|
Activity Number:
|
92
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, July 30, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Health Policy Statistics
|
| Abstract - #308695 |
|
Title:
|
Analysis of Longitudinal Data To Evaluate a Policy Change
|
|
Author(s):
|
Benjamin French*+ and Patrick J. Heagerty
|
|
Companies:
|
University of Washington and University of Washington
|
|
Address:
|
F600 Health Sciences Building, Seattle, WA, 98195-7232,
|
|
Keywords:
|
Generalized estimating equations ; Generalized linear mixed models ; Meta-analysis ; Empirical Bayes
|
|
Abstract:
|
There are challenging aspects of policy change data with respect to analysis and inference that require consideration: defining comparison groups, accounting for heterogeneity in the policy effect, and modeling longitudinal correlation. We compare currently available longitudinal data analysis methods to evaluate a policy change. We also illustrate issues specific to evaluating a policy change via a case study of laws eliminating gun-use restrictions and firearm-related homicide. We obtain homicide rate ratios estimating the effect of enacting a shall-issue law that vary between 0.903 and 1.101. However, using methods that are most appropriate implies that enacting such a law is associated with a non-significant increase in firearm-related homicide. We conclude that in a policy change study it is essential to thoroughly model temporal trends and account for policy effect heterogeneity.
|
