Indirect and Mixed Treatment Comparisons: 3 day course - September 2012
Overview
This course is for health economists, statisticians and decision modellers, and systematic reviewers interested in the extension of pair-wise meta-analysis to indirect and mixed treatment comparisons, in the context of either clinical effectiveness or economic evaluation.
The course focuses on Bayesian methods for statistically combining evidence from networks of trials, integrating statistical estimation within a probabilistic modeling framework. The assumptions underlying both pair-wise meta-analysis and mixed treatment comparisons are critically examined. The course also covers methods for detecting and managing heterogeneity and inconsistency.
This is an informal, hands-on course, based on a mixture of lectures and practical work on published datasets using the Bayesian Markov chain Monte Carlo package WinBUGS. Course tutors are available throughout to answer questions and help with exercises.
It is a collaboration between the Department of Health Sciences, University of Leicester and the School of Social and Community Medicine, University of Bristol.
The methods taught on the course were updated in 2011 to be compatible with the NICE 2008 Methods Guide and the Technical Support Documents on Evidence Synthesis (http://www.nicedsu.org.uk), and with the Report of the ISPOR Task Force on Indirect Comparisons.
Next Course
September 17th - 19th 2012 - Vaughan College, Leicester, UK.
Intended audience
- Anyone undertaking or managing health technology assessments, including in the context of cost-effectiveness analysis,
- Statisticians, familiar with the principles of meta-analysis, who wish to learn about Bayesian methods for evidence synthesis particularly in the context of cost-effectiveness analysis,
- Anyone responsible for managing systematic reviews.
What you will learn
By the end of the course participants will be able to:
- Conduct pair-wise, indirect comparison and mixed treatment comparison (IC/MTC) evidence synthesis using WinBUGS Bayesian software,
- Adjust for covariates,
- Integrate statistical evidence synthesis with probabilistic cost effectiveness analysis,
- Assess the degree of heterogeneity and inconsistency in RCT data,
- Understand the assumptions and potential pitfalls in pair-wise and IC/MTC meta-analysis,
- Participants will also have an introductory understanding of Bayesian methods, hierarchical modeling, and be able to use WinBUGS software.
Course Programme
Day 1: 08.45 - 17.00
- Registration: 08.45 - 9.00
- Introduction to indirect and mixed treatment comparisons
- Introduction to Baysian inference and WinBUGS
- Bayesian meta-analysis: fixed and Random effects
- Heterogeneity
- Meta-regression and adjusting for baseline risk
- Course dinner - 19.30
Day 2: 09.00 - 17.00
- Model critique in meta-analysis: measures of fit, cross-validation
- Integrating meta-analysis and cost-effectiveness modelling
- Mixed Treatment comparison: fixed and random effect models
- Embedding an MTC analysis in a cost-effectiveness model
- MTC with continuous and rate outcomes
Day 3: 09.00 - 15.30
- Inconsistency and heterogeneity: methods and implications
- Ground rules and checklist for MTC and IC
- Participant presentations* or paper critique (to be decided)
- MTC and meta-regression
- Introduction to MTC with multiple end-points
- Course ends 15.30
* Delegates who wish to do so are invited to give brief presentations on MTC datasets, synthesis problems, or policy issues they have encountered. Please consult with Tony Ades (t.ades@bris.ac.uk) or Alex Sutton (ajs22@le.ac.uk) if you are considering doing this.
Course Pre-requisites
- Participants should be familiar with the basic principles of meta-analysis, and have a good working knowledge of logistic regression and statistical interaction,
- Experience with probabilistic decision analysis in cost effectiveness analysis would be an advantage, but is not necessary,
- Participants are expected to have downloaded WinBUGS prior to the course, and to have worked through the on-line tutorial, and explored the on-line BLOCKER example (Help_Examples Vol 1_Blocker). NB you need to use initial values Inits2 for this to run.
Important: Delegates must bring their own laptop PCs downloaded with WinBUGS 1.4.3 software and must have registered and decoded the key. Software and details available at http://www.mrc-bsu.cam.ac.uk/bugs/welcome.shtml. Delegates will also need calculators or spreadsheet software for some of the course exercises.
(Back to top)
Course Faculty
Keith Abrams PhD is Professor of Medical Statistics at the University of Leicester. His primary research focus is on the use of Bayesian methods in bio-medical research and health-care evaluation, especially RCTs, evidence synthesis and decision modeling.
Tony Ades PhD, Professor of Public Health Science, leads the programme in Multi-Parameter Evidence Synthesis in Epidemiology and Decision Making at University of Bristol. His particular interests are in epidemiological applications, particularly in screening and infectious disease.
Deborah Caldwell PhD is a Research Fellow at the University of Bristol. She recently completed a PhD on Indirect and Mixed Treatment Comparisons, with a focus on applications of these methods in Health Technology Assessment and their policy implications.
Nicola Cooper PhD is a Senior Research Fellow at the University of Leicester. Her primary research interest is the integration of statistical evidence synthesis with probabilistic economic decision modeling.
Sofia Dias PhD is a Research Fellow at University of Bristol. She is a statistician with extensive experience of systematic review and meta-analysis, and is statistical editor for the Cochrane Menstrual Disorders and Sub fertility Group. Her recent interests are using MTC evidence structures to estimate bias in RCTs, and in developing methods for assessing inconsistency in MTCs.
Alex Sutton PhD is a Professor in Medical Statistics at the University of Leicester. He has a long-standing interest in methods for evidence synthesis; particularly when used for decision making. His current research includes synthesis methods for adverse events, diagnostic test performance and dealing with publication bias.
Nicky Welton PhD is a Senior Research Fellow at University of Bristol. She is a Bayesian statistician with extensive experience of evidence synthesis including Mixed Treatment Comparisons and extensions to multiple outcomes, synthesis with collapsed categories, methods for potentially biased RCT evidence, and epidemiology of infectious disease synthesis. Expected Value of Information analysis is an area of special interest.
Course fees
Student: £450
Academic, public sector: £700
Commercial sector: £1300
The course fees cover: tuition and course materials, lunch, tea & coffee breaks, course dinner. Fees do not include accommodation.
Registration
If you wish to enrol on the course, please complete the following registration form and post to Katrina Crook at the School of Social and Community Medicine, University of Bristol, Canynge Hall, 39 Whatley Road, Bristol, BS8 2PS or email to katrina.crook@bristol.ac.uk.
Registration form (Word doc 86Kbs)
Cancellation must be received in writing 2 weeks before the course and will be subject to an administration charge of 10% of course fees. If you cancel after this time and we cannot fill your place then you will be liable for the full workshop fee. You can make substitutions at no charge, provided we are notified in writing at least 4 weeks in advance of the course. After that date there will be a 10% administration charge.
In the unlikely event the course is cancelled, our liability is limited to refunding workshop fees,
Course Venue
Vaughan College, University of Leicester, 156-160 St Nicholas Circle, Leicester, LE1 4LB. Tel (44) 116 251 7368. Map / directions / travel to Vaughan College, University of Leicester.
Accommodation
Delegates must arrange their own accommodation. The following hotels are near Vaughan College.
Holiday Inn, 129 St Nicholas Circle, LE1 5LX. http://www.ichotelsgroup.com/h/d/hi/925/en/home (5 minutes).
Travelodge Leicester Central Hotel, Vaughan Way, LE1 4NN. http://www.travelodge.co.uk/ , (7 minutes, rooms not overlooking the main road are quieter).
Ramada Leicester, 79 Granby Street, LE1 6ES. http://www.ramadajarvis.co.uk , (15-20) minutes.
The Belmont Hotel, De Montfort Street, Leicester, Leicestershire, LE1 7GR http://www.belmonthotel.co.uk/ (20-25 minutes)
Further enquiries
Please direct administrative enquiries to Katrina Crook: katrina.crook@bristol.ac.uk or telephone (44) 117 331 4566. For enquires about course content, please contact Tony Ades t.ades@bris.ac.uk or Alex Sutton ajs22@le.ac.uk.