Jun 5, 2018
Eggs Won’t Stop Heart Attacks And Fish Won’t Get You Pregnant
Why studies on individual foods are so often misleading
There’s a cultural phenomenon that we have been buying into for years. Every second day there’s a story in the news: “Broccoli cures cancer! Blueberries prevent aging! Eat these 5 foods to never worry about constipation again!”
The thing is, healthy diets are rarely about individual foods. Yes, there are chemicals found in, say, blueberries that might be useful for human health, but that doesn’t mean that eating more blueberries is necessarily going to do much for you. They might be found in small amounts, they might be poorly bioavailable, or the research might’ve been done only in mice and not replicated in actual people.
Or, often, all of the above.
But we do love a story about how eating more of a single food can save us. It would be so much easier to avoid heart disease if all it took was a few flavanols — drink some tea, eat some chocolate, and you’re suddenly healthy!
Sadly, it’s never that easy.
Eggs and Fish
There are a couple of brilliant examples of this in recent news that paint a perfect example of how we love these miracle food stories: eggs and fish. A few weeks ago, a huge study came out that found that people who ate more eggs had less heart disease, in particular nasty things like strokes and heart attacks. There was also a smaller, but no less well-publicized, study that found that couples who ate more fish were more likely to conceive using IVF treatments.
Cue the braying media chorus: “Eggs prevent heart disease! Fish will get you pregnant! EAT THESE FOODS TO HAVE A HAPPY LIFE!”
Now, both of these studies were pretty interesting. The egg study in particular was a simply amazing piece of research.
But neither of them said quite what you might’ve heard. Eggs don’t necessarily prevent heart disease, and fish almost certainly won’t get you pregnant.
What went wrong?
Studious Science
Both of these studies were observational research. What this means is that identifying what the cause of the outcome — pregnancy or heart disease — is very difficult.
The egg study was impressive. They took 500,000 — yes, you read that right: half a million — people, and looked at their diets. They found that those who ate eggs more often were less likely to have a variety of heart diseases, although this result was not perfectly consistent. They also controlled for a huge variety of external factors, such as income and education level, which means that their result is actually pretty consistent and strong.
The fish study was a bit less impressive, but no less cool. They looked at a few hundred couples who were undergoing IVF, and asked them about what they ate. They found that couples who ate more fish had a shorter time until they got pregnant, and also had more sex. They controlled for some factors in their analysis, but did not have as comprehensive data as was gathered in the egg study.
However, there were some large caveats to both studies. Firstly, as I said, they were observational. They could not control for all variables — there’s a good chance that, even in the amazingly well-done egg study, these results are down to another factor that confounds the relationship*. In particular, the fish study only controlled for a few variables, so it’s likely that something other than eating fish was impacting these couples’ fertility.
There’s also the issue of relative vs absolute benefit. Most of the reporting around the egg study, for example, talked about an 18% reduction in risk from eating an egg a day. This is misleading — the relative difference in risk of having heart disease was 18%, but since heart disease is fairly rare, the absolute difference was less than 0.01%. The same trope was used in the fish study — reporting was all about the relative increase in pregnancies, but the actual differences in risk are somewhat less impressive.
And, of course, while both of these studies were reported as definitive and applicable to your life, nothing could be further from the truth. The egg study was a specific effort aimed at mainland Chinese populations. It’s likely that this is not generalizeable to non-Chinese people, because studies in Europe and Japan have found exactly the opposite result — eating lots of eggs has previously been linked to a higher incidence of heart disease. Similarly, the people in the fish study were all attending IVF clinics — these are not your average couples. More than 2/3 of them had one or both partner earning more than $70,000 a year, which is well above what you’d expect in the general population.
And here we come to the real problem with these studies.
Because they are usually meaningless to your life, but they are almost never reported that way.
Banal and Boring
We all want silver bullets. We want science that speaks to our day-to-day lives. Perhaps because it’s easier to connect to it that way, or perhaps because that’s simply how it’s usually presented in the media.
Sadly, that’s not how this type of science really works.
Epidemiological studies like these on individual foods are interesting, but that’s often all they are. There are too many associations involved with people’s everyday lives to ascertain whether the relationships are causal — say, eating fish=pregnancy — or whether this is just a correlation — eating fish=rich and rich=pregnancy.
There’s unlikely to ever be a study that is so well-controlled, and on such a general population, that it can link a specific food to health outcomes like these. We long for simple solutions, but the bottom line is that diet is as messy as individual people’s lives, and figuring out exactly what is causing a health benefit is incredibly hard to unpick.
Eggs might prevent heart disease, but maybe only in Chinese people. Or maybe it’s not the eggs, and there’s another factor we have yet to discover. We just don’t know.
That’s why dietary guidelines are general — we know that certain broad categories are associated with certain outcomes, and can sometimes even demonstrate these things in a lab, but that doesn’t mean that doesn’t mean we can say that, say, Rustic Wholegrain prevents erectile dysfunction. Eating more whole grains might help, if your diet is currently full of refined breads and the like, but the specific health benefits are extremely hard to define.
The truth is that large populations studies looking at specific foods are mostly really boring. They might mean some minor changes in the guidelines, but realistically an 0.01% reduction in heart disease risk is meaningless to the individual anyway.
What does this mean to you? Well, eating healthily can be hard. Dietary guidelines are based on enormous reviews of the evidence, and tend to be a very good thing to base your diet on.
If you’re worried about your diet, talk to a dietitian. They do long degrees and training so that they can give you the best advice possible.
Just don’t change your life based on a single study reported on in the news.
Chances are it means virtually nothing to your life, and health, after all.
If you enjoyed, follow me on Medium, Twitter or Facebook!
*For example, it could be that people who eat more eggs also eat less fried potatoes. Fried potato consumption has previously been linked to higher rates of heart disease, so maybe it’s just that replacing fries in your diet with boiled eggs is good for your heart. From this study, we just don’t know for sure.
