Early Monday morning, the American Academy of Pediatrics put out an updated policy statement on circumcisions to replace the 1999 version. The new statement comes out mainly in favor of circumcision and says insurances should have to cover it. I’m very disappointed, not only in the conclusion but in the way the authors reached it.
Previously, the 1999 statement said basically “we aren’t sure, there are reasons to circumcise and reasons not to—think about it and decide for yourself.” Since then, the rate of newborn circumcisions in our country has declined gradually, to a little over half of newborn boys. Some say this is a direct result of the AAP statement, leading to decreased coverage by insurers. I suspect it is also because pediatricians and parents became more aware the evidence for cutting wasn’t powerful.
In other developed countries, circumcision is even less common. I don’t think this is all based on finances but likely involves a more reasoned approach to the evidence plus a different ethic regarding the rights of children. Even though the US did sign the UN Convention on the Rights of the Child, we never did ratify it. We, of the developed countries, have one of the strongest tendencies to treat children as the chattel of their parents, rather than offering full protection of human rights from the state.
Without going into every detail of the policy statement, I’ll point out a few elements to give you the general flavor. First, the authors say that based on new evidence, they can state “the health benefits of newborn male circumcision outweigh the risks.” Further on, after a discussion of selected risks like surgical infection, bleeding, disfigurement and sexual function (but ignoring certain conflicting data on sexual satisfaction and pain), they acknowledge it proved impossible to quantify the risks! They say “based on the data reviewed, it is difficult, if not impossible, to adequately assess the total impact of complications, because the data are scant and inconsistent regarding the severity of complications. ” And “financial costs of care, emotional tolls, or the need for future corrective surgery (with the attendant anesthetic risks, family stress, and expense) are unknown.” And “the majority of severe or even catastrophic injuries are so infrequent as to be reported as case reports (and were therefore excluded from this literature review). These rare complications include glans or penile amputation, transmission of herpes simplex after mouth-to-penis contact by a mohel (Jewish ritual circumcisers) after circumcision, methicillin resistant Staphylococcus aureus infection, urethral cutaneous fistula, glans ischemia, and death.”
Now tell me, even without a statistical or medical background, how anyone can say benefits outweigh risks when said risks can’t be well described? The authors probably know that, because in their shorter paper meant for lay persons to review, they admit “health benefits are not great enough to recommend routine circumcision for all male newborns.”
The authors divided up over 1000 medical papers between themselves to study, and they set some rules for which papers to consider. One of their rules was to exclude any studies that failed to meet a certain level of evidence. They used a common scale to describe level of evidence that goes from 1 to 8, with 1 being the best level of evidence and 8 being the poorest. For this paper, they said they would not use any level of evidence worse than 4.
Unfortunately, their cornerstone of evidence favoring circumcision was a recent paper done by the CDC describing potential cost savings of newborn circumcision in the US, based on HIV data from studies in Africa. There was no data on HIV prevention in the US by circumcision included in this study. What’s the problem with that? HIV transmission in the African studies was mostly heterosexual, and the disease happened in men who also had high rates of other STD’s that had caused skin breakdown and/or ulcers. These other STD’s are not as common here. Circumcision does not appear to reduce same-sex HIV transmission. So while it might be true that the same protection from HIV in heterosexual males might happen in the US, we really still don’t know. Medical history is littered with the bad results of reasoning that seemed good at the time.
Because we know reason in place of evidence often fails to hold up under scrutiny, studies that use extrapolated data or data collected for other purposes are labeled category 7 in quality of evidence. The only level poorer is “common sense.” The crux of this new AAP statement is a paper with a level of evidence far lower than what the authors agreed to even look at, and they make no effort to point this out. It should have never gotten past peer review.
How do we weigh risks and benefits in medicine, in general? It can be harder than you’d think! When the outcome involves death or permanent disability, it’s easier. Otherwise, often we wind up looking at things so different they are like comparing vision to taste—which sense organ is your favorite? Would you rather take this medicine that might make you live 3 more months or would you rather have 3 more weeks that will probably be less painful? Would you rather wear a condom or do you place less value on your foreskin? In those cases, it is most reasonable to explain the various outcomes to patients and let them make the call.
Sometimes policy recommendations are made on a cost-benefit basis, when the medical risks aren’t clear. If there is an extreme cost difference, I won’t say that isn’t reasonable—but it should be made excruciatingly clear to individual patients when a given practice is based on cost, not risks and benefits.
When the science is in debate, I think it is ethical for parents to make the call as proxies for their children. I don’t believe it is ethical, if the science says something is too risky, for parental religion or preference to overrule science. What is the situation with circumcisions? I don’t know. Personally, I think there should be an overwhelming evidence of benefit if you are going to permanently remove a body part. I say I don’t know for certain, though, because for specific, individual children the risks could be overwhelming. HIV can be overwhelming, even if it is now more treatable, and it might turn out that circumcision does reduce HIV even in our country. Urinary tract infection, more common in uncircumcised male infants (as common as in girls), can be overwhelming or even deadly. I have seen, with my own eyes, a newborn boy die of overwhelming kidney infection despite desperate and skilled resuscitation efforts (he was circumcised)—but only one, in 16 years of practice. I don’t see evidence that the rates of these tragic events outweigh the occurrence of circumcision-related risks in the population as a whole, but for a specific child, I don’t have a crystal ball to know. So I do think it may be acceptable for that child’s proxy, the parents, to decide.
A couple of years ago, I was explaining this to a group of residents and students while in process of teaching them to do a circumcision on a baby whose parents requested it. I said my gut feeling was that the evidence was going to weigh more against circumcision than for it, and that we would eventually regard it with as much horror as most of the world does female genital mutilation. But I said we were probably not yet in the position to know enough to outlaw it or refuse to do it at parental request. I am very comfortable with refusing to do something to a patient for medical reasons, but not at all comfortable with refusing for personal reasons.
Right afterwards, we got in the elevator, and a little girl of about 4 looked up at me with solemn eyes as she asked, “Why did you cut that baby’s pee-pee?” I was struck dumb—it was as if I’d dropped into a movie like the Sixth Sense. The procedure room is closed, so I knew she couldn’t have seen us. Her mother laughed and told us the girl’s brother had his circumcision that morning. The memory of that question still unnerves me, and I am quite relieved that I personally no longer have to do the procedure, now that a “circumcision team” has formed at my hospital and I’m not on it. I guess that is a cop-out. I still am not sure how to answer the little girl’s question.
Whatever you think about the science and ethics of circumcision and medical decision making, I would think most of us could agree that the AAP policy statement, by dismissing risks it can’t quantify and using a level of evidence it claims not to, has serious flaws. I am embarrassed for a paper like that to be issued from my professional society. It should be revised and made honest.