Are we all “science communicators” now?
An Illustration of a Person's Mind
I help researchers start impactful and satisfying research programs, without the overwhelm or fear of failure. Click here to find out more.
Last week my school hosted Dr. Aaron Carroll, the CEO of Academy Health, for the Hartman Lecture. The title of his talk was, “Science you can trust: leading with evidence in a distrustful time.” He started with a story about data and its presentation in the public sphere. Specifically he referenced a systematic analysis published in The Lancet in 2018, which found that alcohol use was a leading risk factor for global disease burden, causing substantial mortality. Leading the authors to conclude that world-wide policies were needed to reduce overall alcohol consumption.
Not surprisingly this article made headlines, catching the attention of Dr. Carroll. In his piece in the New York Times, he draw particular attention to the article’s treatment of the raw data from the observational studies included in the review. The original article’s abstract shares the percentage of deaths attributable to increased alcohol consumption for men and women across ages and for a variety of causes. After looking at the actual numbers used to calculate the percentages and associated risks Dr. Carroll found that for each 100,000 people who had 1 drink per day 918 could expect to experience one of the 23 alcohol related problems. When compared to those who drank nothing, 914 people could expect to experience one of the 23 alcohol related problems. JUST 4 MORE PEOPLE experienced a negative outcome. Given the limitations of observational research statistical significance, in this case, may not be indicating a causal relationship.
Now Dr. Carroll wasn’t criticizing the percentages shared by the authors from The Lancet article, they were technically correct. What he was questioning were the interpretive implications for the difference between presenting relative risk versus absolute risk to the public.
This idea has been rolling around in my head for the last few days. I understand why relative risk is often presented in research papers: 1) It’s been used regularly in previous works (for right or wrong) and selective inertia in research is just as prominent as clinical inertia; 2) It makes the potential impact of our findings look large which can be important for tenure, promotion, and getting grants; And 3) researchers haven’t historically done research for public consumption.
It’s this last point that I want to unpack a little further. If you are reading this as a non-academic, this seems like a ridiculous comment to make. Health research is (or should be) done to benefit the health of society, right? Technically, yes. But practically, academics (with the possible exclusion of clinician scientists) do research for other researchers to consume. Why? Well for most of us, most of the time, a practical application of any one study we conduct is likely years, if not decades away. There are important reasons for this, not the least of which being that rushing to establish causal relationships can have seriously bad consequences.
Until relatively recently, that meant we scientists didn’t have to think too hard about non-scientists accessing and interpreting our work. Even findings from our projects that involved patients or communities were presented at conferences that don’t invite the public or were published in journals that had pay walls.
Social media and the (maybe) subsequent proliferation of open science has changed this dynamic. It is no longer enough to do our science in obscurity, with no thought about how it connects to the world outside the academy. Now I don’t think this means we all start jockeying to be on TV or someone’s YouTube channel. But I think it is time we start learning and teaching our students how to talk about our work in ways that are consumable (without infantilizing the listener). Come back next week for some best practices in science communication.
I realize this will be scary for some of us, but if we do it together, I think we’ll be okay.
(Words: 638)
Not sure what you need? Complete this 3-minute survey to see if I can help.