BMI Measurements Inaccurate But Still A Government Gold Standard


Everyone needs to read this NY Times article and then think about how inane the concept has become.

The Body Mass Index or “BMI” is used as a measure of a person’s body weight. If your BMI is between 18.5 and 25, you’re normal. More than 25 and you’re overweight. More than 30 and you’re obese. The measurement is based on a person’s weight and height, but it was originally created in the 1800s to measure human growth – not as a measure of a person’s ideal body weight or health. More recent studies show that people considered “overweight” using the BMI measurement are healthier than those who are at the lower end of the “normal” measurement. One study shows that likelihood of death increases with a BMI of less than 23. BMI doesn’t account for the distribution of body fat (abdominal fat is less healthy), BMI falsely classifies muscular individuals as “obese”, and even the CDC has recommended that doctors not use BMI as a diagnostic tool.
Yet what is one of the things our government requires that doctors calculate on every patient’s chart in order to meet “meaningful use” criteria?
You guessed it.
A BMI measurement.

This is what happens when inmates run the asylum.

The reason that we are being required to measure BMI isn’t because a patient’s BMI has any meaningful clinical use … it’s that the BMI can be measured. If it can be measured, it can be tracked. If it can be tracked, then people (essentially health care providers) can be manipulated and penalized if some arbitrary number on a meaningless scale isn’t reached.

Think about it. If we tried to find other substitutes for “health”, they would be difficult to calculate. How many calories does a patient eat? How much alcohol does a patient drink in a day or week? How much exercise does a patient get each day or week? There’s no standard way to objectively quantify or objectively measure any of those criteria.

Instead the government sticks with something easy to measure – even though it has no bearing on a patient’s health. With a little propaganda, the government can make all the patients who don’t know any better think that BMI really is a useful measure of health. Then, if the BMI isn’t calculated and put on the patient’s chart, it gives the government a means to reduce or deny payments to the healthcare providers.

Calculating a BMI and asserting that it is a representation of health is like measuring the number of clouds in the sky at 3PM each day and claiming that a higher number of clouds is an accurate representation of good government.

The scary thing is that another industry has been making similar assertions for years and certain village idiots just continue to believe the misinformation.

Patient satisfaction scores have long been asserted to be a surrogate measure for healthcare quality. Of course, those assertions are made by corporations which receive hundreds of millions of dollars each year from hospitals so that they can compare one hospital to another … on a statistically invalid and entirely misapplied metric. Studies prove that higher satisfaction is associated with higher healthcare costs and almost double the amount of patient deaths. Recall the story about the Texas neurosurgeon who maimed and killed patients yet who had great scores (which were suddenly removed by the Healthgrades staff when the story broke). Healthgrades knows its data are inaccurate, but persists in collecting and disseminating inaccurate and potentially dangerous information.
Junior high statistics classes teach twelve year olds that inadequate sample sizes automatically prevent you from making valid conclusions from the results. Want a real life example? Open up a pack of skittles, take out 5 pieces of candy, note the proportion of colors, and then see if those proportions match the proportions of colors left inside the pack.

Despite the woefully inadequate sample sizes and scientific evidence showing that these measures have no bearing on patient outcomes, the same government that relies upon BMI measurements as a representation of health is going to rely upon patient satisfaction scores as a measure of healthcare quality … and will reimburse hospitals less for care when they have lower satisfaction scores. Hospital administrators and hospital governing boards swallow this obviously inaccurate and misleading information like high school kids sucking beer through a beer bong — all in the name of profits with little regard to the adverse effects on patient health.

It is refreshing to see that hospitals are starting to be held accountable for these decisions. It is easy to prove administrative negligence and hospital board liability when bad faith actions harm patients so that hospitals can earn more money.

After all … the sun is shining. That means that BMI measurements and payment for satisfaction are bad government policies that no one should follow.

I’m a scientist. I know these things.

Review: marijuana and health

potleaf205-thumb-205x2053 out of 5 stars

Adverse Health Effects of Marijuana Use. Volkow ND et al. N Engl J Med 2014 Jun 5;370:2219-2227.

No abstract available

This review article by Dr. Nora Volkow and her colleagues at the National Institute on Drug Abuse is in most respects a rather fair and balanced look at the potential adverse effects of marijuana use. Among the potential negative effects discussed are:

  • impaired driving ability
  • impaired brain development during childhood and adolescence
  • impaired school performance with increased risk of dropping out
  • marijuana addiction and withdrawal

The authors reasonably state that other often-mentioned potential adverse effects, such as progression to using other illicit drugs and increased risk of mental illness, are difficult to evaluate because of multiple confounding factors.

I completely agree with two of the authors’ major points:

  1. Adolescents can suffer negative psychosocial consequences from using marijuana.
  2. Driving while under the influence of THC is dangerous.

However, I think that the impact and coherence of the paper are vitiated by several factors. As this is a broad survey of the medical literature, the authors cite many studies to support their premises and conclusions. Unfortunately, they do not seem to analyze these sources carefully, and — without the reader going back and critically evaluating each reference — it is impossible to determine whether or not the argument is based on good science. In particular, I looked at some of the articles cited to demonstrate that a large number of users become addicted to marijuana (according to dependence criteria in DSM-IV.) This argument seems based on some large-scale surveys with minimal methods and controls.

In addition to the medical review, this article has a political component. The authors make the following simple argument: since marijuana has adverse health effects on young people, and since decriminalizing or legalizing marijuana will increase its use, therefore such decriminalization would increase adverse health consequences. This reasoning is suspect because no state is proposing that marijuana use should legal for children or adolescents. It will be interesting to see what happens in Colorado and Washington in their experiment with legalization.  In addition, the authors do not consider the adverse social and health effects of criminalizing possession of small amounts of marijuana. Certainly having a criminal record or spending time in jail can have significant negative consequences.