SMAA (Stochastic Multi-attribute Acceptability Analysis)

Recommended for further appraisal for use in benefit-risk assessment

1. Description

SMAA is a family of multi-criteria decision methods dealing with statistical uncertainty. The scores and weights, and their combination, is as for MCDA, but that uncertainty about scores can be accommodated through simulations, while uncertainty about weights is explored by assuming they can take on any values. The results are the rank values of the most preferred option for their benefit-risk balance in the simulation runs for different levels of the weights.

2. Evaluation

2.1 Principle
  • Very similar to MCDA
  • Statistical uncertainties of the posterior distributions are propagated from the statistical uncertainties in input values expressed as arbitrary stochastic variables.
  • Confidence intervals around point estimates are used when making decisions.
  • Understanding the principles of SMAA require mathematical understanding of stochastic phenomena and the concept of uncertainty.
  • Value preference in SMAA can be missing.

2.2 Features

SMAA integrates benefits and risks, and ranks the different options in terms of being the best option (second best, third best etc.)

2.3 Visualisation

A grouped bar graph can be used to display the final results of SMAA analyses.

2.4 Assessability and accessibility
  • Interpretation of the final results in terms of ranked probabilities is simple.
  • The implementation requires more mathematical and computational knowledge.
This method was tested in the Rimonabant, Telithromycin and Warfarin case studies.

3. References

[1] Tervonen T, van Valkenhoef G, Buskens E, Hillege HL, Postmus D. A stochastic multicriteria model for evidence-based decision making in drug benefit-risk analysis. Statist Med 2011;30(12):1419-28.
[2] Lahdelma R, Hokkanen J, Salminen P. SMAA - Stochastic multiobjective acceptability analysis. European Journal of Operational Research 1998 Apr 1;106(1):137-43.
[3] Tervonen T, Figueira JR. A survey on stochastic multicriteria acceptability analysis methods. Journal of Multi-Criteria Decision Analysis 2008;15(1-2):1-14.
[4] G. van Valkenhoef, T. Tervonen, T. Zwinkels, B. de Brock, H. Hillege, "ADDIS: a decision support system for evidence-based medicine", In Decision Support Systems, vol. 55, pp. 459-475, 2013.