In the Summer of 2012, articles started appearing saying that thin people with Type 2 diabetes had a higher mortality rate than overweight/obese people with Type 2 diabetes. The media had a field day with this. The sensationalistic headlines (“When It’s Good to be Fat” in SmartPlanet and “'Obesity Paradox': Why Being Thin with Diabetes Is a Dangerous Combo” in Time) were based on Association of weight status with mortality in adults with incident diabetes (JAMA. 2012 Aug 8;308(6):581-90), which was conducted and published by research scientists who didn’t do the most basic first step of medical science: make sure the study subjects actually have the disease you are studying. Dr. Mercedes Carnethon and colleagues analyzed data from five clinical trials to compare mortality rates among subjects who were normal weight with subjects who were overweight/obese at the time of incident adult-onset diabetes. In this case, hidden far down in the article the JAMA article authors stated, “we are unable to determine whether participants who were normal weight at the time of diabetes incidence in our study have LADA.” In other words, the researchers did not exclude people with latent autoimmune diabetes in adults (LADA), which is slow onset Type 1 autoimmune diabetes, from a study on people with Type 2 diabetes. Very telling, in the JAMA article, 11.2% of the people with “Type 2” diabetes were normal weight (BMI less than 25), and studies published worldwide since 1977 indicate that ~10% of people given the label “Type 2 diabetes” are autoantibody positive, have been incorrectly diagnosed, and in fact have Type 1 autoimmune diabetes. Those 10% have a lower average BMI than the true Type 2s.
Dr. Mercedes Carnethon, the lead researcher in the JAMA study, was quoted in the New York Times saying that “sizable numbers of normal weight people develop Type 2 diabetes” and that Type 2 diabetes patients of normal weight are twice as likely to die as those who are overweight or obese. I would say her quote might be corrected to, “sizeable numbers of normal weight people are misdiagnosed as having Type 2 diabetes when they actually have Type 1 diabetes, and those misdiagnosed individuals will die early due to mistreatment and improper medical care.”
When people with Type 1 diabetes are misdiagnosed as having Type 2 diabetes and given Type 2 treatments (usually starting with a sulfonylurea and/or metformin), they are undertreated and denied life-saving exogenous insulin (insulin is the only appropriate therapy for Type 1 diabetes), which leads to rapid onset of complications with the possibility of early death. So of course the “thin Type 2s” had a higher mortality rate—they had Type 1 diabetes, were denied insulin, and died early.
Following this JAMA study and its aftermath, Anne Peters MD (endocrinologist to Gary Hall Jr, who was diagnosed with Type 1 diabetes at age 24, participated in three Olympics and won 10 Olympic medals in swimming) responded to the Carnethon JAMA article with this Medscape video/article in which she discusses how she treats lean adults with new-onset diabetes. Dr. Peters says, “We know that autoimmune type 1 diabetes -- some people also call this latent autoimmune diabetes of the adult (LADA) -- can occur at any age. My oldest patient with new-onset type 1 diabetes was in her 90s. She was 92 or 93 years old and she presented with pretty florid new-onset diabetes. But I have seen adult-onset type 1 diabetes in patients of any age, from their 20s, 30s, or 40s, and the way to measure this is to look for the presence of [autoantibodies]. Not everybody who has adult-onset type 1 diabetes will have a measureable antibody. Some won't but might appear to me to behave so much like a patient with type 1 diabetes that I will still call it adult-onset type 1 diabetes. The reason that it is so important to differentiate between the types of diabetes is that although initially these patients may respond to oral agents because they still have some residual insulin secretion, they are going to progress much more quickly to needing insulin (usually either a multiple daily insulin regimen or an insulin pump) and they will begin to look more and more like a patient with type 1 diabetes.”
So is it a conspiracy? I would say no, but it is certainly neglecting the most rudimentary rules of science. The lack of scientific rigor has extreme consequences for people with diabetes, including early death. Finally, underlying the goofy and absurd headlines “When It Is Good to be Fat” is a truly tragic situation. People’s lives are shortened as a result of probable misdiagnosis, when if they were simply given exogenous insulin they could live long and hopefully happy and productive lives. Not have their lives cut short because of medical malpractice. First do no harm.
The presence of autoantibodies can be used to distinguish between autoimmune diabetes (Type 1a diabetes) and Type 2 diabetes or other non-autoimmune diabetes (for example, monogenic diabetes (aka MODY)). Autoantibodies are not present in Type 2 diabetes; if autoantibodies are present, the person has Type 1a diabetes (according to the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus, “Although the specific etiologies of [Type 2] diabetes are not known, autoimmune destruction of beta cells does not occur.”) The full suite of autoantibodies include GAD, ICA, IA-2, IAA, and ZnT8.