I am amused by this chart on international obesity rates:
Every country in the top 15 uses weights based on measured data. But what happens when you ask people to self-report their weights (red asterisks in the chart)? Their countries all end up in the bottom 20.
This is why anything based on self-reporting should be taken skeptically. It's not always wrong to use self-reported data if it's done carefully, but it's always likely to be biased by all the usual foibles of human nature.
Start making more money weekly. This is valuable part time work for everyone. The best part ,work from the comfort of your house and get paid from $10k-$20k each week . Start today and have your first cash at the end of this week. Visit this article
for more details.. https://createmaxwealth.blogspot.com
I knew it. I dug into the OECD sources and they're going off of BMI for obesity. Page 6 here: https://www.oecd.org/health/health-systems/Obesity-Update-2017.pdf
Fuck that.
BMI has its uses as a very rough look at someone's weight, but it's absolute garbage for something like this. I'm actually a great example: I'm in pretty decent shape, but rate as a 25.8 BMI, which is classified as overweight. I actually have trouble finding pants in my size because I'm too small for most of the pants that get made, not too large.
Sure, not everybody is me - but I'm average height (5'9" to 5'10"). I barely have any "extra" on me at all. Perhaps 10 pounds at the very most (I can pinch a little pudge, but not enough to even make a roll out of). The issue is that I have a moderate amount of muscle. I'm not even bulked up or huge. I can't bench press even 2/3 of my own bodyweight, and I'd probably max out at a 35ish pound bicep curl, and I've never done a squat in my life. I buy medium in everything, and sometimes a small.
BMI is a stupid measurement. I have no doubt that Americans are goddamn fat, in general, but since this is using BMI to say that 40% of Americans are obese, I'm calling bullshit on the actual percentage. For all of the countries.
beat me to it. as soon as i saw they were using height and weight data in the graphic, i figured BMI, but any measure using just H and W is garbage for determining obesity. like you, my BMI is "overweight" currently because i've been getting in better shape for the past year or so and added muscle mass. when i was in worse shape and i'm had a higher bodyfat % is was in the "normal" range.
You folks are missing an important point here-- you do not need the most precise measurement for obesity for this analysis. Put another way, if you used a different metric, do you really think the US wouldn't come out number 1?
When you aggregate data like this, usually the only really important thing is that every item is determined by the same process. Which is why Kevin's point about the asterisked countries is so important. But what is generally unimportant is whether the process that is used is the best possible process for determining the conclusion, because what is important is RELATIVE ranking and not absolute accuracy.
Here's an example I like to use from college football. From 1987 to 2000, Florida State finished in the A.P. Top Five every year. Is the AP Top Five the best way or even a statistically valid way of determining who the top 5 football teams are? No. It's a vote. But nonetheless, the Florida State statistic is meaningful, precisely because it isn't dependent on whether any specific year's FSU team was really the fifth best team in the country but rather portrays the sustained excellence that is necessary to finish in the top 5 year after year.
So it really, really, doesn't matter here if some people object to BMI as the proper measure of obesity. The chart still demonstrates the real phenomenon that America is more full of overweight people than other countries
My point is not that these rankings are wildly off, and we have many other datapoint showing exgtremely high obesity rates in the US. It's a severe public health problem and one I don't think anyone would argue otherwise. My point is that BMI should be retired as a metric for obesity because of it's well documented shortcomings.
It's useful precisely for stuff like this.
This argument reminds me of the wrist measurements used to calculate rates of malnutrition in children. It isn't a good measure for individual cases of malnutrition (I have very tiny wrists, for example, but have never been malnourished), but is a good indicator for whole communities of children. Genetically tiny or sturdy wrists are pretty evenly distributed across the world, so variation in aggregate numbers is likely nutritional.
A lot of people here seem to be excited about BMI accuracy at the border between normal and overweight, when this chart is actually about people who are quite a bit larger than that. Plus, there really isn't any reason to doubt the national ranking of measured BMI. If the measure is off, it's equally off for everyone.
No, we're not missing the point.
We're going off on the clickbaity-ness of charts like this. Like I said - I'm sure the US is number 1 in obesity. I'm also sure the chart would look slightly differently. I also don't know that it really changes the "point" of this post (which is really obvious to anyone with half a brain).
But FFS with the BMI propagation.
We can't get rid of useful statistics that allow us to measure stuff in populations just because it makes some people feel bad, especially since there really isn't a good replacement in this context.
If someone is using BMI to label a healthy person unhealthy, that's the time to complain about it. But not here.
because our topics of conversation are limited to those you approve of?
Because BMI is probably the most useful way of measuring obesity among a population, even if it isn't a perfect measure of individual obesity.
+1
It is NOT stupid if used appropriately: as a statistical measurement for groups or for looking for trends in groups long term. Which is precisely how it is used here.
For individuals it is still useful. If your physician sees 20 people with a body max index of 26, it's likely that 19 of them will need to lose weight to improve their health and reduce their risk for a number of negative things. If he sees a weight lifter with a BMI of 26 he'll think, "The BMI is a little high on paper, but this person is a weightlifter so he's fine because it *says that everywhere BMI is used.*"
Also if you look within the US the rates vary drastically. Colorado at 23.8 is about the same as Germany. While most southern states are off the charts obese.
The biggest difference I noticed when moving back to Washington from Texas was how much thinner everyone was on average.
So true. After 30 years in Texas I moved to Colorado then France. What surprised me the most wasn't the rampant adult and elderly obesity in Texas, but just how rare it was to see obese kids in Colorado and France. Shockingly different, even though I roughly knew these stats when we moved.
Within a year or moving to France me and my wife both dropped several pounds even though we added a decent amount of wine to our regular diet.
Smaller food portion & walkability will do that...
I thought the problem with BMI is that it's the appropriate measure for a population but not for an individual.
The reason being that a population averages over people with different fractions of muscle and fat, neither of which are well defined for any particular person by this measure. We see this in the examples above.
What measure of obesity would you use for a population?
The best measures for individuals and populations would be the one(s) that best predict adverse health developments including heart disease and Type II diabetes, and is/are objectively measurable by a third party, ideally a health professional. Getting a truly random sample would be challenging, particularly in countries without universal healthcare coverage. Some form of stratified sampling with corrections for subgroups might be better if census data is sufficient for adjustments.
Pingback: We’re #1 ! | MADE IN AMERICA
8 of the top 10 highest per-capita sugar consuming countries are in this list. Outliers: Belgium and Netherlands.
am I reading this right?
Japan and Korea are by far the lowest and they use actual data than self reported weight?
They eat more white rice than virtually anyone. Yet they have low weight.
SO is the Atkins diet wrong or is there something genetic in "east Asains, especially those from China, Vietnam, Korea, and Japan"? (This is the way the NY Times describes ORIENTAL people, [people who look like my wife and daughter]. We really need a new word for those few billion people.)
I get so tired of all the "BMI is WORTHLESS! WORTHLESS, I say! Because I" yada yada yada yada.
People who are outliers may benefit from different measurements. Body fat. Aerobic capacity. Blood chemistry. Strength. What. Ever. If you're "exceptional", good for you. But BMI is a pretty good indicator for a huge chunk of the population. It's a height-weight chart in one number!! No wonder it has caught on so fast.
bouncing_b has a good point: it's maybe most useful when applied to populations.
BMI started being used 50 years ago. Yes, it is useful for populations more than for individuals. Even so a lot of individuals with a BMI>25 would, fifty years ago, indeed have been considered overweight. They don't feel so today because everyone around them is even fatter.
Self-reporting doesn't work for that reason too. Our standards have shifted. A lot of people with a BMI>30 don't feel obese at all, just overweight compared to others. All of which goes to say that BMI is actually a pretty good measure. It is cruelly objective and sometimes we need that.
The issue with BMI is that the "Target" and the "Obese level" - were set 200 years ago using data from the population THEN
The result is that the actual "lowest mortality rate" - is firmly in the overweight zone - almost "obese"
Having the "target" set far too low - unattainably low - means that people who could drop a bit of weight don't
I don't think so. The categories overweight and obese were created much later than the BMI itself and shifted over time. What we have now is the 1990s WHO categories (and the 1998 USA ones). (Some Asian countries set the bar for overweight at a lower level.) So there is a time lag yes but not 200 years. Also, while the lowest mortality rate is (depending on the study) partly or largely in the overweight category, it is never "firmly" so.