I figured this topic would come up, as the news story is getting lots of ink around the country. I was interviewed by a couple of large newspapers after it broke and I will briefly state here what I said in those interviews. But first let my just say this was a huge study population-wise and it was huge time-wise. That usually bodes well for the accuracy of conclusions and findings, but not always.

I want to state that I have NOT read word for word the entire study, but I have read many overviews of it by others, both in the news and in the professional community. None of these reviews offered observations on it, but merely stated what they felt the report stated. So I have not picked it apart but intend to do so next week with some science people whose opinions I trust.

I am troubled by a couple things about the study. The first is that the finding is about survival as opposed to incidence. Anytime you speak to survival I guarantee you it is very hard to control for bias. The reason the survival finding (the occurrence rate was not improved/lessened) bothers me is that all the nutrition related research that has been going on in the US over the last few decades, has to do with incidence and chemo prevention through the consumption of micro nutrients. You have likely seen them in the OCF news feed, but the numbers of studies showing that in various models animals and humans less so, that dark colored fruits and vegetables, and other derivatives of them such as chocolate, coffee, wine etc. seem to have some chemo protective properties in relationship to cancer and other diseases, likely from the micro nutrients that are unique to them. There have been dueling results, and not all the studies were ideal in their design, but I believe that the data clearly show that there is a measurable impact. What the positive ones show relates to impact on incidence not survival. But this study comes out of left field and speaks for the first time significantly to survival. This is counterintuitive, and I would have expected the opposite, that like other dark heavily micro nutrient and polyphenol loaded foodstuffs, coffee should be chemo preventative ie. it should impact incidence.

Then there is the issue of caffeinated vs non caffeinated, in which the study found that the caffeinated had impact, the non caffeinated less so. That means one of two things to me. The caffeine is having the impact, which I highly doubt since there are so many other forms of high caffeine consumption that do appear to have positive benefits. Think about long haul truckers that pop caffeine pills in large numbers, members of our military and in particular pilots, that take them in combat theaters to stay alert on long or night time missions, etc. Other than alertness we don't find health benefits associated with caffeine. It could actually mean that the process of decaffeination actually degrades the micro nutrients that seem in other foodstuff to be protective. This was not explored, and the source of the effect was not found or speculated on.

Lastly, when the study was originated decades ago, the authors made some attempt to limit bias by accounting for some confounders, such as tobacco, but the elimination of that type of bias was really incomplete, and this I think is the real weak point in the study.

We all know here that during the actual study period, that HPV16 arose as a very strong player in both incidence of oral and oropharyngeal cancers and survival of them. HPV16 positive cancers in peer reviewed papers have as high as a 35% survival advantage depending on staging at diagnosis etc. That alone throws a wrench in this whole study. Since they did not (they didn't know at the beginning of the study what we know today about HPV) eliminate HPV as a potential cause of survival, nor did they determine if the virus was more common in coffee drinkers than non coffee drinkers. Peer reviewed retrospective published studies (your donations at work as partial funders of this) showed the growing influence of HPV16 specifically, all the way back to 1972. So this is a huge confounding factor in the study that is unresolved.

Finally, anytime that you discuss survival and not incidence, besides the obvious core driver, (stage at time of diagnosis), there are many hard to define confounders. For instance just this week we put a news story up that clearly shows that patient outcomes at smaller regional hospitals are not as good as at the major CCC's, and institution is a factor in the survival numbers. Socio -economic issues also play a huge role in survival, and it is clear that in the US we have very pronounced disparities; between affluent and poor populations, those with and without insurance, those in rural areas vs. those in major cosmopolitan areas etc. This is all a huge issue when we speak to survival and not incidence, because we know in the real world, data clearly shows negative impact to those on the bottom end of the socio economic scale. This is tied to everything from education and awareness, access to healthcare itself let alone screening that might allow for earlier discovery and staging, ability to afford care even if it is available, or insurance to pay for it. Those are the broad areas, but there are smaller confounding issues in this group as well, such as diet and exercise in different populations.

So I am always pleased when a study shows that something impacts survival in oral cancer, but I think my initial gut reaction is that this study has significant unresolved bias in it, and while I like coffee, particularly in the AM, no one could really put up with me if I drank 4 cups a day.

I would like to see what micro nutrients coffee has in common with dark fruits like blackberries, blueberries etc. (all in our news feed) as perhaps there is a simple compound/s that we could take as supplements that would have the same impact.

I don't now if this makes sense to others, but I will post again when some much smarter people than I have a chance to discuss this study next week.