25

One other thing to note here is that in the past chapter or two we followed a different pattern than a lot of web searching. Here we decided who would be the most trustworthy source of medical consensus (the NIH)  and looked up what they said.

This is an important technique to have in your research mix. Too often, we execute web search after web search without first asking who would constitute an expert. Unsurprisingly, when we do things in this order, we end up valuing the expertise of people who agree with us and devaluing the expertise of those who don’t. If you find yourself going down a rabbit hole of conflicting information in your searches, back up a second and ask yourself: whose expertise would you respect? Maybe it’s not the NIH. Maybe it’s the Mayo Clinic, or Medline, or the World Health Organization. But deciding who has expertise before you search will mediate some of your worst tendencies toward confirmation bias.

So, given the evidence we’ve seen in previous chapters about alcohol and cancer–am I going to give up my after-work porter? I don’t know. I really like porter. The evidence is still emerging, and maybe the risk increase is worth it. But I’m also convinced the Washington Post article isn’t the newest version of “eating grapefruit will make you thinner.” It’s not even “Nutrasweet may make you fat,” which is an interesting finding, but a point around which there is no consensus. Instead “small amounts of daily alcohol increase cancer risk” represents a real emerging consensus in the research, and from our review we find it’s not even a particularly new trend. The consensus emerged some time ago (the NIH FAQ dates back to 2010); it’s just been poorly communicated to the public.