The latest problematic study, rooted in spurious correlation and statistical manipulation, has sparked a round of headlines that connect couples’ caffeine consumption to miscarriages. It’s an incredibly painful topic, and reporting on it deserves more scientific scrutiny than it’s gotten. Here’s where the problems lie:
1. Unreliable Methods
You may be tired of hearing it by now, but this study is yet another classic example of correlation does not mean causation. The results were determined from daily journals that recorded lifestyle behaviors, which is not an accurate way to assess what people are eating. Incredibly, the conclusion was based on research that “did not have information on decaffeinated or caffeine free beverages or other aspects of lifestyle such as physical activity, diet, recreational drugs, or sleep habits. As such, we cannot rule out the potential for residual confounding.” Residual confounding means that any of these other factors could be influencing the result instead of caffeine consumption.
Remember, ‘X linked to Y’ is a red flag phrase for scientific studies. As we’ve covered before, correlation is not causation- you can ‘link’ swimming pool drownings to Nicolas Cage movies. You can link the divorce rate in Maine to per capita consumption of margarine. Sometimes we can laugh off spurious correlations, but in this round of media coverage, where it’s being used to generate fear and shame for couples who suffer pregnancy complications, we’re not laughing.
2. Questionable Results
The study also looked at a group that happened to have a significantly higher reported that rate of miscarriage than their peers in the general population, which the study authors noted was a flaw and can make results less reliable.
One major emphasis of news headlines is that the study’s findings related to caffeine were seen in both males and females. The lead researcher, Dr. Buck Louis, specified that “our findings also indicate that the male partner matters, too. Male preconception consumption of caffeinated beverages was just as strongly associated with pregnancy loss as females.” The fact that miscarriages are tied to both male and female pre-pregnancy behaviors means that it’s even more likely there may be other confounding variables resulting pregnancy loss.
3. Unsupported application of study findings
There is a big gap between what the (very large) body of existing research on healthy pregnancy says and what this study found. For example, although the researchers evaluated the impact of cigarette smoking on pregnancy loss, the paper’s results did not confirm the large body of evidence that demonstrates cigarette smoking and consumption of alcohol increases risk for miscarriage. When a study’s findings don’t align with a large body of existing research, it’s a major methodology red flag.
Not only is there no body of science supporting assertions about male caffeine consumption and miscarriages, decades of research evaluated by food and health authorities like the FDA and EFSA (along with March of Dimes, the American College of Obstetricians and Gynecologists (ACOG), and many others) confirm that moderate intake of caffeine is safe for pregnant women- it doesn’t impact fertility, risk of miscarriage, or healthy breastfeeding. (Moderate intake for pregnant women is usually considered to be 200 mg, about 2.5 energy drinks or cups of coffee worth). It’s not ok to use fear-mongering approaches on couples who may be already experiencing a ton of anxiety and worry. Especially given the quality and applicability of this study’s methodology, the alarm bell is not adequately supported or helpful for those couples.
This blog includes contributions from Liz Caselli-Mechael.