Eye 2

Toward a More Definitive Diagnosis of Obesity

One of the few things in obesity that people agree upon is that BMI has serious limitations for the diagnosis of obesity. It’s great for epidemiologists. But it causes problems for clinicians and patients. While it’s useful as a crude measure of weight status, it doesn’t provide much insight into a patient’s clinical status. So for that purpose, two clinical staging systems have emerged. One is the Edmonton Obesity Staging System (EOSS). The other is the Cardiometabolic Disease Staging (CMDS).

This week in Obesity, a new study tells us that CMDS might do a better job of identifying patients at risk for an early death because of obesity.

Different Approaches to the Same Challenge

The EOSS and the CMDS take slightly different approaches  to the clinical challenge of distinguishing high-risk patients from low-risk patients. EOSS uses a clinical assessment of weight-related physical health, mental health, and quality of life to classify patients. But the CMDS focuses more narrowly on measures of cardiometabolic health. It uses blood pressure, lipids, glucose, and waist circumference to generate a risk score. And in fact, it was developed for predicting the diabetes risk. Not as a staging tool for obesity.

Despite this, CMDS did a better job of predicting excess mortality than EOSS. Keisuke Ejima and colleagues used data from NHANES to calculate risk scores for individuals using each of these systems. Both systems worked for predicting mortality. But the CMDS did a better job of predicting mortality and sorting people into classes with different levels of risk.

The Final Word?

This is hardly the final word on these two models. In fact, the authors make it clear that these staging systems serve different purposes. The EOSS is clearly more wholistic. It’s all about the severity of obesity – its effects on mental health, physical health, and quality of life. By contrast, the CMDS is narrowly focused on cardiometabolic risk.

So it’s really not surprising that the CMDS was more precise for predicting mortality. Remember, though, that a patient can live a long time in a miserable state. So looking solely at mortality risk is not good enough. The present study does not tell us that one of these systems is better for a more definitive diagnosis of obesity than BMI alone can provide. It may be that we need both staging systems to deliver the best possible obesity care.

Click here for the study, here for more on the CMDS, and here for more on the EOSS.

Eye 2, photograph © Thomas Tolkien / flickr

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


January 10, 2020

5 Responses to “Toward a More Definitive Diagnosis of Obesity”

  1. January 10, 2020 at 9:10 am, Stephen Phillips said:

    Thanks for recognizing the shortcomings of BMI

    BMI is a blunt instrument. that we use to diagnose obesity
    The AMA has declared obesity a disease.
    Many millions of people that are diagnosed with obesity are absolutely disease free. No hypertension, no diabetes and normal lipid profiles.
    We have been advocating for a diagnosis of obesity with no evidence of disease (Obesity NED)
    Why stigmatize millions people with a disease that they don’t have ? .

  2. January 10, 2020 at 2:06 pm, Allen Browne said:

    There are over 230 complications of the disease of obesity – some present at diagnosis and some predictably coming later in life. I don’t understand the dispute about obesity as a disease. If you have mild form, great. But those with a high BMI deserve a chance at a better quality of life, better economic productivity, and a longer life.

    • January 11, 2020 at 2:26 am, Ted said:

      Amen, Allen.

  3. January 10, 2020 at 5:36 pm, John Dixon said:

    BMI is an excellent metric if it is misunderstood. It is an excellent metric to assess total body fattens if we also know sex, age, and ethnicity. BIA scales essentially use these inputs to derive their result.

    Extreme outliers are often used to justify its failings. No body builders are outliers.

    The relationship between BMI and health risk is Totally misunderstood.

    I contend that BMI is an excellent metric for body fatness but interpreting risk requires additional information. Fat distribution, blood pressure, and basic blood tests provide excellent support. EOSS is an excellent support. But dismissing our best estimate of percent body fat is a mistake.

    • January 11, 2020 at 2:31 am, Ted said:

      Thanks, John. You’re right. BMI is a great index. But it’s only a starting point for clinical assessment. Too many clinicians don’t look beyond it and don’t really provide clinical care for obesity.