Activity Number:
|
18
- New Models, Diagnostics, and Considerations in Evaluating Intervention and Policy Effects
|
Type:
|
Topic Contributed
|
Date/Time:
|
Sunday, August 7, 2022 : 2:00 PM to 3:50 PM
|
Sponsor:
|
Health Policy Statistics Section
|
Abstract #322116
|
|
Title:
|
A Parametric Bootstrapping Approach for Time Series Analysis in Large Claims Databases
|
Author(s):
|
Fang Zhang* and Meghan Mayhew and Frank Wharam and Ladia Albertson-junkans and Arvind Ramaprasan
|
Companies:
|
Harvard Pilgrim Health Care and kaiser permanente center for health research and Duke University and Kaiser Permanente Washington Health Research Institute and Kaiser Permanente Washington Health Research Institute
|
Keywords:
|
bootstrapping;
time series;
claim database;
generalized estimating equations;
generalized linear mixed models;
binary and count outcome
|
Abstract:
|
Large insurance claims databases are frequently used to study the impact of health policies and interventions using time series analysis. Generalized estimating equations and generalized linear mix models are popular statistical tools to estimate the intervention effects by means of odds ratios for binary outcomes and incident rate ratios for count outcomes. To satisfy the need to interpret results using probabilities/incident rates rather than odds ratios/incident rate ratios for a broader audience, we develop a parametric bootstrapping method to estimate the effects of an intervention against the counterfactual at any time point after the intervention. This method can be easily generalized to the functions of the probabilities/incident rates. The methodology is illustrated using simulated data based on a large claim database.
|
Authors who are presenting talks have a * after their name.