Obesity and overweight represent a critical challenge to contemporary public health. Although the body mass index (BMI) has served as the standard tool for the clinical assessment of obesity for decades, medical research confirms that it fails to provide a comprehensive stratification of an individual’s metabolic health and pathological risks.
To address this challenge, researchers at the Precision Healthcare University Research Institute at Queen Mary University of London conducted a large-scale study and developed a more precise model for predicting the risk of obesity-related complications—OBSCORE—which utilizes machine learning algorithms to assess risks with significantly higher clinical accuracy.
The objective of the study was to create an assessment tool capable of classifying individuals with overweight and obesity not merely by body mass, but based on a diverse array of health indicators. Researchers estimate that individuals with identical BMI often possess disparate health risks, a phenomenon that cannot be explained by the weight-to-height ratio alone.
OBSCORE was developed based on the analysis of data from 197,264 participants (median age 58 years) registered in the UK Biobank database, all of whom had a BMI of 27 or greater. Utilizing artificial intelligence and interpretable machine learning methods, the scientists evaluated over 2,000 demographic, clinical, biochemical, and lifestyle-related parameters.
Approximately 48% of the participants were female; nearly 10% had been diagnosed with type 2 diabetes (T2D) at baseline, and 4.4% had a history of major adverse cardiac events.
The final prognostic model incorporated 20 significant indicators (12 of which were also used in the SURMOUNT-1 clinical trial, which assessed the efficacy of tirzepatide for the treatment of obesity), including age, sex, waist-to-height ratio, hypertension status, cholesterol measures, glycated hemoglobin (HbA1c), renal function markers, smoking status, history of cardiovascular disease, long-standing illnesses, self-rated health status, pain (chest, abdominal), and other clinically significant factors.
Based on the analysis of health indicators, the model enabled risk stratification and the prediction of 18 clinical outcomes, including type 2 diabetes, hypertension, coronary artery disease, heart failure, atrial fibrillation, stroke, chronic kidney disease, gout, obstructive sleep apnea, metabolic dysfunction-associated steatotic liver disease, liver cirrhosis, gallbladder disease, gastroesophageal reflux disease, osteoarthritis/arthropathy, venous thromboembolism, obesity-related cancers, chronic obstructive pulmonary disease, and all-cause mortality.
“Overweight and obesity have become a major global health challenge, affecting roughly 60%-70% of adults in many Western countries, with prevalence continuing to rise in most parts of the world. Accurately identifying individuals at the highest risk of developing obesity-related complications is therefore a challenge, but essential to ensure earlier monitoring, interventions, and improved health outcomes. Our study developed and validated a risk prediction model that can stratify individuals into high- and low-risk groups based on their likelihood of developing 18 obesity-related health conditions, including type 2 diabetes, heart disease, or kidney disease,” stated lead author Kamil Demircan, MD, PhD, a postdoctoral researcher at the Precision Healthcare University Research Institute, Queen Mary University of London.
Researchers divided participants into five risk quintiles using OBSCORE predictions. Notably, as all participants had overweight or obesity, even those in the lowest-risk quintile had an elevated BMI. In 12 of the 18 outcomes, OBSCORE demonstrated substantial risk stratification. Specifically, in the highest-risk group, the probability of developing these conditions was 10 times or more higher compared to the lowest-risk group.
Particularly sharp differences were observed in cases of chronic kidney disease (CKD), gout, and type 2 diabetes. The probability of developing these conditions was 89-fold, 36-fold, and 42-fold higher, respectively, compared to the low-risk group.
Regarding cardiovascular mortality, the risk in the highest-risk group was 47-fold higher than in the lowest-risk group. Over a 10-year follow-up period, the cardiovascular mortality rate in the highest-risk quintile was 5.7%, compared to 0.1% in the lowest-risk quintile.
When applied to participants in the SURMOUNT-1 clinical trial, OBSCORE showed that weight loss achieved with tirzepatide was distributed similarly across baseline risk groups; that is, the effect of weight reduction was comparable in both high- and low-risk patients. Furthermore, predicted risk for obesity-related complications decreased across all groups following treatment, indicating that the intervention had a positive impact not only on anthropometric parameters but also on clinical prognosis.
According to the study authors, OBSCORE does not represent an attempt to replace BMI, but rather a significant supplement to it. The model may be utilized as a tool for identifying high-risk individuals and planning timely, targeted, and more intensive preventive and therapeutic interventions.
Despite promising results, the researchers note that before OBSCORE can be implemented in broad clinical practice, its validation in diverse populations is necessary, particularly in younger age groups and other demographically and geographically varied cohorts. However, the presented study underscores the fact that evaluating obesity based solely on BMI is no longer sufficient, and modern medicine is moving toward more individualized, risk-oriented approaches.
Source: medscape.com
