I read a study that kept using the term insulin dependent diabetes mellitus to describe a group of people who seemed unlikely to be type 1 diabetics (they were on average 68 years old diagnosed at age 53, their average BMI was 30, and they were more than 1% of the study area's over-50 population). When I got to where it said they classified people as type 1 diabetics aka insulin dependent based on their answering yes to the question "Are you an insulin dependent diabetic" I was sure that the population was in fact mostly type 2s who were taking insulin.
So I fired off an email to the study authors, telling them that their premise that insulin dependent diabetes has blah blah blah effect was not really valid.
Only then did I read the rest of the study. At the end of the limitations, the authors admit that their insulin dependent population seems unlikely to actually be absolutely insulin dependent because less than 10% of the "insulin dependent diabetics" were diagnosed under the age of 30.
They say that their group of "insulin dependent diabetics" is type 2s who are "relatively insulin dependent". Which, although I think it's a bunch of bull hookie, still means my email was not exactly news to them.