It’s not nice to scare mothers and babies

This week, we saw one of the most inexcusable examples of the misuse of “science” in the war against obesity. Not only were headlines used to scare young women that being fat could mean their babies might be born with hideous birth defects, but both the media and the press release from the journal office of the American Medical Association, failed to fully and accurately report what this study actually found.

Headlines read: “Maternal obesity heightens risk of birth defects” (Washington Post) and “Maternal obesity prior to pregnancy associated with birth defects” (JAMA Press Release). We were told a study found: “Women who were obese before they became pregnant had a higher risk of having babies with certain birth defects, including missing limbs, malformed hearts and underdeveloped spinal cords.”

More accurate reporting (and what women should have heard) would have been: “Maternal "obesity" was not found to be associated with higher risks for birth defects.”

This study led to countless stories of the need for fat women to lose weight before planning a baby. When a study’s findings appear to be used to support a popular belief or government program, we’re most likely to find another epidemiological study because those can most easily conclude whatever researchers set out to find. Sure enough, the study behind the news was not only another observational study, looking for correlations culled from a large database, but it was a “case-control study.”

In its “Evidence-Based Review System for the Scientific Evaluation of Health Claims,” released last month, the FDA said “observational studies lack the controlled setting of intervention studies....[and] cannot determine whether an observed relationship represents a relationship in which the substance caused a ... disease risk or is a coincidence.” The FDA scientists also specifically stated that they consider case-control studies less reliable than other population studies, one reason being that they so poorly control for confounding factors and are also retrospective (looking back in time), meaning selective bias is much more common (researchers look only for things they choose to).

This study, led by Kim Waller, Ph.D., of the School of Public Health at the University of Texas, Houston, was funded by the Centers for Disease Control and Prevention (CDC) cooperatively with the Texas Department of State Health Services Center for Birth Defects Research and Prevention. It opened with a statement about how the “dramatic increase in the prevalence of overweight and obese women of childbearing age is of great public health concern” and referenced supportive data from the CDC National Center for Health Statistics’ latest study published in a 2006 issue of the Journal of the American Medical Association.

This is our first clue something other than good science is at work here. Incredibly, neither the peer-reviewers or anyone reporting on this study bothered to look at the reference, “Prevalence of overweight and obesity in the United States, 1999-2004.” If they had, they would know that the CDC NCHS had reported there had been “no increase in obesity among women” from 1999 through 2004 — in fact, the prevalence dropped slightly from 33.4% to 33.2%. This is another example of the disconnect between the actual evidence (even the CDC’s own findings) and the CDC leadership's massive anti-obesity agenda. But letting it be widely known that women are not becoming fatter, and haven’t for many years, clearly wouldn’t help support the government's war on obesity.

The next point in the opening of this week’s study stated:

A strong association has been demonstrated between a woman’s prepregnancy BMI and risk for offspring with certain birth defects, particularly anencephaly and spina bifida. The potential relation between obesity and other birth defects remains less certain.

The first reference was another case-control study led by Dr. Waller in 1994: “Are Obese Women at Higher Risk for Producing Malformed Offspring?” It found only a higher odds ratio between obesity and neural tube defects, especially spina bifida. This illustrates not only how easy these types of studies can be used to reach whatever conclusion one sets out to find, but the very real dangers of wrongly concluding causation from an association.

This is important, so let’s take a minute to look more closely. Neural tube defects is the only correlation with obesity that even these researchers were able to find reported in the medical literature. It’s been known since the 1960s that neural tube defects (including spina bifida, where the spinal cord is exposed to varying degrees, and anencephaly, where the brain doesn’t develop) appear to have a significant genetic component but that folic acid deficiency may account for about half of the cases. These birth defects develop very early in pregnancy — 18 to 30 days after conception — usually before a woman even knows she’s pregnant, which is why folic acid supplementation and folic acid fortification of foods have been promoted since the 1990s. According to the CDC in a 2004 Morbidity and Mortality Weekly Report, the first year after fortification of white flours and cereal grains went into effect in 1998, the prevalence of spina bifida dropped 31%.

Which women are most likely to be avoiding white breads, baked goods and pastas? Women who are concerned about their weight and dieting, of course. So, they’re also more likely to be heavier women. But the correlation between this birth defect and babies born to fat women isn’t due to the women eating too much folic acid-fortified or folic acid-rich foods, but because they’re trying to avoid them, believing them to be fattening, and eating less overall, trying to cut calories. Misguided health information blaming women's fat will naturally lead more women to diet, which will put their babies at greater risk. It’s not fat or eating too much, but dieting behaviors that increase the risk for neural tube defects. In fact, that is exactly what researchers in California found among 538 cases of neural tube defects. Their study, published in a 2003 issue of the American Journal of Epidemiology, found that restricted food intake, including dieting to lose weight, during the first trimester doubled the risks for spina bifida and fasting diets increased the risk six-fold.

So many health problems resulting from dieting are wrongly attributed to obesity, thanks largely to epidemiological studies looking for associations. That’s why these studies are so popular with those eager to pile on more fears about obesity. And that’s one reason why they can be so harmful.

Coming back to this week’s study, the researchers used data from a database from the Centers for Birth Defects Research and Prevention (CBDRP) project. That’s a joint effort between the CDC’s National Center on Birth Defects and Developmental Disabilities and nine birth defects surveillance registries across the United States (Arkansas, California, Iowa, Massachusetts, North Carolina, New Jersey, New York, Texas and Utah). Since 1998, it has been using computer-assisted telephone interviews to interview 17,676 mothers of babies born with birth defects (as of July 2006), as well as 6,587 control mothers. Births not included are those of babies adopted or in foster care, of deceased mothers, stillborn, or of mothers not residents in the state of the CBDRP registry at the time of delivery.

The mothers were interviewed 6 weeks to 2 years after the delivery of their babies and asked their prepregnancy measurements, ethnicity, education, age, smoking status and if they’d ever taken vitamins during the month before conception. The researchers had no information on the diet or nutritional status of the mothers, any recent weight changes, or any actual clinical health information at all. Going in, the researchers acknowledged that they might see more birth defects among heavier women since they didn’t have data on all of the cases of birth defects because three states didn’t include terminated pregnancies and also obese women may not have terminated their pregnancies since problems “may be less likely to be diagnosed by prenatal ultrasound, as obesity interferes with the quality of the technique.”

These researchers used about 57% of the CBDRP’s total birth defect registry, choosing to look at 16 birth defect categories. To cut to the chase, after a computer logistic regression to look for associations between the mothers’ BMI and birth defects, adjusting for age, education, ethnicity, etc., they were unable to find a single odds ratio that was beyond random chance or statistical error among any of the birth defects and weights. The most significant correlation, also seen by previous researchers, was an 80% lower risk for gastroschesis, a defect of the abdominal wall that lets some of the intestine protrude through, among the most obese mothers. In three categories of birth defects, this study differed from other epidemiological studies, but all of the differences were with tiny odds ratios below tenable numbers and “may be explained by chance,” according to the authers. The only higher risk with an odds ratio even close to 100% was spina bifida among the fat women — and we’ve already examined that.

The researchers stated that “the reasons for an association between maternal obesity and a spectrum of birth defects are unknown.” But, of course, since they were unable to find a tenable association, the lack of any medical rationale isn’t at all surprising.

The bottom line is that despite continued efforts to find ways to blame mothers for their babies’ birth defects, most birth defects are being found to have genetic pathways. Yet, tragically, parents so often blame themselves or fear that they did something to cause their baby harm. Good information about birth defects won’t be found in a computer, with researchers continuing decade after decade to look for the same incomplete correlations among questionnaire data, but in conducting actual clinical research.

To try and scare fat women by threatening them with malformed babies to incentivize them to lose weight is not out of concern for their health or for their babies. It’s to advance an insupportable government agenda. And that’s just plain wrong. Young woman and babies deserve better from healthcare professionals.

© 2007 Sandy Szwarc