Good News! 1927

Happy News Bias on Diabetes Rates

New Cases of DiabetesYahoo Health provides a perfect example of happy news bias in a headline this week: “The Diabetes Rate Is Actually Declining, Says the CDC.”

The problem is that the CDC didn’t say it and it isn’t true. Other reports were not as blatantly wrong as Yahoo Health, but they left the same false impression.

The truth is that new cases of diabetes are still coming, but not at the same blinding rate as was happening in 2008-10. The rate of new cases diagnosed every year is still more than double what it was in the 1980s. And that’s just the new cases.

Americans with DiabetesIf you look deeper, at the prevalence of diabetes, the rate of diabetes in the population is most certainly not going down. The number of Americans with diabetes has more than tripled since the 1980s.

The age-adjusted rate of diabetes (6.2 cases per 100 people) in the U.S. population is more than double what it was in 1980. It peaked in 2010 and didn’t go up or down much in the four years that followed.

That takes us to the question of what’s driving these rates of diabetes. The obvious answer is the dramatic growth in obesity rates. These rates have roughly tripled since 1980. The growth in diabetes rates followed, delayed by the years that it takes obesity to progress to diabetes.

U.S. Adult Obesity RateSuch a lag between the spike in obesity and a resulting spike in diabetes is exactly what one should expect. Likewise, it’s not surprising to see that growth in new diabetes cases slowed down after the growth in obesity rates slowed somewhat in the middle of the last decade.

So look out for happy news bias. Everyone wants good news, but self-deception leads to blind alleys and bad outcomes.

Click here to read the story from Yahoo Health and here to read more from the New York Times. Click here for CDC diabetes statistics.

Happy News! 1927. Photograph © Roadsidepictures / flickr

Subscribe by email to follow the accumulating evidence and observations that shape our view of health, obesity, and policy.


December 3, 2015

8 Responses to “Happy News Bias on Diabetes Rates”

  1. December 03, 2015 at 6:30 am, Joe Gitchell said:

    Thanks, Ted–very helpful. And a fair question whether “happy news bias” is better or worse than “scaremongering/sky is falling/OMG news bias”??

    Or, perhaps, since we’re all biased, what we can strive for as at least noticing and being aware of potential distortions that can result? I think I land there.


    • December 03, 2015 at 8:50 am, Ted said:

      You have good points all. The common thread in good news/OMG bias is a thirst for sensation and simplistic distortion. Only known antidote: critical thinking.

  2. December 03, 2015 at 7:34 am, Morgan Downey said:

    Good observation. Did you also notice that the CDC reports on obesity and this one on t2 diabetes omit Alaskan Natives/Native Americans and Native Hawaiian/Pacific Islanders? Both groups have very high rates of obesity and t2 diabetes. I’ve asked CDC why they don’t include these minority groups. I’ll let you know when/if I get an answer.

    • December 03, 2015 at 8:48 am, Ted said:

      Thanks, Morgan. Sounds like more good news bias. Let me know what you find.

  3. December 03, 2015 at 10:58 am, Stephan Guyenet said:

    Seems to me that the problem is with the imprecision of the word “rate”. Rate can refer to incidence or prevalence. If it’s used to describe incidence, the news reports are correct. The diabetes incidence rate is in fact declining, at least if you take the data at face value. And that is legitimate good news.

    • December 03, 2015 at 3:55 pm, Ted said:

      You are certainly right that “rate” is imprecise terminology, Stephan. Thanks for that observation

  4. December 04, 2015 at 9:51 am, Asheley Skinner said:

    Yes, I was going to say what Stephan said. When I hear “rate” I think “new cases”. A rate includes a time component (incidence) and a ratio does not (prevalence). So, the CDC actually DID say the rate declined (via Edward Gregg), and that was technically true.

    But so many people say “prevalence rate”–I probably have–that’s it’s a silly distinction.

    Even more concerning to me is that a declining incidence rate for diabetes DIAGNOSES does not necessarily have anything to do with the actual rate of diabetes. Notice that the decline began with the most recent economic downturn (~2008). I suspect the fewer incident cases is actually a bad thing, representing delayed diagnosis because people have lost jobs and health insurance.

    • December 04, 2015 at 10:45 am, Ted said:

      Well said, Asheley. Bottom line, talking to the public about new cases without mentioning stubbornly high prevalence is inherently misleading.