International Experiences and Implications of Generative Artificial Intelligence in the Generation of Pharmacoeconomic Evidence

Zhang Jiayi; Tian Furong; Liu Zehui; Zuo Genyong

Abstract：Objective:To summarize the international experience of Generative AI in pharmacoeconomics and evidence generation for health technology assessment,and to provide a reference for the rational use of Generative AI in pharmacoeconomic evidence generation in China.Methods:The position statements and working reports on the rational application of Generative AI published by National institute for Health and Care Excellence(NICE)and International Society for Pharmacoeconomic and Outcomes Research(ISPOR)in the UK were reviewed to conduct a comparative analysis of the similarities and differences in the value and risks generated by generative artificial intelligence in evidence generation,as well as the methods of quality control.Results:NICE paid more attention to the application of Generative AI to the whole process of evidence generation,affirmed its value of improving the automation of the model,and emphasized that the quality control of the generated evidence should pay attention to external supervision and its own autonomy.ISPOR focuses on the application of Generative AI to the core link of pharmacoeconomic evidence generation,affirming that it brings simple and efficient value points to evidence generation,and advocates that the quality control of generated evidence should start from the technology of model training itself.Both of them attach great importance to the risks brought by Generative AI in the generation of evidence from the perspective of security,equity,and transparency.Conclusion:In order to promote the high quality and efficiency of the generated evidence in the field of pharmacoeconomics in China,suggestions are put forward to control the transparency of model algorithms,guard against the risk of data pollution,and improve the quality of training data when applying Generative AI in the field of pharmacoeconomics in China.