Does the Edmonton Obesity Staging System Hold Up in the Real World?

Back in 2008, when I first proposed what is now widely referred to as the Edmonton Obesity Staging System (EOSS) as a means of classifying the severity of obesity beyond BMI, I could not have imagined how readily this concept would be embraced by those of us interested in clinical obesity management. 

A current PubMed search reveals almost 60 published articles on the use of EOSS to determine risk and outcomes in a variety of populations and settings, consistently showing that EOSS does a far better job of predicting morbidity, mortality, surgical complications, fertility rates, occupational function, hospital stay, and even cost than BMI. 

But how practical is the application of EOSS to real-world data in primary care?

This is now the topic of a paper by Rukia Swaleh and colleagues from the University of Alberta, published in CMAJ open.

The paper describes the development of a clinical dashboard that calculates and displays the relation between BMI and EOSS and the prevalence of related comorbidities based on cross-sectional data in over 30,000 patients within the Northern Alberta Primary Care Network and the Canadian Primary Care Sentinal Surveillance Network, who were at least 18 years of age with BMI between 30 and 60 and visited a network clinic between July 2016 and July 2019. 

This EOSS dashboard provides an interactive tool that readily identifies and characterizes patients based on EOSS stage within a primary care practice, thus prompting better surveillance, prevention, and management of patients with obesity and related complications. 

Applying these analyses to the study population revealed several interesting findings.

Firstly, BMI class distribution appeared to be rather similar across EOSS stages, confirming once again that BMI is not a useful or reliable measure of obesity severity. 

Secondly, age alone described 31% of the variation in EOSS stages. In contrast, sex and BMI explained very little of the variation in EOSS stages, together accounting for just over 1% of the variation.

Thirdly, hypertension, dyslipidemia and diabetes (and their complications) were the leading drivers of higher EOSS stages.

The EOSS tool provides several useful features that can be used to further characterize and analyse patients down to seeing tests, medical prescriptions, and last date of visit for individual patients based on EOSS stage. 

Thus, this paper not only shows that one can develop electronic EOSS tools based on real-world practice data but also shows how these data can be used to better manage patients with obesity and related complications in primary practice. 

Berlin, D