SYMPOSIUM ABSTRACT: High powered methods, the big data revolution, and the crisis of replication in medicine and social sciences have prompted new reflections and debates in both statistics and philosophy about the role of traditional statistical methodology in current science.

I invite your constructive comments there as well. I will write some reflections in the “comments” to this post. It was held in Pittsburgh on November 13, 2022. Finally, ethics committee and drug regulatory agencies should see to it that newly gained insights are being put into practice.Below are slides from 4 of the talks given in our Philosophy of Science Association (PSA) session from last month: the PSA 22 Symposium: Multiplicity, Data-Dredging, and Error Control. The investigators can improve the quality of data handling by forming an Investigators Liaison Panel that represents them in the contacts with the sponsor. To meet these requirements, commercial sponsors should have initiatives for independent DSMBs, interim analysis and complete data availability, and conceive corresponding contracts. Regarding accessibility of data, the raw data should be made available to participating researchers and independent scientific institutions, and both positive and negative results should be published, in agreement with a Report on Drug Information by the World Health Organization. Data analysis should be performed in independent institutions, and there is a moral obligation to perform an interim analysis. As to data monitoring, we underline the corporate responsibilities and argue in favour of an independent Data and Safety Monitoring Board (DSMB). We discuss basic ethical principles underlying adequate data handling and suggest some modifications of the Good Clinical Practice guidelines. In this paper we conceive the process of data handling in a broad sense, encompassing monitoring, analysis and publication of data. Data handling is a crucial phase in controlled clinical trials.

Data dredging in clinical trials