Statistical Methodology for Longitudinal Studies in Clinical
Research
IJoineR is a
collaborative research project between the universities of
Liverpool (PI: Professor Paula Williamson), Lancaster (PI:
Professor Peter Diggle) and Newcastle (PI: Professor Robin
Henderson), aiming to bring modern methods of statistical
analysis into mainstream medical research.
Emerging data collection protocols in medical research introduce
complexities which are either not covered by existing generally
available software or, more fundamentally, require further
statistical methodological development. A particular example is
the need for joint modelling of combined repeated measurements
and event-time data. With the increased usage of automatic data
capture information systems, including remote monitoring, it is
now common for repeated measurements of biomarkers to be
available alongside event history data and a major difficulty is
how best to merge information from the two sources, especially
as the biomarker is usually irregularly and imperfectly
observed. The project aims to develop new statistical
methodologies for the analysis of complex longitudinal data
structures, implement the methods in corresponding user-friendly
software in the R language and disseminate the results to the
medical research community
Some key outputs from the project:
Joint modelling of longitudinal and competing
risks data. Williamson P.R. , Kolamnunnage-Dona R. , Philipson
P. and Marson A. G. Statistics in Medicine 2008; 27(30), Pages
6426 - 6438.
Joint Modelling of Repeated Measurements and Time-to-event
Outcomes: the Fourth Armitage Lecture. Diggle P., Sousa I. and
Chetwynd A. Statistics in Medicine 2008; 27(16), 2981–-2998.
Comparative review of methods for handling drop-out in
longitudinal studies. Peter M. Philipson P. M., Ho W. K. and
Henderson R. Statistics in Medicine 2008; 27(30), Pages 6276 –
6298.
Joint modelling of time series measures and recurrent events and
analysis of the effects of air quality on respiratory symptoms.
Zhang H., Ye Y. and Diggle P. and Shi J. Journal of the American
Statistical Association 2007; 103, 48-60.
Analysis of longitudinal data with dropout: objectives,
assumptions and a proposal (with discussion). Diggle P.,
Farewell D. and Henderson R. Applied Statistics 2007; 56,
499-550.
The influence of competing risks setting on the choice of
hypothesis test for treatment effect. Williamson P. R.,
Kolamnunnage-Dona R and Tudur Smith C. Biostatistics 2006; 8:
689-694. DOI: 10.1093/biostatistics/kxl040.
More publications and other outputs:
JoineR workshops:
https://www.liv.ac.uk/medstats/courses.htm
