Method: The study makes use of a research framework developed by Krueger [13] (and subsequently employed in several studies) on the effect of changes in the maximum weekly benefit on injury duration. This research framework is a natural experiment, where the treatment effect is measured as the difference in the “post-reform minus pre-reform” differences between treatment and control groups. Two partial linear regression models (generalized additive regression and quantile regression) are used to validate the measured treatment effect. Further, quantile regression is applied to quantify the effect on injury duration of claim attributes such as age and gender.
Results: Using data sets provided by the Oregon Department of Consumer and Business Services and the New Mexico Workers’ Compensation Administration, it is shown that an increase in the maximum weekly benefit of temporary total disability claims leads to a lengthening of the average benefit duration in the group of affected claimants. This increase in the utilization of indemnity benefits contributes to about 30 percent of the total cost impact of the reform.
Availability: Computing and bootstrapping the difference in the “post-reform minus pre-reform” differences between treatment and control groups is explained in detail and is straightforward to execute. The generalized additive model was implemented using the R package mgcv, which was developed by Simon Wood. (http://cran.rproject.org/web/packages/mgcv/index.html)
The partial linear quantile regression model was implemented using the R package quantreg, which was created by Roger Koenker.(http://cran.rproject.org/web/packages/quantreg/index.html)
Keywords: Generalized Additive Model, Indemnity Utilization, Injury Duration, Legislative Reform, Quantile Regression.
