March 4, 2011
News Flash: Diabetes is Not Good
Type 2 diabetes mellitus has long been known to increase the risk of stroke, heart attack, kidney disease, and eye disease. In the US diabetes is the leading cause of kidney failure requiring dialysis and one of the leading causes of blindness. Diabetes is also increasing in prevalence as people become more overweight.
A study in this week’s New England Journal of Medicine attempted to quantify the risk of premature death associated with diabetes. The results were dramatic, and attracted much media coverage. (See links to articles below.)
The study was a compilation of data from 97 previous studies that were done for entirely different reasons. The studies followed over 800,000 people for an average of 13.5 years. At the time of enrollment 6% of the people had diabetes. The study simply compared the death rates of those who had diabetes at the start of the study with the rest of the subjects. The results were adjusted for age, sex, smoking status and body mass index (BMI).
The subjects with diabetes did much worse. As expected, they were more likely to die of strokes, heart attacks and kidney disease, but they were also more likely to die of liver disease, cancer, infectious disease and even suicide. On average, a 50-year old with diabetes at the start of the study died 6 years earlier than a person without diabetes of the same age, sex, smoking status and BMI.
Well, that’s pretty bad, but not as bad as what the media makes of it. This is not a randomized study. All we’ve learned is that one group dies earlier of a lot of diseases and also has diabetes, and another group dies later and doesn’t have diabetes. That doesn’t mean that diabetes causes the earlier deaths, simply that it is associated with earlier deaths. Lots of factors not measured in the study could have both predisposed to diabetes and caused other life-threatening diseases – family history, diet, different levels of exercise, or a tendency to eat ice cream while driving on the freeway. It doesn’t tell us about diabetes as much as it tells us about the people who happen to have it. So the LA Times headline “Diabetes can take six years off your life…” is completely misleading.
Does this help doctors or patients diagnose or treat diabetes? No. Does it mean that the day a patient is diagnosed with diabetes the date of her death advances 6 years sooner? Not at all. This may provide guidance for scientists designing studies to better clarify the harms caused by diabetes, but for doctors and patients there is no actionable information here. It’s just a reminder that if someone comes to your door offering you diabetes, you should decline.
(Thanks to my patient, Jay F. for pointing me to the LA Times article.)
Washington Post article: Study: 50-year-old with diabetes dies 6 yrs sooner
New England Journal of Medicine article: Diabetes Mellitus, Fasting Glucose, and Risk of Cause-Specific Death
Important legal mumbo jumbo: