Secondary Logo

Journal Logo

Puzzling Observations in a Trial of HIV Partner Notification in Sub-Saharan Africa

Potterat, John J BA

JAIDS Journal of Acquired Immune Deficiency Syndromes: April 15th, 2011 - Volume 56 - Issue 5 - p 381-383
doi: 10.1097/QAI.0b013e318211b451

Correspondence to: John J. Potterat, BA, Independent consultant, Colorado Springs, CO (e-mail:

The author has no funding or conflicts of interest to disclose.

Some studies are diamonds in the rough. Some have emergent properties. The path-breaking trial of HIV partner notification (PN) in Malawi reported by Brown et al in these pages1 has both attributes. Their study is important not only because their prospectively conducted trial is a first for sub-Saharan Africa but also because it points to other public health opportunities. The roughness consists of their study's unpolished aspects, whereas emergent properties are considerations not contemplated by the authors.

From October 2008 through August 2009, Brown et al1 recruited 90% of 267 eligible newly diagnosed HIV-positive men and women attending 2 hospital-based outpatient sexually transmitted infections (STI) clinics in Lilongwe, Malawi. Participant “index patients” able to provide locating information on their sexual partners for the 3-month period preceding HIV diagnosis were randomized to 1 of 3 notification methods: self referral (index patient notifies own partners), contract referral (index patient given 7 days to refer nominated partners; if referral unsuccessful, health worker anonymously notifies partners); or provider referral (health worker anonymously notifies index patient's nominated partners). Index patients were asked to return to the clinic 2 weeks after enrollment to discuss CD4 test results and initiation of HIV care; during this visit, index patients were also queried about notification and referral outcomes of sexual partners. Most index patients were women, married, and named only a single partner. In total, 302 partners were named (1.25 per index patient), of whom nearly three-quarters were main partners; locating information was available for 84% of partners. Index patients also indicated that they intended to resume sex with nearly three quarters of their partners. One hundred seven partners were successfully referred to the clinics, and the proportion of locatable partners presenting to the clinics was 24% in the self referral, 51% in the contract referral, and 51% in the provider referral, arm. Almost all successfully referred persons were main partners: of the 97 main partners, 64% were HIV positive, as were 57% of the 7 casual partners tested, and 3 persons were not tested. The preceding 2 sentences contain the results prompting the focus of this editorial: the missing information “roughness” in their reported PN process and outcomes and the emergent considerations that should accompany the finding of remarkably high proportions of HIV-positive partners.

First, the authors do not discuss reasons for the modest partner referral rates. Considering that the vast majority were stable partners with whom index patients intended to continue their sexual relationship, the observed 51% partner referral success rate in both the contract and provider referral arms needs clarification. Perhaps power dynamics between the sexes influenced these referral outcomes, considering that most index cases were women and their partners, men. Not only were many, if not most, of the uncaptured half of partners presumably known and locatable by index patients, but referral assistance by health aides was the integral operational aspect of each non-self referral arm. Is this glass half full because locatable partners were reluctant to be referred or tested or is it half empty because health aides were inexperienced, insufficiently trained, or passive with demurring partners? Regrettably, the authors are silent on this crucial operational issue. Nor do they supply sufficient descriptions of their notification procedures; such details would be of particular interest because they were implemented in a pioneering context, sub-Saharan Africa. For example, in settings where monogamy is the dominant arrangement, would not partners automatically assume that their solo partner was the HIV-infected party? Were partners, then, explicitly told about exposure to HIV or was this done a bit more obliquely? Was notification done in person or by phone? The authors also do not report if index patients assigned to the self referral arm were more or less likely to appear for the postenrollment visit than those in the other 2 arms nor do they report whether self referral arm sexual partners who had not appeared for testing after 2 weeks were subsequently sought out by health aides (but who would not have been included in the 24% partner referral proportion). Seeking to refer these partners post facto, with patient consent, would not have compromised the study protocol2 and would have been the ethical thing to do.

Second, the very high partner HIV positivity rates observed (64% for main and 57% for casual partners) are a mixed blessing. On the one hand, it shows that PN for HIV with newly diagnosed persons has a high yield and provides opportunities for early treatment interventions, as the authors properly emphasize. On the other hand, the reported rates are remarkably high compared with similar indices in heterosexual settings outside of sub-Saharan Africa. Although few large studies of HIV PN in heterosexual populations have been reported, their yields have generally been considerably lower,3-5 roughly one-quarter to two-fifths of the values reported here.1 Partner populations in these various studies may not be strictly comparable, yet the magnitude of the difference in yield between sub-Saharan Africa findings and those obtained outside of Africa3-5 is so large that it needs further study.

Thus, Brown et al's1 report-and the one from Cameroon6 reporting similar partner HIV positivity rates (53%)-adds to puzzling observations which have, for a quarter century, fueled the question “Why Africa?” Here is not the place to discuss the differing views7,8 of what factors may be driving the turbocharged adult HIV epidemics frequently noted in sub-Saharan Africa: whether these consist almost exclusively of sexual exposures or of some combination of sexual and nonsexual (eg, puncturing) exposures. However, here is the place to point out opportunities that PN offers to elucidate local HIV transmission patterns. In such an investigative mode PN endeavors could, given donor assistance, be augmented by the following: (1) interviewing both HIV-positive and HIV-negative partners for relevant risk factors (ie, sexual and puncturing exposures) and/or (2) analyzing viral sequences of HIV-positive patient-partner pairs to assess linkage.9,10

Such additions to the PN enterprise would also shed light on the authors' unelaborated observation that nearly half (45%) of tested partners were HIV serodiscordant. Indeed if most of their study's participants were married women naming a single partner (how, incidentally, were couples counseled on prevention?) further investigation of why every other sole partner may be HIV negative seems warranted. This suggestion is reinforced by the not infrequent finding of discordant HIV sequences among heterosexual partners studied in different African countries,11-13 including Malawi.14 In any event, scientific caution indicates that neither authors nor readers of their article should automatically assume that HIV-positive partners acquired infection sexually from index patients or vice versa because, unlike classic STI, HIV infection can be transmitted by nonsexual means. For example, seroconcordance in couples could arise through one partner transmitting HIV to the other through blood exposures, such as sharing home medical injection kits, multidose vials, or razors or through both partners sharing a similar profile of blood exposures elsewhere, such as receiving unhygienic medical, dental, or cosmetic care.15,16 Empirically supported assessment of local risk factors for HIV infection would improve the efficacy of risk reduction recommendations for uninfected partners, a priority clearly emphasized by the authors.

Given the authors' data, which support the conclusion that PN for HIV infection in Lilongwe, Malawi, was acceptable, feasible, virtually harmless, and a fine case-finding tool for unrecognized HIV infection and, given that their revolution can be exported to other settings in sub-Saharan Africa, it is surprising that they remain silent on matters of cost. Although it was probably assessment of acceptability-feasibility that primarily motivated their investigation, a sense for what resources might be needed to achieve similar results would have been useful, especially for those intending to imitate their intervention. In the absence of such information, I speculate that much could be accomplished with improvements in the use of available resources coupled with enthusiasm, a solid predictor of PN success.17

Given that PN for garden variety STI has long been shown to be acceptable and feasible in sub-Saharan Africa,18 the question suggests itself: “Why did it take so long to test PN with HIV?” However, at no point should we be anything but grateful that this pioneering trial of HIV PN in a sub-Saharan population was implemented. As a public health tool, PN can fulfill 3 important functions: the ethical obligation to notify exposed persons; reduction of the disease burden; and stimulation of epidemiologic insight.17 Their report, including the emergent epidemiologic insights their data seemingly point to, goes a long way toward satisfying these desirable goals. Finally, their finding that reliance on health aide-assisted PN provided the best results is important and merits special emphasis.

Back to Top | Article Outline


1. Brown LB, Miller WC, Kamanga G, et al. HIV partner notification is effective and feasible in sub-Saharan Africa: opportunities for HIV treatment and prevention. J Acquir Immune Defic Syndr. 2011;56:437-442.
2. Potterat JJ, Rothenberg R. The case-finding effectiveness of a self-referral system for gonorrhea: a preliminary report. Am J Public Health. 1977;67:174-176.
3. Hogben M, McNally T, McPheeters M, et al, and Task Force on Community Preventive Services. The effectiveness of HIV partner counseling and referral services in increasing identification of HIV-positive individuals. Am J Prev Med. 2007;33(2 suppl):S89-S100.
4. Brewer DD. Case-finding effectiveness of partner notification and cluster investigation for sexually transmitted diseases/HIV. Sex Transm Dis. 2005;32:76-83.
5. Malave MC, Shah D, Sackoff JE, et al. Human Immunodeficiency Virus partner elicitation and notification in New York City: public health does it better. Sex Transm Dis. 2008;35:869-876.
6. Tih PM, Forgwei G, Welty T, et al. Integrated HIV contact tracing and partner notification in Cameroon: a feasible HIV infection risk reduction intervention for resource-poor settings. Presented at: Fifth International AIDS Conference on HIV Pathogenesis and Treatment; 2009; Cape Town, South Africa.
7. Schmid G, Buve A, Mugyenyi P, et al. Transmission of HIV-1 infection in sub-Saharan Africa and the effect of elimination of unsafe injections. Lancet. 2004;363:482-488.
8. Gisselquist D. Points to Consider: Responses to HIV/AIDS in Africa, Asia, and the Caribbean. London, United Kingdom: Adonis and Abbey Publishers Ltd; 2007. Available at: Accessed February 10, 2011.
9. Brewer D, Rothenberg R, Potterat JJ, et al. HIV epidemiology in Africa: rich in conjecture, poor in data (reply to letter by Boily et al). Int J STD AIDS. 2004;15:63-65.
10. Brody S, Potterat JJ. Establishing valid AIDS monitoring and research in countries with generalized epidemics. Int J STD AIDS. 2004;15:1-6.
11. Yirrell DL, Pickering H, Palmarini G, et al. Molecular epidemiological analysis of HIV in sexual networks in Uganda. AIDS. 1998;12:285-290.
12. Kaleebu P, Whitworth J, Hamilton L, et al. Molecular epidemiology of HIV type 1 in a rural community in southwest Uganda. AIDS Res Hum Retroviruses. 2000;16:393-401.
13. Trask SA, Derdeyn CA, Fideli U, et al. Molecular epidemiology of human immunodeficiency virus type 1 transmission in a heterosexual cohort of discordant couples in Zambia. J Virol. 2002;76:397-405.
14. McCormack GP, Glynn JR, Crampin AC, et al. Early evolution of the human immunodeficiency virus type 1 subtype C epidemic in rural Malawi. J Virol. 2002;76:12890-12899.
15. Birungi H. Injections and self-help: risk and trust in Ugandan health care. Soc Sci Med. 1998;47:1455-1462.
16. Brewer DD, Rothenberg R, Potterat JJ, et al. Data-free modeling of HIV transmission in sub-Saharan Africa. Sex Transm Dis. 2007;34:54-56.
17. Potterat JJ. Contact tracing's price is not its value. Sex Transm Dis. 1997;24:519-521.
18. Alam N, Chamor E, Vermund SH, et al. Partner notification for sexually transmitted infections in developing countries: a systematic review. BMC Public Health. 2010;10:19.
© 2011 Lippincott Williams & Wilkins, Inc.