The CEAC as implemented in DAMPACK calculates the probability an intervention is preferred (or considered optimal), not the probability the intervention is cost-effective. Multiple interventions can be cost-effective (not dominated and ICER<WTP) but only one can be preferred (maximize NMB or NHB, or equivalently, have the largest ICER below WTP threshold).
Here is an example of a CEAC where the Y axis is appropriately called "Probability cost-effective". Note that unlike the DAMPACK implementation, the sum of the probability each strategy is cost-effective at any given sums to a number greater than one, particularly for larger willingness-to-pay thresholids.
https://www.researchgate.net/publication/306025962/figure/fig2/AS:393531618283544@1470836657983/Cost-effectiveness-acceptability-curves.png
I suggest changing the Y axis in the DAMPACK ceac plot to "Pr Preferred" or "Pr Most Cost-Effective"
The CEAC as implemented in DAMPACK calculates the probability an intervention is preferred (or considered optimal), not the probability the intervention is cost-effective. Multiple interventions can be cost-effective (not dominated and ICER<WTP) but only one can be preferred (maximize NMB or NHB, or equivalently, have the largest ICER below WTP threshold).
Here is an example of a CEAC where the Y axis is appropriately called "Probability cost-effective". Note that unlike the DAMPACK implementation, the sum of the probability each strategy is cost-effective at any given sums to a number greater than one, particularly for larger willingness-to-pay thresholids.
https://www.researchgate.net/publication/306025962/figure/fig2/AS:393531618283544@1470836657983/Cost-effectiveness-acceptability-curves.png
I suggest changing the Y axis in the DAMPACK ceac plot to "Pr Preferred" or "Pr Most Cost-Effective"