Sloppy reporting.
May. 1st, 2007 09:57 am![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
aithneDiabetes cases double in county
Summary of the article: Diabetes diagnoses have doubled since 1996 in King County.
What is wrong with the article? Oh, let me count the things.
1. It notes that there has been a spike in Type 1 diabetes as well as Type 2. This is the most interesting tidbit in the whole article--and it's completely glossed over. What is happening here? Does it correlate with a spike in other autoimmune diseases? Unfortunately, since a spike in Type 1 can't be explained away by saying people are fat and lazy, it's not news.
2. This was a 10-year study. I am willing to bet that the average diet of people in King County did not change significantly from 1996 to 2006; there's an assumption that it has (otherwise why would obesity be rising?) but no proof. A 40-year study, I'd buy, but not a 10-year.
3. The article did not note that the definition of obesity changed during that 10-year period, classifying a whole bunch of people as overweight or obese who had not been before.
4. There are more diagnoses in part because what might have been borderline 10 years ago is considered full-blown diabetes today.
5. I beleive that there is a more complicated relationship between obesity and diabetes than most people think. Who gets diabetes is not simply a matter of who's overweight and sedentary; there are other risk factors. Correlation is not causation, after all. I think that, if you look at the data, you could make a case for diabetes causing weight gain and not the other way around.
6. The bit about diabetes being a disease of ignorance really chaps my hide. "If they knew better, they wouldn't get diabetes!" It's not nearly that easy.
Now that I'm starting to be aware of sloppy science reporting, I'm starting to see it everywhere.
By the way, the report is here: http://www.metrokc.gov/health/datawatch/diabetes-April2007.pdf
Some very interesting things in there--and some interesting omissions. I'm not convinced that the numbers paint a particularly dire picture here.
Summary of the article: Diabetes diagnoses have doubled since 1996 in King County.
What is wrong with the article? Oh, let me count the things.
1. It notes that there has been a spike in Type 1 diabetes as well as Type 2. This is the most interesting tidbit in the whole article--and it's completely glossed over. What is happening here? Does it correlate with a spike in other autoimmune diseases? Unfortunately, since a spike in Type 1 can't be explained away by saying people are fat and lazy, it's not news.
2. This was a 10-year study. I am willing to bet that the average diet of people in King County did not change significantly from 1996 to 2006; there's an assumption that it has (otherwise why would obesity be rising?) but no proof. A 40-year study, I'd buy, but not a 10-year.
3. The article did not note that the definition of obesity changed during that 10-year period, classifying a whole bunch of people as overweight or obese who had not been before.
4. There are more diagnoses in part because what might have been borderline 10 years ago is considered full-blown diabetes today.
5. I beleive that there is a more complicated relationship between obesity and diabetes than most people think. Who gets diabetes is not simply a matter of who's overweight and sedentary; there are other risk factors. Correlation is not causation, after all. I think that, if you look at the data, you could make a case for diabetes causing weight gain and not the other way around.
6. The bit about diabetes being a disease of ignorance really chaps my hide. "If they knew better, they wouldn't get diabetes!" It's not nearly that easy.
Now that I'm starting to be aware of sloppy science reporting, I'm starting to see it everywhere.
By the way, the report is here: http://www.metrokc.gov/health/datawatch/diabetes-April2007.pdf
Some very interesting things in there--and some interesting omissions. I'm not convinced that the numbers paint a particularly dire picture here.
