Reader Comments
Post a new comment on this article
Post Your Discussion Comment
Please follow our guidelines for comments and review our competing interests policy. Comments that do not conform to our guidelines will be promptly removed and the user account disabled. The following must be avoided:
- Remarks that could be interpreted as allegations of misconduct
- Unsupported assertions or statements
- Inflammatory or insulting language
Thank You!
Thank you for taking the time to flag this posting; we review flagged postings on a regular basis.
closeMisrepresentation of sources.
Posted by Sawers on 17 Mar 2014 at 23:56 GMT
The authors introduce their article discussing factors that "made sub-Saharan Africa special" and which could be used to explain the rapid rise and subsequent decline in HIV prevalence in the region. They assert that "undoubtedly, highly connected sexual networks generated by unprotected, often transactional, sex with many different, frequently concurrent, partners are common in Africa" and cites my article with Isaac and Stillwaggon "HIV and concurrent sexual partnerships: Modelling the role of coital dilution," J Int AIDS Soc 14: 44. (They also cite Tanser, Barnighausen, Hund, Garnett, McGrath, et al. "Effect of concurrent sexual partnerships on rate of new HIV infections in a high-prevalence, rural South African population: a cohort study" in Lancet in 2011.) The authors apparently did not read either of the two articles, whose point was to argue that concurrency cannot account for the extraordinarily high HIV prevalence in sub-Saharan Africa.
Nowhere in our article do we assert that "highly connected sexual networks" are unusually prevalent in sub-Saharan Africa. For there to be highly connected heterosexual networks, women's concurrency must be high, and the evidence shows that women's concurrency in sub-Saharan Africa is, if anything, lower than in non-African countries where it has been measured. (See especially Sawers, "Measuring and Modeling Concurrency" in JIAS January 2013.)
Nowhere in our article do we mention transactional sex nor do we explain the high prevalence of HIV in sub-Saharan Africa with unusually common transactional sex. If the authors had read my earlier article with Stillwaggon "Concurrent sexual partnerships do not explain the HIV epidemics in Africa: a systematic review of the evidence" in JIAS September 2010, they would have found an extensive passage debunking the notion that transactional sex is unusually prevalent in sub-Saharan Africa.
Nowhere in our article do we assert that "sex with many different, frequently concurrent, partners are common in Africa." Available evidence for at least a decade has shown that people in sub-Saharan Africa on average report fewer sexual partners (life-time or previous year) than is typical in non-African countries where there have been surveys. (See especially Wellings "Sexual behaviour in context: a global perspective," in the Lancet, November 2006.) The evidence is also clear that both men's and women concurrency in sub-Saharan Africa is not unusually prevalent when compared with non-African countries. (See Sawers, "Measuring and Modeling Concurrency" and Sawers and Stillwaggon "Concurrent sexual partnerships do not explain the HIV epidemics in Africa: a systematic review of the evidence.")
The authors should have been more careful when citing other publications.
Larry Sawers
Department of Economic
American University
Washington, DC
RE: Misrepresentation of sources.
sdevlas replied to Sawers on 24 Apr 2014 at 14:31 GMT
We thank Dr Sawers for his response to our paper. The references to which he refers were for the whole preceding text of the paragraph, not specifically this particular sentence. This is something we should have avoided and we apologize for the confusion.
As regards the substance, we believe that the fact alone that HIV has spread so rapidly in many African countries makes the conclusion that highly connected networks are common in these countries almost a tautology, unless one believes that non-sexual transmission has played a major role. Undoubtedly, sex work (sex workers often reached extremely high HIV prevalences) and concurrency, as well as other factors such as intergenerational sex, have played some role. However, this role is hard to quantify exactly. Whether these factors make Africa “unusual” – whatever that may mean – we cannot say. Self-reported behavior data has consistently been shown to be unreliable, and with few exceptions, such as the Likoma Island Study (which by the way demonstrates the presence of highly connected sexual networks) 1, little information about sexual networks and their role in HIV transmission exists. It is because of this lack of data that we cannot say to what extent the decline in incidence is attributable to behavior change. Other explanations exist, which is all we wanted to bring across.
1 Helleringer S, Kohler HP (2007). Sexual network structure and the spread of HIV in Africa: evidence from Likoma Island, Malawi. "AIDS" 21, 2323-32.
Nico Nagelkerke, Sake de Vlas
Dept Public Health, Erasmus MC, University Medical Center Rotterdam, The Netherlands