In 2014 we reported that the CDC’s Estimate of Diabetes and Its Burden in the United States had some good news in it, because it was beginning to look as though the prevalence of diabetes in the U.S. was beginning to recede, even if measured only by the rate at which new cases were being diagnosed. At the time, the report’s co-author Dr. Ann Albright told Reuters Health that there was reason to be encouraged, yet she cautioned there would be a continued need to watch the numbers “to make sure we can sustain this and ultimately reverse this trend.” The agency’s official comment in the press announcement that accompanied the 2014 report pointed to other surveys of segments of the T2D population indicating that the upward trends, shown in earlier years, of new cases of diabetes and new cases of clinical obesity, were both leveling off.
Here’s a brief 2014-2017 over-and-under comparison:
- This year’s CDC report estimates 30.3 million Americans, or 9.4% of the population, as having diabetes. The 2014 number was 29.1 million, or 9.3%
- In 2014, CDC estimated that 21 million people were diagnosed; that number jumped to 23.1 million in the 2017 report
- In 2014, 8.1 million people were undiagnosed, according to the estimate; but by 2017, the number had dropped to 7.2 million
On the surface of things, yes, it looks like more Americans have diabetes now than three years ago – 1.2 million more people in fact, representing a slightly greater portion of the population. Which sounds bad, doesn’t it? But it might not be. Quoted in CDC’s July 18 news release announcing the new report, Dr. Albright once again saw a silver lining: “Consistent with previous trends, our research shows that diabetes cases are still increasing, although not as quickly as in previous years.”
How Should We Understand What CDC is Telling Us?
CDC gathered its 2017 data from a number of sources, as it did for 2014, such as:
- The National Health and Nutrition Examination Survey (NHANES, which is a CDC program)
- The U.S. Census Bureau
- The Behavioral Risk Factor Surveillance System (BRFSS, also a CDC program)
- State public health databases
- The Indian Health Service (IHS), which serves members of recognized native American tribes and native Alaskans, who have shown to be at a greater risk of developing T2D
BRFSS relies extensively on telephone interviews; NHANES combines patient interviews and reports of physical exams; IHS relies upon reporting by physicians and case managers in rural regions; and state public health agencies employ dozens of different methodologies to collect information about diabetes. There isn’t a method of counting cases common to all the sources that are tapped to prepare the Estimates. And, after all, what’s published is presented as an estimate, designed to inform the development of national public health policy, and to assess whether national policy is making any difference.
Analysts at CDC don’t differentiate between T1D and T2D in assembling most of the tables published in the Estimates. But that doesn’t mean that their statistical method is unsound, or that there’s no conclusions that can be drawn about T2D from the CDC report. As long as the analytical tools are consistently applied from estimate year to estimate year, CDC can be reasonably certain that when there’s a change for the better seen among all people who have diabetes, such as a lessening of the rate at which new cases are showing up, and a drop in the proportion of people likely going undiagnosed, that change is driven more by people who have T2D than people who have T1D.
How Can CDC Know That T2D Is Driving the National Estimates?
In 1998, following a CDC estimate of diabetes prevalence based on the 1995 National Health Interview Survey, the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) began recruiting for a clinical study to be conducted at 27 locations around the country to see whether modest weight loss through dietary change and increased physical activity, or treatment with metformin, the old gold standard oral T2D medication, could better prevent or delay onset of T2D. Of the 3,234 who signed up, 45% were African American, Alaskan Native, Native American, Asian American, Hispanic/Latino, or Pacific Islander — in other words, members of demographic groups underserved by healthcare professionals. The NIDDK study launched at a time when the CDC’s estimate was that 23.6 million Americans had diabetes, 5.7 million of them undiagnosed.
To qualify for the study, participants had to be overweight, and have blood glucose levels higher than normal, but not high enough for a clinical diagnosis of T2D — in other words, all participants had what is now known, by a more precise set of indicators than existed in the 1990’s, as prediabetes. The study was meant to lay the groundwork for a national Diabetes Prevention Program, and to corroborate what doctors had been thinking for years — that controlling weight through regular physical activity and a diet reduced in fat and calories can sharply reduce the odds that a person will wind up having T2D.
Three groups formed — one to take 850 milligrams a day of metformin to lower the amount of glucose produced in the liver and increase the sensitivity of muscle cells to insulin, and another to take a placebo. The third group, the lifestyle intervention group<,em>, received intensive training in diet, physical activity, and behavior modification, and participants were given a goal of losing seven percent of body weight. For a 240 pound person, that’s just under 17 pounds.
The results did indeed corroborate what diabetes treatment professionals had already figured out from their clinical experiences: The lifestyle intervention group achieved the greatest reduction in risk of developing T2D, at 58%. The metformin group came in at 31%. The lifestyle group finding was true across all the ethnic or demographic subgroups, and equally true of men and women. The exercise/diet/behavior modification approach worked especially well for people 60 and older, who reduced their risk by 71%. Risk itself being a statistical creature, it had to be measured, and so it was, by periodic blood testing yardsticks, the same used to measure success in managing diabetes care, and, incidentally, used in clinical trials of new diabetes drugs to assess effectiveness.
People at heightened risk of developing T2D have the ability to delay its onset, for as long as ten years, with easily achievable lifestyle interventions, such as thirty minutes of exercise five days a week and cutting down on dietary fats and calories, and that has continued to be shown through successive NIDDK studies and through participant follow-up contacts.
One outgrowth of the NIDDK study begun almost twenty years ago is a national education program, jointly sponsored by the National Institutes of Health and CDC, offering teaching resources for individuals who may be at risk of developing T2D, and for the professionals who care for them. Building upon this research, for example, the American Diabetes Association has developed a road map for professionals, focusing on how to diagnose prediabetes, even though it’s not yet considered a clinical diagnosis in its own right, and on what to counsel people fitting the prediabetes profile to do to halt their progress toward having T2D.
So it’s not much of a jump at all for CDC to suggest that a change for the better – this year another click down in the rate of growth of all new diabetes diagnoses, comes predominantly from something people can take control of and better protect their health.
Using Your Intuition
It’s still a scary number – almost 725 million people being undiagnosed, and, according to national estimates going back to before 2014, a constant 90% to 95% of those undiagnosed people having T2D. The American Diabetes Association believes that only about 11% of people know that they are in the pre-diabetic stage, which means that 89% of people who are moving toward T2D don’t know it’s happening. It’s hard to consider such numbers and think how anyone at CDC could see even a shred of a silver lining in the 2017 results, isn’t it?
It’s understandable that if you have T2D, and you know it, and you’ve been following good self-care practice, you would wonder about the value of the 2017, and 2014, estimates. After all, whatever CDC finds out about national trends every three years or so doesn’t predict how successful you will be in managing your A1c and your cholesterol and triglyceride counts – all the things your regular lab tests measure and, incidentally, the same things that go into the clinical definition of prediabetes.
So, try applying your own intuition to find the silver lining. It should be easy to recognize that the lifestyle interventions that have been shown, consistently over twenty years or more, to delay or prevent the onset of T2D are the very same lifestyle habits that you can follow to maintain a better quality of life. The CDC’s estimates of prevalence of T2D, and of the burdens on human productivity and the national expenditure for health care, do not on their own determine your own prospects for living well despite having T2D.
Remember that in 2014, 8.1 million people were undiagnosed, according to the CDC’s estimate; but by 2017, the number had dropped to 7.2 million. Among all the conclusions which you might draw, that alone suggests that people are getting better at going to the doctor, and that doctors are getting better at diagnosing diabetes. So if the doctors are doing their jobs, there’s no reason you shouldn’t. Reducing fat intake brings the bad (LDL) cholesterol and triglycerides down, and with that your risk of cardiovascular disease. Weighing more than you should taxes your pancreas and makes your muscles use insulin less efficiently.
When we reported on the CDC’s estimates in 2014 we asked the question “Are we winning the war on obesity?” Turns out we could be.
To read the 2017 CDC Estimates, follow this link.
Want more news on Type 2 diabetes? Subscribe to our newsletter here.