Rhian Daniel MA MSc PhD
- Rhian Daniel's Contacts
- Room G32a
- Keppel Street
- WC1E 7HT
I studied mathematics at Queens' College Cambridge for four years before coming to the LSHTM in 2004 to study for an MSc in Medical Statistics and then a PhD in missing data methods under the supervision of Mike Kenward. Since then I have been working on various projects in the field of causal inference; first I was employed on an MRC-funded grant entitled Quantitative methods for the assessment of systematic error in observational studies: improving causal research. Then I was awarded a Medical Research Council Career Development Award in Biostatistics on the topic: Methods for addressing complex causal questions relating to health research in low- and middle-income countries. Most recently, I have been awarded a Wellcome Trust / Royal Society Sir Henry Dale Fellowship on the topic Statistical methods for studying multidimensional mediators of genetic associations with chronic diseases. This will fund my research until 2020. I collaborate with Bianca De Stavola on mediation analysis, Juan Pablo Casas (UCL) on the analysis of omics data, and Stijn Vansteelandt (Ghent University) on time-dependent confounding, strong confounding and mediation analysis. In 2014 I took one year's maternity leave following the birth of our son, Edryd.
I am co-organiser of the Advanced Statistical Modelling module for the MSc in Medical Statistics, and am responsible for the delivery of the first half of this module which covers various topics in statistical causal inference.
I am a co-organiser of the short course in Causal Inference in Epidemiology: Recent Methodological Developments and teach four sessions (lecture+practical) on this course (on methods based on the propensity score, two on time-dependent confounding, and on sensitivity analysis).
I also teach a session on causal inference for the short course Advanced Course in Epdiemiological Analysis, a session (on missing data) on the MSc module Advanced Statistical Methods in Epdeimiology (ASME), and a session on instrumental variables for the Advanced Research Methods (ARM) MSc module.
I have been involved in external short course teaching, in Nairobi (on missing data) and in Rome, Florence, Turin, Bologna, Stockholm, Uppsala, Oslo, Cardiff, Swansea, Edinburgh, Glasgow, Cambridge, Exeter, York, Manchester, Bristol, the UK-CIM and the RSS in London (causal inference).
I am interested in statistical methods for dealing with time-dependent confounding in longitudinal observational studies and also in the related problem of estimating direct and indirect causal effects, particularly in the presence of intermediate confounding and/or multiple mediators.
My current Sir Henry Dale fellowship, Statistical methods for studying multidimensional mediators of genetic associations with chronic diseases, focusses on the use of causal inference methods, particularly mediation analysis with multiple mediators, for high-dimensional omics data.
- Statistical methods
- Causal Inference
- Centre for Statistical Methodology
- Missing Data
Causal mediation analysis with multiple mediators.
Daniel, R.M.; De Stavola, B.L.; Cousens, S.N.; Vansteelandt, S.;
Biometrics, 2015; 71(1):1-14
Mediation Analysis With Intermediate Confounding: Structural Equation Modeling Viewed Through the Causal Inference Lens
de Stavola, B.L.; Daniel, R.M.; Ploubidis, G.B.; Micali, N.
American Journal of Epidemiology, 2015; 181(1):64-80
On regression adjustment for the propensity score.
Vansteelandt, S.; Daniel, R.M.;
Stat Med, 2014; 33(23):4053-72
Methods for dealing with time-dependent confounding.
Daniel, R.M.; Cousens, S.N.; De Stavola, B.L.; Kenward, M.G.; Sterne, J.A.;
Stat Med, 2013; 32(9):1584-618
Efficient estimation of the distribution of time to composite endpoint when some endpoints are only partially observed.
Daniel, R.M.; Tsiatis, A.A.;
Lifetime Data Anal, 2013; 19(4):513-46
Using causal diagrams to guide analysis in missing data problems.
Daniel, R.M.; Kenward, M.G.; Cousens, S.N.; De Stavola, B.L.;
Stat Methods Med Res, 2012; 21(3):243-56
A method for increasing the robustness of multiple imputation
Daniel, R.M.; Kenward, M.G.
Computational Statistics & Data Analysis, 2012; 56(6):1624-1643
gformula: Estimating causal effects in the presence of time-varying confounding or mediation using the g-computation formula
Daniel, R.M.; De Stavola, B.L.; Cousens, S.N.;
The Stata Journal, 2011; 11(4):479-517
- → View all Rhian Daniel 's publications