External seminar series

The Centre for Biostatistics holds half-day spring and autumn seminars with external guest speakers.  The seminar in October is a joint event with the Royal Statistical Society Manchester Local Group.

Date (provisional): Wednesday 17th October 2018

Time: 14.00-17.00

Venue: Manchester Dental Education Centre (MANDEC), Higher Cambridge Street, Manchester, M15 6FH

Theme: Individual Participant Data Meta-analysis


Internal seminar series

These seminars take place on the second Monday of each month at 14:00 in G306A in the Jean McFarlane Building on the main University campus. Everybody is welcome to attend and there is no need to register in advance.

Next seminar: Monday 11th June

Presenter: Dr Edmore Chamapiwa

Title: Application/Applicability of Marginal Structural Models (MSMs) to irregular healthcare data



The uptake of Marginal Structural Models (MSMs) as a methodological alternative for causal inference particularly when treatments are time varying and confounded by time varying covariates has been increasing and continues to do so. This increase in use of this methodological approach is attributed to its ability to give unbiased treatment estimates when treatments are time varying and confounded by time varying covariates. However, MSMs have been widely used in clinical settings where individuals are observed and have their treatments reviewed at regular time intervals. In healthcare treatment settings, individuals can be observed and treated at irregular time intervals and this irregularity may well be related to previous treatments and it can also influence future treatment. Not much is known about performance of MSMs in irregular data settings. On the basis of the foregoing, this research work sought to investigate application of MSMs to irregular data settings. 

The overall aim of this research work was to investigate applicability and performance of Marginal Structural Models to estimate causal effects in health data settings where individual patients may be observed at different (or “irregular”) time intervals, and these intervals themselves may be related to treatment and confounder histories. A new class of Inverse Probability of Treatment Weighted (IPTW) estimators for irregular data was proposed to address the aim of the thesis. In this talk, I present the proposed IPTW estimator for irregular data and results from a ‘blinded simulation study’ that demonstrate how the estimators can be used in a ‘real life’ healthcare data setting.


Previous internal seminars

10th September 2012:
Richard Emsley: Principal trajectories and advances in mixture modelling
Bo Fu: Statistical solutions: debiased inference about hepatitis C virus-infected communities when analyzing diagnosed patients referred to liver clinics

8th October 2012:
Clare Flach: Selecting instrumental variables in a causal mediation analysis
Kim Hannam: Air pollution in pregnancy: A validation study in North West England evaluating air pollution exposure measurement techniques.

12th November 2012:
Evan Kontopantelis: Software and model selection challenges in two-stage and one-stage (Individual Patient Data) meta-analysis

10th December 2012:
Matthew Carr and Roseanne McNamee: Cardiovascular risk factors for patients with hypertension

11th February 2013: 
Carlo Berzuini: Statistical issues in biomarker-based early detection of cancer

11th March 2013: 
Phil Foden: "Power-up! Does statistical power increase depending on the method of analysis used in a randomised intervention trial with noncompliance?"

8th April 2013: 
Sarah Cotterill: Methodological challenges in behaviour change trials

Jack Wilkinson: Sequential Meta-analysis of rare adverse events

13th May 2013: 
Graham Dunn: Fully integrating biomarker information in trials for Efficacy and Mechanisms Evaluation (EME) in personalised (stratified) medicine

10th June 2013: 
Richard Emsley: Comparison of methods for causal mediation analysis in Stata

1st July 2013:
Jeff Morris: Feature Extraction vs. Flexible Modeling: Automated Modeling of Complex, Irregular Object Data (including Proteomics)

9th September 2013:
David Reeves and David Springate: Using electronic medical record databases to evaluate drug benefits and harms: are the results valid and replicable?

11th November 2013:
Matthew Parkes: Research in Osteoarthritis Manchester: Analysis of Clinical and Structural Parameters in Osteoarthritis Trials

9th December 2013:
Wenhua Wei: Statistics-Genetics synergy: a story of gene-gene interactions

13th January 2014:
Paula Williamson: Improving health by improving trials

10th February 2014:
Evan Kontopantelis: A reanalysis of the Cochrane Library data: the dangers of unobserved heterogeneity in meta-analyses

10th March 2014:
Omar Bouamra: Outcome prediction in trauma

12th May 2014:
Andrew Dowsey: Biomarker discovery for proteomics through statistical modelling of raw mass spectrometry data

9th June 2014:
Chon Zhou: A hierarchical statistical modeling approach to analyze proteomic isobaric tag for relative and absolute quantitation data

8th September 2014:
Hui Guo: A type I interferon transcriptional signature precedes autoimmunity in children genetically at-risk of type 1 diabetes

10th November 2014:
Matthew Sperrin: The Obesity Paradox - Does Fatter Mean Fitter?

8th December 2014:
Chris Roberts: The Analysis and Sample Size of Binary Outcome Measures for Trials with Partial Nesting

9th February 2015:
Neil O'Leary: The Implications of Differential Clustering for the Analysis of Binary Outcomes in Cluster Randomised Trials

9th March 2015:
Richard Emsley: Convex combination of ordinary least squares and two-stage least squares estimators

13th April 2015:
Chris Foley: The use of a random baseline measure in a longitudinal analysis of data from randomised controlled trials

11th May 2015:
Martyn Mcfarquhar: The Analysis of Functional Brain Images: Current Approaches, Limitations, and Future Directions

29th June 2015:
Sally Hollis: Clinical Trial Data Transparency

10th August 2015:
Lesley-Anne Carter, Matthew Gittins, Antonia Marsden: practice conference presentations

14th September 2015:
Xiaoguang Xu: Bayesian nonparametric inference for stochastic epidemic models

12th October 2015:
Susana Conde: Detecting (A) Interactions in Multivariate Binary Comorbidities and (B) Transcriptomic Toxicological Communities

9th November 2015:
Matthias Pierce: Effectiveness of drug treatment on drug related poisoning death and acquisitive crime among opioid users identified in the criminal justice system

14th December 2015:
Hui Guo: From association to causation: identification of causal genes of immune-mediated diseases using statistical colocalisation 

11th January 2016:
David Reeves: Is the “Cohort Multiple Randomised Controlled Trial” a better alternative to the pragmatic RCT?

8th February 2016:
Chris Roberts: The Design and Analysis of Group Therapy Trials

11th April 2016:
Cong Zhou: Modelling longitudinal and survival data of cancer biomarker in plasma

9th May 2016: 
Kayode Ogungbenro: Study Design and Data Analysis of Clinical Pharmacology Experiments

13th June 2016:
Carlo Berzuini: A tutorial on Mendelian Randomisation

11th July 2016:
Jack Bowden: Accounting for heterogeneity and bias in two-sample Mendelian randomization

12th September 2016:
Sarah Cotterill: Methods for evaluating health screening tools: an example of the PAT-POPs tool for screening children in the emergency department for hospital admission

19th September 2016:
Carlo Berzuini: Bayesian Mendelian randomisation

14th November 2016:
Jack Wilkinson: Analysis of multilevel multivariate mixed IVF outcome data in the presence of informative drop out

12th December 2016:
Antonia Marsden: Evaluating predictors of differential treatment response using observational data: some methodological considerations

16th January 2017:
Sarah Rhodes: What is statistical consultancy? Presenting the OSCARSS

13th February 2017:
Glen Martin: Developing Clinical Prediction Models: when and how to utilise existing research

10th April 2017:
Lesley-Anne Carter: Methods for the analysis, reporting and evaluation of experience sampling style data

15th May 2017:
Carlo Berzuini: A Bayesian Framework for Multi-Instrument Mendelian Randomisation

12th June 2017: 
Andy Vail: A career on the front-line: academic or support staff?

17th July 2017:
Matthias Pierce: Identifying and evaluating personalised treatment recommendations

11th September 2017:
Richard Emsley: Evaluating mechanisms in complex intervention trials

13th November 2017:
Emily Granger: Propensity score diagnostics: the challenges we face when providing evidence that a propensity based estimate is unbiased

11th December 2017:
Elizabeth Howarth: Missing data techniques for intention-to-treat analyses

8th January 2018:
Jamie Sergeant: Teaching and learning in medical statistics: helping non-statisticians to help their students learn, teaching medical researchers and enabling others to teach us

12th February 2018:
Jack Wilkinson: Does ovarian stimulation reduce uterine receptivity?

9th April 2018:
Sarah Cotterill: Evaluation of teh NHS Diabetes Prevention Programme: methological challenges of observation health services research

14th May 2018:
Sally Hollis: GCP: An overview with emphasis on the collection and analysis of clinical trial data