Advanced multi-state modelling

Two clocks at once, everything else a fitted model will answer, and the health-economic value layer — the arithmetic of it, not a course in health technology assessment.

Taught by

Michael Crowther, PhD

CEO & Director of Statistical Methodology

Formerly Associate Professor of Biostatistics at the University of Leicester, and the author of the merlin framework.

Research survival analysis · joint models · multi-state models · statistical software

PhD in Medical Statistics, University of Leicester, 2014

Register your interest

What you will be able to do

• Fit and predict a semi-Markov transition whose intensity also depends on a clock that does not reset, and check the entry-node solver that makes it possible
• Get ever-visited probabilities, time spent in each state and standardised occupancy out of a fitted model, and say which question each one answers
• Choose between delta-method and draw-based intervals on the grounds that decide it, and check a whole analysis against a simulated study with a known answer
• Compute discounted life years, QALYs, costs and an ICER, state the discount rate and convention on every one, and read a cost-effectiveness acceptability curve
• Say what the health-economics module does not cover — evidence synthesis, extrapolation, structural uncertainty and submission mechanics — and why that matters more than the arithmetic it does cover

Before you start

• Take first: Multi-state models

Curriculum

13 steps: 11 lectures, 1 practical and 1 quiz.

Provisional — this course is still being written, and the outline will change.

Week 1 · when one clock is not enough

• 01WatchA clock that does not reset26 min
• 02WatchPredicting when two clocks run at once28 min
• 03WatchWhat the entry-node march is doing22 min

Week 2 · more questions for the same model

Opens 7 days after you start.

• 04WatchEver in a state, rather than in it now24 min
• 05WatchTime spent in each state24 min
• 06WatchStandardising over a covariate distribution26 min

Week 3 · uncertainty, and checking the method

Opens 14 days after you start.

• 07WatchDelta method or draws26 min
• 08WatchSimulate the study, then check the answer26 min

Week 4 · putting a value on the model

Opens 21 days after you start.

• 09WatchDiscounted value, and the rate you did not choose26 min
• 10QuizWhat discounting actually did8 min
• 11PracticalPractical · putting a value on the model55 min
• 12WatchIncremental cost-effectiveness, and what the CEAC says26 min
• 13ReadPartitioned survival, and where this course stops24 min