He Drank Coffee. And Did Not Die…

This morning, I read with satisfaction that the World Health Organization (WHO) thinks it’s OK for me to be drinking coffee. Indeed, “moderate” coffee consumption might even be good for me.

That’s nice to know, since I’ve had two cups already today.  And at one time, people worried that coffee might be positively associated with colon cancer.

I thought this summary did a good job of reporting on current coffee meta-studies, differences amongst the underlying work, and the range of likely conclusions.   The results are a mixed bag, depending on things like sampling bias, but most do say the association between coffee and cancer is at least not positive.

Health-related studies are a constant and avidly-consumed feature of news feeds. However when we interpret this information, our human wiring plays a role.  We like to generalize (and we want to live).  Reading about a possible association between coffee and good health, our brains can leap into action: “If I drink more coffee, I’ll be healthier!”

At some level, we know that isn’t what the study says, but what does it say?  I find that it helps me to step through it. So, let’s assume the study applies to people of our general makeup and coffee-drinking habits – that’s not axiomatic, but let’s go with it.

Now, I imagine replicating myself 99 times, so there are now 100 of me, 50 enjoying coffee, and 50 doing without.  The baseline rate for colorectal cancer is about six percent, and one study reported a 24% decrease in colorectal cancer incidence for those drinking at least four cups of coffee a day.  That equates to an expected (but not guaranteed) incidence of 2.24 [(1-0.24)*0.06*50 ] for the 50 coffee drinkers, versus 2.95  [ 0.06 * 50] for the non-coffee drinkers, or a 1.4% difference.

That’s not zero, but very far from certainty (or even from 24%).  Our natural bias can be to overreact – excellent for survival, but sub-optimal for processing facts.  So our doctors will properly tell us what they always have: manage our weight, eat many vegetables, exercise, and stop stressing about things like the daily health report.

The health studies really are interesting – in part because this is one area where good and unbiased data are hard to come by, and as a result the outcomes are often unclear. I’ve often wondered – slightly tongue-in-cheek- whether the stress of trying to accommodate the recommendations of each statistical study negates the potential healthful benefits of following those recommendations.   Now there would be a study I’d like to see.

Meanwhile, it’s bottoms up for the coffee cup- with the WHO’s blessing. 

3 thoughts on “He Drank Coffee. And Did Not Die…

  1. So what is always interesting about medical studies is how well they have controlled other variables. For example, if “regularity” is associated with a decreased risk of colon cancer, has any factor been included to either measure or control the correlation with regularity, I say this because of anecdotal evidence that coffee consumption can increase “regularity”. If so is it the coffee, the regularity or the combination that yields the reduced risk?

    Likewise did the study group with the lower coffee consumption tend to have higher refined sugar or fast food consumption compared to the coffee drinkers? It is an interesting problem in study design to know what other factors to try and control in the data.

    My favorite example of this is the statistical evidence is that the people most likely to be struck by lightening are those people who have already been struck by lightening. Not because they are more naturally magnetic or have blazed a grounding path through their cells, but because the sample set has pretty much demonstrated that they engage in behavior that puts them at high risk of being struck by lightening, and as a group they probably will not change that behavior enough to lower their risk to at or below average.

    Liked by 1 person

  2. I like your writing style. It is direct, a bit humorous, and interesting. The topic is interesting, too, as I tend to drink 2-3 cups of coffee per day and have thought about the effects of it.

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s