A nagging concern has arisen in the wake of a
recently published study [pdf] examining HIV infection rates in female partners of HIV-positive, circumcised men. The study, authored by Wawer et al., was stopped early, reportedly due to "futility." However,
valuable information may have been lost by stopping the study early. The clear trend was towards much greater rates of HIV infection among the female partners of circumcised men. Indeed, HIV infection sero-conversion rates were trending towards 50% or greater among these women.
The researchers called continuing the observation and collection of data an exercise in futility. If the authors were only seeking a benefit to female partners of circumcised HIV-positive men, they would be correct to describe the continued search as futile. However, why were they not also looking for a negative impact?
The relevant facts of the study are these. The male half of the couples recruited were already HIV-positive. The trend was towards higher sero-conversion among women in the couples where the male was circumcised. No additional ethical dilemmas were present in continuing to observe the trend. In fact, offering circumcision to the intact males arguably would have presented the additional ethical dilemma. Therefore, continued observation to the scheduled completion date of the study could only have yielded more and better data, possibly rising to the level of statistical significance.
It is hard to conclude anything other than that the researchers were not looking for a negative impact and perhaps may have desired to avoid a statistically significant adverse finding to male circumcision. Johns Hopkins University, home to Bloomberg School of Public Health, has been a hot bed of research into the efficacy of male circumcision in reducing HIV risk. An adverse finding could jeopardize not only their programs, both research and roll out, centered around the belief that male circumcision has value in the HIV/AIDS fight, but also the entire circumcision industry growing up around this latest fad in HIV/AIDS prevention.
The real reason no statistical significance was observed may be because the observers avoided that possibility.