13 comments

[ 249 ms ] story [ 799 ms ] thread
Is BMI accurate for athletic body types?

I would suspect the risk would be lessened for healthy sportsmen who are adequately rested.

During my Personal Trainer course a few years ago it was stated that an athletic build does not affect BMI
AFAIK some athletes may in fact be (somewhat) overweight according to BMI yet still are unlikely to suffer from health problems related to weight gain. An example would be ice hockey players, a large proportion of whom probably have a BMI in the overweight range but who are very fit. A relatively large proportion of their extra mass is muscle compared to non-athletes.

This is probably the minority of athletes, though, so it might be a fringe case rather than something that should be taught as a norm. I also don't know how that relates to COVID.

Terry Crews is 6'2" and 245lbs. That's an obese BMI at 30.6. Does he look obese to you? Fat is heavy, muscle is even heavier, both will cause your BMI to go up. Obviously most people don't look like him so really the BMI figure kind of stops being useful when you have significant muscle.
There are many different "athletic" body types. A distance runner doesn't look like a rugby player. There are some athletes in certain sports who have an "obese" BMI while maintaining low body fat but those are a tiny fraction of the population. (Most people underestimate their body fat percentage.)

BMI is accurate enough for a general population risk screening tool. We aren't dealing with precise numbers here anyway, just approximate ranges with significant differences between the results of various research studies.

On an individual level excess body fat is a critical risk factor for COVID-19.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7430889/

Body fat % is more accurate than BMI when it comes to gauging health risk, but it is also a lot harder to estimate/calculate so BMI is generally used to estimate risk factors. This is a problem will all BMI-based metrics, and it's worth noting that BMI is not one-size-fits-all. Different body types (muscular vs lean, for example), different ethnicities/cultures, different genders, all have different definitions of a "healthy" body and BMI isn't able to take all of that into account. That's not to say that it can't be useful, but take it with a grain of salt when it comes to interpreting BMI across different fields.
For an activity risk calculator, see the microCOVID project:

https://www.microcovid.org/

I'm curious what people think of these calculators. I wonder if people are surprised by the numbers, given that news headlines don't quantify the risk in detail. Also people may think differently about the same numbers.

I made my own covid mortality calculations last April, and was not scared off by my estimated odds, 1 in 350 per year. Using the linked UK calculator my odds were far better than I had estimated, about 1 in 16,000 per 90 days.

The activity calculator was interesting, also. I had heard that mask-to-mask interactions were based on an hour of nearness while indoors. I thought it to be an overreaction when people would wear masks outside and cross the street to avoid passing on the sidewalk.

So I looked up 10 people with no masks in a room, and it was a 1 in 200 chance per hour of catching covid given the odds of someone from my area being there who had it. Once again the risk was not as high I as thought it would be.

I'm curious what others think of the numbers from these calculators, especially those who find them alarming or surprising.

They don’t differentiate risk and impact.

The Oxford calculator seems to be personally focused, and for me, that’s I what. Reading was relatively low risk. Seems to correlate with what I’ve read.

But the microcovid calc rated something I was previously accustomed to doing regularly as “high”. For me that wasn’t intuitively accurate. I guess for an immediate community of vulnerable or averse people, then high would be right. But not me personally.

Quite difficult to disentangle the personal from community risk. Likely to lead to risk averse behaviour. And that is fine, but only if the mitigation is accurately taken into account.

It seems unlikely that would have been done.

Most useless Stat ever? What is the point of this with no exposure risk?
I think I read a few years ago that a better estimate than the BMI is the ratio of the circumference of the belly to the hips.
My wife carries the majority of her weight in her hips and legs, whereas I carry mine mostly in my abdomen. I have high density fat, and hers is low density. We both have a high BMI. I remain unconvinced that she has zero risk, whereas I have 100% risk.
There is more information here: https://en.wikipedia.org/wiki/Waist%E2%80%93hip_ratio#Indica...

Specifically

> WHR has been shown to be a better predictor of cardiovascular disease than waist circumference and body-mass index.[11] However, other studies have found waist circumference, not WHR, to be a good indicator of cardiovascular risk factors,[12] body fat distribution,[13] and hypertension in type 2 diabetes.[14]