Many statistical practitioners make use of the Bayesian approach because it allows analyses on highly structured data. An important class of models involves the analysis of follow-up studies, i.e. longitudinal-, survival studies or a combination of the two. We will illustrate the Bayesian approach for the analysis of such data, by means of examples, and focus on the analysis of longitudinal studies. For instance, Bayesian implementations will be illustrated on (generalized and non-linear) linear mixed models with non-standard distributions for the random parts, growth curve models, pharmaco-kinetic models, multivariate mixed models, joint mixed models of several random variables, longitudinal models with smooth subject-specific evolutions, longitudinal models with informative measurement times, etc. Finally, we will look at joint modeling of the survival and longitudinal process. Examples will be analysed using WinBUGS/OpenBUGS/JAGS and R-versions of them, but also dedicated R-software.
The 5-day course will consist of theoretical sessions each morning, and practical sessions each afternoon, when participants will work on their laptops with tutor assistance, and (optionally) in small groups. A provisional outline programme is available on request.
The course is designed for applied statisticians and epidemiologists with a solid statistical background. Required skills and knowledge are: programming in R or SAS®, statistical inference, (linear, logistic, Cox) regression models and basic knowledge of Bayesian methodology. It will be advantageous to have practical experience in modelling longitudinal and survival studies.
Course material will be made available in hardcopy as well as electronically, to each participant. Some recommended reading for the course:
- Bayesian Biostatistics, E. Lesaffre and A. Lawson (2012), John Wiley and Sons
- Missing Data in Longitudinal Studies: Strategies for Bayesian Modeling and Sensitivity Analysis, M. Daniels and J. Hogan (2008), CRC Chapman and Hall, Boca Raton
Emmanuel Lesaffre is Professor of Biostatistics at L-Biostat, K.U.Leuven, Leuven, Belgium. His research interests include Bayesian methods, longitudinal data analysis, statistical modelling, analysis of dental data, interval censored data, misclassification issues and clinical trials. He has written more than 350 papers in peer-reviewed statistical and medical journals. He is the founding chair of the Statistical Modelling Society, past-president of the International Society for Clinical Biostatistics and fellow of ISI and ASA.
The course was described as oriented towards an applied audience with a good knowledge of various regression models. It was emphasised that Bayesian concepts would (because of the 2-day format) be introduced only briefly; hence a prior course on the Bayesian approach would be helpful. It was also stated that knowledge of R would be useful for the course, but no prior knowledge of WinBUGS would be assumed, although this also would be helpful, as would some knowledge of classical repeated measurements analysis and classical survival analysis. Two books were recommended as background reading: Lesaffre & Lawson, Bayesian Biostatistics, and Daniels & Logan: Missing Data in Longitudinal Studies.