|
Activity Number:
|
167
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 4, 2003 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Health Policy Statistics
|
| Abstract - #302052 |
|
Title:
|
Imputation of Membership Denominators for Rate-Based Analyses in a Managed Care Population
|
|
Author(s):
|
Evguenia Igorevna Jilinskaia*+ and Cathy Johnson and Stanley Norton
|
|
Companies:
|
Pharmetrics, Inc. and PharMetrics, Inc. and Pharmetrics, Inc.
|
|
Address:
|
150 Coolidge Ave., Watertown, MA, 02472-2815,
|
|
Keywords:
|
imputation ; exponential predictive model ; enrollment data ; rate-based analyses
|
|
Abstract:
|
Rate based Health Care analyses such as per member per month (PMPM) and units of measure per 1,000 are important in understanding cost and utilization of a given population. While retrospective claims data is rich with information regarding utilizers of health care services (patients), it does not provide the necessary information for population-based analyses. The quality of enrollment and claims data is determined based on an exponential predictive model of the percent of enrollees without claims, as a function of the total number of enrollment months within a given year. Based on the "learning sample," values of the average number of member enrollment months per year are estimated for each of 20 age-gender demographic groups. The total number of member months for health plans with incomplete or nonexistent enrollment data is estimated by summarizing counts of patients in each of the demographic categories using estimated enrollment weights and enrollment duration for each health plan, by year, by type of product.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2003 program |
