In this rapidly evolving technological era, ChatGPT represents a significant breakthrough in the field of artificial intelligence (AI), being a pivotal source of virtual content generation.1 Developed by OpenAI, ChatGPT harnesses the potential of natural language processing (NLP) to craft coherent and pertinent responses in natural language. As a knowledgeable virtual assistant, it can provide insights into a multitude of health-related topics, including medical education, with the potential to provide patients with relevant information.2 In the field of skin cancers, advanced AI and healthcare system may act synergically to enhance knowledge in general population. We conducted a survey to assess patients' satisfaction with ChatGPT's responses to common questions about skin cancers. We adopted the US Skin Cancer Foundation's pre-existing 10-question interactive quiz, an educational tool for informing patients about skin cancer prevention.3 The questions were administered to freely available ChatGPT-3.5 and the responses were verified by a consensus panel of 10 dermatologists, affirming the accuracy of 9 of 10 answers. Hence, the questionnaire with ChatGPT's responses (translated in Italian using the same version of ChatGPT), was administered to 220 subjects consecutively accessing dermatology outpatient service for skin cancer screening visits from 1 July 2023 to 1 September 2023; participants were asked to rate their satisfaction with each response according to Likert scale (1—very dissatisfied to 5—very satisfied). Additionally, patients evaluated language clarity, the comprehensiveness of the material, and the knowledge improvement of skin cancer. The questionnaire was correctly completed by 199/220 (90.5%) patients, that represented the sample investigated. Subjects included 99 (49.7%) males and 100 females (50.3%), with a mean age of 50 years (SD = 16.1). A significant percentage of participants had a high educational degree (44.7% upper secondary school, 35.2% university. The majority of patients (40.2%) were employees, while 23.6% were self-employed and/or contractors. Most patients provided a high rate of satisfaction regarded ChatGPT's responses as shown in Table 1. When assessing language clarity and comprehensibility, patients defined themselves as "very satisfied" (47.7%) with the clarity of the language, "satisfied" (44.7%) with the level of information and exhaustiveness of the answers, and "very satisfied" (43.7%) by the increase of knowledge they gained from the questionnaire. Almost half of patients (47.7%) resulted "very satisfied" for language clarity while 44.7% were "satisfied"; "very satisfied" was the preferred response (43.7%) for the increase of knowledge. No associations were found between patients' clinical data and the level of satisfaction expressed. Limit sun exposure: Avoid direct sun exposure during peak hours (from 10:00 AM to 4:00 PM) when the sun's rays are most intense. Seek shade and wear protective clothing, wide-brimmed hats, and sunglasses. Use sunscreen: Apply a broad-spectrum sunscreen with a high SPF before going outdoors. Reapply it every 2 h, especially after swimming or sweating. Avoid tanning beds: Tanning beds emit harmful UV rays that can increase the risk of skin cancer. Completely avoid the use of tanning beds. Perform self-exams: Regularly examine your skin for any moles, freckles, or other lesions that change in appearance or size. If you notice changes, consult a dermatologist. Stay hydrated: Drinking plenty of water helps maintain healthy and resilient skin. Healthy lifestyle: Maintain a balanced diet rich in antioxidants and vitamins, avoid smoking, and reduce excessive alcohol consumption. Protect children: Keep infants under 6 months away from direct sunlight and use protective clothing and sunscreen on older children. Regular checkups: Schedule regular visits to a dermatologist for skin cancer screening, especially if you have a family history of the disease or other risk factors. Remember that prevention and early diagnosis are crucial for reducing the risk of skin cancer. Overall ChatGPT presents as a real opportunity to enhance the accessibility of healthcare information,4-8 and our data suggest a great value of this tool in supporting patient education, together with a high score of satisfaction. This is the first report describing patient assessment of ChatGPT as information source for skin cancer awareness. However, the potential risks of encountering inaccurate information give rise to significant ethical concerns. Our research identified 1/10 question generated by ChatGPT as incorrect (n.8 in Table 1), a finding in line with Johnson et al., which reported an 8.3% inaccuracy rate.9 As ChatGPT can be integrated into healthcare, it should be accompanied by medical oversight to ensure the accuracy of the information provided. Conceptualization: E. De Luca and A. Chiricozzi. Methodology: E. De Luca and G. Coscarella. Original draft preparation: E. De Luca and S. Cappilli. Review and editing: A. Chiricozzi and K. Peris. Supervision: K. Peris. All authors have read and agreed to the published version of the manuscript. K. Peris has received consulting fees and honoraria unrelated to this work from Abbvie, Almirall, Biogen, Celgene, Janssen Galderma, Novartis, Lilly, Novartis, Pierre Fabre, Sandoz, Sanofi and Sun Pharma, outside the submitted work; Chiricozzi has received consulting fees and honoraria unrelated to this work from AbbVie, Almirall, Boehringer-Ingelheim, Bristol Myers Squibb, Leo Pharma, Lilly, Janssen, Novartis, Pfizer, Sanofi Genzyme, outside the submitted work; all the other authors declare that they have no conflicts of interest relevant to this manuscript. This study was conducted in accordance with the Declaration of Helsinki. All patients in this manuscript have given written informed consent for participation in the study and the use of their deidentified, anonymized, aggregated data and their case details for publication.
No takes yet. Share an insight, caveat, or question.
Luca et al. (2024) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: