Denna sida på svenska

Elisabeth Dahlqwist

PhD student

About me

I started my PhD project "The attributable fraction in causal inference and genetics" in June 2015. My supervisors are Arvid Sjölander and Yudi Pawitan.


  • Bachelor in Political science – Uppsala University (2008-2011)
  • Bachelor in Statistics– Uppsala University (2009-2012)                                         
  • MSc in Statistics– Uppsala University (2012-2014)
  • Doctoral student in Biostatistics - Dept. of Medical Epidemiology and Biostatistics, Karolinska Institutet (2015- present)

Research description

My PhD project is to develop the attributable fraction as a causal measure. The attributable fraction measure the proportion of some disease that could have been prevented in a population had some harmful exposure that cause the disease been eliminated. The attributable fraction has been a popular measure in epidemiology and public health since it was defined in the 1950s. However, the attributable fraction is a causal measure since we have to assume that the exposure cause the disease. This is the starting point of my PhD project with the overall aim to:

  • improve the conceptual understanding of the attributable fraction as a causal measure
  • develop methods for using family or clustered data to adjust for unmeasured confounding and estimate absolute effects
  • develop sensitivity analysis for unmeasured confounding for the attributable fraction which allow for continous covariates
  • improve the conceptual understanding of heritability by connecting heritability with the attributable fraction and develop a method for estimating the attributable fraction based on heritability
  • implement my method developments in my R package "AF"

Teaching portfolio

Since 2012 I have regularly been a teaching assistant for different basic statistics courses.  Except the basic statistics courses I have been a teaching assistant in

  • survival analysis (Biostatistics III)
  • scientific theory for medical students (VetU)
  • introduction to R
  • epidemiology (Epidemiology II)



A Note on the Noncollapsibility of Rate Differences and Rate Ratios
Sjölander A, Dahlqwist E, Zetterqvist J
Epidemiology (Cambridge, Mass.) 2016;27(3):356-9

Model-based estimation of the attributable fraction for cross-sectional, case-control and cohort studies using the R package AF
Dahlqwist E, Zetterqvist J, Pawitan Y, Sjölander A
European journal of epidemiology 2016;31(6):575-82