Bookings for this event are currently closed. Please directly with any queries.

Graham Dunn Seminar: 13th May 2026

Name: Graham Dunn Seminar: 13th May 2026
Start: 2026-05-13T13:00:00+01:00
End: 2026-05-13T13:00:00+01:00
Location: Room G107 Alan Turing Building, University of Manchester

About

You are warmly invited to the upcoming Graham Dunn Seminar on Wednesday, 13th May 2026, featuring a series talks on the usage of Electronic Health Records data in healthcare research. Hosted in person at the University of Manchester, this event will both showcase practical experiences in working with electronic health records data (EHR) and discuss how we can make the most of it.

We have three confirmed speakers for this seminar:
Prof Joanne Knight, Professor Joanne Knight, Chair in Applied Data Science, University of Lancaster.
Title: "Insights from NHS data - examples from Neurology"
Abstract:
The Health Data Research Service (HDRS) has been announced with significant funding. Why? I want to walk through a few examples of what NHS data can teach us but also raise a couple of flags about the challenges we still face. The good news stories I will discuss include:
(1) NHS data can be used to understand unmet need - (prevalence of cerebral small vessel disease).
(2) NHS data can be used to investigate things that happen unexpectedly and we are not in a position to randomise (strikes).
(3) NHS data can be used to uncover inequity (differential admission by socioeconomic status).
The challenges I will touch on include the fact that the data that is required to populate the HDRS is not yet complete and requires work in terms of both collection and coding also the data access process is difficult to navigate not least because it is inconsistent.

Dr Matthew Gittins, Lecturer in Biostatistics, University of Manchester
Title: “The Infected Blood Inquiry, helping to unpick a mess.”
Abstract:
In the 1970s and 80s developments in preventative treatments for those with blood borne diseases, such as haemophilia and von Willebrand’s disease, led to a rapid worldwide increase in the need for donated blood. As the need outpaced supply many countries, including the UK, began to source supplies of blood plasma and its derivatives from the United States. Unfortunately, blood supplies in the US during this period were primarily maintained through paid donations and certain targeted campaigns, resulting in an increased proportion of donations coming from the homeless, prisoners, and those with drug addictions. A lack of screening during this time meant that those receiving the preventative treatment were at increased risk of being exposed to contaminated donations resulting in further infection from HIV and Hepatitis C. In 2017 the then prime minister Theresa May announced a long-awaited public inquiry to assess the impact, culpability, compensation, and hopefully identify measure to prevent a repeat of the mistakes made. In this talk I will outline my small contribution to the evidence provided to the inquiry by the statistical expert group. Specifically, this talk will discuss how we explored the long-term health effects of exposure using the limited available data from a national clinical cohort registry using historical Electronic Health Records for patients with blood borne diseases who are identified as being infected with HIV and/or Hepatitis C. In additional, I will look to expand on traditional methods of calculating Years of Life Lost by introducing Royston Palmer flexible parametric survival models to estimate ‘condition specific’ Years of Life Lost.

Dr Fiona Lugg-Widger, Director of Data, Centre for Trials Research, University of Cardiff
Title: "EHR Data in Trials: What we think happens vs what actually happens”
Abstract:
Fiona Lugg-Widger, Director of Data at the Centre for Trials Research, Cardiff University, and Co-Director of the HDR UK Transforming Data for Trials programme, explores the growing role of health systems (routine) data in clinical trials. Drawing on operational experience and methodological insights, this talk highlights key statistical challenges, including data quality, outcome definition, and linkage, alongside emerging approaches such as data utility frameworks and synthetic controls. It also emphasises the importance to share learning and build capability across the trials community to support more robust, data-enabled research.

The seminar will be hosted at the University of Manchester’s Alan Turing building, in room G.107. The seminar will run from 1pm to 4pm, and will include a break for refreshments. Please register in advance using the "Book Now" button above.