When it comes to health, bogus and misleading claims are as common as dirt.
And unfortunately, a lifetime of received wisdom can make it difficult to distinguish between advice grounded in scientific evidence and that grounded in folklore. People are often stubborn, resistant to change, and reluctant to learn new things when new facts contradict their long-held assumptions.
But perhaps we can learn to be more discerning—or at least, maybe our children can.
Dr. Andy Oxman, research director at the Norwegian Institute of Public Health in Oslo, studies how people make decisions about their health and helps them make smarter choices based on scientific evidence and critical thought.
In 2016, Oxman and his team conducted research with 10,000 children from 120 primary schools in central Uganda, teaching some children basic scientific concepts and methods to critically evaluate health claims. Kids who learned these skills were better able to debunk junk science, including a local old wives’ tale claiming that cow dung is good for treating burns. Further research is planned in Africa and Europe.
We recently spoke with Oxman about his research, the importance of evidence-based healthcare, and why we benefit from informed skepticism.
Let’s start at the beginning. What is evidence-based medicine?
Dr. Andy Oxman: The term was coined in 1992. Until then, medical decisions were informed largely by what we already knew from physiology and biochemistry, as well as expert opinion.
Evidence-based medicine prioritizes direct evidence of the risks and benefits of different choices we make about healthcare. It focuses on systematic and transparent evaluation of the certainty of that evidence, something that was not done routinely in the past.
It’s not that clinical experience isn’t relevant. But to really know whether we’re doing more good than harm, we need reliable evidence. As obvious as that may seem, the concept was initially seen as a bit of a threat to many medical practitioners, particularly those in authority, because it challenged their opinions and asked, “What’s the evidence to support them?”
What led you to focus your critical-thinking research on children?
Oxman: A number of years ago, when my two youngest kids were 10, I visited their primary school in Oslo when they were studying the human body. I took a bag of Nonstops [Norwegian candy] to test my hypothesis that children could learn the concepts involved in critical thought.
First, I told them that high-schoolers claimed the red Nonstops helped them study—that the red Nonstops could help them draw straighter lines and do their homework faster, but also that they caused some of them to get dizzy and have stomachaches if they stood up quickly.
I divided the kids into groups, gave each a bag of candies, and told them to figure out whether the high school kids were right or wrong. They very quickly figured out basic principles—that you have to compare like to like, for instance.
In less than an hour they figured out on their own how to do experiments to test the claims about red Nonstops. They determined the high school kids didn’t know what they were talking about: There was no difference between the red ones and all the others.
How difficult was it for kids to understand how to properly test the high schoolers’ claim?
The most striking moment for me was when one of the teachers talked about how to make sure the groups being compared are random, saying, “It’s like what we do in gym class: I pick every other student to make the teams fair.” The kids laughed and explained why that method doesn’t work: “If I want to be with Billy, I’ll make sure that Mary is standing between me and Billy, so that me and Billy end up on the same team.”
The jargon term for that is “unconcealed allocation,” which takes a bit of time to explain to clinicians and researchers. But the kids understood it intuitively, without any explanation.
If your work shows that kids are receptive to these concepts, what’s the basic takeaway for adults?
Oxman: I’ll sometimes ask people, “Would you prefer to make an informed choice or an uninformed choice about things that affect your health?” Virtually everybody prefers informed choices, and that’s what it’s all about.
The term “evidence-based decisions” is now sometimes used beyond how physicians make decisions in clinical situations; it has spread to other disciplines as well.
Evidence-informed policy is something I—and a number of other people—have worked on. Many public and private foundations work toward supporting evidence-informed policy making—not just in healthcare, but across the board.
Why is it so hard for politicians, doctors, and people in general to make evidence-based decisions in our daily lives?
Oxman: The problem is that as we get older we have a lot of responsibilities. Policy makers are busy—most of them, at least—doing what they think is right for their constituents. It’s not their job to be studying how to assess claims and interpret evidence. The same goes for clinicians who are busy taking care of patients. There are lots of barriers that get in the way of critical thinking and making the effort to evaluate evidence in routine practice. We get lots of fixed ideas about things, at least some of which end up being concepts we need to unlearn.
You note that reporters too often mischaracterize scientific findings and help perpetuate bogus health information. How can the media help people differentiate between fact and fantasy?
Oxman: The problem is that there are too few specialized health journalists. You have people with minimal background or relevant training covering important health issues, and they sometimes lack the time to become adequately informed. They also need to get their ideas past editors whose knowledge and expertise may be more general than needed for specific scientific and health stories.
Sites like Health News Review in the U.S. or Behind the Headlines in the U.K. are helpful. HNR uses checklists to assess media reports and rank them from one to five based on how credible and informed they are.
Interview has been edited for clarity and length.