Truace
Subscribe

AI-generated caregiver education about tetralogy of Fallot

Source article: A Comparative Evaluation of ChatGPT and Google Search Output for Caregiver Education in Tetralogy of Fallot

Introduction Caregivers of children with tetralogy of Fallot (TOF) frequently seek medical information online to better understand their child's diagnosis, treatment, and long-term care. Recently, generative artificial intelligence (AI) tools, including Google AI Overview and ChatGPT, have emerged as widely accessible sources of health information. However, the quality, readability, and accuracy of these platforms for caregiver education in congenital heart disease remain poorly characterized. Methods Forty freq…

By the Truvace desk2 min read1 sourcePeer-reviewedTRV-2026-1750· cite & verify
A Comparative Evaluation of ChatGPT and Google Search Output for Caregiver Education in Tetralogy of Fallot
Cardio Surgery 1 by Dominique Cappronnier. CC BY-SA 3.0 · https://creativecommons.org/licenses/by-sa/3.0
Trace impact readingPositive state
P 67The P score combines the specificity and measured human impact of the grounded problem claim with the strength of this Trace’s cited sources.

Both sides are scored from claims and sources, not community votes.

G 72The G score combines the specificity and measured human impact of the grounded gain claim with the strength of this Trace’s cited sources.

In brief

Researchers gathered 40 frequently asked caregiver questions about tetralogy of Fallot using Google's People Also Ask feature. They submitted each to ChatGPT-5.5 and to Google, yielding 80 responses. The Google output was 37 AI Overviews and three organic search excerpts. Responses were scored for length, readability, accuracy, and educational quality.

The results matter because caregivers of children with congenital heart disease commonly look for information online, and generative AI is now an easy source. The study suggests ChatGPT-5.5 may give more accurate and higher-quality information, but its harder reading level could limit usefulness. Whether caregivers actually understand or benefit from the answers was not tested.

Main points

  1. The study used 40 caregiver questions about tetralogy of Fallot drawn from Google's People Also Ask feature, producing 80 responses in total.
  2. ChatGPT-5.5 responses were significantly longer than Google output, averaging 327.9 words versus 192.5 words.
  3. Two independent reviewers rated accuracy on a 5-point Likert scale, and one reviewer assessed quality with a modified EQIP instrument.
  4. ChatGPT-5.5 scored higher on accuracy and modified EQIP, but its responses had higher calculated reading-grade levels.

The gain

Under the study's prompting conditions, ChatGPT-5.5 answers to caregiver questions about tetralogy of Fallot received higher reviewer-rated accuracy and modified EQIP educational quality scores than Google search output.

The problem

ChatGPT-5.5 responses were longer and had higher calculated reading-grade levels than Google search output, which may make them harder for caregivers to read.

The rundown

The comparison pitted ChatGPT-5.5 against Google search output, which consisted of 37 Google AI Overviews and three organic search excerpts. Each of the 40 questions was entered into both systems.

Readability was measured with Flesch Reading Ease and Flesch-Kincaid grade level, alongside word count. These calculated measures showed ChatGPT-5.5 text was longer and sat at a higher grade level than Google's output.

Quality was judged by people: two independent reviewers scored accuracy, and a single reviewer applied the modified EQIP tool for educational quality. The authors frame the findings as applying only under the study's prompting conditions.

What this doesn’t fix

The authors state the findings do not establish improved caregiver comprehension, because comprehension was not directly assessed, and the results apply only under the study's specific prompting conditions.

Sources

  1. Peer-reviewedCardiology and Therapy2026-10-07

The debate