Article: ‘ Recently formed age-disparate partnerships are associated with elevated HIV incidence among young women in South Africa ’

(1)

Supplemental Digital Content 1

Article: ‘ Recently formed age-disparate partnerships are associated with elevated HIV incidence among young women in South Africa ’

Evidence before this study

Evidence is currently mixed on age-disparate partnerships as an HIV-infection risk factor for young women. Evidence from several sources indicates that partnering with older men increases risk of HIV-infection for young women. Cross-sectional studies among women younger than 25 years from several African countries show a positive association between testing HIV-positive and (a) age-difference (number of years partner is older) within

partnerships

^[1]

; (b) age-disparate partnerships, defined as those in which the male partner is five or more years older

^[2-4]

; and (c) intergenerational partnerships (i.e., having a partner 10 or more years older)

^[5]

. One cohort study in Zimbabwe found that intergenerational

relationship increased HIV incidence, but intrageneration relationships (5-9 years age

difference) did not

^[6]

. In South Africa, a phylogenetic study conducted demonstrated that the primary source of HIV infections in young women can be traced to older men

^[7]

. This evidence aligns with epidemiological and behavioural factors associated with age-disparate partnerships that point towards partnerships with older men increasing young women’s risk of HIV infection. First, older male partners are more likely to be HIV-infected with an unsuppressed viral load than younger partners

^[4]

. Second, condomless sex

^[8-11]

, transactional sex

^[11]

, higher frequency of sex

^[12,13]

, and concurrent sexual partnering

^[14]

, are more

prevalent in age-disparate partnerships.

In contrast, three longitudinal studies in South Africa have shown no association between age-disparate partnerships and HIV-acquisition among young women

^[15-17]

. The explanation for null findings in cohort studies remains unclear.

References

1 Gregson S, Nyamukapa C, Garnett G, Mason P, Zhuwau T, Caraël M, et al. Sexual mixing patterns and sex-differentials in teenage exposure to HIV infection in rural Zimbabwe. Lancet 2002; 359:1896–1903.

2 Pettifor AE, Kleinschmidt I, Levin J, Rees HV, MacPhail C, Madikizela-Hlongwa L, et al. A community- based study to examine the effect of a youth HIV prevention intervention on young people aged 15-24 in South Africa: results of the baseline survey. Trop Med Int Health 2005; 10:971–980.

3 Evans M, Risher K, Zungu N, Shisana O, Moyo S, Celentano DD, et al. Age-disparate sex and HIV risk for young women from 2002 to 2012 in South Africa. J Int AIDS Soc 2016; 19:1–16.

4 Maughan-Brown B, George G, Beckett S, Evans M, Lewis L, Cawood C, et al. HIV Risk Among Adolescent Girls and Young Women in Age-Disparate Partnerships. J Acquir Immune Defic Syndr 2018;

78:155–162.

5 Kelly RJ, Gray RH, Sewankambo NK, Serwadda D, Wabwire-Mangen F, Lutalo T, et al. Age differences in sexual partners and risk of HIV-1 infection in rural Uganda. J Acquir Immune Defic Syndr 2003;

32:446–451.

6 Schaefer R, Gregson S, Eaton JW, Mugurungi O, Rhead R, Takaruza A, et al. Age-disparate relationships and HIV incidence in adolescent girls and young women: evidence from Zimbabwe. AIDS 2017;

31:1461–1470.

7 de Oliveira T, Kharsany ABM, Gräf T, Cawood C, Khanyile D, Grobler A, et al. Transmission networks

and risk of HIV infection in KwaZulu-Natal, South Africa: a community-wide phylogenetic study. The

Lancet HIV 2017; 4:e41–e50.

(2)

8 Beauclair R, Kassanjee R, Temmerman M, Welte A, Delva W. Age-disparate relationships and implications for STI transmission among young adults in Cape Town, South Africa. Eur J Contracept Reprod Health Care 2012; 17:30–39.

9 Volpe E, Hardie T, Cerulli C, Sommers M, Morrison-Beedy M. What’s Age Got to Do With It? Partner Age Difference, Power, Intimate Partner Violence, and Sexual Risk in Urban Adolescents. J Interpers Violence 2013; 28:2068–2087.

10 Malema BW. Determinants of Condom use in Botswana: An empirical Investigation of the Role of Gender. Botswana Journal of Economics 2012; 10:59–78.

11 Maughan-Brown B, Evans M, George G. Sexual Behaviour of Men and Women within Age-Disparate Partnerships in South Africa: Implications for Young Women's HIV Risk. PLoS ONE 2016; 11:e0159162.

12 Beauclair R, Dushoff J, Delva W. Partner age differences and associated sexual risk behaviours among adolescent girls and young women in a cash transfer programme for schooling in Malawi. 2018; 18:403.

13 Ritchwood TD, Hughes JP, Jennings L, MacPhail C, Williamson B, Selin A, et al. Characteristics of Age- Discordant Partnerships Associated With HIV Risk Among Young South African Women (HPTN 068). J Acquir Immune Defic Syndr 2016; 72:423–429.

14 Maughan-Brown B, Kenyon C, Lurie MN. Partner Age Differences and Concurrency in South Africa:

Implications for HIV-Infection Risk Among Young Women. AIDS Behav 2014; 18:2469–2476.

15 Harling G, Newell M-L, Tanser F, Kawachi I, Subramanian SV, Barnighausen T. Do Age-Disparate Relationships Drive HIV Incidence in Young Women? Evidence from a Population Cohort in Rural KwaZulu-Natal, South Africa. J Acquir Immune Defic Syndr 2014; 66:433–451.

16 Balkus JE, Nair G, Montgomery ET, Mishra A, Palanee-Phillips T, Ramjee G, et al. Age-Disparate Partnerships and Risk of HIV-1 Acquisition Among South African Women Participating in the VOICE Trial. J Acquir Immune Defic Syndr 2015; 70:212–217.

17 Street RA, Reddy T, Ramjee G. The generational effect on age disparate partnerships and the risk for

human immunodeficiency virus and sexually transmitted infections acquisition. International journal of

STD & AIDS 2016; 27:746–752.

(3)

Table S1: Characteristics of 15-to-24 year-old, HIV-negative women enrolled in the 2015/16 HIPSS cohort (KwaZulu-Natal, South Africa) who were in a heterosexual partnership at baseline

N=830 Weighted estimates

% (n) % (95% CI)

Black African 99.9 (829) 99.9 (99.9-100)

Age

15-19 35.1 (291) 34.8 (31.1-38.4)

20-24 65.9 (539) 65.2 (61.6-68.9)

Education

<Grade 8 1.5 (12) 1.4 (0.1-2.2)

Grade 8-11 44.9 (373) 43.9 (39.9-47.8)

Grade 12 completed 53.6 (445) 54.8 (50.8-58.9)

Resides in urban area 62.7 (520) 58.7 (50.6-66.8)

Household income (monthly)

£R2500 50.0 (415) 46.9 (42.4-51.3)

R2501-R6000 38.0 (315) 39.8 (35.5-44.0)

>R6000 12.0 (100) 13.4 (10.0-16.7)

Drinks alcohol 19.9 (165) 19.4 (16.3-22.6)

Concurrent sexual partners

No 96.1 (798) 96.6 (95.2-98.0)

Yes 3.9 (32) 3.4 (2.0-4.8)

Partner age difference

£ 4 years 72.8 (604) 72.7 (69.0-76.5)

³ 5years 27.2 (226) 27.3 (23.5-31.0)

Partnership duration (years)

0 to 1 22.1 (183) 22.7 (19.5-25.9)

1.01 to 2 22.1 (183) 22.9 (19.5-26.2)

2.01 to 3 14.2 (118) 14.0 (11.3-16.7)

3.01 to 4 11.9 (99) 11.0 (8.7-13.3)

4.01 to 5 10.0 (83) 10.0 (7.5-12.7)

>5 19.8 (164) 19.3 (16.2-22.5)

Partnership duration (quartiles)

Quartile 1 (yrs: 0 to 1.08) 25.2 (209) 25.7 (22.4-29.0) Quartile 2 (yrs: 1.09 to 2.41) 24.9 (207) 25.7 (22.3-29.1) Quartile 3 (yrs: 2.42 to 4.49) 25.2 (209) 24.8 (21.4-28.1) Quartile 4 (yrs: 4.50 to 15) 24.7 (205) 23.9 (20.6-27.2)

(4)

Table S2. Multivariable Cox proportional hazard models of HIV acquisition among 15-to-24 year-old women in KwaZulu-Natal, South Africa (2015-2017)

Adjusted Cox proportional hazard model

Adjusted Cox proportional hazard model

aHR (95% CI) aHR (95% CI) Age-disparate partnership (ref: no) 1.10

(0.55-2.21)

Partnership age-difference 1.04

(in years, 0-27) (0.94-1.14)

Age (15 to 24 years) 1.14** 1.14**

(1.01 - 1.28) (1.01 - 1.28) Grade 12 complete (ref: <grade 12) 0.63 0.64

(0.34 - 1.20) (0.33 - 1.21)

Resides in urban area (ref: rural) 0.73 0.73

(0.38 - 1.37) (0.38 - 1.38) Partnership duration quartiles (ref:

quartile 1)

Quartile 2 0.61 0.62

(0.25 - 1.51) (0.25 - 1.52)

(0.43 - 2.55) (0.43 - 2.54)

(0.21 - 1.25) (0.20 - 1.22)

Concurrent sexual partner (ref: no) 0.41 0.40

(0.09 - 1.94) (0.09 - 1.85) Household monthly income (ref:

£R2500)

R2501-R6000 0.85 0.85

(0.44 - 1.65) (0.44 - 1.66)

>R6000 1.47 1.43

(0.56 - 3.86) (0.53 - 3.87)

Drinks alcohol (ref: no) 1.98* 2.01*

(0.94 - 4.19) (0.94 - 4.28)

Observations 830 830

Notes: 95% CI: 95% confidence interval; aHR: adjusted hazard ratio

* p<0.01, p<0.05, * p<0.1

(5)

Table S3. Crude HIV-incidence estimates by partnership type (age-similar or age-disparate) and partnership duration at baseline among 15-to-24 year-old women (N=830) in KwaZulu-Natal, South Africa (2015-2017)

unweighted data weighted data HIV infections/

person-years

HIV Incidence (per 100 person-years)

(95% CI) (95% CI)

Partnership duration quartile 1

age-similar partner 9/217.62 4.14 (2.15-7.95) 3.59 (1.79-8.23)

age-disparate partner 8/58.50 13.68 (6.84-27.35) 11.83 (5.51-29.51)

age-disparate partner 1/77.58 1.29 (0.18-9.15) 2.26 (na)

Notes: na not available

(6)

Table S4. Multivariable Cox proportional hazard model of HIV-acquisition, with interactions between age-disparate partnerships and partnership duration quartiles

Adjusted Cox proportional hazard model

aHR (95% CI)

Age-disparate partnership (ref: no) 3.13**

(1.02 - 9.65) Partnership duration quartiles (ref: quartile 1)

Quartile 2 0.96

(0.31 - 2.92)

Quartile 3 1.77

(0.62 - 5.05)

Quartile 4 0.90

(0.31 - 2.62) Interactions (age-disparate partnerships and partnerships duration

quartiles)

Age-disparate*quartile 2 0.24

(0.02 - 2.69)

Age-disparate*quartile 3 0.21*

(0.04 - 1.06)

Age-disparate*quartile 4 0.19*

(0.03 - 1.10)

Age (15 to 24 years) 1.14**

(1.01 - 1.28)

Grade 12 complete (ref: <grade 12) 0.64

(0.33 - 1.25)

Resides in urban area (ref: rural) 0.68

(0.36 - 1.31)

Concurrent sexual partner (ref: no) 0.42

(0.08 - 2.08) Household monthly income (ref: £R2500)

R2501-R6000 0.83

(0.44 - 1.58)

>R6000 1.38

(0.50 - 3.77)

Drinks alcohol (ref: no) 2.06*

(0.96 - 4.42)

Observations 830

Notes: 95% CI: 95% confidence interval; aHR: adjusted hazard ratio

* p<0.01, p<0.05, * p<0.1

(7)

An additional analysis was conducted to assess the effect of partner age-differences on HIV-acquisition among 15-to-24 year-old women who were not in a sexual

partnership at baseline, but reported a heterosexual partnership in the 12 months prior to the follow-up survey. This analysis provides an additional assessment of the effect of partner age-disparity on HIV-acquisition within newly formed partnerships (i.e., subsequent to the baseline survey). Cox proportional hazard models were used, controlling for: participant age (years); education (completion of secondary schooling (grade 12) or not); area of residence (urban vs rural); household income (£R2500, vs

>R2500) and alcohol use (whether the respondent drinks alcohol). For this analysis, there was the potential of missing data on sexual partners between the baseline and endline surveys. Data on partnerships were collected at endline for only the 12 months preceding the follow-up interview, but follow-up interviews often occurred more than 12 months subsequent to the baseline survey. There was therefore some uncertainty about the order of exposure and outcome; and it was not possible to control for the number of sexual partners between the baseline and endline surveys.

Table S5. Cox proportional hazard models of HIV acquisition among 15-to-24 year-old women in KwaZulu-Natal, South Africa (2015-2017) who reported a sexual partner in the 12 months prior to the end-line study, but were not in a sexual partnership at baseline (N=209)

Unadjusted model

Adjusted model Unadjusted model

Adjusted model HR (95% CI) aHR (95% CI) HR (95% CI) aHR (95% CI) Age-disparate partnership (ref: no) 2.52 2.71

(0.70 - 9.00) (0.69 - 10.61) Partnership age-difference

(in years, 0-20) 1.15** 1.22**

(1.02 - 1.30) (1.02 - 1.46)

Age (15 to 24 years) 0.63 0.60

(0.33 - 1.21) (0.30 - 1.18)

Grade 12 complete (ref: <grade 12) 4.32 4.64

(0.25 - 73.44) (0.28 - 76.05)

Resides in urban area (ref: rural) 0.31 0.30*

(0.07 - 1.34) (0.07 - 1.20)

Household monthly income >R2500

(ref: £R2500) 0.68 0.67

(0.18 - 2.62) (0.17 - 2.58)

Drinks alcohol (ref: no) 0.83 0.94

(0.05 - 13.20) (0.06 - 13.66)

Observations 209 209 209 209

Notes: 95% CI: 95% confidence interval; HR: hazard ratio; aHR: adjusted hazard ratio

* p<0.01, p<0.05, * p<0.1

(8)

Figure S1. Selection of the analytic sample from participants of the 2015/16 HIPSS survey in KwaZulu-Natal, South Africa.

HIPSS 2015/16 survey (N=10,236)

Ineligible (n=8,375)

• Male (n=3,895)

• Female >=25 years old (n=4,010)

• HIV-positive female <= 24 years old (n=470)

Eligible for cohort study of 15-to-24 year-old women

(N=1,861)

Analytic sample (N=830) Enrolled in cohort study

(N= 1,411)

Non-response (n=450)

Excluded (n=581)

• No sexual partner at enrolment (n=567)

• Homosexual partner at enrolment (n=14)

• Data missing on primary outcome (n=0)

• Data missing on partner age (n=0)

(9)

Figure S2. Adjusted hazard ratios for the association between HIV acquisition and age-disparate partnerships (reported at baseline) among 15-to-24 year-old women (N=830), by partnership duration (tertiles). Hazard ratios estimated using Cox proportional hazard models with interactions between age-disparate partnerships (0/1) and relationship duration tertiles. Models included the following control variables: participant age (years); education (completion of secondary schooling (grade 12) or not); area of residence (urban vs rural); concurrent sexual partner at baseline (no/yes);

household income (£R2500, R2501-6000, >R6000) and alcohol use (whether the respondent drinks alcohol/not).

All

q1

q2

q3

Relationship length (tertiles) ⁰ ¹ ² ³ ⁴ ⁵ ⁶ ⁷ ⁸ ⁹Hazard ratio (95% CI)

Adjusted

(10)

Figure S3. Adjusted hazard ratios for the association between HIV acquisition and age-disparate partnerships (reported at baseline) among 15-to-24 year-old women (N=830), by partnership duration (quintiles). Hazard ratios estimated using Cox proportional hazard models with interactions between age-disparate partnerships (0/1) and relationship duration quintiles. Models included the following control variables: participant age (years); education (completion of secondary schooling (grade 12) or not); area of residence (urban vs rural); concurrent sexual partner at baseline (no/yes);

household income (£R2500, R2501-6000, >R6000) and alcohol use (whether the respondent drinks alcohol/not).

All

q1

q2

q3

q4

q5

Relationship length (quintiles) ⁰ ¹ ² ³ ⁴ ⁵ ⁶ ⁷ ⁸ ⁹

Hazard ratio (95% CI) Adjusted

Article: ‘ Recently formed age-disparate partnerships are associated with elevated HIV incidence among young women in South Africa ’

Supplemental Digital Content 1