TEMPORAL TRENDS IN ICD-10-CM OBESITY CODING ACCURACY, 2021-2025: EVIDENCE FROM THE NORSTELLALINQ LINKED CLAIMS AND EHR DATABASE
Author(s)
Spencer Friedman1, ilan behm, MPH2, Eric Mitchell, BS3, Allison Perry, PhD4.
1Norstella, Yardley, PA, USA, 2Norstella, Englewood, CO, USA, 3Norstella, Brooklyn, NY, USA, 4Norstella, New York, NY, USA.
1Norstella, Yardley, PA, USA, 2Norstella, Englewood, CO, USA, 3Norstella, Brooklyn, NY, USA, 4Norstella, New York, NY, USA.
OBJECTIVES: To evaluate whether ICD-10-CM obesity coding practices improved between 2021 and 2025 comparing annual undercoding, sensitivity, specificity, PPV, and NPV against exact EHR-derived BMI as a reference standard.
METHODS: A retrospective cohort analysis was conducted using NorstellaLinQ’s linked claims and EHR database. Adults (≥18 years) with ≥1 BMI measurement recorded between January 2021 and December 2025 were included. Exact BMI was derived from discrete height and weight fields; highest documented obesity severity during each year was used per patient. ICD-10-CM obesity coding status was assessed within ±6 months of BMI measurement across E66.xx and Z68.xx code families. Sensitivity, specificity, PPV, and NPV were calculated annually using exact BMI as the reference standard.
RESULTS: Between 2021 and 2025, the number of patients with a recorded BMI ≥30 kg/m² grew from 2,159,136 to 2,623,947. Undercoding declined steadily from 69.9% in 2021 to 62.4% in 2025, with sensitivity improving from 30.1% to 37.6% over the same period. Specificity remained stable (89.1% to 89.0%), indicating improved capture of BMI-confirmed obesity without increasing false-positive coding. PPV improved modestly from 64.0% to 67.7%, and NPV from 66.5% to 69.8%. Despite five consecutive years of improvement, over 60% of BMI-confirmed obese patients remained uncoded as of 2025.
CONCLUSIONS: Although obesity coding improved by nearly eight percentage points over five years, sensitivity remained below 40% in 2025, and the majority of BMI-confirmed obese patients still lack a corresponding diagnosis code. NorstellaLinQ's exact EHR-derived BMI provides a substantially more complete approach to patient identification for patient identification than diagnosis codes alone, with ongoing implications for RWE study design, coverage policy, and prior authorization criteria in obesity.
METHODS: A retrospective cohort analysis was conducted using NorstellaLinQ’s linked claims and EHR database. Adults (≥18 years) with ≥1 BMI measurement recorded between January 2021 and December 2025 were included. Exact BMI was derived from discrete height and weight fields; highest documented obesity severity during each year was used per patient. ICD-10-CM obesity coding status was assessed within ±6 months of BMI measurement across E66.xx and Z68.xx code families. Sensitivity, specificity, PPV, and NPV were calculated annually using exact BMI as the reference standard.
RESULTS: Between 2021 and 2025, the number of patients with a recorded BMI ≥30 kg/m² grew from 2,159,136 to 2,623,947. Undercoding declined steadily from 69.9% in 2021 to 62.4% in 2025, with sensitivity improving from 30.1% to 37.6% over the same period. Specificity remained stable (89.1% to 89.0%), indicating improved capture of BMI-confirmed obesity without increasing false-positive coding. PPV improved modestly from 64.0% to 67.7%, and NPV from 66.5% to 69.8%. Despite five consecutive years of improvement, over 60% of BMI-confirmed obese patients remained uncoded as of 2025.
CONCLUSIONS: Although obesity coding improved by nearly eight percentage points over five years, sensitivity remained below 40% in 2025, and the majority of BMI-confirmed obese patients still lack a corresponding diagnosis code. NorstellaLinQ's exact EHR-derived BMI provides a substantially more complete approach to patient identification for patient identification than diagnosis codes alone, with ongoing implications for RWE study design, coverage policy, and prior authorization criteria in obesity.
Conference/Value in Health Info
2026-11, ISPOR Europe 2026, Vienna, Austria
Value in Health, Volume 29, Issue 12S
Code
RWD121
Topic
Clinical Outcomes, Economic Evaluation, Real World Data & Information Systems
Topic Subcategory
Data Protection, Integrity, & Quality Assurance, Distributed Data & Research Networks, Health & Insurance Records Systems, Reproducibility & Replicability
Disease
Diabetes/Endocrine/Metabolic Disorders (including obesity), No Additional Disease & Conditions/Specialized Treatment Areas