Graham's PhD research (1971-74; Department of Brewing and Biological Sciences, Heriot-Watt University, Edinburgh) was on the mechanism of starch breakdown during cereal germination. He then moved to the Microbiology Unit, Department of Biochemistry, University of Oxford (1974-78) to carry out postdoctoral work on the developmental biology of spore formation in the bacterium Bacillus subtilis.
Having abandoned laboratory work, Graham retrained as an applied statistician (Department of Biomathematics, University of Oxford, 1978-79) and worked at the Institute of Psychiatry, University of London, for the next 17 years. He was at first the statistician in the General Practice Research Unit (from 1979 to 1983) and then Lecturer (1983-87) and Senior Lecturer in Statistics (1987- 93) in the Institute's Biometrics Unit. He took over the role as Head of the Unit in 1990, the year in which it became the Department of Biostatistics and Computing. Graham was Reader in Biostatistics from 1993 to 1996. During that period he was also the Institute's Director of Education.
Graham was appointed as Professor of Biomedical Statistics at the University of Manchester at the end of 1996.
He is the author of several applied statistics textbooks and was a founding editor of the international review journal Statistical Methods in Medical Research.
DUNN,G. & EVERITT,B.S.(1982; Dover reprint 2004) An Introduction to Mathematical Taxonomy. Cambridge: Cambridge University Press.
DUNN,G., EVERITT,B.S. & PICKLES,A.(1993). Modelling Covariances and Latent Variables Using EQS. London : Chapman and Hall.
DUNN,G. & EVERITT,B.S.(1995). Clinical Biostatistics: An Introduction to Evidence Based Medicine. London : Edward Arnold.
DUNN, G. (2000). Statistics in Psychiatry. London: Arnold .
EVERITT,B.S. & DUNN,G.(2001). Applied Multivariate Data Analysis (2nd Edition). London : Edward Arnold.
DUNN, G. (2004). Statistical Evaluation of Measurement Errors (2nd Edition). London: Arnold.
Professor of Biomedical Statistics, Centre for Biostatistics, Institute of Population Health, University of Manchester.
Graham’s research is primarily focussed on the design and analysis of randomised trials of complex interventions, specialising on the evaluation of cognitive, behavioural and other psychological approaches to the treatment of psychosis, depression and other mental health problems. Of particular interest is the design and analysis of multi-centre explanatory trials from which it is possible to use instrumental variable (IV) methods to test for and estimate the effects of mediation and moderation, and for the effects of dose (sessions attended) and the quality of the therapy provided (including therapist effects).
Much of the above methodological research using instrumental variable methods has direct relevance to trial designs fully integrating biomarker information for the evaluation of treatment-effect mechanisms in stratified (personalised) medicine.
He also has interests in the design and analysis of measurement reliability studies. A key methodological component of both of these fields of applied research is the development and implementation of econometric methods such as the use of instrumental variables.
He is a member of the UK Mental Health Research Network (MHRN)-funded Methodology Research Group and the MRC North West Hub for Trials Methodology Research.