ArticleThe Permanente journal2024
Evaluating ChatGPT's Utility in Medicine Guidelines Through Web Search Analysis.
Article in The Permanente journal, 2024. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Not yet cited in PubMed.
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
0 citing papers in PubMed.
No citing paper in PubMed yet.
Corrections and comments
PubMed lists nothing against this paper. Absence here is not a guarantee, only a check that was made.
Authors and funding
8 authors.
Funding
No grant is acknowledged in the PubMed record.
Abstract
introductionWith the rise of machine learning applications in health care, shifts in medical fields that rely on precise prognostic models and pattern detection tools are anticipated in the near future. Chat Generative Pretrained Transformer (ChatGPT) is a recent machine learning innovation known for producing text that mimics human conversation. To gauge ChatGPT's capability in addressing patient inquiries, the authors set out to juxtapose it with Google Search, America's predominant search engine. Their comparison focused on: 1) the top questions related to clinical practice guidelines from the American Academy of Family Physicians by category and subject; 2) responses to these prevalent questions; and 3) the top questions that elicited a numerical reply.
methodsUtilizing a freshly installed Google Chrome browser (version 109.0.5414.119), the authors conducted a Google web search (www.google.com) on March 4, 2023, ensuring minimal influence from personalized search algorithms. Search phrases were derived from the clinical guidelines of the American Academy of Family Physicians. The authors prompted ChatGPT with: "Search Google using the term '(refer to search terms)' and document the top four questions linked to the term." The same 25 search terms were employed. The authors cataloged the primary 4 questions and their answers for each term, resulting in 100 questions and answers.
resultsOf the 100 questions, 42% (42 questions) were consistent across all search terms. ChatGPT predominantly sourced from academic (38% vs 15%, p = 0.0002) and government (50% vs 39%, p = 0.12) domains, whereas Google web searches leaned toward commercial sources (32% vs 11%, p = 0.0002). Thirty-nine percent (39 questions) of the questions yielded divergent answers between the 2 platforms. Notably, 16 of the 39 distinct answers from ChatGPT lacked a numerical reply, instead advising a consultation with a medical professional for health guidance.
conclusionGoogle Search and ChatGPT present varied questions and answers for both broad and specific queries. Both patients and doctors should exercise prudence when considering ChatGPT as a digital health adviser. It's essential for medical professionals to assist patients in accurately communicating their online discoveries and ensuing inquiries for a comprehensive discussion.
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.