ArticleBMC musculoskeletal disorders2025
Let's chat (GPT) about adolescent idiopathic scoliosis: accuracy and reliability of chat responses to frequently asked questions.
Article in BMC musculoskeletal disorders, 2025. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Cited by 5 papers.
What it found
Each row is one number read from the abstract, on the scale the paper reported it, with its interval. Left of the dashed line favours the treatment, right favours the comparator. Under each row is the sentence it came from. New to these charts? A ten-minute tutorial.
The abstract states no effect estimate the extractor could read, or names no intervention and outcome on the map, so this paper lights no cell and moves no belief. It is still indexed, cited and linked below.
The trial behind it
Trials whose registry record cites this paper, or whose number appears in the abstract. A trial that started after this paper was published is citing it as background, not reporting it.
Neither the registry nor the abstract names a trial number. If this is a trial report, that itself is worth knowing.
Who cites it
5 citing papers in PubMed.
- Research priorities in Spine Deformity: a machine learning-based topic analysis of the journal's first decade (2013-2026).Spine deformity · 2026Article
- From Algorithms to Clinics: Recent Progress in AI for Scoliosis Diagnosis and Management.Journal of clinical medicine · 2026Review
- An expert-led benchmark using common patient questions: evaluating large language models for adolescent idiopathic scoliosis education.European spine journal : official publication of the European Spine Society, the European Spinal Deformity Society, and the European Section of the Cervical Spine Research Society · 2026Article
- Clinical Evaluation of ChatGPT-5.3 Responses to Patient-Oriented Questions on Scoliosis: A Multidimensional Expert Analysis.Healthcare (Basel, Switzerland) · 2026Article
- Evaluating large language models as tools for public health education on scoliosis.Frontiers in public health · 2026Article
Corrections and comments
PubMed lists nothing against this paper. Absence here is not a guarantee, only a check that was made.
Authors and funding
6 authors.
Funding
No grant is acknowledged in the PubMed record.
Abstract
backgroundPatients increasingly seek online medical information, with artificial intelligence (AI) chatbots like ChatGPT emerging as potential resources for adolescent idiopathic scoliosis (AIS); however, their accuracy and reliability need assessment. This study aimed to evaluate the accuracy and reliability of ChatGPT, an AI model, in answering questions related to AIS.
methodsSixty-four questions across four categories (general information, diagnosis and screening, treatment and follow-up, and quality of life [QoL]) were adapted from FAQs on professional association websites, the SOSORT consensus article, and QoL questionnaires. Two reviewers rated ChatGPT's responses on a scale from 1 (correct and comprehensive) to 4 (completely incorrect). Descriptive statistics were calculated to demonstrate the percentages of responses per score as well as the percentages of scores across categories. Each question was entered twice to assess reliability and response similarity. The percentage of responses that differed when the same query was entered twice into the system was calculated. The Cohen's Kappa statistic was utilized to assess the level of agreement between the two reviewers.
resultsOf all the responses, 53.1% were rated as "correct and comprehensive," while 34.4% were rated as "correct but not comprehensive." ChatGPT performed best in the QoL category, with 13 out of 15 (86.7%) responses rated as correct. The second-best performance was in the diagnosis and screening category, with 7 out of 13 (53.8%) correct responses, followed by the general information category, with 9 out of 17 (52.9%) correct responses. The lowest performance was in the treatment and follow-up category, with 5 out of 19 (26.3%) correct responses. Consistency in ChatGPT's responses when questions were entered twice was 76.6%. Agreement between the reviewers' scores was excellent, as indicated by Cohen's Kappa statistic (Kappa: 0.82, 95% CI: 0.59 to 1.04; p = 0.0001).
conclusionsChatGPT demonstrated strong accuracy in addressing questions related to QoL in AIS, but its accuracy in treatment-related areas remains insufficient. Therefore, patients and parents are advised to consult medical professionals rather than rely solely on AI-generated information for AIS treatment and management.
Indexed as
Identifiers
What OpenQuestion holds
Registered trials
Read under generation 80e0d062 · epoch 390. Bibliography from PubMed, PubMed Central and OpenAlex; grants from NIH RePORTER; trial links from ClinicalTrials.gov; estimates, votes and beliefs from the OpenQuestion graph.