In this model the state government administrative data and machine learning methods are combined to examine whether the risk of future opioid dependence, abuse, or poisoning can be predicted in advance of an initial opioid prescription. This can be used as a tool to restrict illegal usage of opioids. The sample for this model was more than 80,000 individuals who received and opioid prescription or an injection according to the Medicaid records.An adverse opioid-related outcome is defined as receiving a diagnosis of opioid dependence, abuse, or poisoning or receiving treatment for an opioid use disorder.Data from Medicaid, insurance and prescription data are used for feature generation.Three models ie. a regularized regression, an ensemble, and a neural network are implemented to find the perfect fit.
Input variables : Combined features from Medicaid,Insurance and Prescription data
Output Variables : Risk of opioid dependence
| Statistical | : | Somers D | Accuracy | Precision and Recall | Confusion Matrix | F1 Score | Roc and Auc | Prevalence | Detection Rate | Balanced Accuracy | Cohen's Kappa | Concordance | Gini Coefficent | KS Statistic | Youden's J Index |
| Infrastructure | : | Log Bytes | Logging/User/IAMPolicy | Logging/User/VPN | CPU Utilization | Memory Usage | Error Count | Prediction Count | Prediction Latencies | Private Endpoint Prediction Latencies | Private Endpoint Response Count |
Visit Model : github.com
Additional links : pnas.org
| Model Category | : | Public |
| Date Published | : | January, 2020 |
| Healthcare Domain | : | Provider |
| Code | : | github.com |
| Member Experience |
| Behavioral Health |