In 2006 the WHO published a study claiming the following, “As a result of this study we have, for the first time, evidence that deliveries among women who have been subject to FGM are significantly more likely to be complicated and dangerous,” said Joy Phumaphi, Assistant Director-General, Family and Community Health, WHO. Read more here.

The WHO was right about one assertion: “for the first time.” Otherwise, the study claimed by Professor Saad M El Fadil, the study Principal Investigator in Sudan, to be “high-quality research” had many problems that are hard to believe were mistakes of oversight and lack of knowledge and experience. Some of them do not require an expert to point them out, but only a reader who refuses to be intimidated by big names into resigning his/her common sense.

But before I mention my critique of the study, I must say that it is sad that now we can’t use this study and some of its credible results regarding the severer forms of female genital cutting to stamp out the harmful and criminal forms of those practices. That is the price science pays when it gets politicized, it loses its credibility.

Let us only take the division of the sample as an example. Here is what the WHO said about the division of the sample:

1. No FGM: no evidence of any genital mutilation
2. FGM I: excision of the prepuce, with or without excision of part or all of the clitoris
3. FGM II: excision of the clitoris with partial or total removal of the labia minora
4. FGM III: excision of part or all of the external genitalia and stitching or narrowing of the vaginal opening (infibulation)

The second sample is rather interesting.

• How could they lump together three different practices into one sample?
• How could they do that when they know that the tissues of the clitoral hood and the clitoris are different, embryologically, histologically, anatomically, and physiologically?
• How could they do that when the WHO itself said before about the removal of the clitoral hood that it is like male circumcision, medically speaking.[1]
• Why can’t anyone with some common sense ask them how they can prove that any statistically significant results in sample 2 are not all attributed to the removal of the complete clitoris?
• Moreover, were there statistically significant results? I will let John Tierney, editor of the TiernyLab, answer this question: “In that WHO study, not a single statistically significant difference was found between those who had a “type 1” genital surgery versus no surgery; no statistically significant differences were found between those who had no genital surgeries and those who had type 1, 2 or 3 genital surgeries for the best predictor of infant health, namely birth weight; the perinatal death rate for the actual women in the sample who had a “type 3” surgery was in fact lower (193 infant deaths out of 6595 births) than those who had no surgery at all (296 deaths out of 7171 births) and only became statistically significant in a negative direction through non-transparent statistical manipulation of the data; the study collected data on women across six nations but never displayed the within nation results; there was no direct control for the quality of health care available for “circumcised” versus “uncircumcised” women; the sample was unrepresentative of the whole population; and in general any reported increased risk for genital surgery was astonishingly small and hardly a mandate for an eradication rather than a public health program.”[2]
• Please read also Carla Obermeyer’s comprehensive and critical reviews of the medical and demographic evidence on the topic (published in the journal Medical Anthropology Quarterly). She said, “On the basis of the vast literature on the harmful effects of genital surgeries, one might have anticipated finding a wealth of studies that document considerable increases in mortality and morbidity. This review could find no incontrovertible evidence on mortality, and the rate of medical complications suggest that they are the exception rather than the rule.”[3]

Even if they could argue any statistical significance, I don’t know of many medical procedures that have zero harm. In medicine, it is about the balance between the harm and benefit. In fact, even the social acceptance of medical decisions must be taken in consideration, according to a report from the American College of Chest Physicians Task Force that published an article on Grading Strength of Recommendations and Quality of Evidence in Clinical Guidelines. [4]

But if they could prove some harm, and we also believed them without any proofs that there is no benefit, one thing remains, and I challenge them or anyone else to show respond to it. Why in the world did you divide your sample in such manner?

Just wondering!

Hatem al-Haj, MD, PhD