More evidence against the Obesity Epidemic

Interesting WSJ article on a new JAMA Obesity study.

Long story short, BMIs of 25-30 and 30-35 show no increase in mortality vs BMI 20-25 (which is considered "normal" weight).  In fact they showed a slight decrease in mortality, but perhaps not statistically significant.  Note that these BMI classes are "overweight" and "obese" respectively.  This is actually great news for the vast majority of Americans that occupy these BMIs.  No doubt organizations like CDC will still say "40% of US population is overweight or obese", but for the most part the data say there's *no health risk* associated with that statement!

The article also talks about the "Obesity Paradox":
The new report is the latest, and largest, to document what scientists call the "obesity paradox." Other studies have shown that people with heart disease, diabetes and other chronic health conditions tend to live longer if they carry excess pounds even though excess weight is associated with heightened risk of heart disease, type 2 diabetes and several cancers which in turn raise the risk of premature death
Imagine how silly it would be for a WSJ article to talk about a "long pants" paradox- that long pants are strongly associated with tall people, but putting long pants on short people or short pants on long people didn't make people shorter or taller!  The obvious conclusion is correlation, not causation.  Yet the article (and presumably medical literature) present this as some profound mystery, not one easily explained by a common statistical effect.

Note how careful the reporter and quoted researchers are not to offend conventional wisdom.  Even with this powerful evidence that having moderately high BMI is no health danger, saying so is openly considered anathema.  In the end, this is not science- this is ideology.  It seems like no matter what the data say, being overweight is "bad" and losing weight is "good".


Popular posts from this blog

Why Tesla's FSD Approach is Flawed

What Smart Tesla fans Get Wrong about FSD

Modelling Q4 Model 3 Deliveries