David Baxter PhD
Late Founder
What's with all the studies on coffee, and why can't we get a straight answer?
CBC Second Opinion
November 25, 2017
Yet another study about coffee — this one pointing to potential health benefits —
grabbed headlines this week. But some researchers caution that these studies actually
don't have much value. (Eric Vidal/Illustration/Reuters)
Coffee lovers rejoiced this week as media outlets around the world pounced on a news release headline from the BMJ (British Medical Journal) that declared, "Moderate coffee drinking 'more likely to benefit health than to harm it,' say experts."
It's just the latest in a parade of often-contradictory studies over many years about the health effects of that cup of java — some claiming coffee is good for you, some claiming it can be harmful. In this case, the study went so far as to suggest that drinking three cups a day was associated with lower risks of heart disease, diabetes and some types of cancers.
But the researchers themselves acknowledged that the data was mostly observational, and essentially concluded that it was likely safe — except for pregnant women and people at risk of fractures — to proceed to clinical trials to get more definitive answers.
So what does this widely covered study really add to our understanding about coffee and its health effects?
Nothing, according to two nutritional scientists at the University of Toronto.
"It is useless," Ahmed El-Sohemy, who specializes in research about genetics and metabolism, and has conducted several studies on coffee, said bluntly.
That's because these types of studies rely on analyzing data taken from large populations of people and looking for correlations — such as whether or not people who reported drinking coffee also developed heart disease or diabetes or liver disease or whatever it is the researchers are looking for.
But, as anyone who has taken a basic statistics course knows, "correlation is not causation," said Richard Bazinet, an associate professor in U of T's department of nutritional science.
"We look at these population-type studies where we say, 'Ah! People who drink two coffees a day, what's their health like?'" Bazinet said. "These correlations show up, and then when you change the population a bit, they disappear, probably because often they're not real effects."
Although researchers try to account for other lifestyle variables that could explain why someone does or doesn't develop a particular disease — including diet, smoking, exercise or other risk factors — there's only so much they can do with the data they have.
Also, coffee contains different components. Caffeine, for instance, like any other drug, can have different effects — positive, negative, or neutral — on different people, El-Sohemy said. The antioxidants contained in coffee, meanwhile, are widely considered to be beneficial for people's health.
If it's so hard to get a definitive answer on coffee's health benefits — or harms — why are there so many studies about it?
One reason is that coffee is such an ingrained part of our culture and so widely consumed, people have an insatiable desire to know its health effects, said Bazinet. It's a similar story with topics like red wine and chocolate.
That public appetite, in turn, could drive medical journals to accept more studies on the subject.
Or perhaps there aren't actually more studies on coffee than other nutrition topics, but media outlets gravitate to them because they know it's what their audiences want to read about.
In the end, El-Sohemy said, the focus needs to shift away from these large population-based studies and attempts to come up with a one-size-fits-all recommendation toward an approach that acknowledges coffee's health effects can vary by individual.
CBC Second Opinion
November 25, 2017
Yet another study about coffee — this one pointing to potential health benefits —
grabbed headlines this week. But some researchers caution that these studies actually
don't have much value. (Eric Vidal/Illustration/Reuters)
Coffee lovers rejoiced this week as media outlets around the world pounced on a news release headline from the BMJ (British Medical Journal) that declared, "Moderate coffee drinking 'more likely to benefit health than to harm it,' say experts."
It's just the latest in a parade of often-contradictory studies over many years about the health effects of that cup of java — some claiming coffee is good for you, some claiming it can be harmful. In this case, the study went so far as to suggest that drinking three cups a day was associated with lower risks of heart disease, diabetes and some types of cancers.
But the researchers themselves acknowledged that the data was mostly observational, and essentially concluded that it was likely safe — except for pregnant women and people at risk of fractures — to proceed to clinical trials to get more definitive answers.
So what does this widely covered study really add to our understanding about coffee and its health effects?
Nothing, according to two nutritional scientists at the University of Toronto.
"It is useless," Ahmed El-Sohemy, who specializes in research about genetics and metabolism, and has conducted several studies on coffee, said bluntly.
That's because these types of studies rely on analyzing data taken from large populations of people and looking for correlations — such as whether or not people who reported drinking coffee also developed heart disease or diabetes or liver disease or whatever it is the researchers are looking for.
But, as anyone who has taken a basic statistics course knows, "correlation is not causation," said Richard Bazinet, an associate professor in U of T's department of nutritional science.
"We look at these population-type studies where we say, 'Ah! People who drink two coffees a day, what's their health like?'" Bazinet said. "These correlations show up, and then when you change the population a bit, they disappear, probably because often they're not real effects."
Although researchers try to account for other lifestyle variables that could explain why someone does or doesn't develop a particular disease — including diet, smoking, exercise or other risk factors — there's only so much they can do with the data they have.
Also, coffee contains different components. Caffeine, for instance, like any other drug, can have different effects — positive, negative, or neutral — on different people, El-Sohemy said. The antioxidants contained in coffee, meanwhile, are widely considered to be beneficial for people's health.
If it's so hard to get a definitive answer on coffee's health benefits — or harms — why are there so many studies about it?
One reason is that coffee is such an ingrained part of our culture and so widely consumed, people have an insatiable desire to know its health effects, said Bazinet. It's a similar story with topics like red wine and chocolate.
That public appetite, in turn, could drive medical journals to accept more studies on the subject.
Or perhaps there aren't actually more studies on coffee than other nutrition topics, but media outlets gravitate to them because they know it's what their audiences want to read about.
In the end, El-Sohemy said, the focus needs to shift away from these large population-based studies and attempts to come up with a one-size-fits-all recommendation toward an approach that acknowledges coffee's health effects can vary by individual.
