

S51–S60.

Glynn JR, Carael M, Buve A, Musonda RM, Kahindo M, Study Gruop on the

an

with high prevalence of HIV infection. JAIDS Journal of Acquired Immune Deficiency

11

M

Syndromes 2003; 33: 526–35.

Lurie MN, Williams BG, Zuma K, et al. Who infects whom? HIV-1 concordance and

52.

pt ed

discordance among migrant and non-migrant couples in South Africa. AIDS 2003; : 2245–

12 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. Loutfy MR, Wu W, Letchumanan M, et al. Systematic Review of HIV Transmission

ce

13

between Heterosexual Serodiscordant Couples where the HIV-Positive Partner Is Fully

Ac

Suppressed on Antiretroviral Therapy. PLoS ONE 2013; 8: e55747.

14

Tanser F, Bärnighausen T, Grapsa E, Zaidi J, Newell M-L. High Coverage of ART Associated with Decline in Risk of HIV Acquisition in Rural KwaZulu-Natal, South Africa. Science 2013; 339: 966–71.


Heterogeneity of HIV Epidemics in African Cities. HIV risk in relation to marriage in areas

17 15 Cohen MS, Smith MK, Muessig KE, Hallett TB, Powers KA, Kashuba A. Antiretroviral treatment of HIV-1 prevents transmission of HIV-1: where do we go from here? The Lancet

cr ipt

2013; 382: 1515–24.

16 Vandormael A, Newell M-L, Bärnighausen T, Tanser F. Use of antiretroviral therapy in

households and risk of HIV acquisition in rural KwaZulu-Natal, South Africa, 2004–12: a

17

us

prospective cohort study. The Lancet Global Health 2014; 2: e209–15.

Hontelez JAC, Lurie MN, Bärnighausen T, et al. Elimination of HIV in South Africa

an

Medicine 2013; 10: e1001534.

M

18 Grinsztejn B, Hosseinipour MC, Ribaudo H, et al. Effects of early versus delayed initiation of antiretroviral treatment on clinical outcomes of HIV-1 infection: results from the phase 3

19

pt ed

HPTN 052 randomised controlled trial. The Lancet Infectious Diseases 2014; 14: 281–90. Mutevedzi PC, Lessells RJ, Heller T, Bärnighausen T, Cooke GS, Newell M-L. Scale-up of a decentralized HIV treatment programme in rural KwaZulu-Natal, South Africa: does rapid expansion affect patient outcomes? Bull World Health Organ 2010; 88: 593–600. Del Romero J, Castilla J, Hernando V, Rodriguez C, Garcia S. Combined antiretroviral

ce

20

treatment and heterosexual transmission of HIV-1: cross sectional and prospective cohort

Ac

study. BMJ 2010; 340: c2205–5.

21

Donnell D, Baeten JM, Kiarie J, et al. Heterosexual HIV-1 transmission after initiation ofantiretroviral therapy: a prospective cohort analysis. The Lancet 2010; 375: 2092–8.


through Expanded Access to Antiretroviral Therapy: A Model Comparison Study. PLoS

18 22 Reynolds SJ, Makumbi F, Nakigozi G, et al. HIV-1 transmission among HIV-1 discordant

cr ipt

couples before and after the introduction of antiretroviral therapy. AIDS 2011; 25: 473–7. 23 Wang L, Ge Z, Luo J, et al. HIV Transmission Risk Among Serodiscordant Couples: A

Retrospective Study of Former Plasma Donors in Henan, China. JAIDS Journal of Acquired Immune Deficiency Syndromes 2010; 55: 232–8.

Jia Z, Mao Y, Zhang F, et al. ArticlesAntiretroviral therapy to prevent HIV transmission in

us

24

serodiscordant couples in China (2003–11): a national observational cohort study. The

an

25

Birungi J, Min JE, Muldoon KA, et al. Lack of Effectiveness of Antiretroviral Therapy in

M

Preventing HIV Infection in Serodiscordant Couples in Uganda: An Observational Study. PLoS ONE 2015; 10: e0132182.

pt ed

26 Kennedy CE, Fonner VA, Armstrong KA, O’Reilly KR, Sweat MD. Increasing HIV serostatus disclosure in low and middle-income countries. AIDS 2015; 29: S7–S23. 27

Tanser F, Hosegood V, Bärnighausen T, et al. Cohort Profile: Africa Centre Demographic Information System (ACDIS) and population-based HIV survey. International Journal of

Naicker N, Kharsany ABM, Werner L, et al. Risk Factors for HIV Acquisition in High Risk

Ac

28

ce

Epidemiology 2008; 37: 956–62.

Women in a Generalised Epidemic Setting. AIDS Behav 2015; published online Feb 8.

DOI:10.1007/s10461-015-1002-5.

29

Bärnighausen T, Tanser F, Newell M-L. Lack of a Decline in HIV Incidence in a Rural


Lancet 2013; 382: 1195–203.

19 Community with High HIV Prevalence in South Africa, 2003–2007. AIDS Patient Care and

30

cr ipt

STDs 2009; 25: 405–9. Bärnighausen T, Hosegood V, Timaeus IM, Newell M-L. The socioeconomic determinants of HIV incidence: evidence from a longitudinal, population-based study in rural South Africa. AIDS 2007; 21: S29–S38.

Mossong J, Grapsa E, Tanser F, Bärnighausen T, Newell M-L. Modelling HIV incidence

us

31

and survival from age-specific seroprevalence after antiretroviral treatment scale-up in rural

an

32 Karim QA, Kharsany AB, Frohlich JA, et al. Stabilizing HIV prevalence masks high HIV

M

incidence rates amongst rural and urban women in KwaZulu-Natal, South Africa. International Journal of Epidemiology 2011; 40: 922–30. Zaidi J, Grapsa E, Tanser F, Newell M-L, Bärnighausen T. Dramatic increase in HIV

pt ed

33

prevalence after scale-up of antiretroviral treatment. AIDS 2013; 27: 2301–5. 34

Bor J, Tanser F, Newell ML, Bärnighausen T. In A Study Of A Population Cohort In South Africa, HIV Patients On Antiretrovirals Had Nearly Full Recovery Of Employment. Health

Hosegood V, McGrath N, Moultrie TA. Dispensing with marriage: Marital and partnership

Ac

35

ce

Affairs 2012; 31: 1459–69.

trends in rural KwaZulu-Natal, South Africa 2000-2006. DemRes 2009; 20: 279–312.

36

Lindsey JC, Ryan LM. Tutorials in biostatistics: Methods for interval-censored data. Statistics in Medicine 1998; : 219–38.


South Africa. AIDS 2013; 27: 2471–9.

20 37 Sparling YH. Parametric survival models for interval-censored data with time-dependent

38

cr ipt

covariates. Biostatistics 2006; 7: 599–614. Jain V, Sucupira MC, Bacchetti P, et al. Differential Persistence of Transmitted HIV-1 Drug Resistance Mutation Classes. J INFECT DIS 2011; 203: 1174–81. 39

Muessig KE, Cohen MS. Advances in HIV Prevention for Serodiscordant Couples. Curr

Melo MG, Santos BR, De Cassia Lira R, et al. Sexual Transmission of HIV-1 Among

pt ed

M

2008; 35: 912–5.

an

Serodiscordant Couples in Porto Alegre, Southern Brazil. Sexually Transmitted Diseases

FIGURE LEGEND

Figure 1. Adjusted hazards ratio for each partnership category compared to co-habiting couple,

ce

HIV+ partner not on ART (reference category). Interval-censored time-updated proportional hazards models adjusted for age, sex, educational attainment, household wealth, concurrent

Ac

partners, and condom use.


40

us

HIV/AIDS Rep 2014; published online Aug 22. DOI:10.1007/s11904-014-0225-9.

ce

pt ed

M

an

cr ipt us


Ac

21

sc

2,153 (36.8%) 3,456 (59.0%) 249 (4.3%)

Ac c

ep t

ed

M an u

Mean age at baseline (SD) Educational attainment (N=15,684) None or primary (0-7 years) 6,402 (40.8%) Secondary (8-12 years) 8,668 (55.3%) Tertiary 614 (3.9%) Household wealth quintile (N=14,278) Lowest 2,809 (19.7%) nd 2 lowest 3,813 (26.7%) Middle 3,462 (24.3%) nd 2 highest 2,351 (16.5%) Highest 1,843 (12.9%) Multiple partners, past 12 months 464 (3.3%) (N=14,293) Any inconsistent condom use, past 12 5,342 (37.4%) months (N=14,295) Knows own HIV status (N=11,177) 3,279 (29.3%) SD: Standard deviation. N: Number of respondents.

Male (N=6,355) 28.6 (17.9)

1,010 (19.0%) 1,379 (26.0%) 1,312 (24.7%) 879 (16.6%) 729 (13.7%) 414 (7.2%)

1,677 (29.1%) 842 (20.9%)


All (N=17,016) 33.2 (20.5)

rip t

Table 1. Baseline descriptive characteristics by respondent sex Female (N=10,661) 36.0 (20.5) 4,294 (43.2%) 5,212 (53.0%) 365 (3.7%) 1,799 (20.1%) 2,434 (27.1%) 2,150 (24.0%) 1,472 (16.4%) 1,114 (12.4%) 50 (0.6%) 3,665 (42.9%) 2,437 (34.1%)

Table 2. HIV incidence by partner status and sex

1,632

1,003 1,243

25,335 39,586

ep t Ac c

Men No. Events 12 7 0 7

PY

rip t

64

sc

8,492

1.7 (1.5 to 2.0)

49

3.9 (3.0 to 5.0)

14

4.0 (3.7 to 4.2) 3.1 (3.0 to 3.3)

294 376

M an u

148

ed

Negative Partner Positive Partner On ART Not on ART Unknown partner 197 10,713 1.8 (1.6 to 2.1) status In non-cohabiting 78 2,171 3.6 (2.8 to 4.5) relationship Not in relationship 1,297 40,114 3.2 (3.1 to 3.4) Overall 1,619 60,349 2.7 (2.6 to 2.8) No. Events: Number of seroconversions; PY: Person-years

No. Events 8 20 4 16

Women PY Incidence Rate per 100 PY 3,720 0.2 (0.1 to 0.4) 406 4.9 (3.0 to 7.6) 179 2.2 (0.6 to 5.7) 227 7.0 (4.0 to 11.4)


No. Events 20 27 4 23

Overall PY Incidence Rate per 100 PY 6,644 0.3 (0.2 to 0.5) 707 3.8 (2.3 to 5.6) 294 1.4 (0.4 to 3.5) 413 5.6 (3.5 to 8.4)

2,923 302 116 186

Incidence Rate per 100 PY 0.4 (0.2 to 0.7) 2.3 (0.9 to 4.8) 0 (0 to 3.2) 3.8 (1.5 to 7.8)

2,221

2.2 (1.6 to 2.9)

538

2.6 (1.4 to 4.4)

14,779 20,763

2.0 (1.8 to 2.2) 1.8 (1.6 to 2.0)

Table 3. Association between partner serostatus and HIV acquisition

ep t

ed

1.00 0.99 (0.86 to 1.13) 0.94 (0.73 to 1.22)

Ac c

MODEL 3 Hazards Ratio (95% CI) 1.00 10.04 (5.51 to 18.32) 4.36 (2.67 to 7.13) 6.26 (3.71 to 10.58) 6.42 (3.97 to 10.40) 2.24 (1.98 to 2.54) 1.11 (1.08 to 1.12) 0.998 (0.997 to 0.999)

rip t

M an u