Health and nutrition news outlets are at it again: another day, another misleading and alarmist study about the potential dangers of low-carb diets during pregnancy. This is nothing new. We reported on a similar situation a while back, and not much has changed. It seems that as word spreads about the efficacy of low-carb diets for myriad health conditions as well as for fat loss, the nutritional “old guard” and the powers that be are circling their wagons ever tighter, lest the public begin to understand that no, fat really, really isn’t bad for them, and maybe basing their diets on starchy carbohydrates—as specifically recommended by decades of government nutrition advice—wasn’t the best way to go.
What sensationalist headlines are we talking about here?
- Low carb diets like Atkins, Paleo or Keto linked to risk of birth defects including spina bifida, study claims (Daily Mail, Jan 25, 2018)
- Low carb diets like Paleo and Keto increase risk of birth defects (New Zealand Herald, Jan 26, 2018)
- New UNC-Chapel Hill study links low carbohydrate intake to increased risk of birth defects (University of North Carolina at Chapel Hill, Jan 25, 2018)
But is this true?
Is it possible that a diet that’s highly effective for gout, type 2 diabetes, metabolic syndrome, dyslipidemia, PCOS, GERD, and more, could be harmful for a developing baby? Let’s take a closer look at the study.
Owing to decreased consumption of carbohydrate-based processed foods that are fortified with folic acid, researchers speculated that women who follow low carb diets might have lower folate status compared to women who do not restrict carbohydrates. And considering the relationship between inadequate folate and increased risk for neural tube defects (NTDs) and other birth defects, the researchers sought to determine if women who follow low carb diets are at greater risk for producing offspring with NTDs than are women who don’t restrict carbs.
According to the abstract:
“To assess the association between carbohydrate intake and NTDs, we analyzed data from the National Birth Defects Prevention Study from 1,740 mothers of infants, stillbirths, and terminations with anencephaly or spina bifida (cases), and 9,545 mothers of live born infants without a birth defect (controls) conceived between 1998 and 2011. Carbohydrate and folic acid intake before conception were estimated from food frequency questionnaire responses.”
Houston, we have a problem. Food frequency questionnaires (FFQs) are extremely shaky evidence on which to come to solid conclusions. The authors acknowledge as much:
“As nutrient intakes were estimated based on self-reported frequency of specific food items using an abbreviated FFQ, we further caution against interpreting the estimated nutrient values in this study as absolute values; we cannot assume, for example, that all women classified in our study as having restricted carbohydrate intake actually consumed fewer than 95 g/day.”
So we can’t be certain these women even were restricting carbohydrates.
Another glaring issue:
“An important limitation of our case-control study is that we lacked biospecimens collected during pregnancy. Therefore, we cannot directly assess the association between carbohydrate intake and RBC folate concentrations, nor can we assess the degree to which the association between NTDs and restricted carbohydrate intake is mediated by low folate status versus other pathways.”
That’s right: maternal folate levels were not measured. Folate intake was estimated based on the food frequency questionnaire.
It’s true that women who consume grain products fortified with folic acid may be getting more synthetic folic acid in their diets than women who follow low carb diets and consume little to no grain. But low carb diets are hardly lacking in folate. Egg yolks, a fairly common staple among low carb dieters, are a good source of folate, as are the leafy greens that typically make up a significant portion of low carb dieters’ vegetable intake. Spinach, broccoli, romaine lettuce, and brussels sprouts are just a few examples.
The FFQ used, which contained only 58 items, is not provided in the full text study nor in the supplemental material, so it’s impossible to know what foods were included, and whether or not they encompassed a large selection of folate-rich foods that are low in carbohydrate.
Since folate levels were not directly measured, any relationship between a possible increased incidence of NTDs and other birth defects in offspring of women following low carb diets compared to standard (high carb) diets cannot be attributed to folate status. (To their credit, the researchers acknowledge this and speculate there might be an as-yet unidentified aspect of low carb diets that contribute to increased risk for NTD.)
And the hits just keep on comin’:
“During the maternal interview, usual consumption of foods in the year before pregnancy was ascertained using a modified 58-item food frequency questionnaire with additional questions about cereal consumption within 3 months of conception.”
The FFQ did not pertain to the diet consumed during pregnancy. It pertained to the year before pregnancy. The researchers also acknowledge this shortcoming: “Another limitation is that the FFQ captured usual diet in the year before conception, and the extent to which diets may have differed around the time of conception or during the first few weeks of pregnancy is unknown.” At best—at best—the FFQ used in this study is a very loose guess as to which foods, and in which amounts, the women consumed during the stage of pregnancy during which the neural tube is formed.
To be fair, the researchers pointed out that neural tube development takes place before many women are even aware they’re pregnant, so it’s unlikely that they would have made dramatic changes to their habitual diets during that time. Even ceding this point, this is still incredibly unreliable evidence on which to come to conclusions about low carb diets being harmful for fetal development.
Additionally, based on the data provided, there were a total of 1559 cases of NTD: 93 among the low carb group and 1466 among the non-carb restricted diet, so 94% of the occurrences of NTD were in the group not following a low carb diet.Granted, even though low carb diets have gained in popularity, relative to the population as a whole, the number of people following low carb diets is still very small. So it’s logical that there would be more cases of NTD among offspring of women not eating low carb, simply because there are so many more women not eating low carb. Nevertheless, if low carb diets were a serious risk factor for NTDs, you would expect low carb moms to account for more than just 6% of the NTDs.
Funny how what was emphasized in the news was “low carb diets cause birth defects,” rather than a much more cautious and measured interpretation of weak and highly suspect conclusions based on unreliable data. Sadly, we will likely only see more of this in the age of clickbait headlines and media outlets competing for pageviews.
If you’d like to see more of the problematic aspects of this study, UK-based nutrition researcher Zoe Harcombe wrote a letter to the study authors and to the journal editor requesting clarification on a number of points.