BSc, MSc, PhD, CStat
OCTRU Lead Statistician
Expertise in biostatistics, multilevel/hierarchical statistical modelling, statistical data-science, multi-parameter evidence synthesis, and randomized trials.
I am a biostatistician working on the design, conduct, analysis, and reporting of clinical trials at the Oxford Clinical Trials Unit and the Centre for Statistics in Medicine. My main interest is creating standardized methodology to support the production of evidence in real-world medical practice.
Standardization methodology can be used to contextualize RCT evidence with real-world data and to use these data in the design of clinical trials to make them more efficient, improve their external validity, and support the translation of findings in clinical practice. Improving trials in this way can create a virtuous circle with the potential to improve the quality of real-world data and in turn support more efficient and relevant clinical trials. More broadly, I am interested in the application of hierarchical statistical modelling, evidence synthesis, and causal methodology in observational studies and clinical trials to solve practical problems.
I have developed methods in comparative effectiveness and mapping between patient-reported and clinical outcomes. These methods allow for simultaneous synthesis and mapping of treatment effects across different measurement scales, giving us estimates of instrument relative responsiveness. I have also developed different approaches for estimating the minimum clinically important difference for depression. My research portfolio is broad and spans from theoretical work linking multilevel modelling with measurement error and misclassification, to more applied spatio-temporal modelling, and multi-parameter evidence synthesis to inform policy decisions and program evaluation. More recently, I co-led a project (MAPPED) to develop and validate clinical prediction tools for psychosis using routine primary care data (Clinical Practice Research Datalink).
I joined OCTRU in May 2023. Before that, I spent two years in NHS Blood and Transplant working on registry-based randomized clinical trials and almost 15 years working in biostatistics, epidemiology, and evidence synthesis at the University of Bristol, Bath, and Manchester. I am a chartered Statistician (RSS), trained in Mathematics (BSc) and Computational Statistics (MSc). The MRC supported my PhD on life-course statistical epidemiology at the Environment Epidemiology Unit in Southampton, which was followed by an ESRC post-doctoral award at the Centre for Multilevel Modelling at Bristol University.
My ORCID: https://orcid.org/0000-0002-6392-6690
Developing and internally validating a prognostic model (P Risk) to improve the prediction of psychosis in a primary care population using electronic health records: The MAPPED study.
Sullivan SA. et al, (2022), Schizophrenia research, 246, 241 - 249
How much change is enough? Evidence from a longitudinal study on depression in UK primary care
Kounali D. et al, (2022), Psychological medicine, 52, 1875 - 1882
Is there a role for Desmopressin as an alternative to platelet transfusion prior to procedures or Radiological InterVEntions in thrombocytopenic critically ill patients (DRIVE)? Results of a randomised double-blind placebo-controlled feasibility trial
Desborough M. et al, (2022), British journal of haematology, 197, 40 - 41