Perceptions of a Group of Experts on the Integration of Artificial Intelligence in the Management of Atopic Dermatitis

Abstract

Therapeutic patient education (TPE) plays a key role in the management of chronic skin diseases such as atopic dermatitis (AD), where long-term outcomes depend heavily on patient engagement and adherence [1, 2]. Initiatives such as hospital-based “atopy schools” supported by multidisciplinary teams have demonstrated their value in improving patient knowledge, adherence, and outcomes [3, 4]. However, access to such programs remains limited in many regions due to logistical, financial, and geographic barriers [5]. To explore how digital health innovations might help address these challenges, the Oriented Patient Education Network in Dermatology (OPENED) [6], under the auspices of the International Society of Atopic Dermatitis (ISAD), conducted an international survey among healthcare professionals. The aim was to assess perceptions of integrating artificial intelligence (AI) into AD management, with a particular focus on TPE as a conceptual framework. We sought to capture perceived benefits, risks, and readiness for adoption. An online questionnaire of nine items (multiple choice, Likert-scale, and open-ended questions) was distributed to 90 members of OPENED and the French Therapeutic Group (GET). The survey was reviewed by experts for clarity. A total of 38 responses were received (response rate 42%), including 33 dermatologists (87%) and 5 other healthcare professionals (13%). Most respondents were hospital-based (84%), with the remainder in private practice or other settings. Table 1 summarizes respondent demographics and practice characteristics. Notably, 86% reported using therapeutic action plans, 90% employed patient follow-up tools, and 80% routinely practiced shared decision-making. Multidisciplinary collaboration was reported by 40%, and 30% were involved in structured educational programs, such as atopy schools. Respondents expressed considerable interest in AI applications. As shown in Table 2, 86% agreed that AI could facilitate patient–clinician interaction, and over 90% felt it could improve consultation preparation. Many also perceived potential benefits in predicting disease flares and providing personalized responses. However, significant concerns were reported: 89% worried about inaccurate or misleading information, 69% about dangerous or biased responses, and 67% about possible negative impacts on adherence. Subgroup analysis revealed broadly similar patterns among dermatologists and other healthcare professionals, though numbers were too small for robust comparisons. While enthusiasm for AI in AD care is clear, respondents highlighted important risks. Concerns about misinformation, patient confusion, and erosion of clinician authority underscore the need for caution. Interestingly, nearly all respondents (37/38) indicated they would be willing to use an AI-integrated tool in practice (see Table 3, Supplementary Appendix). In addition, it is important to distinguish between different types of AI applications, as their potential roles and risks vary. Chatbots and conversational agents may facilitate patient engagement and provide accessible educational support; predictive algorithms may assist in anticipating disease flares and guiding proactive management; and AI-driven educational tools can help structure patient learning in a personalized way. Each of these modalities requires different safeguards and validation approaches, underscoring the need for tailored evaluation before integration into routine care. This survey has limitations. The response rate was modest and may reflect survey fatigue, workload, or timing, raising the possibility of selection bias. We did not collect demographic data such as age, gender, nationality, or prior experience with AI, which limits interpretation and prevents generational or cultural comparisons. The small number of non-dermatologist respondents further restricts subgroup analysis. Nonetheless, the respondents represent an experienced, international group of AD specialists, offering valuable early insights. In conclusion, healthcare professionals recognize AI's potential to enhance AD management and patient education, but also express substantial reservations. Moving forward, targeted training for clinicians is essential, alongside transparent guidelines to ensure AI tools complement rather than replace clinician judgment. Including patients in discussions about AI use in care will also be critical to maintain trust and shared decision-making [7-9]. All authors contributed to the conception and design and analysis of the study. Jean-Francois Stalder, David Troya, and Alain Taieb coordinated the survey. David Troya organized data collection. Jean-Francois Stalder and David Troya drafted the first version of the manuscript. All authors revised the manuscript critically for intellectual content and approved the final version. The authors have nothing to report. V.A. has been an investigator and/or consultant to Abbvie, Eli Lilly, Pfizer. S.B. has been an investigator and consultant for Astrazeneca, Almirall, Sanofi-Genzyme, Abbvie, Galderma, Alexion, Novartis, Janssen, Leo-Pharma, Pfizer, Eli Lilly, UCB Pharma, Incyte. M.D. has been an investigator and consultant for AbbVie, Eli-Lilly, Regeneron, Sanofi Genzyme, Pfizer, Pierre Fabre, Novartis, Almirall, Arena Pharmaceuticals, ASLAN Pharmaceuticals, Incyte, Kymab. L.E. has served as a consultant for Almirall, Arcutis, Asana, Biersdorf, Celgene, Dermavant, Dermira, DS Biopharma, Forte, Galderma Labs, Kyowa Hakkin Kirin, Incyte, Leo, Lilly, Matrisys, Menlo Therapeutics, Novan, Ortho Derm/Valeant, Novartis, Otsuka/Medimetriks, Pfizer/Anacor, Regeneron/Sanofi and TopMD. He has served as an investigator for Abbvie, Leo, Pfizer/Anacor and Regeneron/Sanofi; an advisory board member of Ortho Derm/Valeant; a speaker for Almirall; and on the Data Safety Monitoring Board for Glenmark. R.L.O. has been an investigator and/or consultant for Bayer, Eli Lilly, Abbvie, Sanofi and Amgen. MEH reports fees from Pierre-Frabre and Sanofi. M.T. has been a speaker and/or consultant an/or investigator and/or participant of Advisory Board for Abbvie, Bausch Health, Bioderma, Eli Lilly, La Roche posay, Leo pharma, Mead Johnson, Novartis, Pfizer, Pierre Fabre, Pfizer, Mead Johnson, Sanofi Genzyme. P.L. reported receiving grants from AOBiome, Regeneron/Sanofi Genzyme, and AbbVie; personal fees from Regeneron/Sanofi Genzyme, Leo, Eli Lilly, Pfizer, Galderma, L'Oreal, Almirall, ASLAN Pharma Advisory board, Dermavant, Pierre Fabre, Menlo Therapeutics, IntraDerm, Exeltis, AOBiome, Arbonne, and Amyris; stock options from Micreos and; other royalties from patented product from Theraplex. M.S.A. is an employee at Pierre Fabre Group. A. TAIEB is the president of ISAD. A. TORRELO declares consulting fees and honoraria from Lilly, Viatris, Sanofi and Pfizer. AW has been an inversitator and/or consultant for Abbvie, Almirall, Beiersdorf, Bioderma, Galderma, Glenmark, Leo Pharma, Eli Lilly, L'Oreal, Maruho, Novartis, Pfizer, Pierre Fabre, Regeneron, and Sanofi-Aventis and a board member of ISAD, ESPD and vice-chair of ETFAD. The other authors declare no conflicts of interest. All data is available from the corresponding author upon reasonable request.

Ähnliche Arbeiten