48/100

Emerging evidence

Cite
More
EmailLinkedInX

ChatGPT answers on endometriosis surgery and fertility rated more complete than Google's, study finds

Study title: ChatGPT versus Google Search-linked responses to surgical and fertility-related endometriosis questions: a cross-sectional content analysis

Evidence type
Benchmark or lab evaluation
Sample
n = 80
Sources
1 sourcePeer-reviewed
Published
2 min read

Study title: ChatGPT versus Google Search-linked responses to surgical and fertility-related endometriosis questions: a cross-sectional content analysis

A cross-sectional analysis of 40 prespecified patient-style questions about endometriosis found that ChatGPT's answers earned a higher overall clinical quality score than Google Search-linked answers: 4.91 versus 4.53, a paired mean difference of 0.38 (95% CI 0.32 to 0.45).

Read the study abstract

Objective: To compare guideline concordance, safety, actionability, and readability of ChatGPT and Google Search-linked responses to patient-facing surgical and fertility-related endometriosis questions. Methods: We performed a cross-sectional comparative digital content analysis of 40 prespecified questions across diagnosis/imaging, minimally invasive surgery, deep organ-involving disease, and fertility/ovarian reserve. For each question, one ChatGPT response and one Google Search-linked response archived on Ma…

The gain

In a content analysis of 40 endometriosis surgery and fertility questions, two blinded physicians gave ChatGPT a higher overall clinical quality score than Google Search-linked answers (4.91 vs 4.53), with no response judged harmful or misleading.

In brief

The gap came mainly from completeness (0.94 difference) and actionability (0.60 difference). Safety scores were similarly high on both platforms, and no response was judged harmful or misleading. ChatGPT responses were also somewhat longer, at 52.6 words against 38.6.

Readability was a shared weakness: average grade-level estimates for both platforms were above the sixth-grade reference benchmark. The authors conclude the findings support clinician-reviewed educational use and further assessment of whether patients actually understand the answers.

Main points

  1. The study covered 40 prespecified questions on diagnosis and imaging, minimally invasive surgery, deep organ-involving disease, and fertility and ovarian reserve.
  2. Eighty responses were analysed: one ChatGPT answer and one Google Search-linked answer per question, archived on May 31, 2026.
  3. Two physicians scored the responses independently, with platform and source identifiers concealed.
  4. ChatGPT's overall clinical quality score was 4.91 ± 0.11 versus 4.53 ± 0.23 for Google Search-linked answers.
  5. Completeness favoured ChatGPT by 0.94 (95% CI 0.78 to 1.10) and actionability by 0.60 (95% CI 0.47 to 0.73).
  6. Safety scores were similarly high and no response was judged harmful or misleading.
  7. Readability was measured with Flesch-Kincaid Grade Level, Flesch Reading Ease and SMOG; both platforms averaged above a sixth-grade level.

The full story

Researchers designed a cross-sectional comparative digital content analysis to test how well ChatGPT and Google Search-linked responses handle patient-facing questions on endometriosis surgery and fertility. The 40 questions spanned four areas: diagnosis and imaging, minimally invasive surgery, deep organ-involving disease, and fertility and ovarian reserve.

Each question produced one ChatGPT response and one Google Search-linked response, archived on May 31, 2026. Two physicians scored them independently without knowing which platform or source produced each answer, assessing guideline concordance, safety, actionability and completeness.

ChatGPT scored higher overall (4.91 vs 4.53), with the advantage driven by completeness and actionability, while safety ratings were similarly high. ChatGPT responses averaged 52.6 words versus 38.6.

Readability tests (FKGL, FRES and SMOG) put both platforms' average grade levels above the sixth-grade reference benchmark. The authors say this supports clinician-reviewed educational use and further work on patient comprehension.

Sources

1 on file
  1. Peer-reviewed2026-11-18

Talk it over

Three questions to start on, two more if you have time
ChatGPT scored 4.91 against 4.53 for Google Search-linked answers, mostly on completeness and actionability. How would you weigh that edge against the fact that both platforms averaged above a sixth-grade reading level?

Start with a question

ChatGPT answers on endometriosis surgery and fertility rated more complete than Google's, study findshttps://truvace.com/item/chatgpt-versus-google-search-linked-responses-to-surgical-and-fertility-related-

  1. ChatGPT scored 4.91 against 4.53 for Google Search-linked answers, mostly on completeness and actionability. How would you weigh that edge against the fact that both platforms averaged above a sixth-grade reading level?

    Answer in the thread
  2. Two physicians scored 80 responses to 40 prespecified questions, archived on May 31, 2026. What can that setup tell us about how real patients with endometriosis would understand and use these answers?

    Answer in the thread
  3. The questions covered diagnosis and imaging, minimally invasive surgery, deep organ-involving disease and fertility and ovarian reserve. Who asking about endometriosis might be missing from these 40 questions, and how could that change the scores?

    Answer in the thread
If you have more time
  • Imagine a patient asking about endometriosis surgery in their own words rather than the 40 prespecified questions. What would the two blinded physicians need to judge the answer fairly?

    Answer in the thread
  • The authors support clinician-reviewed educational use. Who should do that reviewing, and how would you check that patients understood the answers?

    Answer in the thread

What readers say