Thursday, October 20, 2016

The missing piece in diabetes prevention

Some of my posts, like this one on problems with proliferating systematic reviews, have very brief gestational periods. I read something that stirs my thinking on a topic, I am inspired, and I write about it in one sitting. Others can take longer, such as this post about direct primary care entering the mainstream of medicine, which came together after a few months of reading and speaking with others about the topic. And sometimes I have a topic in mind to write on but just can't decide where it fits best: this blog, the AFP Community Blog, my Medscape commentary series, or a "professional" publication such as a medical journal.

What I had in mind was a sequel of sorts to last summer's post about the folly of screening for prediabetes, which I believed was more likely to cause harm by labeling more people as being diseased than it was to motivate them to take action to prevent diabetes. As the American Medical Association and the Centers for Disease Control and Prevention stepped up their efforts to identify people "at risk" for prediabetes (and an analysis published in JAMA Internal Medicine calculated that 60% of U.S. adults under age 60, and 80% of adults age 60 or older, would fall into this category), I wanted to say more about this over-medicalized approach to diabetes prevention and what a better alternative could look like. Several times I started writing, but couldn't quite decide where the post was leading.

My thoughts finally gelled in a Medscape commentary published earlier this week, where I pointed out three problems with preventing diabetes by screening for prediabetes:

1) The evidence suggests that this approach doesn't help people live longer (mortality) or better (quality of life).

2) Screening (especially with the CDC's tool) will lead to lots of overdiagnosis.

3) By ignoring social determinants of health, such as poverty and inequality, that resist lifestyle changes, this strategy may worsen health disparities.

Here is what I concluded:

Changing unhealthy environments can be a far more effective and long-lasting intervention than one-on-one clinical counseling. In the late 1990s, the US Department of Housing and Urban Development randomly assigned 4500 women with children in high-poverty urban areas to no housing vouchers, unrestricted traditional vouchers, or vouchers that could only be redeemed for housing in low-poverty areas. Ten to 15 years later, the group receiving traditional vouchers was no healthier than the control group, but the group receiving low-poverty vouchers had significantly lower body mass index and glycated hemoglobin levels.


Although the Medicare DPP as proposed should improve the health of many of our patients, in order to prevent diabetes without worsening health disparities, family physicians also require resources to address social determinants of health. To this end, the American Academy of Family Physicians recently published a position paper describing strategies for collaborating effectively with public health partners to lead the prevention of chronic diseases in our communities.