Does diabetes -- or poorly-controlled diabetes -- leads to dementia?   (February 3, 2009)

I just read an OMG news story. That is to say, my initial reaction was "oh, my g-d, that's scary!" In the story, the writer indicates that "people who are diagnosed with diabetes before age 65 have more than double the chance of developing Alzheimer's disease." After reading the abstract of the study that was being discussed, however, my initial OMG reaction changed to "seems like just another lousy research study."

The abstract [Mid- and Late-Life Diabetes in Relation to the Risk of Dementia A Population-Based Twin Study] does conclude that diabetes with onset in mid-life increases the risk of Alzheimer disease and vascular dementia. But I have a huge issue with this conclusion: is it the mere presence of diabetes that led to the increased risk, or was it poorly-controlled diabetes that led to the increased risk? There's no data available in the abstract about whether any attempt was made to measure A1C levels or any other measure of diabetes control in the population that was studied. If the authors had done so, and had found that people with diabetes (PWD) who had higher A1C levels had a higher risk of dementia, and found that PWD with near-normal A1C levels had near-normal risk, then we could all assume that it's uncontrolled diabetes that leads to increased risk.

To claim that diabetes is associated with dementia, without any comment about whether the risk was found to be associated with differing levels of hyperglycemia, is inappropriate. And for PWD like myself with normal A1C levels shouldn't have to wonder about this: the authors should have clearly spelled out whether it was "diabetes" (their conclusion) or "poorly-controlled diabetes" (my surmise) that was associated with dementia. Without such information, the publication wasn't worth publishing.