Actionable companion · Assessment and evidence of learning

From evidence to district action: Evaluation of the Quality and Reliability of ChatGPT‐4's Responses on Allergen Immunotherapy Using Validated Instruments for Health Information Quality Assessment

AI-assisted editorial implementation guide. The district actions are proposed applications, not findings or recommendations attributed to the original authors. No additional peer review is asserted.

Evidence and context

Health-professions education context: test transfer to K–12 rather than assuming it.

Original authors’ abstract, reproduced under the linked source license:

ABSTRACT Background Chat Generative Pre‐Trained Transformer 4 (ChatGPT‐4) represents an advancing large language model (LLM) with potential applications in medical education and patient care. While Allergen Immunotherapy (AIT) can change the course of allergic diseases, it can also bring uncertainty to patients, who turn to readily available resources such as ChatGPT‐4 to address these doubts. This study aimed to use validated tools to evaluate the information provided by ChatGPT‐4 regarding AIT in terms of quality, reliability, and readability. Methods In accordance with EAACI clinical guidelines about AIT, 24 questions were selected and introduced in ChatGPT‐4. Independent reviewers evaluated ChatGPT‐4 responses using three validated tools: the DISCERN instrument (quality), JAMA Benchmark criteria (reliability), and Flesch‐Kincaid Readability Tests (readability). Descriptive statistics summarized findings across categories. Results ChatGPT‐4 responses were generally rated as “fair quality” on DISCERN, with strengths in classification/formulations and special populations. Notably, the tool provided good‐quality responses on the preventive effects of AIT in children and premedication to reduce adverse reactions. However, JAMA Benchmark scores consistently indicated “insufficient information” (median = 0–1), primarily due to absent authorship, attribution, disclosure, and currency. Readability analyses revealed a college graduate–level requirement, with most responses classified as “very difficult” to understand. Overall, ChatGPT‐4 demonstrated fair quality, insufficient reliability, and difficult readability for patients. Conclusions ChatGPT‐4 provides generally well‐structured responses on AIT but lacks reliability and readability for clinical or patient‐directed use. Until specialized, reference‐based models are developed, healthcare professionals should supervise its use, particularly in sensitive areas such as dosing and safety.

A 90-day district pilot

1. Days 1–15 · Appraise. Have the curriculum lead and a teacher read the methods and limitations. Record the study design, sample, comparator, outcomes and evidence gaps. Define a local question that the study can help inform.
2. Days 16–30 · Design. Redesign one low-stakes assessment: require a first attempt, an AI-assisted revision, a source-check log and a short oral explanation. Use the same rubric for the comparison task. Name an accountable owner and set the baseline, success threshold and budget before implementation.
3. Days 31–60 · Test. Start with a small voluntary cohort, preserve an accessible non-AI route and require teacher review. Use synthetic or de-identified examples; approve data handling and tool terms through district processes.
4. Days 61–90 · Decide. Blind-score reasoning, factual accuracy and unaided transfer on a fresh task. Record suspected misconduct separately from demonstrated evidence. Compare with baseline, document uncertainty and stop or revise if learning worsens, access is unequal or critical errors persist.

The board decision brief

Report the learning problem, evidence quality, participating population, total cost including review time, measured benefit, errors, access differences and remaining uncertainties. Make a dated recommendation to stop, revise or expand. A positive perception survey alone does not establish learning improvement.

Attribution and reuse

Ivan Cherrez‐Ojeda; Torsten Zuberbier; Gabriela Rodas‐Valero; Jorge Sanchez; Michael Rudenko; Stephanie Dramburg; Pascal Demoly; Davide Caimmi; René Maximiliano Gómez; German D. Ramon; Ghada E. Fouda; Kim R. Quimby; Herberto Chong‐Neto; Oscar Calderon Llosa; Jose Ignacio Larco; Olga Patricia Monge Ortega; Marco Faytong‐Haro; Oliver Pfaar; Jean Bousquet; Karla Robles‐Velasco. Evaluation of the Quality and Reliability of ChatGPT‐4's Responses on Allergen Immunotherapy Using Validated Instruments for Health Information Quality Assessment. Clinical and Translational Allergy (2025-12-01). https://doi.org/10.1002/clt2.70130

Companion prepared 2026-10-30. Original authorship and DOI are preserved. The abstract is an attributed source extract; the implementation plan is new AI-assisted editorial material, offered under the same CC BY 4.0 license. This guide has not undergone independent peer review.