Agenda

Session Coordinator(s)

Facilitator(s)

Presenter(s): Ian Duncan, FSA, CSPA, FCA, FCIA, FIA, MAAA

Session Coordinator(s)

Facilitator(s)

Presenter(s): Sreenivas (Si) Konda, Ph.D

Basic statistics (means/medians/variance/standard error of estimates);

Test of significance and their interpretation (p values; R²; F, Chi-sq, and t-tests; ROC curves);

Hypothesis testing and interpretation of ANOVA results;

Session Coordinator(s)

Facilitator(s)

Presenter(s): Sreenivas (Si) Konda, Ph.D

Hands-on Exercise

• Introduction to R;
• Review of data examples;
• Data visualization and Exploration;
• Graphical applications in R.

Materials: Prescription Cost Dataset

Software: R

Session Coordinator(s)

Facilitator(s)

Presenter(s): Ian Duncan, FSA, CSPA, FCA, FCIA, FIA, MAAA

• Introduction to medical claims;
• Introduction to prescription drug claims;
• Introduction to clinical values and their interpretation;
• Introduction to medical algorithms (e.g. how to identify members with a particular condition from claims data; how to distinguish between claims generated by different provider and service types, for example).

Materials: Lecture notes; Chapter 1; Health care Dataset; Sample EMR data

Software: R

Session Coordinator(s)

Facilitator(s)

Session Coordinator(s)

Facilitator(s)

Presenter(s): Sreenivas (Si) Konda, Ph.D

• Linear Regression Models (LRM)
• Generalized Linear Models (GLM)
• Bootstrap, K-fold Cross-validation (CV)

Session Coordinator(s)

Facilitator(s)

Presenter(s): Sreenivas (Si) Konda, Ph.D

Materials:  Lecture notes
Medicare dataset

Session Coordinator(s)

Facilitator(s)

Session Coordinator(s)

Facilitator(s)

Presenter(s): Ian Duncan, FSA, CSPA , FCA, FCIA, FIA, MAAA; Nhan Huynh, MS

Session Coordinator(s)

Facilitator(s)

Presenter(s): Ian Duncan, FSA, CSPA, FCA, FCIA, FIA, MAAA; Nhan Huynh, MS

The CMS Hospital Readmission Reduction Program (HRRP) applies a penalty (reduced DRG reimbursement) to those hospitals whose readmissions exceed a benchmark. Hospitals have an incentive to apply case management to patients at high risk of readmissions.   Using patient level data with more specific patient information (with 13 explanatory variables) we will build logistic regression models for re-admissions and compare with the industry standard LACE index. In addition, for practical implementation purposes will apply an   economic model to determine the optimal level of intervention.

Chapter 20

Medicare Dataset

Materials: Lecture notes
Dataset

Session Coordinator(s)

Facilitator(s)

Session Coordinator(s)

Facilitator(s)