On the recent 2KetoDudes podcast HEAL Thyself with Melanie Miller the guest mentions (and the show links to) some recent study that found that West Virginia has the 2nd highest rate of Obesity (in adults anyway, I believe).
Well, this reminded me of a post from fathead-movie.com from a few years ago which talked about much publicized earlier studies that showed the South had higher rates of Obesity, and lead to much speculation as to why, only to be somewhat refuted by better studies.
You see, the problem was the cited studies were just doing telephone calls to random people and asking them for information to run through a BMI calculation (granted, BMI isn’t the best measurement to begin with, but it is a quick and easy general indicator). However, when another study actually went around and did the more time intensive task of actually having people measured rather than self-report, they found that the regions previously reporting the highest rates still had pretty similar rates, but they simply were not the highest by a long shot. Instead, it at least appears that people talking on the telephone in certain regions were more likely to lie (about height or weight) which pushed down the prevalence of obesity found in that area.
The paper on that is found here: The geographic distribution of obesity in the US and the potential regional differences in misreporting of obesity
Well, that made me wonder whether the West Virginia stat mentioned had a similar problem. And from what I could dig up from the article mentioning this, the data was taken from CDC data. Here is one of the state pages from the CDC (West Virginia just for relation):
https://nccd.cdc.gov/dnpao_dtm/rdPage.aspx?rdReport=DNPAO_DTM.ExploreByLocation&rdRequestForwarding=Form
A few things of note from this, one being the notes indicating how many types of people were simply excluded from these reports just for the adult measurements:
Respondents reporting weight < 50 pounds or ≥ 650 pounds; height < 3 feet or ≥ 8 feet; or BMI: <12 or ≥ 100 were excluded. Pregnant respondents were also excluded.
I get that they are trying to rule out major outliers, but I’m not sure how reasonable it is in this instance. Particularly with that BMI rating of <12 being ruled out, a state could have a major body builder community, and body builders commonly try to get into the single digits for BF%. Granted, they also often don’t hit that with regards to BMI (too much extra muscle) so maybe that doesn’t matter. On the other side though, they are ruling out those who are the most overweight as well, which is something you’d think we’d very much want to look for. Honestly, the only one there I think really makes much sense to rule out is the pregnant respondents as their current BMI will not be representative of much.
But anyway, of greater interest is where this data is coming from, which it links to as from here: Behavioral Risk Factor Surveillance System (BRFSS)
Well, right off the bat there we find this description:
The Behavioral Risk Factor Surveillance System (BRFSS) is the nation’s premier system of health-related telephone surveys that collect state data about U.S. residents regarding their health-related risk behaviors, chronic health conditions, and use of preventive services.
(emphasis added)
So this is likely the same source and method that gave the data from a few years ago that was found in the above study to be inaccurate due to misreporting.
Now, unfortunately for West Virginia, while this may mean that it’s questionable that they actually have the 2nd highest obesity rate, it doesn’t mean they don’t have at least the Obesity rate they are reported to have from the report. In all likelihood, it’s a matter of everyone else (perhaps even West Virginia) happens to have even worse obesity rates than they think, and potentially there are other regions of the U.S. that are worse off relative to other states than is indicated from that report.
This does still drive home how much of a problem Obesity is in the country in general, but we should be a bit cautious about the actual reported numbers based on telephone surveys, and many regions that do not seem to have such a bad problem may be worse than it seems.