I am Professor of Biostatistics at the Department of Medical Epidemiology and Biostatistics (MEB) at Karolinska Institutet in Stockholm, where I’ve been employed since March 1999. My primary research interests are developing and applying statistical methods for population-based cancer survival analysis, particularly the estimation and modeling of relative/net survival. I also have general interests in epidemiology, particularly cancer epidemiology. My MEB home page contains details of my research. I am head of the MEB Biostatistics group and a member of the board of Karolinska Institutet.
PhD (Statistics), 1997
University of Newcastle, Australia
B.Math (honours class 1), 1992
University of Newcastle, Australia
1-6 June 2020:
Paul Lambert and I will teach a 1-week course on Statistical methods for population-based cancer survival analysis in Veneto, Italy. Summer school cancelled due to covid-19. We hope to return in 2021.
13-17 July 2020: I will be teaching at the course Cancer Survival: Principles, Methods and Applications at the London School of Hygiene & Tropical Medicine.
I am fortunate to have had the opportunity to teach at courses organised by the summer school in Italy and the Cancer Survival Group at the London School of Hygiene & Tropical Medicine each summer for more than a decade. Each of the courses has evolved each year and I have improved my knowledge through interaction with both participants and co-teachers. Potential participants occasionally ask me about the differences between the the courses. Following is my assessment of the differences between the two courses so that potential participants may make an informed decision as to which course suits them best (assuming resources do not allow attending both).
The courses cover similar content, but different aspects are emphasized due to the different backgrounds and experience of the teachers. The course in Italy is taught by Paul Lambert and myself, who are statisticians, so we devote more time to the statistical methods. The course in London also covers the statistical methods but not to the same extent. The cancer survival group in London are world-leading in international comparisons of cancer patient survival so their course devotes more time to specific features of the analysis and interpretation of such studies (including data quality and quality control, choice of method, analytic approach, intepretation, and impact on policy). Paul Lambert and I discuss the role of cancer survival research in policy, but do not have the experience of Michel Coleman who delivers two lectures on this topic during the London course.
In summary, the Italy course devotes more time to the statistical methods. Paul Lambert and I work more in clinical cancer epidemiology and our course devotes more time to methods and applications in clinical epidemiology whereas the London group are stronger in descriptive epidemiology and devote more time there. The Italy course devotes considerably more time to statistical modelling, especially the theory and application of flexible parametric models. There is broad knowledge among both faculties, but we have different areas of specific expertise.
The teaching styles are somewhat different. The London course is more structured, with a greater number of teachers who are each allocated a fixed amount of time. The Italy course is less structured and more student-focussed. Lectures are planned in advance, but the schedule is flexible with the pace and content tailored to the audience. The style of the labs are very different; the aim in London is that all participants complete the same exercises whereas in Italy the focus in on providing tailored exercises and one-on-one help to suit each participant’s interests.
I have co-authored, with Enzo Coviello, Stata commands for estimating and modelling relative survival. Enzo is responsible for all of the nice code.
Some Stata tutorials, most of which are about survival analysis
SAS code for estimating and modelling relative survival. I wrote this code in the 1990s and have since moved from SAS to Stata, so this code is not maintained.
Some SAS tips and tricks (primarily on data management)