GW HTA lecture 6

Question 1

Models in HTA are used to:

Answer 1

• compare all the data
• use all evidence
• extrapolate in the future

Question 2

Aim of deciion models

Answer 2

inform decision makers

Question 3

key elements in a decision model

Answer 3

• input parameters
• expected values (QALY, effectinvess, ICER)
  Modelling aims to predict the unkown time cycles –> from intermediate to final cycle or observing until end of time horizon

Question 3

name three assumptions of extrapolating

Answer 3

• one-time-benefit
• rebound-effect
• continuous effect

Question 4

assumption one-time-benefit

Answer 4

rate of death becomes the same over the long term

Question 5

assumption of rebound effect

Answer 5

curve converge more rapidly after the trial follow-up (conditional rate of death becomes even higher with treatment)

Question 6

assumptions continuous treatment

Answer 6

curves continue to diverge in the longer term

Question 7

Decision tree

Answer 7

illustrates the proabilities of oucomes driven by the treatment alternatives and the events that can happen.

Question 8

Modelling pros

Answer 8

• makes explicit definition of relevant patient group, clinical events, patient outcome, costs etc. necessary
• makes is possible to examine the impact of input uncertainty and outcome
• Releatively fast and sample (compared to empircal research) and relatively cheap

Question 9

Modelling cons

Answer 9

• Oversimplifies the complicated, real world
• Model structure subject to bias
• Model input subject to bias
• Misinterpretation of results is easy

Question 10

conclusion about models

Answer 10

Modelling ia powerful tool to synthesize available data and information in order to (economically) evaluate health care programmes.

Question 11

Modelling often used in economic evaluations –> limited evidence

Answer 11

• structure
• evidence; combines data sources and extrapolates (time horizon)
• uncertainty

Question 12

Markov model

Answer 12

• reflects coninuous risk over longer period
• Organized around health states rather than pathways
• Probabilities relate to transistions between health states
  Cycle strength defines period of transition

Question 13

Markov assumption

Answer 13

States that the probability of moving from one state to the other depends on the current state.
- previous health states are disregarded
- time spent in current state is disregarded

Question 14

cycle length

Answer 14

Length of time interval between transitions

Question 15

Timing transition; half -cycle correction (markov?)

Answer 15

Instead of assuming all patinets swithc from health state at the end or very beginning of cycle, assume all patient switch half-way the cycle