Ideal body weight

What is an ideal body weight? How could it be determined?

One approach to determine ideal body weight is to use one's personal opinion. But the problem is that everyone's opinion is subjective. As the Dude says in the Big Lebowski: "That's just, like, your opinion, man." As an engineer and scientist, I would prefer an approach that has more scientific rigor and quantifiable numbers.

So the next reasonable approach to determine an ideal body weight would be to statistically measure some objective or goal such as minimizing mortality in a population. One can then also measure potentially relevant factors such as a person's height, weight, and frame size. In this way, one might determine statistical correlations between the the goal and the other factors. Of course, correlation does not imply causation. Nonetheless, my personal takeaway could be that if I can control my weight, and I am reasonably near the body weight that minimizes statistical risk, then this might be beneficial. (Do check with your doctor before making any such changes!)

Naturally, one could also optimize for other target goals, such as minimizing illness or maximizing job performance, sports performance, or appearance. So the meaning of the term "ideal" depends on the goal that one is trying to achieve. But let's suppose for the sake of argument that the goal is to minimize statistical mortality risks.

Interestingly, the scientific research for determining the ideal weights with respect to mortality was done in the 1970s and 1980s. See this article from 2000 for a detailed discussion of the history of this research. In summary, researchers extracted data for height, weight, and body frame size from tables collected by the Metropolitan Life Insurance Company. Researchers then fit different equations to these data, which are believed to represent the optimal weight in a statistical sense with respect to mortality. These include formulas such as the Robinson (1983), Miller (1983), and Devine (1974) formulas. The results predicted by each formula are not identical, but generally fairly close to each other. This is because of different math modeling assumptions made by each of the authors.

These formulas have even been implemented by many enterprising websites such as this one. One simply needs to enter one's height and gender, and the calculator will determine an ideal weight estimate.

My personal interpretation of these results is that if I am significantly outside the range of weights estimated by the formulas then this would not be optimal. Therefore, I work to modify my weight to be close to or within the range.