PRC Dialogue Series

Felix Gerber andProf. Niklaus Labhardt (International Clinical and Health Services Research (ICHSR), University of Basel)

Fan Li, Associate Professor of Biostatistics (Yale School of Public Health)

Mediation analysis is widely used in health science research to evaluate the extent to which an intermediate variable explains an observed exposure-outcome relationship. However, the validity of mediation analysis can be compromised when the exposure is measured with error. Motivated by the Health Professionals Follow-up Study (HPFS), we investigate the impact of exposure measurement error on assessing mediation with a survival outcome, based on the Cox proportional hazards outcome model. When the outcome is rare and there is no exposure-mediator interaction, we show that the uncorrected estimators of the natural indirect and direct effects can be biased into either direction, but the uncorrected estimator of the mediation proportion is approximately unbiased as long as the measurement error is not large, or the mediator-exposure association is not strong. We develop ordinary regression calibration and risk set regression calibration approaches to correct for the exposure measurement error-induced bias when estimating mediation effects and study their empirical performance in simulations. We apply the proposed approaches to the HPFS (1986-2016) to evaluate extent to which reduced body mass index mediates the protective effect of vigorous physical activity on the risk of cardiovascular diseases and consider several additional sensitivity analyses.

Event co-organized by Population Research Center (PRC), Center for Reproducible Science (CRS), and Jacobs Center for Productive Youth Development.

PRC_Dialogue_FanLi (PDF, 162 KB)

Dr. Andreas Edel (Population Europe / Max Planck Institute for Demographic Research)

Demography stands as one of the oldest empirical social sciences. The seminal work, "Natural and Political Observations Made upon the Bills of Mortality," published in 1662 by John Graunt, a London haberdasher and early Fellow of the Royal Society, marks the inception of demographic and epidemiological studies. Graunt's aim was to contribute to evidence-based policy-making, particularly addressing mortality during the plague pandemic and other demographic phenomena such as fertility and regional disparities. Today, demographic research remains integral to policy-making at regional, national, and European levels, underscored by the recent appointment of the first Vice-President for Democracy and Demography by the European Commission. This lecture will delve into the historical and contemporary dimensions of the discipline, while also examining the interface between science and policy. It will explore how 'impact' increasingly shapes funding proposals, while simultaneously presenting challenges for researchers operating at the science-policy nexus.

Andreas Edel has served as the Executive Secretary of Population Europe, the network comprising Europe’s foremost demographic research centers, since 2009. Previously, he held the position of Akademischer Geschäftsführer (Chief Research Co-ordinator) at the Max Planck Institute for Demographic Research from 2004 to 2009, and served as the Deputy Director of the Max Planck International Research Network on Ageing from 2007 to 2009. Prior to his tenure at the Max Planck Institute, he worked as a coordinator for European projects at the Wissenschaftskolleg zu Berlin – Institute for Advanced Study from 2001 to 2004, and as a historian at the Universities of Munich and Regensburg.

Flyer (PDF, 211 KB)

Prof. Laura C. Rosella (Canada Research Chair in Population Health Analytics, University of Toronto)

The role of data and analytics has never been more important in society and, in particular, for informing public health decisions. There has been a rapid change in the nature of data used in epidemiologic research and an increased focus on the use of Artificial Intelligence (AI) in the analysis of those emerging data sources. There are many conflicting points of view on the utility of AI in an epidemiologic context and a lack of clarity on the implications for the discipline and public health more generally. This talk will cover the myths, the debates, and the skepticism (from all sides) and outline a proposed direction for the role of AI in the future of public health. We will additionally cover the intersection of AI with a precision public health frame and the potential impact and unintended consequences of this framing. The talk will include implications related to bias, health equity, surveillance and causal inference. The talk will conclude by summarizing the implications for epidemiology research and training.

Tuesday, September 19, 2023, 4:30PM until 5:30PM, Rämistrasse 69, Zürich, SOC E-006
This event is associated with the DSI AI week.

Flyer (PDF, 276 KB)