Dr Joie Ensor Bsc, MSc, PhD

Dr Joie Ensor

Department of Applied Health Sciences
Associate Professor in Biostatistics

Contact details

Address
Public Health Building
University of Birmingham
Edgbaston
Birmingham
B15 2TT
UK

Joie’s research interests span the complete workflow of clinical prediction models (CPMs) from initial design through post-implementation monitoring and adaptation. He is passionate about applying right-to-left, implementation-first thinking to the design of CPMs as decision support tools in healthcare.

Qualifications

  • Fellow of the Royal Statistical Society (Oct 2021 – Present)
  • PhD in Biostatistics (“Evidence synthesis for prognosis and prediction: Application, methodology and use of individual participant data”), University of Birmingham, 2017
  • MSc in Medical Statistics (with specialisation in HTA), University of Leicester, 2012
  • BSc (Hons) in Mathematics, University of Leicester, 2010

Biography

Joie is an Associate Professor in Biostatistics working within the BESTEAM research group (Biostatistics, Evidence Synthesis, Test Evaluation and prediction Modelling). He has worked on numerous methodological and applied research projects and developed widely used statistical methodologies and software tools to implement these in practice. He is dedicated to improving patient outcomes through rigorous biostatistical research and education.

Joie has extensive expertise in prediction modelling and evidence synthesis and has held previous positions at the University of Birmingham and the Centre for Prognosis Research at Keele University. His work focuses on the interface between statistical methodology, practical application, and education – each pillar reinforcing and informing the others.

Dr. Ensor has broad experience in prognostic and diagnostic research. His earlier work focussed on the development of methodology for the initial phases of the CPM workflow and in reporting of CPM research. His recent research is particularly focused on the implementation of prediction models and their impact on behaviour change and patient outcomes.

As developer and maintainer of the widely used 'pm-suite' software packages for prediction modelling (https://github.com/JoieEnsor), he has substantial technical expertise in statistical programming, simulation methods, and computational approaches to solving complex statistical problems. He believes that research methodology should be easy to implement in practice, for example through user-friendly software implementations and tutorials.

Teaching

Under/Post-graduate modules:

  • Public Health MPH/Diploma/Certificate –
    • Systematic Reviews and Evidence Synthesis (SRES),
    • Epidemiology, Statistics and Research Methods (ESRM),
    • Practical Epidemiology and Statistics (PEaS),
    • Dissertation supervision.
  • Medicine and Surgery (MBChB) — Medical Statistics in Professional and Academic Skills (Module lead)
  • Artificial Intelligence Implementation (Healthcare) MSc 

Continuing Professional Development

Postgraduate supervision

Joie supervises several PhD students undertaking applied and methodological research on clinical prediction modelling projects with a focus on implementation of these tools in practice.

Joie would be interested in supervising research students on projects in the following areas of clinical prediction modelling with a mixture of methodological and applied themes:

  • Evaluation (design, targeted validation, evidence thresholding, decision analytic methods)
  • Implementation (design, practicalities, barriers, efficiency, communication)
  • Impact (study design, metrics, decision analytic approaches)
  • Behaviour (UX design, interaction with CPM tools, education, communication and understanding for end-users)

Research

Research interests

Joie’s research interests primarily focus on prognosis research (overall prognosis, prognostic factor, prognostic model and predictors of treatment effect research) and evidence synthesis (univariate/multivariate meta-analysis, IPD meta-analysis, network meta-analysis methods, systematic reviews of prognosis studies) and often explore the intersection between these two topics (e.g., exploiting meta-analysis methodologies for prognostic modelling using clustered datasets).

Joie is interested in the complete workflow of clinical prediction models and is particularly focussed on i) increasing the impact of CPMs as decision support tools on patient and system outcomes, and ii) reducing the vast research waste seen in the CPM field. His ambition is to embed right-to-left thinking in all CPM research, with the aim of reducing research waste through a more targeted approach from the outset - evaluating implementation requirements and end-user/system needs at inception to increase the rate of successfully implemented and impactful CPMs as decision support tools.

Current projects