The mainstream media is buzzing with the news that C-sections purportedly increase the risk of obesity in offspring by 15%.
Here’s the LA Times:
[pullquote align=”right” cite=”” link=”” color=”” class=”” size=””]There’s just one problem. An increase of 15% is essentially nothing.[/pullquote]
Your very first moments of life can influence your risk of obesity for years, a new study shows.
Babies delivered via cesarean section were 15% more likely to be obese as kids, teens and young adults than were babies who made the trip through the birth canal, according to the report in JAMA Pediatrics…
Nutritional epidemiologist Changzheng Yuan of the Harvard T.H. Chan School of Public Health and her study coauthors had good reason to suspect that a C-section put a baby on the path to obesity. Two recent reports that pooled data from other studies found that cesarean birth was associated with a 22% increased risk of obesity.
There’s just one problem. An increase of 15% is essentially nothing.
The study is Association Between Cesarean Birth and Risk of Obesity in Offspring in Childhood, Adolescence, and Early Adulthood, and appears to be methodologically excellent, correcting appropriately for confounding variables like maternal weight.
The problem is the importance they place on a very tiny difference that may be no difference at all.
Consider a different relationship, smoking and lung cancer. Smoking increases the risk of lung cancer by more than 2000%.
How about homebirths? Homebirth advocates are fond of claiming that the increased risk of neonatal death at homebirth is trivial, but CDC statistics indicate that it is in the range of 200% and the most definitive statistics, from Oregon, show that homebirth increases the risk of perinatal death by 800%.
Several years ago Gary Taubes wrote a piece for the New York Times Magazine explaining how lay people can judge the results of epidemiological studies, Do We Really Know What Makes Us Healthy? He was writing in the wake of new revelations about estrogen replacement therapy that showed that the benefits of estrogen had been vastly overstated. He pointed out that the estrogen fiasco was a foreseeable result of using weak epidemiological data to make sweeping pronouncements. It was a cautionary tale similar to many cautionary tales in epidemiology, particularly those concerning lifestyle behaviors.
…[T]he perception of what epidemiologic research can legitimately accomplish — by the public, the press and perhaps by many epidemiologists themselves — may have run far ahead of the reality. The case of hormone-replacement therapy for post-menopausal women is just one of the cautionary tales in the annals of epidemiology. It’s a particularly glaring example of the difficulties of trying to establish reliable knowledge in any scientific field with research tools that themselves may be unreliable.
Taubes offered lay people rules of thumb for evaluating claims based on epidemiological data.
…[H]ow should we respond the next time we’re asked to believe that an association implies a cause and effect, that some medication or some facet of our diet or lifestyle is either killing us or making us healthier? We can fall back on several guiding principles, these skeptical epidemiologists say. One is to assume that the first report of an association is incorrect or meaningless, no matter how big that association might be… Only after that report is made public will the authors have the opportunity to be informed by their peers of all the many ways that they might have simply misinterpreted what they saw…
If the association appears consistently in study after study, population after population, but is small — in the range of tens of percent — then doubt it. For the individual, such small associations, even if real, will have only minor effects or no effect on overall health or risk of disease. They can have enormous public-health implications, but they’re also small enough to be treated with suspicion until a clinical trial demonstrates their validity (my emphasis).
The authors of the C-section paper acknowledge that similar studies have found no difference in obesity rates, or small differences, and many studies that claimed to find differences in obesity rates did not correct for confounding variables:
Despite inconsistent findings from individual studies, two recent meta-analyses reported a 22% increased odds of adult obesity associated with cesarean delivery. However, many of the studies included in these meta-analyses—particularly in the meta-analyses for adult obesity—failed to account for important potential confounders, most importantly for maternal prepregnancy BMI.
Let’s apply Taubes’ principles to the claim that C-section increases the risk of obesity in offspring by 15%.
1. Assume that the first report of an association is incorrect or meaningless: This is not the first report of an association.
2. If the association appears consistently in study after study, population after population: This finding does not appear consistently. A number of studies have found no association between C-section and obesity
3. If the association appears … small — in the range of tens of percent — then doubt it: An increase of only 15% is very small, essentially no difference at all.
C-sections increase the risk of obesity in offspring by 15%?
May so, maybe no, but either way the difference is so small that it doesn’t tell us whether C-sections have any impact on obesity at all.
Addendum: Here’s the relevant chart from the paper.
