London - Advanced Markov/DES (2017)

2017/11/02 - 2017/11/02 | London, UK

The course will be presented in London, UK on November 2, 2017.
The course will run from 8:30 AM to about 5:15 PM.

The course location is:

Cavendish Conference Centre
22 Duchess Mews
London
W1G 9TD
United Kingdom
Tel: 01392580484

Course Flow

This one-day course extends the two-day training into advanced modeling techniques and tools.

Module 1 shows how to make models more realistic when a combination of factors – treatment, health state, patient history – affect the probability of future events. The events then drive long-term changes in probabilities, costs and utilities.

Module 2 introduces Discrete Event Simulation (DES) models, where Markov transition probabilities are replaced by time-to-event distribution sampling. We start with DES models using fixed risks, then extend this technique with more complex time-to-event calculations.

Modules 3 & 4 introduce techniques to validate your models by reviewing patient pathways and internal model calculations.

Modules 5 & 6 introduce the advanced modeling techniques to handle population dynamics and patient interaction.

Module 7 demonstrates how to examine results for a subgroup within overall results for a heterogeneous patient population.

Course Agenda

  1. Complex Markov Simulation Models
    1. Calculate event probabilities based on a complex combination of treatment, health state and patient history
    2. Apply the impact of patient history to long-term costs and utilities
  2. Discrete Event Simulation Models
    1. Build a Discrete Event Simulation (DES) model, replacing Markov event probabilities with DES time-to-event distribution sampling
    2. Integrate non-fixed risk time-to-event distributions into the model
    3. Run simulation and sensitivity analyses on DES models
  3. Patient Tracking Reporting
    1. Track patients in model to generate patient tracking reports that follow each patient through the entire model
    2. Generate cohort-level “Markov-like” reporting from patient-tracking data
  4. Review Internal Calculation Details
    1. Output internal trace data to the console to validate model calculations
  5. Population Dynamics and Budget Impact
    1. Integrate dynamic cohort size into a Markov model to study population dynamics and budget impact
  6. Patient Interaction
    1. Run patients in a synchronized time environment to represent patient interaction and/or resource constraints
  7. Subgroup Analysis
    1. Integrate real patient data for simulated patient characteristics through bootstrapping
    2. Filter simulation results to study subgroups within a heterogeneous cohort

Prerequisites

Note that experience with TreeAge Pro is required as a prerequisite for the course. You should be comfortable with all topics covered in the Two-Day Healthcare Modeling course.