• Research
  • Publications
  • Talks
  • Books
    • BMHE
    • BCEA book
    • BMPR
  • Software
    • BCEA
    • survHE
    • blendR
    • bmhe
    • SWSamp
    • bmeta
    • INLA
  • Blog
  • Teaching
    • Summer school

Summer school: Bayesian methods in health economics

Last updated

January 17, 2025

Aims, objectives and intended audience

This summer school aims at providing an introduction to Bayesian analysis and Markov Chain Monte Carlo (MCMC) methods using R and MCMC sampling software (such as OpenBUGS and JAGS), as applied to cost-effectiveness analysis and typical models used in health economic evaluations.

We will also focus on more recent methods for Probabilistic Sensitivity Analysis and research prioritisation, including Value of Information calculations.

The summer school is intended for health economists, statisticians, and decision modellers interested in the practice of Bayesian modelling and will be based on a mixture of lectures and computer practicals. The emphasis will be on examples of applied analysis: software and code to carry out the analyses will be provided and presented/discussed in details.

Important

Participants are encouraged to bring their own laptops for the practicals. We shall assume a basic knowledge of standard methods in health economics and some familiarity with a range of probability distributions, regression analysis, Markov models and random-effects meta-analysis. However, statistical concepts are reviewed in the context of applied health economic evaluations in the lectures.

We shall also assume at least basic familiarity with R, although, once again, we shall have a specific focus on the practicals and explain in details the practice of (Bayesian) modelling for health technology assessment.

Lectures and practicals are based on the following textbooks: Bayesian Methods in Health Economics (BMHE), Bayesian Cost-Effectiveness Analysis with the R package BCEA (BCEA), The BUGS Book (BB) and Evidence Synthesis for Decision Making in Healthcare (ESDM).

The summer school essentially is made of two halves — the first main topic is Bayesian modelling; the second is its application to health-economic evaluation. These two will be intertwined throughout — we will switch back and forth to the description of the methods and their application to data on cost-effectiveness of (mainly, but not exclusively) pharmaceutical interventions.

In any case, the main learning objective of this summer school is for you to be able to perform a Bayesian analysis (specifically on health care data and with the objective of a full economic evaluation). We will be formal in the exposition of the technical concepts, while not fixating with proofs and theorems (but rather trying to provide the rationale and the intuition behind the use of the various methods).

Important

In order to achieve this goal, much of the summer school will concentrate on practical skills, such as programming and doing data analysis in R and Bayesian software such as BUGS. Both pieces of software are publicly available and you can install them on your machine. You probably have used R in your previous work/studies. In any case, you will be given a lot of material to make up for any lack of knowledge. We expect you to go through the exercises and try to improve your computational skills throughout the summer school. As for BUGS, you probably have not seen it before — but again, you will be provided with a wealth of material and so, by the end of the week, you should be able to write, run and debug BUGS code proficiently!

Details

Faculty

The summer school is taught by Gianluca Baio, Anna Heath, Nathan Green, Howard Thom and Natalia Kunst.

Acknowledgements

🙏 Special thanks go to Mark Strong, Chris Jackson and Nicky Welton, who have all been involved in previous versions of the summer school (and might be again, in future ones…) and have been instrumental in creating some of the material we present here.

Daily schedule

The following is the tentative daily scheduled for the next edition — topics and lecturers may slightly vary, but by and large the list of topics and the setup of the summer school will remain very close to this. Click on the banners below to expand the full schedule for each day.

Day 1 — Monday
Start End Topic
10:00 11:00 Lecture 1: Introduction to Bayesian reasoning, computation and BUGS
11:00 11:15 Coffee break
11:15 12:00 Practical 1. Monte Carlo analysis
12:00 13:00 Lecture 2: Learning from data using MCMC and BUGS
13:00 14:00 Lunch
14:00 15:15 Practical 2. MCMC analysis
15:15 15:30 Coffee break
15:30 16:30 Lecture 3: Introduction to health economic evaluation
16:30 17:00 Practical 3. Introduction to R and cost-effectiveness analysis using BCEA
Day 2 — Tuesday
Start End Topic
09:00 10:00 Lecture 4: Individual level data in health economics
10:00 11:00 Practical 4. Cost-effectiveness analysis with individual-level data
11:00 11:15 Coffee break
11:15 12:15 Lecture 5: Aggregated level data
12:15 13:15 Practical 5. Evidence synthesis and decision models
13:15 14:15 Lunch
14:15 15:15 Lecture 6: Evidence synthesis and network meta-analysis
15:15 15:30 Coffee break
15:30 16:30 Practical 6. Network meta-analysis
Day 3 — Wednesday
Start End Topic
09:00 10:00 Lecture 7: Survival analysis in HTA
10:00 10:45 Practical 7. Survival analysis
10:45 11:00 Coffee break
11:00 12:00 Lecture 8: Population adjustment: Multilevel regression-NMR
12:00 13:00 Practical 8. Population adjustment: Multilevel regression-NMR
13:00 14:00 Lunch
14:00 15:00 Lecture 9: Population adjustment: Multiple imputation marginalisation
15:00 15:15 Coffee break
15:15 16:00 Practical 9. Population adjustment: Multiple imputation marginalisation
Day 4 — Thursday
Start End Topic
09:00 10:00 Lecture 10: Missing data in cost-effectiveness modelling
10:00 10:45 Practical 10. Missing data
10:45 11:00 Coffee break
11:00 12:00 Lecture 11: Markov models
12:00 13:00 Practical 11. Markov models
13:00 14:00 Lunch
14:00 15:00 Lecture 12: Model error and structural analysis
15:00 15:15 Coffee break
15:15 16:00 Practical 12. PSA to structural uncertainty
Day 5 — Friday
Start End Topic
09:00 10:15 Lecture 13: Introduction to Value of Information
10:15 10:30 Coffee break
10:30 11:15 Practical 13. Computing the EVPPI in BCEA and SAVI
11:15 12:00 Lecture 14: Calculating expected value of sample information
12:00 12:45 Practical 14. Computing the EVSI using Monte Carlo simulations
12:45 13:45 Lunch
13:45 14:15 Lecture 15: Regression-based EVSI
14:15 15:00 Practical 15. Computing EVSI using regression

Next edition — London, 23-27 June 2025

This year, the summer school will be held at University College London (UCL), Department of Statistical Science, room 113, on the first floor of 1-19, Torrington Place, WC1E 7HB.

Travelling to London

London is of course a major hub and is served by several airports. Heathrow and Gatwick are the main ones, but there are connections also to London City, Stansted and London Southend. Most of them are fairly distant from the city centre, but all have quite good connections using Transport for London (TfL), which includes the London Underground (or, “The Tube”) as well as metropolitan rail and bus services. There are also coaches going directly to Central London from any of the airports – some options are detailed here. Train tickets can be booked and bought at Trainline.

UCL is bang in the middle of Central London, close to landmarks such as Piccadilly Circus, Oxford Street and the West End, with plenty of restaurants and amenities (not that you’ll have time to enjoy them, given the packed schedule… 😉).

The campus is within walking distance from at least three Tube stations:

  1. Goodge Street, on the Northern Line;
  2. Tottenham Court Road on the Central, Northern and Elizabeth Lines;
  3. Warren Street on the Victoria and Northern Lines

and very close to major railway stations (London Euston and London King’s Cross/St Pancras).

TfL has a very handy Journey Planner that can be used to… well: plan your journey from anywhere within the network.

Hotels

There are a million hotels around UCL although you may choose to stay in different areas of (Central) London and use TfL to get there. Here are some hotels nearby.

  • hub by Premier Inn London Goodge Street hotel. This is pretty minimal, but literally in front of UCL.
  • Radisson Blu Hotel, London Euston Square. Slightly more distant (but only 6 minutes away…).
  • Four Points Express by Sheraton London Euston. If you fancy walking 15 minutes to UCL…

Off term (which the Summer School is), UCL also offers accommodation in the Student Halls. This is a website providing more information. This may be a cheaper and more cost-effective alternative.

Registration

Participation fees

There are three types of partecipation fees:

  1. Students (£500)
  2. Public sector (£1,200)
  3. Private sector (£1,800)

These course fees include tuition and access to the course material (slides, R/BUGS code/scripts for the practicals, relevant papers, etc). Accommodation and dinners are excluded and participants should arrange separately.

Registration process

Registration is now open and available through UCL Online Store. The deadline for registration is set to 31 May 2025 and there are 20 places available – so hurry!


         

© Gianluca Baio 2022-2024