Bad science by the AAP.

The AAP report on circumcision: Bad science + bad ethics = bad medicine

For the first time in over a decade, the American Academy of Pediatrics (AAP) has revised its policy position on infant male circumcision. They now say that the health benefits conferred by the procedure just slightly outweigh the risks and harms. Not enough to come right out and positively recommend circumcision (as some media outlets are erroneously reporting), but just enough to suggest that whenever it is performed—for cultural or religious reasons, or sheer parental preference, as the case may be—it should be covered by government health insurance.

That turns out to be a very fine line to dance on. But fear not: the AAP policy committee comes equipped with tap shoes tightly-laced, and its self-appointed members have shown themselves to be hoofers of the nimblest kind. Their position statement is full of equivocations, hedging, and uncertainty; and the longer report upon which it is based is replete with non-sequiturs, self-contradiction, and blatant cherry-picking of essential evidence. Both documents shine as clear examples of a “lowest common denominator” mélange birthed by a divided committee, some of whose members must be well aware that United States is embarrassingly out of tune with world opinion on this issue.

On a global scale, medical experts are steadfastly divided on the question of whether the circumcision of male minors confers any – let alone significant – health benefits. Indeed, child health experts in Britain, Germany, Scandinavia, Australia, New Zealand, and Canada are firmly of the view that non-therapeutic circumcision (NTC) confers no meaningful benefits, and that it should be neither recommended to parents nor funded by health insurance systems.

Nota bene: these cosmopolitan physicians and the medical boards on which they sit have access to the very same data as the AAP. They just don’t draw the same conclusions.

In view of this empirical uncertainty on the medical question, it is problematic to assert, as the AAP does in its new report, that a person does not have the right to decide whether he wishes to keep his own healthy foreskin at birth, and that the right belongs instead to his parents. A more reasonable conclusion is that the owner of the foreskin should be allowed to consider the evidence (in all its murkiness) for himself—when he is mentally competent to do so—and make a personal decision about what is, after all, a functional bit of his own sexual anatomy and one enjoyed without issue by the vast majority of the world’s males.

According to the Seattle-based physicians group Doctors Opposing Circumcision, there is neither a medical nor an ethical case for removing healthy genital tissue from baby boys. They can’t consent to the procedure in the first place, and the bulk of the claimed—yet heavily disputed—health benefits don’t even apply to them: babies are not sexually active, yet circumcision is supposed to protect chiefly against sexually-transmitted infections and related diseases. In any case, these are afflictions whose prevention is much more soundly assured by the use of a condom in adulthood than by genital surgery in infancy. And before you bring up urinary tract infections in early childhood, remember these are profoundly rare for boys, and can be easily treated with antibiotics if and when they do occur—no surgery required. So how did the AAP reach its much-hyped, yet ultimately fallacious, and certainly medically unjustified conclusion?

* * *

First, let’s be clear about what their position is. “This is not really pro-circumcision,” explains one of the authors of the technical report behind the new analysis. You wouldn’t know that from reading the week’s headlines, which have taken the “health benefits” narrative and gone running impetuously on to town, but there it is from the horse’s mouth. Instead, the AAP believes that the purported benefits of circumcision are merely “sufficient” to “justify accessto this procedure for families choosing it” and to “warrant third-party payment for circumcision of male newborns” if and when it does occur.

Here they depart from their 1999 statement in asserting that (1) the benefits of the surgery definitively outweigh the risks and costs and (2) that it is therefore justifiable to perform the operation without the informed consent of the patient. This does not follow. In medical ethics, the risk/benefit rule was devised for therapeutic procedures aimed at treating an extant pathological condition, and for minor prophylactic interventions such as vaccination. It has no relevance to nonessential amputative surgery, especially when it involves the painful removal of healthy, functional erogenous tissue from the genitals, and when safer, more effective substitute strategies exist for achieving the same ends.

You may be surprised to learn that the word “condom” does not appear even once in the 28 page AAP report.

In making their risk/benefit calculations, then, the AAP simply leaves out a critical bulk of factors relevant to the equation, including the existence of a range of proven healthcare tools like condoms, vaccines (including an effective HPV vaccine), and antibiotics. If they had bothered to consider human rights and bodily integrity issues, the function of the foreskin, its value to the individual, and his possible wishes in later life, as well, their computations would quite plainly yield a very different answer.

Now, some readers will be unaware that the AAP is not a dispassionate scientific research body, but rather a trade association for pediatricians. Those among its members and stakeholders who perform NTCs stand to profit from the procedure, to the collective annual tune of $1.25 billion. Given the yawning potential for a financial conflict of interest, then, there needs to be a very strong, independent medical case for circumcision; and the AAP had better be able to show that it’s both the safest and most cost effective means of promoting infant health. Both of these propositions fail—and fail dramatically—as I will continue to show in what follows.

* * *

The AAP has been tossing and turning on the question of circumcision since 1971, when it announced that “There are no valid medical indications for circumcision in the neonatal period.” From 1999 until August 27th of this year, the AAP had maintained that the “health benefits” of circumcision were perhaps neck-and-neck with the costs, at best, so that it could not recommend the procedure from a therapeutic perspective. This policy was in line with the still-current official position of everyother major medical association in the world. Except, actually, those that now actively campaign against circumcision, such as the Royal Dutch Medical Association in Holland and the Canadian Paediatric Society.

For the AAP to revise its stance, then, it stands to reason that something must have changed—either human biology has altered, or some new evidence must have cropped up—to justify tipping the cost-benefit scales away from their recently prior equilibrium. Indeed, the AAP circumcision “task force” makes much ado of a small, hand-picked collection of studies conducted in Africa between 2005 and 2007 purporting to show a link between circumcision and a reduced risk of becoming infected with HIV.

According to the New York Times, these studies include 14 publications “that provide what the [AAP] characterizes as ‘fair’ evidence that circumcision in adulthood protects men from HIV transmission from a female partner.”

Notice the phrase in adulthood. The AAP policy, by contrast, is concerned with circumcision in infancy, a procedure for which there is literally no evidence of a protective effect against HIV. Notice also “fair” rather than “good” evidence and that the findings apply exclusively to (heterosexual) (African) (adult) males. Never mind the poor females, for whom circumcision of the male partner is a risk factor for becoming infected with HIV. The New York Times continues:

“Three of the studies were large randomized controlled trials of the kind considered the gold standard in medicine, but they were carried out in Africa, where H.I.V. — the virus the causes AIDS — is spread primarily among heterosexuals.”

There are a number of things to say about these “randomized controlled trials.” First, the trials were “randomized” and “controlled” in name only, as this exhaustive analysis carefully demonstrates. Clinically relevant flaws included “problematic randomization and selection bias, inadequate blinding, lack of placebo-control … inadequate equipoise, experimenter bias, attrition …  not investigating male circumcision as a vector for HIV transmission, not investigating non-sexual HIV transmission, as well as lead-time bias, supportive bias … participant expectation bias, and time-out discrepancy (restraint from sexual activity only by circumcised men).”Hence, as I explained in this earlier post, the “Africa studies” were anything but a “gold standard” in medicine.

Critics have also pointed out that the “60%” figure that is typically sold as the relationship between circumcision and reduction of HIV infections is the misleading output of a statistical sleight-of-hand: the absolute reduction between the circumcised and intact groups in these flawed studies was a mere 1.3%.

The next thing to highlight is the part of the quote that comes after the “but” – a very important “but” – namely that “[the trials] were carried out in Africa” where, as the article goes on to explain, HIV is mainly a heterosexual phenomenon. Outside of Africa, it is mainly not—it is largely transmitted among injecting drug users and gay men—which means that even if we were to accept the erroneous data from the poorly-designed “randomized controlled” studies, we would have no evidence that circumcision could be useful in the United States. Or Australia, or New Zealand, or anywhere else in the developed world. The epidemiological and social environments are just flatly non-analogous — as this study painstakingly shows.

Hence, as even the authors of the AAP report acknowledge, “the degree of benefit, or degree of impact [of circumcision], in a place like the U.S. will clearly be smaller than in a place like Africa.” Well, indeed.

Of course, we already knew that circumcision does not present a serious obstacle to heterosexual HIV-transmission in the U.S., since the U.S. has both the highest rates of infant circumcision and the highest rates of heterosexually transmitted HIV among industrialized nations.

But wait, there’s more! The AAP report completely ignores both (1) additional evidence from Africa that contradicts the findings they do include and (2) multiple studies of heterosexual men in North America looking at whether there is an association between circumcision status and HIV status. Several were large enough to show a difference if one existed. Only one showed a difference. And a recent study from Puerto Rico found that circumcised men were at significantly greater risk for HIV as well as a host of other diseases including genital warts. So matters have, in fact, been studied closer to home, yet the niggling implications of these data are left unacknowledged by the AAP task force.

But let’s put all that to the side. For even if it were true that circumcision offered some minor protective effect against HIV/AIDS or other STDs such as HPV (for which, as I stated before, there is an effective vaccine)—despite the best evidence to the contrary, and against all the points I have just laid out—it would still not follow that the procedure could be ethically performed on infants. This is crucial. Given that there is a cheaper, more effective, less invasive, less coercive alternative—namely condom-use in adulthood—it cannot be considered even remotely consistent with biomedical ethics to endorse the risky genital cutting of a pre-verbal child toward the same ostensible end.

As pediatrician, statistician, and professor of clinical medicine Robert Van Howe showed methodically in this recent cost-benefit analysis—also conspicuously missing from the AAP report—infant circumcision is more costly and does more harm than leaving the baby alone, even based on models that start from very generous premises about the potential health benefits of foreskin-removal. If the AAP wants to justify “third party payments” it cannot plausibly claim them for a procure that is more perilous, more ethically problematic, less effective and less cost effective than available alternatives. The government dime is clearly better spent elsewhere.

So let’s review:

1. The AAP used to say that circumcision could not be recommended on health grounds, which was, and remains, the only scientifically credible position for it to maintain.

2. In 2012, the AAP revised its position in light of “new evidence” suggesting that the health benefits could now be said to “outweigh” the harms and risks of the procedure.

3. The “new evidence” consists almost entirely of data collected in Africa between 2005 and 2007 suggesting that circumcision in adulthood may reduce the risk of HIV through unprotected heterosexual intercourse.

4. These data, however, are of poor quality, and show an absolute reduction of HIV infections of only 1.3% between the treatment and control groups. They are conflict with abundant other data not mentioned by the AAP. And even if they were taken seriously on their own terms, the would only apply to adult heterosexual males in Africa – not to infants in the United States.

Astonishingly, the AAP report itself makes essentially this same last point: “… the task force recommends additional studies to better understand the impact of male circumcision on transmission of HIV and other STIs in the United States because key studies to date have been performed in African populations with HIV burdens that are epidemiologically different from HIV in the United States.” Emphasis mine.

Yes, and until those studies are run – and run properly, with consenting populations, under strict ethical controls, I might add – it would be prudent for the AAP to abstain from making unsubstantiated claims about the benefits of circumcising infants in the United States. Especially since, as they concede on page 772 of their report “the true incidence of complications after newborn circumcision is unknown.” It should go without saying that if one doesn’t know how often complications occur, then one is plainly ill-equipped to assert that the benefits outweigh the risks.

* * *

It took the AAP circumcision “task force” several years to choreograph its latest tap-dance routine. Why it has produced a document so far out of line with both world opinion and the most basic of bioethical principles is a fascinating—and disturbing—question, but one which I cannot hope to answer in a single blog post. Whatever the reason, however, one can be sure that it has far more to do with culture than with science. As medical historians and cultural analysts have meticulously documented, circumcision as a birth ritual remains deeply, and uniquely, embedded in American medical culture and in the naïve expectations of doctors and parents alike. This sets the U.S. apart from everywhere else in the developed world—certainly outside of religious communities for whom the ritual is still self-consciously sacramental, and by whom it is performed without needing the rationalization of “health benefits.” Like any ritual, American proponents of circumcision are loath to give it up, for dread of the unknown consequences.

* * *

Long ago, the Aztecs feared that the sun would fail to rise if they did not make the annual sacrifice of human hearts. But we know better than to mistake our cultural habits for science or medicine. At least we should know better. The AAP committee—unable to remove its culture-colored lenses before conducting its review of the available health data—puts a sorry mark of shame upon my country of birth. I call upon the AAP to retract its ill-considered policy without delay.

http://blog.practicalethics.ox.ac.uk/2012/08/the-aap-report-on-circumcision-bad-science-bad-ethics-bad-medicine/

One thought on “Bad science by the AAP.

Leave a Comment

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s