Skip to main content
Deutsch

Detail

  1. Home
  2. Research
  3. Researcher Profiles
  4. Researcher Profiles
  5. Detail
Susanne Strohmaier
Susanne Strohmaier

Center for Public Health (Department of Epidemiology)
Position: Research Associate (Postdoc)

T +43 1 40160 - 34705

Further Information

Keywords

Biostatistics; Epidemiologic Methods; Epidemiology

Research group(s)

Research interests

My main methodological interests concern methods for time-to-event analysis and causal inference, particularly mediation analysis. My previous EU-project SCOUT (Supporting Causal Conclusions from Observational Survival Studies https://sustrohmaier.wixsite.com/scout2 ) focused on evaluating different properties of multiplicative and additive hazard models in real-life scenarios and the comparison of methods for mediation analysis in the presence of rare (binary) mediators or outcomes.

My more applied projects involve analysing large scale epidemiological studies, including the Harvard cohorts (particularly the Nurses’ Health Study 2 (NHS2) and the Growing Up Today Study (GUTS)) and the Austria based Vorarlberg Health Monitoring and Promotion Programme (VHM&PP) data. Particularly, I'm interested in intergenerational analyses – e.g. linking the NHS2 and GUTS cohorts – to study associations between maternal lifestyle (diet quality) and occupational factors (rotating night shift work) during pregnancy and adverse outcomes in their offspring during childhood and adolescence, including obesity, depression/anxiety or alterations in stress marker responses. 

Techniques, methods & infrastructure

 As large scale cohort studies result in complex data structures involving longitudinal observations and time-to-event outcomes and aim at answering causal research questions, most of my methodological research so far has focused on causal inference methods that allow to deal with these data set-ups. In particular, I was involved in developing different methods to estimate direct and indirect effects in settings with a repeatedly measured mediator and a time-to-event outcome. Other ongoing work in the context of mediation analysis focuses on the handling of rare (binary) mediators or outcomes.

A general interest for time-to-event analysis techniques as well as applications in clinical epidemiology has inspired research on the comparison between additive and multiplicative hazard models in practice. Specifically, we are comparing different aspects of Cox's proportional hazards model and Aalen's additive hazard model in context of causal modeling on the one hand, but also predictive and descriptive modelling on the other hand. 

Grants

  • SCOUT - Supporting Causal Conclusions from Observational Survival Studies (2018)
    Source of Funding: EU, H2020-MSCA-IF-2017
    Principal Investigator

Selected publications

  1. Strohmaier, S. et al., 2019. Night shift work before and during pregnancy in relation to depression and anxiety in adolescent and young adult offspring. European Journal of Epidemiology, 34(7), pp.625,
  2. Aalen, O.O. et al., 2019. Time-dependent mediators in survival analysis: Modeling direct and indirect effects with the additive hazards model. Biometrical Journal. Available at: http://dx.doi.org/10.1002/bimj.201800263.
  3. Strohmaier, S. et al., 2019. Maternal rotating night shift work before pregnancy and offspring stress markers. Physiology & Behavior, 207, pp.185–193. Available at: http://dx.doi.org/10.1016/j.physbeh.2019.05.007.
  4. Strohmaier, S. et al., 2015. Dynamic path analysis - a useful tool to investigate mediation processes in clinical survival trials. Statistics in Medicine, 34(29), pp.3866–3887. Available at: http://dx.doi.org/10.1002/sim.6598.
  5. Aalen, O.O. et al., 2017. Feedback and Mediation in Causal Inference Illustrated by Stochastic Process Models. Scandinavian Journal of Statistics, 45(1), pp.62–86. Available at: http://dx.doi.org/10.1111/sjos.12286.