TCS Daily

Standard Deviance

By Jonathan Robison, PhD, MS - March 18, 2004 12:00 AM

Well, it's Spring Break time and the promotion of weight loss with its associated array of products and services is reaching a fever pitch ("Lose 10 pounds in just two weeks"; "Melt away fat while you sleep"). Amid all the misrepresentation and hype, how can anyone really know how much they should weigh?

In our culture, whether a person's weight is considered acceptable or healthy is most often determined by comparing that weight to a set of established standards. The original source for these standards was the Height/Weight Tables developed by the Life Insurance Industry in the late 1890's in an attempt to gauge the relationship between weight and mortality for managing risk.

Today, the most widely used standard for determining what is and is not an acceptable weight is an extension of the weight-tables concept, the BMI or Body Mass Index. BMI is calculated fairly simply from a formula that uses a person's height and weight. According to the latest recommendations, people with BMI's under 25 are considered to be at an optimal/healthy weight, people with BMI's between 25 and 30 are considered to be overweight and people with BMI's over 30 are considered to be obese.

Despite their almost ubiquitous acceptance by the health establishment over the last century, strong arguments have been made that the height/weight tables were flawed from the beginning. Some of these arguments concern issues of scientific measurement and statistics that are beyond the scope of this article and have been covered in depth in other places. (1,2)

But some of the problems with these tables are so obvious that they border on the ridiculous. For example, the tables showed that the modest increases in weight that typically occurred as people aged were associated with increased mortality. So it was assumed that the weights that corresponded to the lowest death rates should stay the same as people age (i.e. people should not gain weight as they get older). Unfortunately, the data on which the tables are based do not support this assumption. In fact, it turns out that weights associated with the lowest mortality actually increase as people get older. So, while it is true that older people gain weight and die earlier, by failing to acknowledge the change in weights that are associated with lower mortality as people age, the most important thing we learn from these tables is that as we get older we are more likely to die -- not exactly rocket science!

Another example has to do with the use of the notorious frame sizes. When looking at the actuarial data, statisticians from the life insurance industry noticed that for any given height there was a considerable range of weights (30-40 pounds) that was associated with low mortality. Because this did not fit with the concept of a specific ideal or desirable weight, they concluded that the wide range of weights must be explained by differences in skeletal frame sizes. So, they arbitrarily split the weight range into thirds with smaller frames being connected with the lighter end of the weight range, medium frames with the medium weights and large frames with the heavier weights. Though this may seem logical, the process suffers from a serious lack of validity most clearly stated by one of the world's leading researchers in the area of nutrition and health, Dr. Ancel Keys, who described the height/weight tables as:

"Arm-chair concoctions starting with questionable assumptions and ending with three sets of standards for 'body frames' that were never measured or even properly defined." (3)

The BMI is fraught with many of the same problems that are associated with the height/weight tables. Although support for the use of this measurement is often based on the supposed relationship between BMI and total body fat, it turns out that BMI is not a good predictor of total body fat in individuals. (4) Perhaps most importantly and contrary to traditional dogma, a growing body of research supports that the relationship between a person's health and their weight or their BMI is unclear at best. This sentiment was reinforced in a recent editorial in the New England Journal of Medicine, when the two senior editors of the Journal, both physicians, stated:

"the data linking overweight and death, as well as the data showing the beneficial effects of weight loss, are limited, fragmentary, and often ambiguous." (5)

Much of the scientific reality of these standards and measurements has been lost in our thin-obsessed culture. This is not to suggest that there are no health risks associated with increased weight. But it does suggest that determining what you should weigh is more complicated than commonly believed.

In reality, the answer to the question, 'How much should you weigh?' is that it's largely an illegitimate question. The notion that you should weigh a specific amount as determined by any standardized chart or table is an oversimplification since it ignores too many variables and distracts from the larger question which is, 'How do I become healthy and maintain health?'


1. Schroeder, C. R. Fat is not a four-letter word. Minneapolis, MN: Chronimed, 1992.

2. Gaesser, G. A. Big fat lies: The truth about your weight and your health. Carlsbad Ca.: Gurze Books, 2002.

3. Keys, A. (l980). Overweight, obesity, coronary heart disease and mortality. Nutrition Reviews, 38, 297-307.

4. Kline, G. Analyzing BMI: Can It Measure Individual Risk? Healthy Weight Journal 2001;Jan./Feb:10-13.

5. Kassirer, J. P., & Angell, M. (1998). Losing weight - An Ill-fated new year's resolution. New England Journal of Medicine, 338(1), 52-54.


TCS Daily Archives