A sessão iniciará às 15h00 com um conjunto de conferências proferidas por Henrique Barros, Presidente do ISPUP, Michel Hof, da Academisch Medisch Centrum Universiteit van Amsterdam e Kasia Czabanowska, Presidente da Association of Schools of Public Health in the European Region (ASPHER).
O encerramento das atividades contará com a presença de Fernando Araújo, Secretário de Estado Adjunto e da Saúde.
14:30 - 15:00 | Boas-vindas e Abertura
15:00 - 15:30 | The making of a Public Health Problem
Henrique Barros, Instituto de Saúde Pública da Universidade do Porto (ISPUP)
15:30 - 16:15 | Record linkage with time-to-event data
Michel Hof, Academisch Medisch Centrum Universiteit van Amsterdam
16:15 - 17:00 | Leading for Health – an imperative for change in the 21st century
Kasia Czabanowska, Association of Schools of Public Health in the European Region (ASPHER)
17:00 | Encerramento
Fernando Araújo, Secretário de Estado Adjunto e da Saúde
Porto de Honra
Henrique Barros was born in Porto, in 1957. He graduated in Medicine in 1981 and teaches at the School of Medicine of Porto University (FMUP) since 1979. From 1988-89 was research fellow at the Internal Medicine department of Lund ́s University Hospital. In 1991 became Gastroenterology specialist and finished his PhD, with a research in the epidemiology of viral hepatitis. He is Full Professor of Epidemiology since 1999. He led the organization of the Master Programs in Public Health (coordinator), Epidemiology, Health Education and Sociology and Health, as well as the doctoral program in Public Health at the University of Porto and the doctoral program in Global Public Health (program director), a joint program with Nova University of Lisbon, financed by the Portuguese Research Foundation (FCT). He has developed research in national and international projects, in areas such as clinical and perinatal epidemiology, cardiovascular, infectious and cancer diseases, which resulted in (co) authorship of more than 200 scientific publications in international journals. He is editor of the journal Arquivos de Medicina, and member of the Editorial Board of the European Journal of Epidemiology. He was member of the Scientific Council for Health Sciences Foundation for Science and Technology (2004-12). He was National Coordinator for Infection HIV/AIDS (2005-11) and President of Portuguese Association of Epidemiology (2013-15). Currently, he is the President of the Institute of Public Health, University of Porto, member of the Executive Commission of the National Ethical Committee for Clinical Investigation (CEIC) and member of the Medical Sciences (MED) Scientific Committee of Science Europe and President- Elect of the International Epidemiological Association (2014-17).
Michel took the bachelor and master in Medical Informatics at University of Amsterdam and the PhD in Biostatistics at the Department of Clinical Epidemiology, Biostatistics and Bioinformatics of the University of Amsterdam, in 2015.
His main research line is the development of statistical methods to combine data from multiple files using partially identifying variables (e.g. gender, birth-date). With a record linkage procedure, these variables can be used to determine whether observations from different files belong to the same individual (a match) or to different individuals (a non-match). Generally, the combined strength of these variables is too low leading to mistakes. Separation of record linkage and data analysis of the linked data is common practice. As a result, most analyses of data derived from record linkage are secondary; the analyst only has access to imperfectly linked data consisting of the observation pairs that were identified as matches. A problem of this approach is that external knowledge (e.g. clerical review) is needed to correct subsequent analyses of the linked data. Instead of separating the record linkage procedure and the data analysis, he showed that it is better to perform both tasks simultaneously.His approach to this problem not only leads to more accurate estimates, it also does not require external knowledge. Currently, he proposed methods to correct for (generalized) linear regression models and time-to-event models.