Cost-effectiveness version created by F. Javier Díez, based on the unicriterion version created by Manuel Luque. Revised in November 2015.

References

M. Luque. Probabilistic Graphical Models for Decision Making in Medicine. PhD thesis, UNED, Madrid, 2009.

M. Luque, F. J. Díez and C. Disdier. A decision support system for the mediastinal staging of non-small cell lung cancer. In Proceedings of the 8th Bayesian Modelling Applications Workshop, (Barcelona, Spain).

Description of the problem

A doctor has to decide how to treat a patient that suffers from non-small cell lung cancer.

The key to select a treatment is whether there is metastasis at the mediastinum or not. A CT scan is always performed. Then the doctor may decide to perform TBNA, PET, mediastinoscopy (MED), EBUS, and/or EUS, and finally chooses a treatment. The decision process must take into account the morbidity and economic cost of each test and each treatment, the quality adjusted life expectancy, probabilities of survival, and the willingness-to-pay (lambda).

Structure of the network

The structure of chance and decision variables reflects that the target and unknown variable is N2N3, which indicates whether there is metastasis or not. Chance variables representing the results of tests have a probabilistic relation with the N2N3 variable and result of the TAC. We must also note that the results of MED, EBUS and EUS depend on the result of the PET.

The structure of utility nodes has been built by using super-value nodes [Tatman and Shachter, 1990]. It consists of two main parts. The node Total_Economic_Cost and its ancestors represent the economic cost of the medical tests and treatments. The node Total_QALE and its ancestors represent the quality adjusted life expectancy of the patients. Both parts are summed by weighting the Total_Economic_Cost by the node C2E (cost to effectiveness), which represents the inverse of the lambda parameter (willingness to pay) used in cost-effectiveness analysis. The node Net_Effectiveness represents the global utility.

The Quality Adjusted Life Expectancy (QALE) in utility nodes is measured in Quality Adjusted Life Years (QALYs). Costs are in euros.

]]>

Survival to mediastinoscopy.

]]> 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0

Survival to the treatment.

]]> 0.7000000000000001 0.29999999999999993 0.81 0.18999999999999995 0.44999999999999996 0.55 1.0 0.0 0.0 1.0 0.0 0.0 0.0 0.971 0.029 0.0 0.9950000000000001 0.004999999999999893 1.0 0.0 0.0 1.0 0.0 0.0 0.0 0.96 0.04 0.0 0.21999999999999997 0.78 1.0 0.0 0.0 1.0 0.0 0.0 0.0 0.93 0.06999999999999995 0.0 0.78 0.21999999999999997 1.0 0.0 0.0 1.0 0.0 0.0 0.0 0.25 0.75 0.0 0.08999999999999997 0.91 1.0 0.0 0.0 1.0 0.0 0.0 1.0 0.0 0.0 1.0 0.0 0.0 0.0 0.9950000000000001 0.004999999999999893 0.0 0.9950000000000001 0.004999999999999893 0.0 0.9950000000000001 0.004999999999999893 0.0 0.9950000000000001 0.004999999999999893 1.0 0.0 0.0 1.0 0.0 0.0 1.0 0.0 0.0 1.0 0.0 0.0 0.0 0.308 0.692 0.0 0.07499999999999996 0.925 0.0 0.308 0.692 0.0 0.07499999999999996 0.925 1.0 0.0 0.0 1.0 0.0 0.0 0.0 0.9950000000000001 0.004999999999999893 0.0 0.9950000000000001 0.004999999999999893 1.0 0.0 0.0 1.0 0.0 0.0 0.0 0.9950000000000001 0.004999999999999893 0.0 0.9950000000000001 0.004999999999999893 1.0 0.0 0.0 1.0 0.0 0.0 0.0 0.42000000000000004 0.58 0.0 0.09999999999999998 0.9 1.0 0.0 0.0 1.0 0.0 0.0 0.0 0.42000000000000004 0.58 0.0 0.09999999999999998 0.9 1.0 0.0 0.0 1.0 0.0 0.0 0.0 0.999 0.0010000000000000009 0.0 0.999 0.0010000000000000009 1.0 0.0 0.0 1.0 0.0 0.0 0.0 0.53 0.47 0.0 0.17 0.83 0.0 1.0 0.001 0.999 0.0 0.0 2.08 0.42 2.64 1.25 5.75 1.17 0.0 -1.08E-4 0.0 0.0 0.0 -8.33E-4 0.0 -1.25E-4 0.0 -1.25E-4 0.0 0.0 -2.1E-5 -2.1E-5 199.00000000000003 0.0 80.0 0.0 0.0 620.0000000000001 620.0000000000001 0.0 620.0000000000001 0.0 620.0000000000001 0.0 3000.000000000001 0.0 1290.0000000000002 0.0 4142.6 9764.4 0.0 1.0 0.02 0.98 0.022 0.978 0.0 1.0 0.02 0.98 0.04 0.96 UNICRITERION €