Secondary Logo

Journal Logo

HIV Positivity and Referral to Treatment Following Testing of Partners and Children of PLHIV Index Patients in Public Sector Facilities in South Africa

Joseph Davey, Dvora PhD, MPHa,b,c; Wall, Kristin M. PhD, MPHd; Serrao, Claire MDc; Prins, Marlien BAc; Feinberg, Madaline MAc; Mtonjana, Ntokozo MDe; Hlophe, Khanyo MDf; Zuma, Lindiwe MDf; Sejake, Senate MDg; Malone, Todd MPH, MEdc

JAIDS Journal of Acquired Immune Deficiency Syndromes: August 1, 2019 - Volume 81 - Issue 4 - p 365–370
doi: 10.1097/QAI.0000000000002048
Rapid Communication

Background: There is an imperative need for innovative interventions to identify people living with HIV and initiate them on antiretroviral therapy. The objective of this study was to determine the feasibility of providing index partner/child testing of people living with HIV.

Methods: We trained 86 nurses and counsellors in 56 public health facilities in 6 high HIV burden Districts in South Africa 2017 to provide index partner/child testing (tracing and testing of partners/children of people living with HIV). We collected programmatic data including index partner/child HIV positivity by age, gender, and location of testing. In subanalyses, we evaluated factors associated with identifying HIV-positive partners and children in separate models using multivariable logistic regression.

Results: We tested 16,033 partners and children of index patients between October 2017 and June 2018. Most of those tested were women (61%) and 20–39 years old (39%). Overall, 6.4% were 10–14 years old, 9.5% were 15–19 years, and 8% were ≥50 years. HIV positivity was 38% [95% confidence interval (CI) = 36% to 40%]. In children ages 10–14 years, 13% were HIV-infected (95% CI = 11% to 14%). In subanalyses, HIV positivity in partners was associated with their increased age [adjusted odds ratio (aOR) for increase in 5-year age category = 1.21; 95% CI = 1.04 to 1.42], female gender (aOR = 1.38; 95% CI = 1.04 to 1.82), and index partner bringing the partner in for HIV testing vs. referring the partner through the provider or recommending testing to the partner (aOR = 1.94, 95% CI = 1.43 to 2.63), adjusting for location of testing. Almost all patients diagnosed (97%) were referred to antiretroviral therapy.

Conclusions: Providing index partner/child testing was feasible and we identified a very high yield when testing partners and children of index patients. Index partner and children testing should be offered to all patients living with HIV to improve case finding.

aDepartment of Epidemiology, Fielding School of Public Health, University of California, Los Angeles, CA;

bDivision of Epidemiology and Biostatistics, School of Public Health and Family Medicine, University of Cape Town, South Africa;

cDepartment of Epidemiology, BroadReach Healthcare, Cape Town, South Africa;

dSchool of Public Health, Emory University, Atlanta, GA;

eDepartment of Health, Alfred Nzo District, South Africa;

fDepartment of Health, Harry Gwala District, South Africa; and

gDepartment of Health, Sedibeng District, South Africa.

Correspondence to: Dvora Joseph Davey, PhD, MPH, Department of Epidemiology, School of Public Health, University of California, Los Angeles, 650 Charles E Young Drive S, Los Angeles, CA 90095 (e-mail:

Supported by the American people through the United Stated Agency for International Development (USAID) under Cooperative Agreement No. AID-674-A-12-00016; Systems Strengthening for Better HIV/TB Patient Outcomes. The contents are the responsibility of BroadReach and do not necessarily reflect the views of USAID or the United States Government.

The authors have no conflicts of interest to disclose.

Received October 12, 2018

Accepted February 27, 2019

This is an open-access article distributed under the terms of the Creative Commons Attribution-Non Commercial-No Derivatives License 4.0 (CCBY-NC-ND), where it is permissible to download and share the work provided it is properly cited. The work cannot be changed in any way or used commercially without permission from the journal.

Back to Top | Article Outline


In South Africa, the country with the largest HIV epidemic and antiretroviral therapy (ART) program in the world, an estimated 85% of HIV-infected adults knew their status and 71% of those are on ART in 2017. Of those on ART who had a recent viral load test, 88% had achieved viral load suppression (viral suppression is considered less than 1000 copies per mL by the South African Department of Health) as of 2017.1 Innovative, cost-effective, and scalable approaches such as partner notification and index partner-testing strategies are needed to meet UNAIDS 90-90-90 goals, which refer to 90% of people living with HIV knowing their HIV status, 90% of those on ART, and 90% of people on ART who are virally suppressed (VLS).1 A recent systematic review demonstrated the dearth of research on HIV partner notification strategies, with only 10 studies identified.2 Our review highlighted the effectiveness of partner assisted referrals, mainly assisted by health care providers, compared with passive referral, not assisted by health care providers, in improving partner testing, notification of serostatus, yield of new positives, and linkage to ART.

In addition, the same HIV testing and treatment cascade is poorer in men with only 78% of men living with HIV diagnosed, 67% of those on ART, and 82% VLS.1 There is a current, imperative need for innovative interventions to identify men living with HIV, engage them in ART, and promote retention and adherence to achieve VLS. Strategies are needed to reach men who are at-risk and untested (in past 12-months), and men previously diagnosed, but not adhering to treatment. Clinic service delivery was defined as being unfriendly to men, especially adolescent men.3–5

Index partner testing strategies could be a high-yield way to identify men and link them to treatment. One randomized controlled study from Rosenberg et al in Malawi compared passive invitation with passive invitation plus contact tracing of male partners.6 In that study, contact tracing was used to recruit men to get tested jointly with their partner, thereby using couples HIV counseling and testing (CHCT) as a mechanism for partner notification and mutual serostatus disclosure in the context of antenatal care (ANC).7

Our study evaluated the impact of implementing a novel testing strategy, index partner/child testing, in which nurses and counselors were trained to track and trace partners and children of index patients living with HIV (recently diagnosed or on ART) implemented in public health hospitals and clinics in South Africa. We evaluated the impact of index partner/child testing on case finding and referral to ART.

Back to Top | Article Outline


To improve case finding, ART initiation, and HIV serostatus disclosure, we adapted the CDC/WHO CHCT guidelines8,9 to focus on tracing and testing adult sex partners (15 years or older) and children (10–15 years old) of HIV-infected individuals in 6 high HIV prevalence districts (Alfred Nzo, King Cetshwayo, Gert Sibande, Ugu, King Cethswayo, and Sedibeng districts) in 4 provinces (Eastern Cape, Gauteng, Kwa Zulu Natal, and Mpumalanga). In our evaluation, we excluded children less than 10 years old who were tested and reported through prevention of mother-to-child transmission interventions.10 During our study, we trained nurses and lay counsellors working in HIV testing, ANC, ART, and TB services in 56 large primary health care public facilities between March and December 2017. Clinics were primarily urban and peri-urban, but we included some large, rural clinics. As we did not collect routine data from all sites before October 2017, this analysis includes data collected from October 2017 to June 2018 when facilities had been trained and data collection was systematic. In addition to partner/child tracing and testing, our protocol included specific training on how to conduct CHCT for serodiscordant and concordant HIV-positive couples and their children, facilitate disclosure of serostatus results where one individual is HIV-infected, and how best to refer HIV-infected patients (or couples) to ART, or same-day ART services where available. The counselors were trained to screen for intimate partner violence or other potential social harms as a result of partner notification and HIV status disclosure. Anyone who reported IPV was referred to a social worker for counseling and legal support.

Back to Top | Article Outline

Data Collection

During index patient post-HIV test counseling or ART adherence counseling (for patients already on ART) in ANC, TB, and ANC services, counsellors and nurses asked index patients to refer their partner or children who did not know their HIV status or were not known to be on ART for testing. The provider would give the index patient three options for testing of their partner/child:

  • the index patient invited their partner/child to test via a written invitation provided at the clinic which contained the index patient's folder number or ID (which could lead to immediate testing if the partner/child was already in the clinic), but the partner/child returned alone for testing,
  • the clinician called the partner to invite them for individual testing (without mentioning the index partner), or
  • the index patient returned for testing with their partner or child, at which time they could test together or separately.

The above-mentioned options were not mutually exclusive, though the provider collected data only on the method that was effective at bringing the partner or child in for testing.

Trained counsellors and nurses entered data in standardized government clinic HIV-testing service logbooks after testing each participant and would note if the partner or child was referred by an index patient (self-reported). The partner/child was linked to the index patient if the counsellor or nurse received the written invitation with the index patient's folder number or ID, or if the provider knew the index patient who was bringing their child or partner in for testing.

Back to Top | Article Outline


In a subanalysis of a subsample of 10% of partner/children tested identified through a convenience sample of the total patients, we collected additional data on partner/child testing (including where was testing provided, how was the partner/child traced, and whether they were referred for ART). This data was collected on an electronic form as part of the patient electronic data from participants who tested following referral for partner/children testing. Participants were selected during a 2-month time period to compile additional information about the intervention and evaluate factors associated with HIV-positive diagnosis and referral to ART initiation.

Back to Top | Article Outline

Quality Control of Data and Data Management

Each facility had a supervisor who would observe the quality of the HIV counseling and testing services and review the HIV testing logbooks and electronic data for missingness or inconsistencies on a monthly basis. The supervisor would provide mentorship and retraining as necessary and clean the data with the provider to ensure that missing data were collected and inconsistencies were corrected. The supervisors would review the HIV testing logbooks and compare them with the index partner testing reports to ensure consistency.

Back to Top | Article Outline

Data Analysis

We analyzed routinely collected data from trained staff to report on findings between October 2017 and June 2018 including positivity by age and gender. Descriptive analysis of categorical variables using proportions summarized characteristics of partners and children of index patients who tested for HIV in the subsample of patients using χ2 tests for differences in HIV positivity. We then modelled the outcome of HIV status of partner and child tested using univariate and multivariable logistic regression for partners and children in separate models adjusting for a priori confounders including gender and age of the partner/child tested.

Back to Top | Article Outline

Human Subjects Considerations

Program activities were implemented as part of routine primary health care and HIV testing and counseling. All data were retrieved from a deidentified retrospective analysis of patients' electronic charts. Names, dates of birth, and ID numbers were removed from the dataset before analysis by the Department of Health staff. Participants provided informed consent to undergo HIV testing and counseling and for partner notification as part of the standard of care. University of California Los Angeles's Institutional Review Board provided exemption (UCLA IRB#19-000227).

Back to Top | Article Outline

Role of the Funding Source

This pilot program was funded by the United States Agency for International Development (USAID) under Cooperative Agreement AID-674-A-12-00016; managed by BroadReach Healthcare. The donor had no involvement in the study design, data collection or analysis, interpretation of the results, or writing of the report. The corresponding authors had full access to all data in the study and had final responsibility for the decision to publish without involvement of the donor.

Back to Top | Article Outline


During our pilot study, we trained 34 nurses and 52 lay counsellors working in HIV testing, ANC, and TB services in 56 large primary health care public facilities between March and December 2017. Following training, providers tested 16,033 partners and children of index patients between October 2017 and June 2018. Most of those tested were women (61%; n = 9710) and between 20 and 39 years old (39%, n = 6263). Overall, 6.4% were 10–14 years old (n = 1022), 9.5% were between 15 and 19 years old, and 8% were 50 years or older. In total, 6038 of those tested were HIV-infected [38%; 95% confidence interval (CI) = 36% to 40%]. In children aged 10–14 years, 13% were HIV-infected (95% CI = 11% to 14%), 16% in girls vs. 10% in boys (P < 0.05). In women, highest positivity was in 30–34-year-old women (44%; 95% CI = 42% to 47%), followed by 25–29-year-old women(43%; 95% CI = 41% to 46%). In men, highest positivity was in 35–39-year-old men (55%; 95% CI = 52% to 58%), followed by 40–49-year-old men (53%; 95% CI = 50% to 56%) (Fig. 1).



In the subset analysis of a convenience sample of 9.7% of patients (n = 1554) in which we collected additional data, most patients were tested in public health clinics (98%) and 2% in public hospitals (n = 34). Most partners and children were tested in voluntary counseling and testing services in the facility (65%), followed by the outpatient department provider-initiated testing (33%), ANC (1%), tuberculosis care (0.5%), and voluntary male medical circumcision (0.5%) (Table 1). Participants who tested after invitation from their partner (or with their partner) were identified in the HIV-testing log from the counsellor, though they may have tested in TB or ANC services as part of the standard of care.



In how the partner came into testing, 52% of index patients chose to ask their partners to get tested (using an invitation), but did not return to test with them, 24% were tested following phone tracing by the health care provider, 21% of index patients brought in their partner for individual or joint testing, and 3% were unknown. For children, 48% came in with their parent, 28% were tested following phone tracing by the health care provider, 21% of parents asked their children to come in for testing (and came in alone), and 3% was unknown. Overall, 40% of index patients were counselled and tested together with their partner via CHTC. Almost all clients were referred for ART (97%), some receiving same-day initiation if they demonstrated they were ready to start ART (Table 1).

Odds of diagnosing an HIV-positive partner increased with their age [adjusted odds ratio (aOR) for increase in 5-year age category = 1.21; 95% CI = 1.04 to 1.42], female gender (aOR = 1.38; 95% CI = 1.04 to 1.82), and bringing the partner in for HIV testing vs. referring the partner through the provider or recommending testing to the partner (aOR = 1.94, 95% CI = 1.43 to 2.63), adjusting for location of testing. Odds of diagnosing an HIV-positive child (under 15 years old) was highest with increased age (aOR = 2.20, 95% CI = 1.90 to 2.55). In location of testing, outpatient department provider-initiated testing had the highest odds of positivity (aOR = 1.80, 95% CI = 1.38 to 2.34) versus voluntary counseling and testing adjusting for age and gender. Referral for ART initiation (97% were referred) did not differ by gender, age, or location of testing service.

Back to Top | Article Outline


Testing of partners and children of index patients living with HIV, referred to as index partner testing, is an important high yield testing strategy that is essential to reach people living with HIV who may not know their status or know their status, but are not yet on ART, especially men who remain disproportionately underdiagnosed in South Africa.1,10 The positivity in index partner testing was very high; almost half of partners of index cases were HIV-infected. Furthermore, children 10–14 years old had a very high positivity of 13%. In South Africa, children aged 12 years and above can consent for HIV testing on their own if the provider deems them to be sufficiently mature,18 see Figure 2 for HIV testing guidelines that we developed for children of HIV-infected mothers. Most index partner testing occurred in voluntary counseling and testing services, yet testing of index partners in the outpatient departments in public health facilities yielded more HIV-infected partners. The most successful mode of invitation was the index patient inviting their partner/child to return for testing (or they were recruited/tested during an index partner clinical visit which they were attending). Integrating index partner/child testing into public health facilities was feasible and effective at increasing the positivity of HIV testing services from 8% before the pilot to 41% in sex partners of index patients.



Our pilot reached 2605 HIV-positive males (41% of 6323 males tested) in 9 months and referred almost all men to initiate ART. Index partner testing in the outpatient departments in public health facilities yielded more HIV-infected adult male partners compared with other services including VCT. In Tanzania, successful partner referral was 2.2 times more likely among male compared with female index clients.11 As in Tanzania, women may need additional support to overcome challenges in the partner notification process, especially in a context of high interpersonal violence in South Africa.15 Several studies have integrated male partner and couples HIV testing into ANC, yielding mixed results.12,16,17 Similarly, we had very low uptake of male partner testing in ANC. HIV self-testing may be another option to improve HIV testing uptake in male partners of HIV-infected women.14

Almost half of index patients who brought their partners in for testing chose to invite their partners to get tested themselves (similar to studies in Malawi where index-partner initiated strategies were preferred over provider-initiated contact7,12), and one-third of index cases came with their partners for individual or joint testing. However, the recent systematic review of partner notification found that provider-initiated strategies were most effective in yield and linkage.2 This has implications for the role out of index partner testing in that it may be less expensive and simpler to empower index patients to disclose their HIV status and refer their partner to come in for testing. However, female index cases and those with limited power may be left out if they are unable to discuss HIV with their partner(s) and/or children for fear of reprisal, abuse, or abandonment. We advocate for providing both provider and index patient led strategies to improve disclosure and index partner testing, including couples' testing and counseling.6

PEPFAR released a toolkit on index and partner notification in April 2018, which includes job aids, talking points, and tools for documenting and monitoring partner notification services.13 However, the toolkit does not discuss the potential for couples' counseling and testing to conduct index partner testing, which our study found to be a popular option. We found that couples' counseling was also popular and described by participants to receive accurate information and reduce potential conflict between partners. The logistics of how couples' counseling occurs when one partner already knows their serostatus (and may be on ART) is challenging and may require retesting of the HIV-infected partner if they have not yet disclosed their status. Future studies should explore how best to invite, jointly test, and link to care the partners of index patients living with HIV.3

Limitations of our study include that we collected programmatic data with counsellors and nurses. We did not collect data on index patients who declined to invite their partners/children for testing or were not successful in partner tracing and testing. We did not collect data on ART initiation and only report on referrals to ART. Furthermore, we did not confirm whether the partner or child had been tested before and already knew their HIV status. This may have biased our results to overestimate the true positivity of those tested. Furthermore, there is potential for bias in the results and positivity due to convenience sampling.

Back to Top | Article Outline


Integrating index partner testing into HIV care was feasible and we identified a very high yield when testing partners and children of index patients. Index testing should be provided to all people diagnosed with HIV, whether or not they have initiated ART, to provide opportunities to diagnose those at highest risk of being infected with HIV who may still be unaware of their status. There is an urgent need for policies and training to be rolled out to implement index partner and children testing at scale. Additional implementation research is needed to evaluate which implementation methods (eg, comparing the effectiveness and costs for clinic-based, community-based, and self-testing strategies) are most effective and cost-effective at reaching partners and their children of index patients and linking those diagnosed to treatment.

Back to Top | Article Outline


1. Human Sciences Research Council. South African National Prevalence, Incidence, Behaviour and Communication Survey, 2017. Presentation to HIV Think Tank, August, 2018.
2. Dalal S, Johnson C, Fonner V, et al. Assisted HIV partner notification services: a systematic review and meta-analysis. AIDS. 2017;31:1867–1876.
3. Smith P, Marcus R, Bennie T, et al. What do South African adolescents want in a sexual health service? Evidence from the South African Studies on HIV in Adolescents (SASHA) project. South Afr Med J. 2018;108:677–681.
4. Dickson-Tetteh K, Pettifor A, Moleko W. Working with public sector clinics to provide adolescent friendly services in South Africa. Reprod Health Matters. 2001;9:160–169.
5. Zanoni BC, Archary M, Buchan S, et al. Systematic review and meta-analysis of the adolescent HIV continuum of care in South Africa: the cresting wave. BMJ Glob Health. 2016;1:e000004.
6. Joseph Davey DL, Wall KM. Need to include couples' HIV counselling and testing as a strategy to improve HIV partner notification services. AIDS. 2017;31:2435–2436.
7. Rosenberg NE, Mtande TK, Saidi F, et al. Recruiting male partners for couple HIV testing and counselling in Malawi's option B+ programme: an unblinded randomised controlled trial. Lancet HIV. 2015;2:e483–e491.
8. Centers for Disease Control (CDC). Couples HIV Counseling and Testing Intervention and Training Curriculum. 2011. Available at: Accessed September 21, 2018.
9. World Health Organization. Guidance on Couples HIV Testing and Counselling Including Antiretroviral Therapy for Treatment and Prevention in Serodiscordant Couples: Recommendations for a Public Health Approach. 2012. Available at: Accessed September 21, 2018.
10. The South African National AIDS Council. South Africa National Strategic Plan on HIV, TB and STIs 2017-2022. Available at: Accessed September 21, 2018.
11. Plotkin M, Kahabuka C, Christensen A, et al. Outcomes and experiences of men and women with partner notification for HIV testing in Tanzania: results from a mixed method study. AIDS Behav. 2018;22:102–116.
12. Kamanga G, Brown L, Jawati P, et al. Maximizing HIV partner notification opportunities for index patients and their sexual partners in Malawi. Malawi Med J. 2015;27:140–144.
13. PEPFAR Solutions Platform. Index and partner notification testing toolkit. 2018. Available at: Accessed October 8, 2018.
14. Matovu JKB, Kisa R, Buregyeya E, et al. “If I had not taken it [HIVST kit] home, my husband would not have come to the facility to test for HIV”: HIV self-testing perceptions, delivery strategies, and post-test experiences among pregnant women and their male partners in Central Uganda. Glob Health Action. 2018;11:1503784.
15. Okafor CN, Barnett W, Zar HJ, et al. Associations of emotional, physical, or sexual intimate partner violence and depression symptoms among South African women in a prospective cohort study. J Interpers Violence. 2018:886260518796522. doi: . PubMed PMID: 30160637.
16. Mohlala BKF, Boily MC, Gregson S. The forgotten half of the equation: randomized controlled trial of a male invitation to attend couple voluntary counselling and testing. AIDS. 2011;25:1535–1541.
17. Msuya SE, Mbizvo EM, Hussain A, et al. Low male partner participation in antenatal HIV counselling and testing in northern Tanzania: implications for preventive programs. AIDS Care. 2008;20:700–709.
18. Human Science Research Council. Legal, ethical and counselling issues related to HIV testing of children HIV testing of children: legal guidelines for implementers. 2012. Available at: Accessed February 18, 2019.

index partner testing; positivity; HIV testing and counseling; couples; partners; children

Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.