aceassessing clinical remission in Japanese asthma patients using EMR data extraction including LLM
Source article: Is It Feasible to Assess Clinical Remission in Asthma via a Customized Data Extraction Approach from a Real-World Clinical Electronic Medical Records Database in Japan? A Retrospective Real-World Study
Abstract: Objective : Clinical remission (CR) is an ambitious and attainable treatment goal for patients with asthma. This retrospective, observational study determined the feasibility of assessing CR using an electronic medical record (EMR) database. Methods : Data from the JAMDAS database were analyzed for patients with asthma initiating fluticasone furoate/umeclidinium/vilanterol single-inhaler triple therapy between August 18, 2020 and December 31, 2024. Asthma Control Test (ACT) scores were extracted using structured…
Contested: both sides are scored from claims and sources, not community votes.
Hospital Universitari Doctor Peset, València 08 by 19Tarrestnom65. CC BY-SA 4.0 · https://creativecommons.org/licenses/by-sa/4.0
Researchers tested whether clinical remission in asthma could be measured from routine electronic medical records in Japan by analyzing 32,258 patients starting fluticasone furoate/umeclidinium/vilanterol triple therapy in the JAMDAS database from 2020 to 2024, extracting ACT scores with SQL and exacerbation-related hospitalizations and emergency transports with a large language model.
The work matters because it demonstrates a practical AI-assisted method to monitor an ambitious treatment goal in real-world care, yet it also reveals that sparse documentation of ACT, FEV1%/FVC, and FeNO in everyday practice sharply reduces the number of patients who can be evaluated, leaving uncertainty about how remission rates would look in the full population.
- Study analyzed 32,258 patients with asthma initiating fluticasone furoate/umeclidinium/vilanterol single-inhaler triple therapy between August 18, 2020 and December 31, 2024 from the JAMDAS EMR database.
- ACT scores were extracted using structured query language while hospitalization and emergency transport data for exacerbations were extracted using a large language model.
- Among patients with 3-month ACT data (2.9% [n = 934/32 258]), 98.6% had no exacerbations and 81.0% had no OCS use, with 41.1% meeting all CR components at 3 months and 69.3% at 18 months among evaluable patients.
Researchers were able to extract asthma control and exacerbation data from routine EMRs using SQL and a large language model, suggesting it may be feasible to assess clinical remission in Japanese patients initiating triple therapy.
Routine clinical recording of remission components was sparse, with only a small fraction of eligible patients having ACT, FEV1%, or FeNO data, limiting availability of EMR data for remission assessment and generalizability.
The rundown
The retrospective study used the JAMDAS database to evaluate patients initiating single-inhaler triple therapy, applying structured query language for ACT scores and a large language model for hospitalization and emergency transport as exacerbation proxies, then applying the Japanese Practical Guidelines for Asthma Management 2024 definition of OCS-free, no exacerbations, ACT ≥ 23 at 3-month intervals to 18 months.
Results showed high rates of no exacerbations and no OCS use among the small evaluable subset, but overall data capture was low across follow-up, prompting the authors to conclude feasibility is suggested while noting recording gaps for ACT and lung function tests constrain EMR-based remission assessment.
CR assessment was constrained by infrequent routine recording of key components in clinical practice, with very low availability of ACT, lung function, and FeNO data, limiting generalizability of findings to the broader eligible population.
Sources
- Peer-reviewedJournal of Asthma2026-09-15
How should this claim be treated?
The debate