Aaron Carroll, a pediatrician, is president and chief executive of AcademyHealth.
I recently attended a meeting where a prominent health expert talked about a program her organization runs that sends medical students into schools to teach kids how to cook, believing that it will lead to better nutrition, less obesity and healthier lifestyles. The audience ate it up. The idea is wholesome, it’s cheap, and you can almost picture the photo that would run with a news story.
About 20 years ago, I had a similar idea. I wrote up a grant proposal focused on researching whether teaching people to cook would lead to better health. I didn’t get the funding, to my disappointment, but it turns out it didn’t really matter. Lots of researchers had already tried the same thing — with disheartening results.
Just two years ago, a systematic review published in the journal Appetite looked at the results of 21 studies, six of them randomized controlled trials, of school-based cooking classes for schoolkids ages 4 to 12. Overall, such classes yielded only small improvements in the children’s cooking ability and vegetable intake. A year later, a large study of more than 600 students ages 6 to 9 in 26 schools in Britain found no improvements in food literacy, cooking skills, or vegetable and fruit intake.
We have been telling kids to eat better and exercise for what seems like forever. A massive Cochrane review of more than 170 studies involving almost 190,000 children found that focusing on diet and nutrition did little to nothing to change their BMI. Programs pushing diet and activity produced a small reduction that was gone after 15 months.
As a pediatrician, I spent decades telling parents and children to eat better and exercise more. They would nod. Some tried, but few saw results that they could sustain. Eventually, they’d become frustrated, and the next year we’d enact the same scene at the annual visit.
There are few Americans alive today who do not know that eating more fruits and vegetables is a good idea. They know that being more active is good. What we have is not a knowledge deficit problem, even though we keep funding it as if it is.
When we want to determine whether a new drug works, we conduct a trial of its effectiveness. If it fails, it’s done. Behavior programs get trials, too. But when they fail, they just get rebranded. Worse, we sometimes just disbelieve the results and try again.
What makes this situation even more warped is that we know what does work when telling people to change their behavior doesn’t: changing what’s around them. In 2010, school meal standards were changed, and later research found lower obesity risk among children in poverty. Placing taxes on soda reduces consumption. And if your goal is to get people to lose weight, nothing can really compete with GLP-1s.
None of these interventions involve advice. They all cost money, or encounter resistance from industry or institutions, while offering cooking classes doesn’t. We fund advice because it’s cheap and uncontroversial, and we resist the things that work because they aren’t.
Food-as-medicine and lifestyle messaging has recently become federal policy in the Department of Health and Human Services. The wellness industry, worth around $2 trillion a year, sells it at retail. But the moment “eat better and exercise” becomes a government program is the moment when the evidence for what works matters most. It’s unfortunate that the agency that is most likely to provide that evidence — the Agency for Healthcare Research and Quality — has been hobbled by grant cancellations and impoundments.
We don’t have to guess at what works, because AHRQ has already paid someone to find out. When it was preparing the evidence review for the U.S. Preventive Services Task Force’s 2024 recommendation on childhood obesity, it reviewed every available trial of behavioral treatment for kids with a high BMI. It found that behavioral programs worked only when they involved 26 or more hours with professionals, such as dietitians, psychologists and exercise specialists, over a year. Even then, they resulted in only a few pounds of weight loss on average. Interventions with less contact were not reliable. That’s the bar any proposed cooking class or educational intervention needs to clear.
If we are going to keep relying on advice to make a difference, we owe it to people to test whether that advice works in the real world, at the “dose” people actually get, focusing on the outcomes we really care about. Because if even well-run pilot programs that send nutrition educators into schools for 20 hours, measured and supported by an enthusiastic staff, can’t make a difference, then it’s unlikely that medical students doing so as an elective will.
When a vast amount of research shows that education and advice don’t achieve the results we want, the answer isn’t to give it a makeover and try again. The appropriate next step is to fund what does work and test what we don’t yet know.
Cooking lessons for kids, community gardens or other initiatives to promote healthy eating get funded because they photograph well and make for great human interest stories. They aren’t obesity policy. My fear is that when they fail to move the needle on chronic disease or obesity, in a few years we’ll just see someone else proposing them all over again.
The post This solution to obesity doesn’t work. America still eats it up. appeared first on Washington Post.
