Dr. Lisa Lix - Response Shift Methods for Longitudinal Patient-Reported Outcomes
Response shift, a change in the meaning of an individual’s self-evaluation of a health measure, is known to be a potential source of bias in longitudinal analyses of health-related quality of life (HRQOL) data and other patient-reported outcomes. This presentation will examine different methods to test for response shift, including a new method based on measures of the change in the relative importance of HRQOL domains over time. As well, some recent developments in the use of latent variable scoring methods to adjust for the effects of response shift will be explored.
Lisa Lix is Professor and Manitoba Research Chair in the Department of Community Health Sciences, University of Manitoba. Prior to returning to Manitoba in September 2012, she was Centennial Research Chair in the School of Public Health at the University of Saskatchewan. In Saskatchewan she led the development of new MSc and PhD programs in biostatistics, served as Site Director of the Western Regional Training Centre (WRTC) in Health Services Research, and established the Population Health Data Laboratory. Her research interests include health services research methodology, quality of administrative health databases, the analysis of repeated measures and longitudinal data, and robust statistical methods for the analysis patient-reported outcomes. Lisa has served on the Board of the Statistical Society of Canada in a variety of different capacities since 2005 and was President of the Biostatistics Section in 2010.