Last week, the scary health story du jour plastering the airwaves concerned a research study that found a link between high caffeine consumption and miscarriage. While some news outlets were relatively objective in their reporting with headlines like "Best evidence yet of caffeine-miscarriage link" and "caffeine is linked to miscarriage risk", others distort the truth a bit and proclaim "Too much caffeine increases the risk of miscarriage" and "Caffeine consumption raises miscarriage rate", ignoring the fact that correlation and cause are not the same thing. In other words, if women who drink more caffeine end up having more miscarriages, it doesn’t prove that the caffeine was the cause of the increase in miscarriage rate. Many of my patients asked for my comment on this and I didn’t know how to answer them because I didn’t have the actual study available to read. I searched the American Journal of OB/Gyn website and still couldn’t find the article even though I have login access. Fortunately, I was able to ask the editor of the journal who is a friend and respected mentor of mine. He was kind enough to send me a preview copy last week, which I just got done reading.

This was a two-year study which recruited patients from Kaiser Permanent in the San Francisco area. Every patient in their system who had a positive pregnancy was invited to join the study. Starting out with the 2729 women who were eligible to be in the study, in the end only 1063 (39%) actually participated. The rest were too busy, too far along in the pregnancy or refused for other reasons. While this study has many strong points, this vulnerability to selection bias is one of the weak points. The study consisted of detailed interviews about caffeine intake, demographic information, alcohol and smoking habits, Jacuzzi use and exposure to magnetic fields as well as data on nausea and vomiting.

Here are some of the interesting findings regarding the population studied:

• Of the 1063 women in the study, 172 (16%) of them suffered a miscarriage. This is in agreement with the common belief that about 1 out of 6 pregnancies naturally end in miscarriage.
• General breakdown of caffeine use. Of the 1063 women in the study, 25% consumed no caffeine during pregnancy and 60% were light caffeine users taking in less than 200mg daily. High caffeine users, defined as having intake greater than 200mg daily, comprised 15%. By the way, 200 mg was described as being equal to about two cups of coffee or five caffeinated sodas.
• The oldest group of women (over 35) were more likely to be high caffeine users than the youngest group (under 25). In the younger group, the no-caffeine group outnumber the high-caffeine group 4 to 1. While in the older group, there were just as many in the high-caffeine group as the no-caffeine group.
• Racial differences in caffeine intake. White women were proportionately more likely to be high-caffeine users than blacks, Hispanics or Asians.
• Previous miscarriage history. Women who never had a miscarriage before were twice as likely to be no-caffeine than high-caffeine. On the contrary, women who had 1 or more miscarriages in the past were a bit more likely to be high-caffeine than no-caffeine.
• Morning sickness and caffeine. Of all the women, 40% had vomiting during the pregnancy while the other 60% had no vomiting. In the yes-vomiting group, the no-caffeine users outnumbered high-caffeine users 2:1. In the no-vomiting group, there were 155 no-caffeine users to 116 high-caffeine users.
• Smoking. Smokers were 4.5 times as likely to be high-caffeine as no-caffeine, while nonsmokers were twice as likely to be no-caffeine as high-caffeine. This strongly confirms the previous idea that smokers drink caffeine more than non-smokers.

The bottom line is that the high-caffeine group were not identical to the no-caffeine group. They were associated with many characteristics that could be miscarriage risk factors. They tended to be older, have a history of previous miscarriage, and have habits of smoking, alcohol consumption and Jacuzzi use, so it would not be surprising if the high-caffeine wound up having a higher miscarriage rate.

Here are the results with respect to miscarriage:

• First of all, BEFORE accounting for the differences in risk factors, the miscarriage rates were different in the three groups. It was 12% in the no-caffeine group vs 15% in the low-caffeine group vs 25% in the high-caffeine group. This, in itself, is not really that exciting, precisely because of the confounding factors. It’s clear that the high-caffeine users had a lot more risk factors for miscarriage independent of their caffeine intake.
• I have to admit, I don’t know enough about the math of the statistics used for this, but the authors stated that they recalculated the relative risk of miscarriage using formulas that adjusted for maternal age, race, education, family income, miscarriage history, vomiting, smoking, alcohol and Jacuzzi use and exposure to magnetic fields. Even after adjusting for all this, they found that compared to the no-caffeine group, the low-caffeine group had a higher ratio of miscarriage that was likely somewhere between .93 to 2.2 times higher. The high-caffeine group had a miscarriage ratio that was likely somewhere between 1.3 to 3.7 times higher.
• It would have been nice to see if high-caffeine users could show a lower miscarriage risk if they actively cut down on their caffeine use once they found out they were pregnant, but the authors stated that there weren’t enough people in this group to offer any statistically significant results.
• The impact of caffeine on miscarriage risk was less in patients who had other risk factors. This makes sense because those are patients who are more likely to miscarry anyway, for other reasons. So being caffeine-free will not help them. Bearing in mind that over 50% of miscarriages are due to something wrong with the fetus, so no amount of being caffeine-free is going to save those pregnancies.

I think I’ll wrap things up a little differently by saying what I advise you NOT to conclude from all these and other scary news stories:

• MYTH: Pregnancies are so fragile that a cup of coffee will kill your baby!
• MYTH: Miscarriages usually happen because of something you did or ate, like lifting too heavy a weight, or getting into that argument with your boss.
• MYTH: Miscarriages are rare, so if you have one, there must be something that you did wrong.
• MYTH: If you have had four or more miscarriages, it is pretty much impossible that you’ll ever have a healthy baby.
• MYTH: If you have a miscarriage during your first pregnancy, you should go see an RE and get a full workup.
• MYTH: You can prevent a miscarriage by going on complete bedrest.

All of the above are common false beliefs that I’ve heard from patients.

My take on all this is as follows:

• Reevaluate your life before you get pregnant. In general it is always better to be healthier. How can you modify your diet, your habits, your stress level, your weight, your exercise routine and your smoking and alcohol habits?
• Once you are pregnant, do not exceed two cups of coffee per day. Ideally, stopping coffee completely would be best, but if it’s too difficult to do so, then at least cut down to one cup daily or less.
• Remember it is the job of journalists to sell newspapers and gain TV viewers and they do this best by inciting emotions, including fear. Keep this in mind if you are being pulled into a sense of panic by what you see in the media.
• Do reality testing. Is caffeine really that deadly to pregnancies? Out of the millions of healthy babies being born, how many of their mothers do you think are heavy coffee drinkers? I don’t know the exact answer, but I’m pretty sure it’s not zero.