On the Level

Objective Numbers for Health, Nutrition, and Obesity?

Objectivity is something that gets a lot of attention here. We prize, pursue, and advocate for it. Especially in matters of health, nutrition, and obesity. Conventional wisdom suggests that objective numbers for health outcomes are important tools for this pursuit. But are we kidding ourselves when we rely on numbers for objectivity?

Professor Deborah Stone says we should pay attention to the subjectivity of what we count and measure. Because what we decide to count is a value judgment. And if we don’t count or measure something, then it doesn’t count. The word count has two distinct meanings, she reminds us:

“In fact in every language I know, the verb to count has both the sense of tallying up things and of being important. So counting is a way we decide what’s important.”

Thus, she says, numbers are not purely objective facts. They reflect human judgments.

The Eternal Problem of Measuring Obesity

This perspective helps when we consider the eternal struggles we have with measuring health, nutrition, and obesity. For example, BMI stands out. This simple measure is a pretty good screening tool for obesity. Epidemiologists love it, because it gives them a practical measure for estimating the prevalence of obesity.

In the clinic, it’s not so hard and fast, though. At a BMI above 30, a person usually has clinically significant obesity. But some do not. It takes more than this simple, singular number to evaluate a person’s metabolic health.

On top of that, we have the problem of weight bias and healthism. Because of cultural bias that tends to blame and shame people with obesity, this diagnosis can take on a significance that that is more than just medical. And thus, BMI becomes a hated number. It can even get swept into arguments about racism.

Similar problems plague us when we try to measure dietary patterns and the healthfulness of different foods. Global measures of health are hardly simple, either.

No Substitute for Rational Discourse

Nonetheless, we persist in the belief that objectivity is important. An explicit discussion of both facts and values can help to promote objectivity and better policy making. We all have biases and values. But when they remain implicit and unspoken, consensus is difficult to build. Policies may not work so well as a result.

In the words of Professor Stone, “Counting is no substitute for talking.”

Click here for a recent interview with Stone about her new book, Counting: How We Use Numbers to Decide What Matters. For her 2017 James Madison Award Lecture on this subject, click here.

On the Level, photograph © Clint Budd / flickr

Subscribe by email to follow the accumulating evidence and observations that shape our view of health, obesity, and policy.


October 14, 2020

One Response to “Objective Numbers for Health, Nutrition, and Obesity?”

  1. October 14, 2020 at 9:05 am, Mary-Jo said:

    I learned a whole new appreciation for ‘numbers’ and counting doing my degree in Epidemiology. There’s a reality to measurement that lies between subjectivity and objectivity -what I shall call, ‘reasonability’. I learned the intense value of a measurable robust parameter, something that can capture truth and reality as best as possible. One just needs to have a careful, defendable rationale for the parameter. BMI, for ex.,is a single number that encapsulates height and weight and their relationship, can be a neat approximation of POSSIBLE obesity. It is used to ‘count’ cases in and across populations, but cannot be well defended to use as a measure of obesity within an individual. Another measure Is HEI, healthy eating index. It’s a neat, measurable parameter of diet quality that captures many aspects about dietary intake per ‘definitions’ from evidence-based research: recommended servings of fruit and veg, fat intake, sodium intake, for ex. It still misses a lot, though: does your food come from processed or unprocessed sources, home cooked or restaurants, etc. So, HEI can be used to assess population dietary quality, to compare between people or groups, but for analysis of individual diet, it can be lacking.