Assessing Claims-Based Body Mass Index in Comparative Obesity Effectiveness Research

By
Karan R. Chhabra
Babak J. Orandi
October 27, 2026

Jama Network Open

Objective:

The article examines whether Medicare claims capture baseline BMI among beneficiaries receiving bariatric surgery or initiating GLP-1 receptor agonist therapy, and discusses the implications for comparative obesity effectiveness research.

Approach:

Study population: Hider et al examined a 20% sample of Medicare claims from 2017–2023, identifying 173,344 beneficiaries with documented obesity who underwent bariatric surgery or initiated GLP-1 receptor agonist therapy.
BMI ascertainment: The study assessed BMI recorded through ICD-10-CM Z68 codes at treatment initiation.

Key Findings:

Median BMI was 47.5 among bariatric surgery patients and 39.5 among patients initiating GLP-1 RA therapy.
ICD-10-CM Z68 BMI codes were recorded for nearly all surgical patients and approximately 84% of patients initiating GLP-1 RA therapy.

Interpretation:

The findings indicate that claims may capture baseline obesity severity more often than previously assumed, particularly for surgical patients. More consistent Z68 coding could support adjustment for baseline BMI in treatment comparisons.

Limitations:

The study established baseline BMI ascertainment but did not show that BMI changes during treatment can be tracked using Z68 codes.
Claims-based BMI was not validated against a criterion standard such as electronic medical records.
BMI is an imperfect measure of obesity severity and does not capture body composition, symptoms, or functional status.
Claims may not establish whether GLP-1 RAs were prescribed for obesity or diabetes, and residual confounding may remain.

Conclusion:

The article argues that accurate baseline severity measurement is important for credible comparisons of obesity treatments and that more consistent Z68 coding could strengthen claims-based research.