Truace
Subscribe

Readability, reliability, and quality of AI-Generated responses on pelvic congestion syndrome: A comparative analysis of ChatGPT, gemini, and perplexity

ObjectivePelvic congestion syndrome (PCS) is a vascular cause of chronic pelvic pain arising from ovarian and pelvic venous insufficiency, typically presenting as dull, aching pelvic pain that worsens with prolonged standing, physical activity, and the premenstrual period, and is particularly common among multiparous women of reproductive age. Online health information on this condition is frequently accessed by both patients and healthcare professionals. This study aimed to evaluate the readability, quality, an…

By the Truvace desk2 min read1 sourcePeer-reviewedTRV-2026-1856· cite & verify
Readability, reliability, and quality of AI-Generated responses on pelvic congestion syndrome: A comparative analysis of ChatGPT, gemini, and perplexity
"Chatbots linked" by Ashhstarr is licensed under CC BY-SA 4.0. To view a copy of this license, visit https://creativecommons.org/licenses/by-sa/4.0/.

In brief

Researchers took 18 commonly searched keywords on pelvic congestion syndrome, a vascular cause of chronic pelvic pain, and submitted them to ChatGPT-4o, Gemini and Perplexity. They then scored the responses with several readability, quality and reliability instruments.

Patients and clinicians often look for health information online, so the readability and quality of chatbot answers affect how usable that information is. The authors describe chatbots as promising but found the content too complex and of suboptimal quality. The study covers only three freely accessible platforms and does not measure patient outcomes.

Main points

  1. All chatbot responses significantly exceeded the sixth-grade reading level recommended by the AMA and NIH (p < .001).
  2. Perplexity produced the most complex responses but scored highest on JAMA, DISCERN and GQS.
  3. Gemini scored best on EQIP and on most readability indices, while ChatGPT-4o did best on CLI and ARI.
  4. The 18 keywords were derived from 25 Google Trends terms after removing duplicates and irrelevant ones.

The problem

Responses from ChatGPT-4o, Gemini and Perplexity to common pelvic congestion syndrome questions were too complex for the recommended sixth-grade level, and overall quality scores were suboptimal.

The rundown

The study began with 25 PCS-related keywords from Google Trends. After duplicates and irrelevant terms were removed, 18 remained and were submitted separately to each of the three chatbots.

Readability was calculated with indices including Gunning Fog, SMOG and Flesch Reading Ease, using two independent tools. Quality was measured with the Global Quality Score and the EQIP tool, and reliability with the JAMA benchmark criteria and modified DISCERN.

Platform results differed by measure. ChatGPT-4o led on CLI and ARI readability, Gemini on the remaining readability indices, and Perplexity gave the most complex text yet led on several quality and reliability scores. The authors conclude that clearer, more balanced and more reliable content could improve these tools' contribution to public health.

Sources

  1. Peer-reviewedPhlebology: The Journal of Venous Disease2026-10-18

The debate