8
votes
What physicians get wrong about the risks of being overweight
Link information
This data is scraped automatically and may be incorrect.
- Authors
- Katia Savchuk, Mandy Erickson, Krista Conger, Nina Bai, Hanae Armitage
- Published
- Sep 5 2023
- Word count
- 863 words
Very important to consider that this article is talking about overweight, but not obese individuals. Obesity is still clearly linked to reduced life expectancy, and almost 40% of Americans aren't just overweight but obese.
My main concern about normalising obesity is that it's so strongly linked to socioeconomic status, poor people are much more likely to be overweight and have diabetes. If we stop seeing obesity as a problem, then we in part sweep away concerns about poorer people having access to healthy diets and active lifestyles.
Note that loss of appetite and losing weight unexpectedly often means there’s something wrong with you and you should see a doctor. You can also lose weight just by being sick and losing your appetite. I’ve lost a few pounds by getting the flu. (To flip that, gaining some weight might sometimes mean you weren’t getting sick and were enjoying life?)
This makes weight changes a confusing signal by itself, if you don’t know its cause. I assume the statisticians are aware, but I wonder how well they handled it?
Most longitudinal studies involve a large enough cohort and a long enough time-frame that this kind of noise just averages out or ends up affecting the results very little. Changes of a few lbs are unlikely to move what category you fall into unless you're on the fringe of two categories. A good statistician will be looking at the effect of BMI on mortality as both a continuous and categorized variable to fully understand what's going on and to minimize any bias that may occur due to methodology.
Big issue I see with this claim immediately: the article exclusively focuses on BMI, rather than body fat percentage.
Of course it does, we're talking about physicians here. Not only is body fat percentage typically not measured, it's not used in the clinic because BMI is good enough for most patients. Nearly all medicine guidelines and general rules on healthiness are based on large studies which are abstracted from individual genetic makeup and expression. Perhaps more importantly, existing literature suggests that a slightly above average (well, average for when the studies were conducted at least) body fat percentage is often associated with lower overall mortality because a mix of factors bring it both up and down. Your risk of coronary events goes up, but your risk of dying to trauma such as in a car accident goes down - the net result is that most metastudies show a small negative mortality rate with the overweight BMI category, which is what this article is focused on.
It's important to note that studies of weight-associated mortality may not be wholly accurate, as individuals tend to lose a considerable amount of weight in the time leading up to their death. If weight is measured shortly before or after their demise, it may lead to them being placed in a weight category unrepresentative of their condition prior to their illness.
Really depends on the study itself. Many of these studies (if not most) involve surveys or longitudinally tracking individuals in which the average BMI over a time-frame is used and mortality is determined via registered death databases. The BMI is not assessed at the time of death, or even near the time of death. As always, however, we should be paying attention to how studies are designed.