Male Circumcision and HIV Prevention: Insufficient Evidence and Neglected External Validity

Lawrence W. Green, DrPH, John W. Travis, MD, MPH, Ryan G. McAllister, PhD,
Kent W. Peterson, MD, FACPM, Astrik N. Vardanyan, MA, Amber Craig, MA

Published in American Journal of Preventive Medicine ~ Full article here

General Population Correlates
Effectiveness in real-world settings rarely achieves the
effıcacy levels found in controlled trials, making predictions
of subsequent cost-effectiveness and population health
benefıts less reliable. The following related concerns
deserve further scrutiny:
1. The three RCCTs were terminated early because results
had reached signifıcance showing reducedHIVinfections
in experimental compared with control groups; however,
it was too soon to gauge long-term effectiveness.
2. The results have no relevance for women or for men
who have sex with men.
3. Far more participants were lost to follow-up than were
reported to have contracted HIV.
4. A substantial number of participants appeared to have
contracted HIV from nonsexual sources: 23 of the 69
infections reported in the South African trial and 16 of
the 67 in the Ugandan study.11
5. Participants received continuous counseling, free condoms,
and monitoring for infection, which was unlikely
in real-world campaigns.
6. The sanitary conditions of the surgeries would be diffıcult
to replicate on a mass scale in many parts of Africa where
HIV infection rates and prevalence are highest.

Correlation between HIV prevalence and male circumcision
prevalence in observational studies12,13 is inconclusive.
Substantial evidence contradicts the RCCTs’
results and suggests that real-world population benefıts
from male circumcision might be minimal:

1. An analysis14 of HIV prevalence compared to circumcision
status in sub-Saharan Africa concluded that male
circumcision is not associated with reduced HIV
2. Another study15 on circumcision prevalence compared
to HIV in the general South African population
concluded: “Circumcision had no protective effect on
HIV transmission.”
3. When commercial sex worker patterns are controlled,
male circumcision is not signifıcantly associated with
lower HIV prevalence.16
4. Mathematical impact modeling of circumcision, antiretroviral
therapy (ART), and condom use for South Africa
concluded: “Male circumcision was found to have considerably
lower impact than condom use or anti-retroviral
therapy on HIV infection rates and death rates.”17
5. Both the U.S. and sub-Saharan Africa have relatively
high incidence rates of HIV infection, considering that
about 75% of U.S. men and about 70% of sub-Saharan
African men are circumcised—higher percentages
than in most other regions or countries with lower
prevalence of HIV (Demographic and Health Surveys,


Related Posts Plugin for WordPress, Blogger...