Member Outreach Model to Predict Target Responders
This ML model is used to predict if a member will respond to outreach campaign. It helps prioritise members who are more likely to respond to an outreach. It is used for improving CMS Stars ratings for health plans. This model can be used for focusing upon appropriate personnel for customer engagement as well. It uses demographics, SDoH, clinical data and historical outreach data to determine whether or not a person will respond to the outreach.It is trained on a dataset contained more than 34000 members data. The performance metrics show Accuracy of 93%,F1 score 73% while 83% and 66% for Recall and Precision respecively.
Input variables : Demographic, SDoH, Clinical data, Historical outreach data
Output Variables : Will respond to outreach (Yes/No)
Metrics to Monitor
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
|
Business
|
:
|
Outreach Campaign Efficacy |
Members Engaged
|
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 :
medictiv.citiustech.com