MAS6062 Bayesian Methods and Clinical Trials

Both semesters, 2019/20 20 Credits
Lecturer: Dr Miguel Juarez uses MOLE Timetable Reading List
Aims Outcomes Assessment Full Syllabus

The module addresses two areas of statistical thinking and methodology, both important for medical applications. The first is the Bayesian approach to inference and decision-making, in which uncertainty in our knowledge is described by probabilities and combined optimally with observational information. The module introduces both the fundamental concepts of Bayesian inference and practical computational methods for implementation. The second part introduces a variety of clinical trial designs found in both commercial companies and in technology evaluation. The range is from laboratory early Phase trials, through pharmaceutical trials to evaluation of health technology, including evaluation of the economic component of trials.

There are no prerequisites for this module.
No other modules have this module as a prerequisite.

Outline syllabus

  • Semester 1: Bayesian Statistics
    • Subjective probability.
    • Inference using Bayes Theorem. Prior distributions. Exponential families. Conjugacy. Exchangeability.
    • Predictive inference.
    • Utility and decisions. Tests and interval estimation from a decision-theoretic perspective.
    • Hierarchical models.
    • Computation. Gibbs sampling. Metropolis-Hastings. Case studies.
    • Linear regression.
  • Semester 2: Further Clinical Trials
    • Basics
    • Cluster trials
    • Phase I/II
    • Bayesian Mixed Treatment; further survival analysis
    • Sample size for survival
    • Monitoring in clinical trials


  • To familiarize the student with the Bayesian approach to inference.
  • To describe computational implementation of Bayesian analyses.
  • To extend knowledge of clinical trials.

Learning outcomes

  • understand and apply Bayesian ideas of prior-posterior updating,
  • understand the concepts of utility and maximization of expected utility in decision making,
  • understand the idea of Gibbs sampling and apply it to practical problems in Bayesian inference,
  • understand aspects of the nature and design of clinical trials,
  • offer advice on the design and size of clinical trials for a variety of objectives, and will:
  • be familiar with the different phases of clinical development,
  • have some appreciation of health economics in the context of economic evaluations alongside clinical trials,
  • understand the role of clinical trials in health technology assessment,
  • have knowledge of the main regulatory guidelines for the design, conduct and analysis of clinical trials,
  • have a basic understanding of pharmacokinetics and pharmacodynamics,
  • be familiar with concepts such as clustering in randomized trials and complex interventions,
  • have awareness of issues associated with multiperiod trials and bioequivalent evaluation.

34 lectures, no tutorials, 6 practicals


Semester 1: One project worth 15% and an 85% exam in May/June.

Semester 2: Two projects worth 30% and 70%, each. (The contents remain the same, only the assessment has changed.)

Full syllabus

Bayesian theory

  • The subjective interpretation of probability. Constructing subjective probabilities.
  • Independence and exchangeability.
  • Inference using Bayes Theorem. Discrete examples.
  • Prior distributions. Exponential families. Conjugacy.
  • Continuous examples: normally distributed data with known variance, binomial data.
  • Continuous examples: poisson and normal distributions with unknown variance.
  • Predictive inference.
Decision theory and its role in inference
  • Utility and decisions. Maximising expected utility.
  • Point estimation, interval estimation and hypothesis testing from a decision-theoretic perspective.
Bayesian modelling
  • Hierarchical models
  • Model checking. Robustness. Sensitivity.
Bayesian computation with Markov Chain Monte Carlo (MCMC) methods
  • Gibbs sampling.
  • MCMC using R
  • R practicals: case studies. (practical sessions)
(inter-semester break)
Further Clinical Trials: basics
  • Overview of clinical development
  • Overview of ICHE9 (1 session)
  • Introduction to GCP (1 session)
  • Data structures and storage (1 session)
  • Sample size calculations (1 session)
  • Adaptive designs (1 session)
  • Introduction to cross-over trials (1 session)
  • Multi-period crossover trials (1 session)
Cluster trials
  • Introduction to cluster trials (1 session)
  • Simple analysis of cluster trials (1 session)
  • General methods of analyzing cluster trials (1 session)
  • Sample size calculations for cluster trials (1 session)
Phase I/II
  • Intro to Pharmacokinetics (1 session)
  • First time in man (1 session)
  • Trials for Bioequivalence (1 session)
Guest lectures
  • Bayesian Mixed treatment comparisons 1 (1 session)
  • Bayesian Mixed Treatment comparisons 2 (1 session)
  • Further Survival 1 Cox models (1 session)
  • Further Survival 2 Parametric models (1 session)
Further topics
  • Sample size for survival (1 session)
  • Monitoring in Clinical Trials (1 session)

Reading list

Type Author(s) Title Library Blackwells Amazon
A Gelman, Carlin, Stern and Rubin Bayesian Data Analysis
A Julious, Tan and Machin An introduction to statistics in early phase trials
B Campbell and Walters How to design, analyse and report cluster randomised trials in medicine and health services research
B Campbell, Machin and Walters Medical Statistics: A Textbook for the health sciences
B Lee Bayesian Statistics: An Introduction

(A = essential, B = recommended, C = background.)

Most books on reading lists should also be available from the Blackwells shop at Jessop West.

Timetable (semester 2)

Mon 10:00 - 10:50 lecture   Hicks Lecture Theatre 10
Wed 10:00 - 10:50 lecture   Hicks Seminar Room F24