Skip Navigation LinksHome > January 2007 - Volume 34 - Issue 1 > Response to Gray's Letter
Sexually Transmitted Diseases:
doi: 10.1097/01.olq.0000253217.82634.f5
Letters to the Editor: Author's Response

Response to Gray's Letter

Orroth, Kate K. PhD*; White, Richard G. PhD†; Hayes, Richard J. Dsc†

Free Access
Article Outline
Collapse Box

Author Information

*Infectious Disease Epidemiology Unit, Department of Infectious and Tropical Diseases, London School of Hygiene and Tropical Medicine London WC1E 7HT, United Kingdom; and †Infectious Disease Epidemiology Unit Department of Epidemiology and Public Health, London School of Hygiene and Tropical Medicine, London WC1E 7HT United Kingdom

Back to Top | Article Outline

To the Editor:

Gray et al question our simulation estimates of the proportions of new HIV infections in the Mwanza and Rakai trial populations attributable to chancroid.

One of our main findings was that the simulated PAF for chancroid in Mwanza was much higher than in Rakai, and we attributed this to higher-risk sexual behavior in Mwanza,1 resulting in a higher prevalence of chancroid,2 as well as a more mature HIV epidemic in Rakai, where a large proportion of HIV transmission occurred within stable partnerships with a low prevalence of curable STDs.3 Gray et al claim the model “seriously exaggerates” the role of chancroid due to unsupported assumptions about the prevalence of chancroid in Mwanza and the high cofactor effect assumed for chancroid.

Our simulated PAF for the effect of chancroid on HIV acquisition in Rakai was 2.3%, which is low and only slightly higher than the empirical estimate of <1%.4 The simulated prevalence of chancroid in Mwanza was 1%. We acknowledge in our paper that empirical data on chancroid prevalence in the general population were not available. However, the reported incidence of genital ulcer disease (GUD) in the Mwanza trial population was high, and data from STD patients in Mwanza town showed that chancroid was an important cause of GUD. In a consecutive series of 200 patients with GUD, chancroid was suspected in 20 to 55% and confirmed by culture in 20 to 25% (H. Grosskurth, personal communication), in sharp contrast with the 2.5% of GUD cases that were PCR-positive for chancroid in Rakai. To assume zero prevalence for chancroid in Mwanza in these circumstances would seem to us a much less defensible assumption.

Gray et al also criticize the “extremely high cofactor effect (a relative risk of 25)” assumed by the model. However, this is not a “relative risk” in the traditional epidemiologic sense as it refers to the cofactor effect during a single sexual contact between HIV-infected and HIV-uninfected partners. Relative risks measured during an extended follow-up period are lower than the per-contact cofactor effect because the STD will not be present throughout the follow-up period.5 In Rakai, for example, Gray et al report an adjusted relative risk for HIV acquisition in subjects reporting GUD during the 10-month period between survey rounds of 3.14. The mean reported duration of GUD was 1.2 months.4 Using these data, the per-contact cofactor effect for GUD can be estimated as about 20.

In our sensitivity analysis, we examined the effect of reducing the cofactor effect for chancroid from 25 to 12.5, and the simulated PAF for chancroid only decreased from 39% to 30%. Thus, although there is considerable uncertainty about the exact size of cofactor effects, our qualitative conclusions remain unchanged.

Our study does not “distort strategies for future STD control efforts.” On the contrary, it demonstrates how the effects of STDs on HIV transmission vary between populations, depending on underlying STD prevalences and HIV epidemic stage. There is now clearer understanding of the need to tailor HIV intervention strategies to the local epidemiologic context.

Back to Top | Article Outline

References

1. Orroth KK, Korenromp EL, White RG, et al. Higher risk behaviour and STD rates in Mwanza compared to Rakai and Masaka: A possible explanation for outcomes of HIV prevention trials? AIDS 2003; 17:2653–2660.

2. White RG, Orroth KK, Korenromp EL, et al. Can population differences explain the contrasting results of the Mwanza, Rakai and Masaka HIV/STD intervention trials? A modelling study. J Acquir Immune Defic Syndr 2004; 37:1500–1513.

3. Orroth KK, White RG, Korenromp EL, et al. Empirical observations underestimate the proportion of HIV infections attributable to sexually transmitted diseases in the Mwanza and Rakai STD treatment trials: Simulation results. Sex Transm Dis 2006; 33:536–544.

4. Gray RH, Wawer MJ, Sewankambo NK, et al. Relative risks and population attributable fraction of incident HIV associated with STD symptoms and treatable STDs in Rakai District, Uganda. AIDS 1999; 13:2113–2123.

5. Hayes RJ, Plummer FA. The cofactor effect of genital ulcers on the per-exposure risk of HIV transmission in sub-Saharan Africa. J Trop Med Hyg 1995; 98:1–8.

© Copyright 2007 American Sexually Transmitted Diseases Association

Login