Coffee Is Good for You—Unless It’s Not!

I love coffee—and I’m not alone.

Americans drink 400 million cups a day (yes, you read that correctly) and we spend $30 billion on it every year.

The good news is that there’s a lot of research that links drinking coffee with health benefits, including lower risk of cancer, Parkinson’s disease, obesity, diabetes, metabolic syndrome, and heart disease. I covered that research in detail in a recent podcast.

Here’s the bad news: while coffee is undoubtedly beneficial for some people, it may be harmful for others.

I talked about some of the factors, including sleep, stress, and intolerance to proteins in coffee beans, that determine individual response to coffee in another podcast a while back.

But there’s another important factor to consider: genotype.

Coffee is the primary source of caffeine for Americans. Caffeine is metabolized by an enzyme in the liver that is encoded for by the CYP1A2 gene. Unfortunately, about 50 percent of the population has a variant in the CYP1A2 gene that leads to slow processing of caffeine.

For these “slow metabolizers,” drinking coffee:

• Is associated with a higher risk of heart disease (1)
• Is associated with a higher risk of hypertension (2)
• Is associated with impaired fasting glucose (3)
• May not have the protective effects against some cancers that it appears to for “fast metabolizers” (4, 5)

That said, in some cases coffee appears to be beneficial even for slow metabolizers. For example, caffeine is neuroprotective and reduces the risk of Parkinson’s disease in both slow and fast metabolizers. (6) Other studies have even shown that fast—not slow—metabolizers of caffeine may be at higher risk of bone loss. (7)

Is coffee good for you? That depends.

Adding to the confusion, many of the large, observational studies I reviewed in my podcasts found that the overall effect of coffee intake in the populations studied was positive. If 50 percent of people are at higher risk of disease from drinking coffee because of their genotype, then why isn’t this showing up in these epidemiological studies?

What are we to make of these conflicting data?

The most obvious conclusion is that it’s impossible to make a general statement about the health impacts of coffee. The answer to the question, “Is coffee good for me?” is: “It depends.”

This shouldn’t come as a surprise. If you’ve been following my work for any length of time, or you’ve read my book, you’ve probably heard me say, “There’s no one-size-fits-all approach when it comes to diet.”

The most recent research on nutrition, including these studies on coffee and caffeine, confirm that this is true. While we share a lot in common as human beings, we also have important differences: genes, gene expression, metabolic activity, gut microbiome, lifestyle, activity level, and numerous other factors will differ from person to person, and all of these will impact how we respond to a particular food (or beverage, like coffee).

For example:

• There is wide variation in post-meal blood sugar among people eating identical meals, and diets that are personalized on the basis of dietary habits, physical activity, and gut microbiota are more successful than “standardized” diets. (8)
• Response to low-carb and low-fat diets in overweight people varies considerably and may depend on their insulin sensitivity and other factors that are not yet fully understood. (9)
• Caffeine consumed in the afternoon or evening significantly disrupts sleep in some people, but not in others. (10)

These studies represent the future of nutrition research. In fact, I’m quite sure that in a relatively short period of time we’ll consider the current notion that there’s a single, optimal diet for everyone as an outdated and ignorant idea.

Another conclusion that we might infer from the conflicting data on coffee is that even within a particular genotype the effects are variable. In other words, some slow metabolizers might be adversely affected by caffeine where others aren’t, and the opposite might be true for fast metabolizers. This also makes sense because there are a huge number of factors above and beyond the CYP1A2 genotype that would influence how coffee and caffeine affect an individual, from their baseline diet to their stress levels and sleep to their gut microbiota. It’s also true that being a fast or slow metabolizer of caffeine isn’t binary (i.e., two possible speeds: fast or slow), but more of a spectrum (ranging from very slow to very fast).

Now that we’ve established that coffee and caffeine can be both beneficial and harmful, how do you know how it affects you? Here’s what I suggest:

1. Listen to my podcast called Is Drinking Coffee Good for You? in which I discuss some of the non-genetic factors that determine individual response to coffee.
2. If you haven’t already done this, titrate yourself off coffee (reduce your consumption slowly until you’re off it completely) and other sources of caffeine for at least 30 days. Then add it back in and see how you respond.
3. Find out whether you’re a “slow” or “fast” metabolizer. You can get this kind of genetic data through companies like 23andme and SmartDNA. If you’ve done 23andme, log in, go to “My account,” select “Browse raw data,” and type “CYP1A2” into the “Jump to a gene” search box. Once on the search results page, find the rs762551 SNP. In the far right column, it will give your variant of that SNP. If you are AA, you’re a fast metabolizer. If you are AC or CC, you’re a slow metabolizer (with CC being slower than AC).

We live in exciting times. At some point in the future, we’ll be able to create much more precise nutritional recommendations based on our genotype and epigenetic factors, in addition to all of the other factors I’ve discussed in my book—such as health status, lifestyle, physical activity, and goals.

Right now, we’re only scratching the surface and still have a lot to learn. But we already know enough to stop asking questions like “Is coffee (or carbohydrate, fat, etc.) healthy” and start asking questions like “Is coffee (or carbohydrate, fat, etc.) healthy for me?”