Scholarlink

Background <p>Artificial intelligence (AI) is transforming inflammatory bowel disease (IBD) care, but concerns over data quality, privacy, and provider familiarity remain.</p> Aims <p>To examine international health care providers’ (HCPs) perceptions, clinical use of, and barriers to applying AI in IBD care.</p> Methods <p>An anonymous 17-item survey was developed and pretested by an international team. The survey was fielded internationally from January to May 2025 via social media, international IBD newsletters and email listservs.</p> Results <p>Of 323 IBD professionals from North America (35.3%), Europe (26.0%), Asia–Pacific (19.2%), 207 (64.1%) viewed AI as very or extremely promising. AI-assisted endoscopy was reportedly adopted by 131 (40.6%) of respondents, but only 69 (21.4%) reported applying AI in IBD care or research. Males supported AI in IBD endoscopic scoring (89.9% vs. 79.1%, <i>p</i> = 0.009) and radiomics more than females (86.6% vs. 75.2%, <i>p</i> = 0.022), while females endorsed AI for patient education more than males (81.5% vs. 70.6%, <i>p</i> = 0.026). 145 (44.9%) respondents were concerned about AI usage in IBD care. Females were less concerned about cost (18.5% vs. 30.1%, <i>p</i> = 0.019) and more concerned about oversight (28.7% vs. 14.7%, <i>p</i> = 0.003). Concerns regarding AI usage decreased with age (OR 0.021 for age &gt; 55 compared to age 25–34&#xa0;years old; 95% CI 0.13–0.84). Exposure to IBD-focused AI-based tools was associated with more favorable perceptions of AI use in IBD care (OR 2.63, 95% CI 1.25–5.53).</p> Conclusion <p>Based upon survey results, HCPs had concerns over insufficient education, regulatory ambiguity, and perplexing implementation strategies for AI. Proven clinical benefit (74.9%) and adequate AI training (72.1%) are the top solutions for enabling the appropriate adoption of AI tools in IBD.</p> Graphical Abstract <p></p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Healthcare Providers' Perspectives on the Role of Artificial Intelligence in the Care of Patients with Inflammatory Bowel Disease: An International Survey

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年5月24日

查看全文
  • Jalpa Devi,
  • Colleen Y. Colbert,
  • Sophia Haroon Dar,
  • Beatriz Gros,
  • Ashkan Rezazadeh Ardabili,
  • Agnes H. Y. Ho,
  • Anita D. Misra-Hebert,
  • Joseph Sleiman

摘要

Background

Artificial intelligence (AI) is transforming inflammatory bowel disease (IBD) care, but concerns over data quality, privacy, and provider familiarity remain.

Aims

To examine international health care providers’ (HCPs) perceptions, clinical use of, and barriers to applying AI in IBD care.

Methods

An anonymous 17-item survey was developed and pretested by an international team. The survey was fielded internationally from January to May 2025 via social media, international IBD newsletters and email listservs.

Results

Of 323 IBD professionals from North America (35.3%), Europe (26.0%), Asia–Pacific (19.2%), 207 (64.1%) viewed AI as very or extremely promising. AI-assisted endoscopy was reportedly adopted by 131 (40.6%) of respondents, but only 69 (21.4%) reported applying AI in IBD care or research. Males supported AI in IBD endoscopic scoring (89.9% vs. 79.1%, p = 0.009) and radiomics more than females (86.6% vs. 75.2%, p = 0.022), while females endorsed AI for patient education more than males (81.5% vs. 70.6%, p = 0.026). 145 (44.9%) respondents were concerned about AI usage in IBD care. Females were less concerned about cost (18.5% vs. 30.1%, p = 0.019) and more concerned about oversight (28.7% vs. 14.7%, p = 0.003). Concerns regarding AI usage decreased with age (OR 0.021 for age > 55 compared to age 25–34 years old; 95% CI 0.13–0.84). Exposure to IBD-focused AI-based tools was associated with more favorable perceptions of AI use in IBD care (OR 2.63, 95% CI 1.25–5.53).

Conclusion

Based upon survey results, HCPs had concerns over insufficient education, regulatory ambiguity, and perplexing implementation strategies for AI. Proven clinical benefit (74.9%) and adequate AI training (72.1%) are the top solutions for enabling the appropriate adoption of AI tools in IBD.

Graphical Abstract