Integrating patient preferences and clinical trial data in a
Bayesian model for quantitative benefit-risk assessment
Introduction:
Regulators increasingly incorporate patients’
view on benefit-risk tradeoffs but little is known on how to integrate elicited preferences into the quantitative models. There is little knowledge on how to integrate these preferences with clinical performance data and how to use knowledge about the uncertainty surrounding both types of parameters (preference and performance).
Methods:
A multi criteria decision analysis (MCDA) model was developed that integrates clinical trial data, elicited patient preferences and
uncertainty surrounding these estimates. Stochastic
characteristics of preference and drug performance parameters can be approximated from stated preference studies and
performance data from systematic reviews or RCT’s. Risk and
benefit scores of drugs are then simulated with Monte Carlo methods using approximated distributions.
Henk Broekhuizen, MSc
1; Karin Groothuis, PhD
1; Brett Hauber, PhD
2; Maarten IJzerman, PhD
1(1) University of Twente, dept. Health Technology and Services Research, Enschede, the Netherlands (2) RTI Health Solutions, Research Triangle Park, NC
Results:
The model was applied to an anti-depressants case. We included two benefit and one risk criteria (figure 1). Preference data was derived from an analytical hierarchy process study with 12 major depression disorder patients
who were currently in remission and the performance data (pooled odds ratio’s compared to placebo) were derived
from a systematic review. The distribution around preferences was approximated with a bootstrap method, the distribution around performance data was approximated with a normal distribution in the log domain. The simulations
show all drugs have high (≈1) acceptabilities (figure 2). The problem is more sensitive to performance information
than to preference information and most sensitive to the adverse events performance criterion.
Correspondence about this poster to: h.broekhuizen@utwente.nl
Benefit
-risk
plane
Benefits
Risks
Response Remission
Adverse
events
Drug
A
Drug
C
Drug
B
Figure 1: MCDA structure for the
antidepressants case.
Conclusion:
Using this MCDA model it is possible to include patient preference in a quantitative risk-benefit assessment model. The model allows integration of stochastic uncertainty concerning preference and performance. It demonstrates that comprehensive presentation of data is possible.
Figure 2: Risk-benefit plane. All simulation runs are
below the μ=1 threshold and are thus acceptable.
Figure 3: A value sensitivity graph for the adverse events weight and drug A (left). A decision sensitivity graph
for drug A (middle). A ranking sensitivity graph for drug B and its performance on the response criterion
(right). In the ranking sensitivity graph, the green line is drug B’s probability of being rank=1, blue of rank 2
and red of rank 3. In all graphs, the two black vertical lines denote the 95% CI of the parameter on the x-axis.
0.15 0.20 0.25 0.30 2 3 4 5 6 7
Adverse events w eight
D ru g A 's b e n e fit -r is k r a tio 0.15 0.20 0.25 0.30 0 .0 0 .2 0 .4 0 .6 0 .8 1 .0
Adverse event w eight
p ro b a b ili ty o f a c c e p ta n c e a t m u = 3 1.6 1.8 2.0 2.2 2.4 2.6 0 .0 0 .2 0 .4 0 .6 0 .8 1 .0
Drug B's performance on the response criterion
P ro b a b ili ty
