Activity Number:
|
212
|
Type:
|
Contributed
|
Date/Time:
|
Tuesday, August 13, 2002 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Biometrics Section*
|
Abstract - #300552 |
Title:
|
Statistical Adjustment for Lying about Seat Belt and Alcohol Use in the Analysis of Motor Vehicle Accident Data
|
Author(s):
|
Haifeng Guo*+ and Kent Eskridge+ and Daniel Christensen and Qu Ming and Thomas Safranek
|
Affiliation(s):
|
Minneapolis Medical Research Foundation and University of Nebraska, Lincoln and Nebraska Department of Health and Human Service and Nebraska Department of Health and Human Service and Nebraska Department of Health and Human Service
|
Address:
|
914 S. 8th St. , Minneapolis, Minnesota, 55404, USA 103 Miller Hall, Lincoln, NE , Nebraska, 68583-0712, USA
|
Keywords:
|
Seat belt usage ; misclassification ; classification error ; odds ratio ; medical costs
|
Abstract:
|
Due to stiff state laws, occupants in motor vehicle accidents often lie to police officers about their seat belt and alcohol use. Biased conclusions result when using accident data from police reports to describe the effectiveness of seat belts in preventing injuries and reducing medical costs. We investigated the effects of lying about seat belt and alcohol use on the odds ratio of injury as well as medical costs. A statistical method and a SAS program were developed to adjust data for lying about seat belts and alcohol use. The method was used on a Nebraska motor vehicle accident data set to evaluate the odds ratios and medical costs estimates. Our results showed that the bias of the odds ratio of injury and medical costs due to lying about seat belts and alcohol use depends both on the amount of lying and the reported frequencies. For the Nebraska data set, lying about seat belt and alcohol use had a small impact on the odds ratio estimates but a large impact on some medical cost estimates.
|