Comparison of artificial intelligence and multidisciplinary team assessment for evaluating biologic decisions in severe uncontrolled asthma

Philipp Suter; Robert Greig; Rory Chan; Thomas Fardon; Paul Cadden; Arlene Shaw; Brian Lipworth

doi:10.1016/j.jaip.2026.07.049

Comparison of artificial intelligence and multidisciplinary team assessment for evaluating biologic decisions in severe uncontrolled asthma

Philipp Suter (Lead / Corresponding author)
, Robert Greig
, Rory Chan
, Thomas Fardon
, Paul Cadden
, Arlene Shaw
, Brian Lipworth

Research output: Contribution to journal › Article › peer-review

Abstract

Background
The management of severe asthma with biologics has become more complex involving multidisciplinary team meetings. Artificial intelligence has been proposed as an adjunct to aid healthcare professionals in other areas of the medical field.

Objective
Our aim is to determine if there is a role for artificial intelligence, as an adjunct to the severe asthma multidisciplinary team meeting, to offer improved consistency and efficiency in biologic prescribing.

Methods
This is a retrospective, unblinded single-centre, multisite cohort study including patients with uncontrolled severe GINA defined stage 5 asthma for biologic consideration who were biologic naïve. They were discussed at joint multidisciplinary team meetings between secondary care referral centres in the health board NHS Tayside, Scotland between January 2024 and August 2025. The multidisciplinary team biologic decisions were compared to five large language model recommendations (ChatGPT, Gemini, CoPilot, PerplexityAI and DeepSeek) using Cohen’s kappa with the multidisciplinary team meeting as the reference.

Results
Among 114 patients, the mean (SD) age was 54·5 (15·0) years, asthma control questionnaire score 2·98 (1·1), annualised exacerbation rate 5·21 (2·1). Multidisciplinary team decisions and artificial intelligence recommendations showed poor agreement (κ = 0·143–0·213). Agreement between artificial intelligence models was weak to moderate (κ = 0·348–0·586) with frequent decision making errors identified with all artificial intelligence models.

Conclusion
At the time of analysis, artificial intelligence models show poor concordance highlighting their unsuitability as either a standalone or adjunctive decision maker for biologic therapy. Multidisciplinary team meetings remain the reference standard.

Original language	English
Journal	The Journal of Allergy and Clinical Immunology: In Practice
Early online date	31 Aug 2026
DOIs	https://doi.org/10.1016/j.jaip.2026.07.049
Publication status	E-pub ahead of print - 31 Aug 2026

Keywords

artificial intelligence
asthma
multidisciplinary team meeting
biologics

Access to Document

10.1016/j.jaip.2026.07.049Licence: CC BY

Cite this

@article{ffe69be2a8354ec2abf1363445848cac,

title = "Comparison of artificial intelligence and multidisciplinary team assessment for evaluating biologic decisions in severe uncontrolled asthma",

abstract = "Background The management of severe asthma with biologics has become more complex involving multidisciplinary team meetings. Artificial intelligence has been proposed as an adjunct to aid healthcare professionals in other areas of the medical field. Objective Our aim is to determine if there is a role for artificial intelligence, as an adjunct to the severe asthma multidisciplinary team meeting, to offer improved consistency and efficiency in biologic prescribing. Methods This is a retrospective, unblinded single-centre, multisite cohort study including patients with uncontrolled severe GINA defined stage 5 asthma for biologic consideration who were biologic na{\"i}ve. They were discussed at joint multidisciplinary team meetings between secondary care referral centres in the health board NHS Tayside, Scotland between January 2024 and August 2025. The multidisciplinary team biologic decisions were compared to five large language model recommendations (ChatGPT, Gemini, CoPilot, PerplexityAI and DeepSeek) using Cohen{\textquoteright}s kappa with the multidisciplinary team meeting as the reference. Results Among 114 patients, the mean (SD) age was 54·5 (15·0) years, asthma control questionnaire score 2·98 (1·1), annualised exacerbation rate 5·21 (2·1). Multidisciplinary team decisions and artificial intelligence recommendations showed poor agreement (κ = 0·143{\textendash}0·213). Agreement between artificial intelligence models was weak to moderate (κ = 0·348{\textendash}0·586) with frequent decision making errors identified with all artificial intelligence models. Conclusion At the time of analysis, artificial intelligence models show poor concordance highlighting their unsuitability as either a standalone or adjunctive decision maker for biologic therapy. Multidisciplinary team meetings remain the reference standard.",

keywords = "artificial intelligence, asthma, multidisciplinary team meeting, biologics",

author = "Philipp Suter and Robert Greig and Rory Chan and Thomas Fardon and Paul Cadden and Arlene Shaw and Brian Lipworth",

note = "{\textcopyright} 2026",

year = "2026",

month = aug,

day = "31",

doi = "10.1016/j.jaip.2026.07.049",

language = "English",

journal = "The Journal of Allergy and Clinical Immunology: In Practice",

issn = "2213-2198",

publisher = "American Academy of Allergy, Asthma and Immunology",

}

TY - JOUR

T1 - Comparison of artificial intelligence and multidisciplinary team assessment for evaluating biologic decisions in severe uncontrolled asthma

AU - Suter, Philipp

AU - Greig, Robert

AU - Chan, Rory

AU - Fardon, Thomas

AU - Cadden, Paul

AU - Shaw, Arlene

AU - Lipworth, Brian

PY - 2026/8/31

Y1 - 2026/8/31

N2 - Background The management of severe asthma with biologics has become more complex involving multidisciplinary team meetings. Artificial intelligence has been proposed as an adjunct to aid healthcare professionals in other areas of the medical field. Objective Our aim is to determine if there is a role for artificial intelligence, as an adjunct to the severe asthma multidisciplinary team meeting, to offer improved consistency and efficiency in biologic prescribing. Methods This is a retrospective, unblinded single-centre, multisite cohort study including patients with uncontrolled severe GINA defined stage 5 asthma for biologic consideration who were biologic naïve. They were discussed at joint multidisciplinary team meetings between secondary care referral centres in the health board NHS Tayside, Scotland between January 2024 and August 2025. The multidisciplinary team biologic decisions were compared to five large language model recommendations (ChatGPT, Gemini, CoPilot, PerplexityAI and DeepSeek) using Cohen’s kappa with the multidisciplinary team meeting as the reference. Results Among 114 patients, the mean (SD) age was 54·5 (15·0) years, asthma control questionnaire score 2·98 (1·1), annualised exacerbation rate 5·21 (2·1). Multidisciplinary team decisions and artificial intelligence recommendations showed poor agreement (κ = 0·143–0·213). Agreement between artificial intelligence models was weak to moderate (κ = 0·348–0·586) with frequent decision making errors identified with all artificial intelligence models. Conclusion At the time of analysis, artificial intelligence models show poor concordance highlighting their unsuitability as either a standalone or adjunctive decision maker for biologic therapy. Multidisciplinary team meetings remain the reference standard.

AB - Background The management of severe asthma with biologics has become more complex involving multidisciplinary team meetings. Artificial intelligence has been proposed as an adjunct to aid healthcare professionals in other areas of the medical field. Objective Our aim is to determine if there is a role for artificial intelligence, as an adjunct to the severe asthma multidisciplinary team meeting, to offer improved consistency and efficiency in biologic prescribing. Methods This is a retrospective, unblinded single-centre, multisite cohort study including patients with uncontrolled severe GINA defined stage 5 asthma for biologic consideration who were biologic naïve. They were discussed at joint multidisciplinary team meetings between secondary care referral centres in the health board NHS Tayside, Scotland between January 2024 and August 2025. The multidisciplinary team biologic decisions were compared to five large language model recommendations (ChatGPT, Gemini, CoPilot, PerplexityAI and DeepSeek) using Cohen’s kappa with the multidisciplinary team meeting as the reference. Results Among 114 patients, the mean (SD) age was 54·5 (15·0) years, asthma control questionnaire score 2·98 (1·1), annualised exacerbation rate 5·21 (2·1). Multidisciplinary team decisions and artificial intelligence recommendations showed poor agreement (κ = 0·143–0·213). Agreement between artificial intelligence models was weak to moderate (κ = 0·348–0·586) with frequent decision making errors identified with all artificial intelligence models. Conclusion At the time of analysis, artificial intelligence models show poor concordance highlighting their unsuitability as either a standalone or adjunctive decision maker for biologic therapy. Multidisciplinary team meetings remain the reference standard.

KW - artificial intelligence

KW - asthma

KW - multidisciplinary team meeting

KW - biologics

U2 - 10.1016/j.jaip.2026.07.049

DO - 10.1016/j.jaip.2026.07.049

M3 - Article

C2 - 42674108

SN - 2213-2198

JO - The Journal of Allergy and Clinical Immunology: In Practice

JF - The Journal of Allergy and Clinical Immunology: In Practice

ER -

Comparison of artificial intelligence and multidisciplinary team assessment for evaluating biologic decisions in severe uncontrolled asthma

Abstract

Keywords

Access to Document

Fingerprint

Cite this