{"version":"1.0","type":"rich","provider_name":"Acast","provider_url":"https://acast.com","height":250,"width":700,"html":"<iframe src=\"https://embed.acast.com/$/64b7f867cf6b1700110bbf54/6a296cb632e30dceaf03447e?\" frameBorder=\"0\" width=\"700\" height=\"250\"></iframe>","title":"Patti's People - Patti Peeples speaks with Dalia Dawoud","thumbnail_width":200,"thumbnail_height":200,"thumbnail_url":"https://open-images.acast.com/shows/64b7f867cf6b1700110bbf54/1781099622582-aa99d548-d9da-4706-9365-4cdad41242fc.jpeg?height=200","description":"<p><strong>In this ‘Patti’s People’ episode, Patti Peeples of The Peeples Collaborative speaks with Dalia Dawoud (Research Principal, HTA Policy and Strategy, Cytel) about the real-world readiness of AI in HTA, from which use cases are mature enough to trust, to why most HTA agencies are still reacting rather than leading.</strong></p><p><br></p><p><strong>Questions:</strong></p><p><br></p><ul><li><strong>00:00:</strong> Introduction</li><li><strong>01:30: </strong>You’ve said AI is moving faster than anything HTA has faced, even faster than real-world evidence. What feels fundamentally different about this moment, and what’s driving that acceleration?</li><li><strong>05:49:</strong> Where do you fall on the view that HTA bodies are ‘reacting’ to AI in real time rather than leading its development and adoption?</li><li><strong>09:31:</strong> The way we generate and evaluate evidence may not be fit for purpose for AI. How do we overcome that?</li><li><strong>12:46: </strong>Are we slowing the adoption of AI by demanding too much certainty too early?&nbsp;&nbsp;</li><li><strong>14:23:</strong> We often talk about bias and decision risk, but those exist in traditional evidence too. What are the deeper risks that HTA bodies may not be saying out loud?</li><li><strong>17:44</strong>: You mentioned de-skilling and job displacement. Where could disruption actually hit first in HTA?&nbsp;</li><li><strong>24:51:</strong> You highlighted literature reviews as a ‘safe’ use case. What makes something safe enough for AI adoption?</li><li><strong>30:02:</strong> Can the methods and parameters we use to validate traditional evidence be transferred to AI, or do we need entirely new approaches for AI validation?</li><li><strong>31:07:</strong> You mentioned transforming unstructured data into usable variables as a promising area. What level of accuracy is good enough for HTA?</li><li><strong>36:32:</strong> Can you tell us about NICE’s HTA lab?&nbsp;</li><li><strong>37:49:</strong> Have HTA bodies, such as NICE, sufficiently defined failure in these pilots?&nbsp;</li><li><strong>41:34:</strong> ELEVATE-GenAI introduced reporting standards for AI in HEOR. Was the goal primarily guardrails, acceleration, or improving quality?</li><li><strong>44:02:</strong> Rapid Fire Questions</li></ul><p><br></p><p><strong>Dalia Dawoud</strong></p><p>Research Principal, HTA Policy and Strategy, Cytel</p><p><br></p><p><strong>Dalia Dawoud, PhD, is a Research Principal specializing in Health Technology Assessment (HTA) policy and strategy, with a focus on artificial intelligence in evidence generation. She contributes to ISPOR’s AI Working Group and co-developed ELEVATE-GenAI and the Taxonomy of Generative AI. Her work focuses on opportunities for integrating AI into HTA while maintaining rigor, transparency, and trust.</strong></p><p><br></p>","author_name":"The Evidence Base"}