Table 2 Calibration approach [30]
Calibration Step | Approach for MYRIAD |
|---|---|
1. Identify parameters to vary in calibration process. | • Monthly transition probabilities between low-risk, medium-risk, and OUD states. |
2. Select calibration targets. | • Fatal opioid overdose rates, 2015–2019 |
3. Select a goodness of fit (GOF) measure. | • Pearson’s Chi-squared |
4. Select a search strategy | • Latin hypercube sampling |
5. Identify acceptable GOF parameter sets (convergence criteria). | • Model outputs are within 99% confidence interval of the observed calibration target values |
6. Identify the termination of the calibration process (stopping rule). | • 100 number of sample runs |
7. Identify how the model calibration results and economic parameters are integrated. | • Use parameter set that minimizes GOF measure |