I am an admitted coffee aficionado. I love the taste. I love the warmth. No, I don't drink those maddening caramel strawberry double shot nonfat iced frappucinos – I like my coffee hot with a splash of cream and some very safe aspartame. Nevertheless, I've never thought much about coffee health advantages – it never seemed relevant to me.
I remember directing a clinical trial at a large teaching hospital in Seattle, WA back in the late 1980s, and there was a Starbucks kiosk in the lobby. I know most of you would think "what's so great about that?" But, it was nearly 30 years ago, and Starbucks wasn't a thing that it is today – I know some coffee snobs hate them, but 30 years ago, good coffee was unknown to most of the USA. Well, unless you lived in Seattle, apparently.
That kiosk started my love of coffee. I tried different coffee makers and methods of brewing coffee. Over the years, I've settled on a French press (as it is known in Canada and the USA, a coffee plunger in Australia and New Zealand, or a cafetière in France and the UK) for my coffee, which probably makes me a coffee snob.
I've written about the coffee health effects previously. And, in August 2018, a new paper was published that seemed to indicate that drinking lots of coffee lowered your risk of mortality. And, of course, websites across the internet chimed in with the great news. But did it really say that? Of course, your coffee-addicted ancient dinosaur will take a look.
That new paper on coffee health benefits
The paper, by Dr. Erikka Loftfield of the Division of Cancer Epidemiology and Genetics at the National Cancer Institute, was published in JAMA Internal Medicine. The researchers examined the relationship between coffee consumption and all-cause mortality in nearly 500,000 individuals. Although they focused on all causes of death, they also looked at mortality as a result of various cancers, cardiovascular disease (including stroke), and other diseases.
They also employed a genetic analysis examining whether genes influenced mortality while drinking coffee.
The researchers found that as individuals drank more coffee, they appeared to die a little less frequently. They also found that this result was not changed regardless of the presence of genes that are associated with caffeine metabolism. And the reduction in mortality showed a dose-response relationship up to 8+ cups of coffee per day (that's a lot).
The hazard ratios (HR), which is a statistical measurement of increase or decrease of risk of an event (such as mortality) from a particular cause (like a dose of coffee), at various average levels of coffee consumption per day were as follows:
- <1 cup of coffee, HR = 0.94 (HR=1.0 means no effect, less than 1.0 means a negative effect, or lower mortality risk, and greater than 1.0 means a positive effect, or higher mortality risk).
- 1 cup, HR =0.92
- 2-3 cups, HR=0.88
- 4-5 cups, HR=0.88
- 6-7 cups, HR=0.84
- >6 cups, HR=0.86
In addition, the researchers showed similar correlations for instant (who drinks this swill?), ground, or decaffeinated coffee.
The researchers also found no reduction in risk for lung cancer, colorectal cancer, breast cancer, pancreatic cancer, cardiovascular disease, ischaemic heart disease, stroke, and respiratory disease. And they also found that the genes associated with caffeine had little influence over the results.
But do these results give us any indication of coffee health benefits? Let's look at it from two different perspectives.
What don't I like about this coffee health study?
The most important point is that it is an observational type of study. That is, it requires patients to remember how much coffee is consumed, and then attempt to correlate it to mortality. There are too many variables here, especially for such a small, and barely statistically significant, effects. On the hierarchy of biomedical evidence, epidemiological studies are of moderate value.
Furthermore, observational studies have a fairly critical issue – it can be difficult to determine confounding variables, that is, a variable that can influence both the cause and the effect. A classic example of confounders is that ice cream consumption is strongly correlated with drownings – but the confounding variable is summer heat which increases both ice cream consumption and increased drownings (since pools, lakes, rivers and such are popular).
It's possible that healthier individuals drink more coffee and have a lower risk of all-cause mortality. Or individuals who drink more coffee exercise more. Or eat less. Or have less stress. It's difficult to tease out confounders in these type of studies, thus making it really difficult to make any claims about coffee health effects.
It is theoretically practical to set up a study that controls for many of these confounders. But it is difficult to control them all, because you may not know what they all are. The ice cream example is easy, but trying to find all of the confounders for coffee may be impossible.
Finally, short of providing us with a lot of evidence, like biological plausibility, it is very difficult to jump from correlation to causation. The anti-vaccine religion and other pseudoscience pushers in alternative medicine are renowned for their inability to understand this simple point.
It'll probably difficult to actually show causality between coffee consumption and mortality. Coffee contains over 100 different chemical compounds, so if this study is an indicator that coffee reduces mortality, it's going to take decades before we can determine which, if any, of those compounds have a beneficial health effect.
The authors themselves stated that "limitations include the study’s low participation rate, approximately 5.5%; consequently, the cohort is not demographically representative of the general UK population, with evidence of a “healthy volunteer” selection bias."
What do I like about this coffee health study?
Epidemiological studies like this have more value with more data points. This study included a half million individuals – that makes it more powerful than taking a survey at your local Starbucks. Or Peet's. Or Coffee Bean & Tea Leaf. Yes, I have a certain California bias to my coffee.
Second, the study did show a modest dose-response relationship. In any medical study, we would like to see a greater effect with a greater dose, and this study generally (though kind of weakly) shows that.
However, my scientific critique of this study is that it is underwhelming, but it does lead me to believe that there's enough data to keep looking.
Summary
The authors concluded:
...coffee drinking was associated inversely with all-cause mortality, including in those drinking at least 8 cups per day, in both slow and fast metabolizers of caffeine, and in consumers of ground, instant, and de- caffeinated coffee. Our results are based on observational data and should be interpreted with caution. Nevertheless, these results provide further evidence that coffee drinking can be part of a healthy diet and may provide reassurance to those who drink coffee and enjoy it.
Despite the blaring headlines that say "Six coffees a day could save your life," this study shows no such a thing. At best, it shows that consuming a lot of coffee is probably safe.
But is there a coffee health benefit? Probably not, but more research could answer that question.
Citations