What The Science Says About Circumcision: Part 1 — The Benefits

Note: This should go without saying, but the article below refers to male circumcision. Female circumcision is a completely separate practice, occurring for the express purpose of destroying sexual function. That the two practices share a name in common is unfortunate and misleading.

This is part 1 of a three-part series. Part 2 will deal with the scientific evidence as it pertains to risks associated with circumcision, and part 3 will deal with making sense of the circumcision-related rhetoric. It’s taken me a while to get up the guts to write an article on what the science says about circumcision, because it’s such an emotionally charged issue. I didn’t have to deal with the circumcision decision personally, since W is a girl. However — and perhaps at least in part because I didn’t have to deal with the decision — I can approach the science in a completely unemotional manner, and I think there’s value in so doing.

Here’s the thing; unlike many other parenting decisions that can be made based solely upon the science, circumcision is a decision that is (nearly by necessity) based upon emotion, morals, and values. Why? Two reasons. First, in the U.S., there’s no medical justification for doing the procedure, but there’s a major values-based justification for circumcision among many prevalent religious groups. Second, there’s no scientific evidence that routine circumcision causes harm and should be avoided, but many parents feel emotionally uncomfortable with the idea of removing a part of their baby’s body.

In the end, there are three points I’m trying to make with this article:

1) Circumcision is a very sensitive issue BECAUSE values are so wrapped up in the decision.

2) It’s fine to make a decision based upon values, but it’s important to recognize when a decision is values-based rather than evidence-based, and it’s critically important to avoid pretending that there is evidentiary support for a decision if, in fact, there’s not.

3) There is no strong evidence either FOR or AGAINST routine circumcision of boys in the U.S.

On to the science. There are several arguments that have historically been given in support of circumcision. Among these, there’s the argument that circumcision helps reduce the risk of HIV infection. There is actually reasonable scientific support for this assertion; a variety of studies (mostly conducted in Africa) have found that circumcision helps reduce the risk of acquiring the disease by somewhere in the range of 50-60% (see, for instance, Auvert et al, Bailey et al, Gray et al). In the Auvert, Bailey, and Gray studies, participants were recruited from among uncircumcised men. Those randomly assigned to the treatment group were circumcised (with their permission), while those randomly assigned to the control group were left uncircumcised (though they were offered the option of circumcision at the conclusion of the study). The randomization of the subjects allows for drawing causal (rather than correlational) conclusions; that is, because of the study design, we can say circumcising men helps reduce the risk of HIV. Correlationally, Bongaarts et al found that across 409 African ethnic groups, circumcision practices were 90% correlated with prevalence of HIV, indicating a significantly reduced risk of HIV among circumcised males.

Meta-analysis of studies on circumcision and HIV infection also reveal a correlation between circumcision and reduced risk of HIV (a meta-analysis is a study of the results of many different studies). Weiss et all looked at 27 studies of HIV and circumcision in Africa, and found that the vast majority of studies showed a significantly (about 50%) reduced risk of HIV. The reduced risk was even more significant (about 70%) in men at high risk for HIV infection. Similarly, Moses et al noted that a “substantial” body of evidence links circumcision to reduced risk of HIV.

Why would circumcision reduce the risk of HIV infection?  A study published in the British Medical Journal suggests that there is a high concentration of specialized cells that happen to have HIV receptors on the inside of the foreskin. These cells are a potential entry-point for the HIV infection (Szabo et al). The authors recommend routine circumcision in areas with high HIV prevalence. The World Health Organization (WHO) is convinced by the preponderance of evidence that circumcision is recommended in areas with high HIV prevalence. It’s worth noting, incidentally, that WHO is the same organization that recommends breastfeeding until at least 2 years of age, lest the reader be tempted to label the organization as “overly Western” too quickly.

Unfortunately, while there’s lots of evidence to support male circumcision in areas of epidemic HIV, the evidence doesn’t help inform scientific decision-making on circumcision in the U.S. The lower prevalence of HIV in the U.S. reduces the risk of HIV to such an extent that it’s no longer clear whether circumcision is justified solely as a means of HIV risk-reduction. The U.S. Centers for Disease Control and Prevention (CDC), for instance, doesn’t recommend circumcision solely for the purpose of HIV risk-reduction (though they are considering recommending the procedure for uncircumcised homosexual men).

There’s also been some suggestion that circumcision reduces the risk of contracting HPV (human papillomavirus), which is linked to cervical cancer in women, and which can be transmitted to a woman through sexual intercourse with an HPV-infected man. Several studies have found a correlation between circumcision and reduced risk of HPV infection. Castellsagué et al found in a sample of men that those who were uncircumcised were four times more likely to have HPV than circumcised men. Further, monogamous women with uncircumcised male partners were more likely to contract cervical cancer than monogamous women with circumcised male partners, even when the circumcised males had a history of six or more sexual partners. Similarly, Tobian et al found a reduced risk of HPV in circumcised men. Neither of these studies was randomized or controlled, however, meaning that while it’s possible to say that being circumcised is associated with reduced risk of contracting HPV, the studies don’t allow us to say that being circumcised causes reduced risk of contracting HPV.

However, there are also randomized studies that shed causal light on the relationship between circumcision and HPV. Auvert et al, Gray et al and Wawer et al all demonstrated that when HIV-negative, uncircumcised men were randomly assigned to be circumcised (with their consent) or remain uncircumcised, those who were circumcised were less likely to contract HPV. Gray noted that circumcision not only decreased the likelihood of contracting HPV, it increased the likelihood of clearing the infection from the body in the case of contraction. Wawer examined both the men and their female partners at a 24-month follow-up; the partners of the circumcised men were significantly less likely to have HPV infection at the time of the follow-up than the partners of the uncircumcised men. On the basis of the evidence, it’s reasonable to say that male circumcision reduces the risk of HPV in both males and in their female partners. Still, with an HPV vaccine available on the market, male circumcision isn’t the only way to achieve a reduced risk of HPV, so it can’t be recommended purely on that basis.

There are a number of studies that suggest male circumcision also reduces the risk of penile cancer, which is relatively rare in the U.S. Typically, cancerous changes begin on the glans (tip) of the penis or on the foreskin, and infection with HPV increases the risk of cancer development, according to the U.S. National Library of Medicine. Penile cancer is much more prevalent in countries with a low circumcision rate, including Africa and South America, explains the American Cancer Society. While the exact cause of penile cancer isn’t known, the accumulation of smegma under the foreskin of an uncircumcised man may increase the risk. Several studies have correlated increased risk of penile cancer with having an intact foreskin (see, for instance, Maden et al, Schoen et al, Tseng et al). The Tseng study suggests that the risk of penile cancer is most strongly associated with phimosis, which is a condition in which the foreskin doesn’t retract appropriately.

While the relationship between circumcision and reduced risk of penile cancer is well established, there isn’t enough data at the present time to recommend routine circumcision for the sole purpose of preventing penile cancer. In part, this is because penile cancer is so rare in the U.S. However, the cancer is most common in men age 60 and older, and circumcision rates have historically been quite high in the U.S., with 80-90% of men born in the 1940s through 1970s circumcised (Laumann et al, Xu et al). Recently, rates of circumcision in the U.S. have been falling somewhat, with the CDC estimating just over 50% of males circumcised in-hospital in 2010 (MMWR). It remains to be seen whether the penile cancer rate will increase in the coming decades concomitantly with these men reaching the prime age for development of penile cancer.


Science Bottom Line:* There is no scientific evidence that strongly supports circumcision in the United States for the sole purpose of preventing disease.**


**Obviously, this is not to say that there’s no reason to circumcise, nor is it to say that there’s no SCIENTIFIC reason to circumcise in areas with epidemic HIV, etc. The point here is that the argument FOR circumcision in the U.S. can’t be made on the basis of scientific evidence, and must instead be made on the basis of values and beliefs.


What do you think about what the science shows?



Auvert et al. Randomized, controlled intervention trial of male circumcision for reduction of HIV infection risk: the ANRS 1265 Trial. PLoS Med. 2005 Nov;2(11):e298. Epub 2005 Oct 25.

Auvert et al. Effect of Male Circumcision on the Prevalence of High-Risk Human Papillomavirus in Young Men: Results of a Randomized Controlled Trial Conducted in Orange Farm, South Africa. J Infect Dis. 2009 Jan 1;199(1):14-9.

Bailey et al. Male circumcision for HIV prevention in young men in Kisumu, Kenya: a randomised controlled trial. Lancet. 2007 Feb 24;369(9562):643-56.

Bongaarts et al. The relationship between male circumcision and HIV infection in African populations. AIDS. 1989 Jun;3(6):373-7.

Castellsagué et al. Male circumcision, penile human papillomavirus infection, and cervical cancer in female partners. N Engl J Med. 2002 Apr 11;346(15):1105-12.

CDC. Trends in in-hospital newborn male circumcision–United States, 1999-2010. MMWR Morb Mortal Wkly Rep. 2011 Sep 2;60(34):1167-8.

Gray et al. Male circumcision for HIV prevention in men in Rakai, Uganda: a randomised trial. Lancet. 2007 Feb 24;369(9562):657-66.

Gray et al. Male circumcision decreases acquisition and increases clearance of high-risk human papillomavirus in HIV-negative men: a randomized trial in Rakai, Uganda. J Infect Dis. 2010 May 15;201(10):1455-62.

Laumann et al. Circumcision in the United States. JAMA 1997;277(13):1052-7.

Maden et al. History of Circumcision, Medical Conditions, and Sexual Activity and Risk of Penile Cancer. J Natl Cancer Inst. 1993 Jan 6;85(1):19-24.

Moses et al. The association between lack of male circumcision and risk for HIV infection: a review of the epidemiological data. Sex Transm Dis. 1994 Jul-Aug;21(4):201-10.

Schoen et al. The Highly Protective Effect of Newborn Circumcision Against Invasive Penile Cancer. Pediatrics. 2000 Mar;105(3):E36.

Szabo et al. How does male circumcision protect against HIV infection? BMJ. 2000 Jun 10;320(7249):1592-4.

Tobian et al. Male Circumcision for the Prevention of HSV-2 and HPV Infections and Syphilis. N Engl J Med. 2009 Mar 26;360(13):1298-309.

Tseng et al. Risk Factors for Penile Cancer: Results of a Population-based Case–Control study in Los Angeles County (United States). Cancer Causes Control. 2001 Apr;12(3):267-77.

Wawer et al. Effect of circumcision of HIV-negative men on transmission of human papillomavirus to HIV-negative women: a randomised trial in Rakai, Uganda. Lancet. 2011 Jan 15;377(9761):209-18. Epub 2011 Jan 6.

Weiss et al. Male circumcision and risk of HIV infection in sub-Saharan Africa: a systematic review and meta-analysis. AIDS. 2000 Oct 20;14(15):2361-70.

Xu et al. Prevalence of circumcision and herpes simplex virus type 2 infection in men in the United States: the National Health and Nutrition Examination Survey (NHANES), 1999-2004. Sex Transm Dis. 2007 Jul;34(7):479-84.


Measuring Pain Sensation In Infants

(c) The Journal of Visualized Experiments, from Electrophysiological Measurements and Analysis of Nociception in Human Infants

For those who are interested in scientific articles but don’t like to read them, there’s a great resource online. It’s called JoVE, the Journal of Visualized Experiments, and the idea is that they do open source (meaning you don’t have to pay to access the information) video-format publication. That is to say, instead of reading a scientific paper, you can watch one!

Check out this video of research on how newborns sense pain during necessary medical procedures (no worries; the medical procedure they’re testing is the “heel stick” that they do on all newborns. The baby in the video grimaces a little, but doesn’t cry, and you don’t see any blood.) This research is important because it helps to inform our understanding of how newborns sense and respond to painful stimuli (as opposed to non-painful touch), which will eventually help medical practitioners fine-tune pain-control measures during necessary medical procedures.


What do you think of this research?



Fabrizi et al. Electrophysiological Measurements and Analysis of Nociception in Human Infants. J Vis Exp. 2011 Dec 20;(58). pii: 3118. doi: 10.3791/3118.

Have You Read Any Good Books Lately?

So, I’m going to start this post by making an embarrassing and uncomfortable admission. I’m not a TV-watcher; we don’t even have a TV in the house (that’s not the embarrassing part). However, because W essentially insists that I go to bed when she does and lie there nursing her on and off for the 2 or so hours until she falls deeply asleep, I find myself looking for things I can do lying on my side. There are lots of good things I could do. I could do research for my next SquintMom. I could get caught up on emails. I could read a book. But typically, I do none of these things. Typically (here’s the embarrassing part), I watch TV shows on my computer. I have Netflix, and furthermore, I recently learned that ABC puts episodes of their current shows online. I’ve discovered shows that I’m pretty sure everyone else has known about for years, but they’re new to me…like Grey’s Anatomy, and Private Practice, and The Bachelor (it really hurts to admit to watching that last one). And to be honest, my TV-watching is Driving. Me. Nuts. It wouldn’t bother me to watch it once in a while, but for several hours a night…I hate how much time I’m wasting.

So why do I do it?

Well, at the end of the day, I’ve spent 16-or-so odd hours dealing with a teething toddler, have answered dozens and dozens of student emails, have done research for my science writing and/or SquintMom, have read the latest releases from science blogs, medical journals, etc, etc, etc. I’m pretty tired. Note: this does not make me special. I’m a mom. We’re all tired at the end of the day. I’m just saying that, like so many of us, I’m no longer functioning on all cylinders by 8 pm. Anyway, I’m pretty tired, and I really don’t want to do any more work. I’d have no trouble, however, relaxing with an interesting book. Unfortunately, I’ve been burned a few too many times in the last year with poor choices of reading material (for anyone who’s curious, Steven Pinker’s Better Angels of Our Nature is nearly as long as War and Peace, but not nearly as satisfying. Oh, and it takes some real liberties with data interpretation. And it’s expensive. You really let me down on this one, Science Friday. Not that I’m resentful.) Anyway, the only thing I’m more resentful of than wasting time reading a book that ultimately sucks is buying an ultimately sucky book on Kindle (which then means I can’t trade it in). And for this reason, I find myself turning more than I should to instant TV. But I want to change.

To that end, I want to start a discussion about good books (on any topic you like…science, parenting, novel, or whatever). If you’ve read a great book lately — one that you think is worth reading — please post the author, title, a very brief description, and a sentence or two on why you thought it was so fabulous in the comments. Together, we can avoid sucky reading material.

Oh, and if you see a post someone else has made about a book and you agree, say so. If you totally disagree, say that too (we all have different opinions, and that’s fine)…but be nice about it, ok?

I’ll start us off. I highly, HIGHLY recommend Harvey Karp’s Happiest Toddler On The Block. It’s basically about how to communicate with a 1-4 year old, including how to promote good behavior and avoid tantrums. I thought it was well supported and was easy to read, without being condescending. It also included lots of insight into the developing toddler brain, and common sense communication tips.


What good book(s) have you read lately?


New Study Suggests Some Children May “Grow Out” Of Autism Diagnosis

There are several different diagnoses that all fall under the umbrella of autism spectrum disorders (ASD). While ASD isn’t fully understood, its prevalence is on the rise in the U.S. in recent years; as many as 1% of children may be diagnosed with some form of ASD (per the CDC), though researchers debate whether the rate increase is the result of increased prevalence or simply increased diagnosis. One factor that complicates ASD, both with regard to diagnosis and therapy, is that several other conditions often co-occur with autism. These include ADHD, learning disability, speech delay, and psychiatric conditions. It’s not clear whether ASD increases the likelihood of having a co-occurring condition, whether the opposite is true, or whether a third factor affects both the autism and the co-occurring condition. Regardless, awareness and understanding of co-occurring conditions is important to diagnosing and treating ASD effectively.

A new study published in Pediatrics (Close et al) suggests that co-occurring conditions may help health care practitioners to predict whether a child who is diagnosed with ASD will maintain the diagnosis as they age. It appears that some children “grow out” of their autism (or more accurately, grow out of the diagnosis of autism). According to the National Survey of Children’s Health, 40% of children who had at one time been diagnosed as autistic were no longer considered to have ASD at the time the parent responded to the health survey (Kogan et al). This can’t be taken to mean that 40% of children eventually outgrow autism; it’s possible (probable) that some of these children were originally misdiagnosed. It’s even possible that some of the children were incorrectly diagnosed as no longer having ASD. Still, the results of the Kogan study are interesting and are part of what prompted Close and colleagues to look at the differences between children whose diagnoses changed over time and those whose diagnoses did not change.

Looking at a group of 1366 children (who had been included in the National Survey of Children’s Health) between the ages of 3 and 17 years, Close and colleagues catalogued the co-occurring diagnoses that accompanied ASD, and determined whether co-occurring conditions were associated with changing ASD diagnosis status. They found that children aged 3-5 with current ASD were 11 times as likely to have a learning disability and 9 times as likely to have a developmental delay as children age 3-5 with a “past, but not current” (PBNC) autism diagnosis (meaning children who were once diagnosed with autism, but were no longer considered autistic at the time of the survey). These results are interesting, but it’s worth noting that because autism is quite difficult to diagnose in very young children, the sample size for these youngest individuals was very small. This suggests (albeit weakly) that children with learning disability and/or developmental delay are less likely to “outgrow” their autism diagnosis than children without these co-occurring conditions.

For children aged 6-11, Close and colleagues found that those with current ASD were nearly 4 times as likely to have had a past speech problem, 3.5 times as likely to have a current anxiety problem, and only one-fifth as likely to have had a past hearing problem as PBNC children. This suggests that children with speech and/or anxiety problems are less likely to “outgrow” their autism diagnosis than children without these conditions. Further, children with hearing problems and an autism diagnosis are five times MORE likely to “outgrow” their autism diagnosis than those without hearing problems. It may be that hearing difficulties can lead to autistic-appearing behavior (lack of communication, lack of engagement, lack of appropriate social interaction) without true underlying autism. Correcting the hearing problem could then reverse the “autistic” behavior and change the diagnosis.

Adolescents age 12-17 with current ASD were nearly 4 times more likely to have speech problems, 10 times more likely to have seizures and/or epilepsy, and only one-tenth as likely to have had a past hearing problem as those with PBNC ASD diagnoses. This suggests that children with speech problems and/or seizures/epilepsy are less likely to “outgrow” their autism diagnosis than children without these conditions. Further, it strengthens the association between resolving hearing problems and eventually “outgrowing” an autism diagnosis.

It’s important not to make too much of any one study, but this research certainly highlights some of the complexities associated with the diagnosis of autism. As researchers learn more about ASD, health care practitioners will be able to diagnose children more accurately. It remains to be seen whether children can ACTUALLY outgrow autism; it’s entirely possible that this study indicates that some children are incorrectly diagnosed with autism early in life, and that it becomes clear later on that they don’t actually have ASD. Time and more research will provide more robust answers.


Science Bottom Line:* It’s not possible to draw firm conclusions from this study, but it does shed some light on which children might eventually see some improvement in autistic behavior and/or no longer be considered autistic. These children include those who do not have speech, developmental, or psychological problems, and who DO have hearing problems (as long as those hearing problem are corrected).


Do you think some kids can “outgrow” autism?



Centers for Disease Control and Prevention (CDC). Prevalence of autism spectrum disorders – Autism and Developmental Disabilities Monitoring Network, United States, 2006. MMWR Surveill Summ. 2009 Dec 18;58(10):1-20.

Close et al. Co-occurring Conditions and Change in Diagnosis in Autism Spectrum Disorders. Pediatrics 2012 Jan 23;129(2): DOI: 10.1542/peds.2011-1717.

Kogan et al. Prevalence of parent-reported diagnosis of autism spectrum disorder among children in the US, 2007. Pediatrics. 2009 Nov;124(5):1395-403. Epub 2009 Oct 5.

Science Under The Influence (Of Motherhood)

This week, my husband and I are attending the Science Online conference in North Carolina. It’s a great opportunity to network and to learn about trends in science education (my day job). Because W is still nursing (and because, truth be told, there’s no part of me that’s ready to be away from her at night yet), she came with us. The plan was that my husband and I would take turns attending sessions and watching W, which is perfect…in theory. In reality, I am finding it hard to present myself as a SCIENTIST (…and mother) when I feel more like a MOTHER (…and scientist).

Granted, I wear both hats comfortably at home, and after some initial bumps, I’m now quite good at doing my science writing in 2-minute increments between verses of “Old McDonald.” Those times when I have to interact with others in a professional manner and for a period longer than W’s attention span, I’m at school and W is at home. I think about her, but I can compartmentalize. Having her here, in the conference building, makes things different. I’m fine during sessions while she’s with her father, but when we meet up in the halls between sessions (which is when the most interesting conversations take place and the most networking opportunities present themselves), my attention is definitely pulled in her direction. There’s no doubt I’m doing science under the influence of motherhood.

The complete realization of how “under the influence” I am has come upon me slowly, probably because I’m so used to multitasking (a word that once meant doing several things well at once, and now means desperately trying to accomplish a task one-handed while juggling a diaper bag, a baby, and a cup of Cheerios). Still, the evidence is mounting. In the break room today, I sought out another woman holding a baby, and the conversation was more along the lines of how old is she and is she walking yet than what are your preferred modalities in the undergraduate classroom? Furthermore, I may have sniffed W’s bottom (you know, poop check) mid-sentence while networking with another conference attendee. I say “may have” because the recollection is vague; I’m so used to this commonplace and now-unremarkable action that I didn’t fully register having (allegedly) performed it until afterward.

While being “under the influence” may have come to the forefront of my attention at the conference, I think the truth is that I’m always at least partly influenced by motherhood, because it has changed me so much in such pervasive ways. I’m still me, but compared to the old me, I’m running a little later, look a little less put-together, and need a little more coffee in the morning. In the end, though, I think that doing science under the influence of motherhood is a good thing. I think it makes me, if nothing else, better at my job. Motherhood has taught me the importance of repeating myself over and over again (and in different ways) to promote comprehension. It’s softened my approach and deepened my empathy. Most of all, it’s given me a concrete understanding of and appreciation for the fact that everyone is under the influence of something, be it motherhood, or a job, or simply life itself. We’re all (mothers or not) doing our own version of juggling Cheerios one-handed, and I really get that now.


What are you doing under the influence of motherhood, and how’s that working for you?



Night Nursing and Cavities

Extended nursing is loosely defined. In the United States, where only about a third of babies are exclusively breastfed until 3 months of age and fewer than a sixth are exclusively breastfed until 6 months of age (per the CDC), one could reasonably claim that breastfeeding beyond a year is “extended.” The American Academy of Pediatrics recommends breastfeeding for at least a year (with complementary foods after six months of age), while the World Health Organization recommends at least two years. It goes without saying, then, that a baby breastfed per the recommendations of these organizations will still be breastfeeding when teeth have come in. Some lucky parents have babies who start sleeping through the night at only a few months of age, while other mothers find themselves nursing once, twice, or even multiple times per night well beyond a baby’s first birthday. Certain sources, including La Leche League, suggest that breast milk isn’t cariogenic (cavity-causing), and even protects the teeth. Others, however, suggest that breast milk pooling in a baby’s mouth leads to early cavities, which can have significant ramifications for later oral health. What does the science say about night nursing and cavities?

One problem with finding a scientific answer to this question is that it’s difficult research to do. Case studies — reports of medical findings in a given individual — provide a limited amount of information, but aren’t a strong platform from which to derive inductive generalizations. This is because it’s difficult or impossible to establish causality in the case of an individual. As such, while there are reports in the literature of nursing caries associated with breastfeeding, these don’t support the conclusion that night nursing causes cavities.

Stronger evidence that night nursing either is or is not associated with cavity formation comes from population-level analysis. Dentist Brian Palmer, who studies ancient human skulls, concludes that there’s no connection between breastfeeding and night nursing on the grounds that 1) there isn’t evidence of cavities in ancient skulls of children, and 2) these children were probably breastfed for an extended period of time. Unfortunately, there are several problems with his theories. First, he has no proof that children were nursed at night (yes, they probably were…but he has no proof). Second, he does not take into account other aspects of diet that could significantly impact dental health. The conclusion he can reasonably draw from his research is that nursing didn’t cause cavities in children 500-1000 years ago, but it’s not possible to generalize this conclusion to today’s children because of significant dietary and lifestyle differences.

A few studies have looked at populations of modern children in an attempt to determine whether night nursing correlates with cavities. A study of children in Tehran found an association between bottle-feeding with milk at night and cavity development, but no association between breastfeeding at night and cavity development (Mohebbi et al). A study of Swedish children found that it was the intake of cariogenic food that was most associated with early cavity formation (Hallonsten et al). This finding weakens the findings of the Mohebbi study where they apply to children in the U.S. and other Western industrialized nations, because of significant dietary differences. Hallonsten also found, interestingly enough, that children who engaged in extended breastfeeding were more likely to consume cariogenic foods and have other cavity-promoting dietary habits than those who weaned at younger ages. A study of Dutch children found that frequency of breastfeeding and lack of fluoride were most associated with development of cavities (Weerheijm et al).

Note that the experimental design in the studies above is not one that allows determination of causality, only correlation. It’s possible that parents who breastfeed at night also engage in or encourage behavior x (whatever that might be), which predisposes their children to (or helps prevent) cavities.

Cavities are, of course, complicated things. There are a multitude of factors that make them more likely (bacterial colonization of the mouth, intake of cariogenic foods), as well as factors that make them less likely (dental hygiene, fluoride). Perhaps the most important question to answer in order to inform the night nursing/cavities association is whether human milk itself is cariogenic. La Leche League claims it is not, but this appears not to be supported by any particular scientific evidence, as they cite no direct research on the cariogenicity of human milk. Research evidence, in contrast, suggests that human milk is mildly cariogenic, though far less so than sugar water or soda (Bowen et al). The researchers ranked the cariogenicity of various tested substances as follows: table sugar, 1; soda (cola), 1.16 (the acid probably contributed to the increased cariogenicity as compared to table sugar); honey, 0.88; human breast milk, 0.29; cow’s milk, 0.01; distilled water, 0. The authors speculated that the increased cariogenicity of human milk as compared to cow’s milk may be due to the greater concentration of lactose in human milk, and (likely more important) the much lower concentration of dental health-supporting minerals (such as calcium and phosphate) in human milk. Based upon this research, it is unreasonable to suggest that human milk is non-cariogenic.


Science Bottom Line:* Human milk is approximately 1/3 as cariogenic as table sugar, and should be treated as a mildly cariogenic food. It’s probably reasonable to consider brushing a child’s teeth after a night nursing session, or at least wiping them off with gauze.


What do you do to help prevent nursing cavities in your night-nursing baby or toddler?



Bowen et al. Comparison of the cariogenicity of cola, honey, cow milk, human milk, and sucrose. Pediatrics. 2005 Oct;116(4):921-6.

Hallonsten et al. Dental caries and prolonged breast-feeding in 18-month-old Swedish children. Int J Paediatr Dent. 1995 Sep;5(3):149-55.

Mohebbi et al. Feeding habits as determinants of early childhood caries in a population where prolonged breastfeeding is the norm. Community Dent Oral Epidemiol. 2008 Aug;36(4):363-9.

Palmer; B. Breastfeeding and infant caries: No connection. ABM News and Views 2000; 6(4): 27,31.

Palmer B. The Influence of Breastfeeding on the Development of the Oral Cavity: A Commentary. J Hum Lact 1998;14:93-98.

Weerheihm et al. Prolonged demand breast-feeding and nursing caries. Caries Res. 1998;32(1):46-50.

More Non-Vax Nonsense

This post is the continuation of last week’s rebuttal of a recent article published on Mothering.com by Jennifer Margulis, entitled “Pregnancy and the Flu Shot.”

Margulis’ article continues with another anecdote (calling to mind a quote I love, often attributed to Roger Brinner, which states: “The plural of anecdote is not data”). This time, the story is about a woman who got the flu shot, and then proceeded to get influenza. No medical intervention or treatment — not ONE — is 100% effective. This is true of all diagnostic techniques, all medications, all vaccinations, all surgeries. In the case of the influenza vaccination, there’s the additional complication that the CDC must make an educated forecast (which Margulis refers to as a “guess,” as though it’s made by throwing dice rather than based upon sound scientific techniques) as to which strains of flu are likely to be most prevalent in a given year. The simple fact is that the CDC doesn’t — can’t — always forecast accurately, but most times they do. Even those years in which the flu shot doesn’t prevent the flu, it attenuates the severity. In the end, deciding to get the flu shot (like so many other decisions) involves doing a risk-to-benefit analysis. The question is, quite simply, do the benefits of getting the shot outweigh the risks.

In order to do a proper risk-to-benefit analysis, it’s important to know both the risks and the benefits. Strong scientific evidence supports that the flu shot is effective in pregnant women (see, for instance, Sumaya et al, Murray et al, Zaman et al). The Sumaya and Zaman studies further examined the efficacy of a maternal influenza vaccination in preventing influenza in newborns. The studies found that, up until about 6 months of age, the infants had influenza antibodies (indicating some degree of immunity, Sumaya et al) and reduced incidence of flu (Zaman et al). One complication in determining the effectiveness of the flu shot is that the vaccine doesn’t prevent non-influenza respiratory illness, and the only way to be sure that a case of illness is influenza is to test for it. As a result, some studies (for instance, Black et al) measure the impact of the influenza vaccine by looking at the rate of medical visits and/or hospitalizations for ANY respiratory illness in vaccinated versus unvaccinated individuals. Studies that don’t measure the effectiveness of the flu vaccine by looking at actual reduction in flu are less likely to show that the vaccine is effective, but to the credit of the researchers, many — including Black and colleagues — admit this shortcoming and weakness of the research. In the end, organizations and researchers who review the scientific literature feel that the shot is effective (that is to say, confers a definite benefit) in pregnant women (see, for instance, Tamma et al, Mak et al, and the recommendations of the CDC).

Then there is the matter of analyzing the risks associated with the influenza vaccine. A large number of scientific studies have shown that routine influenza vaccination is not associated with any severe side effects in individuals who are not allergic to its ingredients (egg, for instance). Pregnant women are not at any greater risk of side effects than non-pregnant individuals, and research suggests no increased risk of pregnancy-related complications (see, for instance, Tamma et al, Munoz et al, Mak et al).

One point in particular made by Margulis is that thimerosal, a mercury-containing preservative in flu vaccinations, causes harm (because mercury is a potent neurotoxin). However, no studies have found any evidence of harm from thimerosal exposure in flu vaccines (see, for instance, Tamma et al). Moreover, a study that looked specifically at neurodevelopment (researchers have wondered if thimerosal could be responsible for increasing the risk of autism and ADHD in infants) found no increased risk of these disorders in thimerosal-exposed infants (Verstraeten et al). Margulis makes the point that the thimerosal studies are weak because they do not include a control group. However, here, she demonstrates her ignorance of the scientific method and of study design. Human experimentation, particularly when it comes to determining whether a substance is dangerous to infants, isn’t ethical. Thimerosal studies aren’t done by injecting some babies (or pregnant women) with thimerosal-containing vaccine and others (the control group that Margulis is looking for) with a vaccine that doesn’t contain the thimerosal. Instead, studies are done by retrospectively (afterward) comparing the health of infants who were exposed to thimerosal to that which is considered normal or expected (“normal” is the control group, even though there isn’t an explicit control in the experimental design). This sort of design is common in medicine and allows ethical determination of toxicity; there is no flaw here.

In the end, while there is good reason to continue to study the effects of thimerosal in vaccines, evidence suggests that in small and occasional doses, it’s not likely to be a problem. Wait, LIKELY? You mean I’m taking a RISK? Yes, you are. You take a risk every time you eat food (it could contain pesticides, mercury, Listeria…), breathe outdoor air in the city (you are breathing pollutants, including benzene and formaldehyde, that cause illness and cancer), enjoy the sunshine (it’s breaking your DNA and possibly causing cancer), or take a tablet of over-the-counter medication (you could be allergic, or could have an adverse reaction). Life is all about risks. There is nothing you do that is without risk. The idea is not to avoid all risk — because that would be impossible — it’s to take MEASURED and APPROPRIATE risks that are WORTH THE BENEFITS. I doubt anyone would argue (at least, not successfully) that you shouldn’t eat, breathe, walk in the sun, or take over-the-counter medication because of the risk. The benefits in each of those cases (and in the case of influenza vaccination) outweigh the risks.

The problem here, as in so much of the anti-vax rhetoric and other harmful nonsense propagated on the Internet and elsewhere in the media, is that Margulis simply isn’t qualified to make judgments on scientific issues. She’s a journalist, according to her byline, and is without a scientific background (possible intro courses in college notwithstanding). As Alice at Science of Mom so beautifully put it in this wonderful guest-post at The Mother Geek, “It would be naïve for me to think that I could understand the vaccine field better than the committees of scientists and doctors who have made this their life’s work.” And that’s coming from a PhD in science! Heck, I’m on that same page; there’s no WAY I would have the hubris to think that I could know better than vaccine scientists whether a vaccine is a good idea, is safe, or confers benefits. That’s why I leave making those decisions to the vaccine scientists. My question, then, is this: if PhDs in science feel it’s important to turn to the TRUE EXPERTS to make vaccination decisions, why doesn’t a journalist? Why don’t the celebrities? Why don’t the moms with great instincts, but no scientific background, who do their vaccination “research” on sketchy Internet sites and by listening to Jenny “Formerly Host of Singled Out” McCarthy instead of by turning to medical experts who’ve made investigating vaccines their life’s work?


Science Bottom Line:* The influenza vaccine is safe and effective in all trimesters of pregnancy, and research supports getting it.


Do you trust the experts? Why or why not?



Black et al. Effectiveness of influenza vaccine during pregnancy in preventing hospitalizations and outpatient visits for respiratory illness in pregnant women and their infants. Am J Perinatol. 2004 Aug;21(6):333-9.

CDC. Prevention and Control of Influenza, Recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Morb Mortal Wkly Rep. 2011 Aug 26;60(33):1128-32.

Mak et al. Influenza vaccination in pregnancy: current evidence and selected national policies. Lancet Infect Dis. 2008 Jan;8(1):44-52.

Munoz et al. Safety of influenza vaccination during pregnancy. Am J Obstet Gynecol. 2005 Apr;192(4):1098-106.

Murray et al. Antibody response to monovalent A/New Jersey/8/76 influenza vaccine in pregnant women. J Clin Microbiol. 1979 Aug;10(2):184-7.

Sumaya et al. Immunization of pregnant women with influenza A/New Jersey/76 virus vaccine: reactogenicity and immunogenicity in mother and infant. J Infect Dis. 1979 Aug;140(2):141-6.

Tamma et al. Safety of influenza vaccination during pregnancy. Am J Obstet Gynecol. 2009 Dec;201(6):547-52. Epub 2009 Oct 21.

Verstraeten et al. Safety of thimerosal-containing vaccines: a two-phased study of computerized health maintenance organization databases. Pediatrics. 2003 Nov;112(5):1039-48.

Zaman et al. Effectiveness of Maternal Influenza Immunization in Mothers and Infants. N Engl J Med. 2008 Oct 9;359(15):1555-64. Epub 2008 Sep 17.

Previous Older Entries