Biography

I am a biostatistician working at the Cancer Registry of Norway. I also work part-time as a Guest Professor in the Biostatistics Group in the Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden. Previously, I worked at the Biostatistics Research Group at the University of Leicester, UK.

I have a variety of interests, mainly in the area of survival analysis in epidemiology. Much of my work focuses on the analysis from data from national cancer registries. I particularly enjoy writing software to enable transferal of new methods into practice.

I enjoy teaching and have developed some interactive graphs to help with communication.

Interests

  • Survival Analysis
  • Epidemiology
  • Cancer
  • Statistical Software

Education

  • PhD in Biostatistics, 1999

    University of Leicester, UK

  • MSc in Medical Statistics, 1992

    University of Leicester, UK

Publications

A full list of my publications can be found here

I have written a book with Patrick Royston titled Flexible parametric survival models using Stata: Beyond the Cox model.

A review of the book can be found here

In the software section of my webpage you will find some tutorials on using these models.

Recent & Upcoming Talks

A list of selected older talks can be found here

Recent (and not so recent) Talks:

10 September 2024 Improving the speed and accuracy when fitting flexible parametric survival models on the log hazard scale 2024 Northern European Stata Conference, Oslo, Norway

30 August 2024 A practical approach to fitting cancer survival models when data can’t move across borders Association of Nordic Cancer Registries (ANCR) Symposium 2024, Bodø, Norway

7 June 2024 Recent developments in the fitting and assessment of flexible parametric survival models 2024 German Stata Conference

12 October 2022 [Improving fitting and predictions for flexible parametric survival models](Improving fitting and predictions for flexible parametric survival models) Northern European Stata Conference

9 September 2022 Improving fitting and predictions for flexible parametric survival models 2022 UK Stata Conference

6 August 2021. Making Stata estimation commands faster through automatic differentiation and integration with Python. 2021 Stata Conference

18 February 2021. Regression standardization with time-to-event data to estimate marginal measures of association and causal effects using the standsurv command.

22 September 2020. Standardised and reference adjusted all-cause and crude probabilities in the relative survival framework. Advances in Survival Analysis. International Biometric Society - British and Irish Region.

August 2020 A marginal model for relative survival.. Intenational Biometric Society 2020

26 September 2019. Issues in Standardization. Symposium for statisticians working in register-based cancer epidemiology 2019, Stockholm, Sweden

29 August 2019. Standardised crude probabilities of death to improve understanding of national and international cancer survival comparisons. Association of the Nordic Cancer Registries meeting 2019, Stockholm Sweden

30 August 2019. Marginal estimates through regression standardization in competing risks and relative survival models. Nordic and Baltic Stata Users Group meeting 2019, Stockholm, Sweden.

Software

I have developed a number of Stata commands. I have added some examples of using this code and intend to add to these over time.

  • stpm2 - flexible parametric survival models
  • stpm3 - flexible parametric survival models
  • standsurv - standardized survival curves and more after fitting various types of survival models.
  • mrsprep - prepare data to directly fit marginal relative survival models.
  • rcsgen - generate restricted cubic splines
  • stpm2_standsurv - standardized survival curves after fitting an stpm2 model
  • stpp - Non-parametric marginal relative (net) survival using Pohar Perme estimator
  • mlad - Maximum likelIhood estimation with automatic differentiation using Python
  • stpm2cif - cause-specific cumulative incidence function after fitting a stpm2 competing risks model
  • stcrprep - data-preparation command to fit a range of competing risks models.
  • partpred - partial predictions
  • strcs - flexible parametric model on log hazard scale

I have worked with Paul Dickman, who has written some excellent Stata tutorials, many of which use my commands.

Teaching

When I worked at the University of Leicester my main teaching was on the MSc Medical Statisics course. I was a student on this course way back in $1991 / 1992$.

I also teach specialist courses. Below are some upcoming course.

Contact