Related Experiment Videos
Statistical evaluation of standardized field sobriety tests.
Michael P Hlastala1, Nayak L Polissar, Steven Oberman
1Division of Pulmonary and Critical Care Medicine, Department of Medicine, University of Washington, Seattle, WA 98195-6522, USA. hlastala@u.washington.edu
Journal of Forensic Sciences
|June 4, 2005
Summary
Standardized Field Sobriety Tests (SFSTs) show variable accuracy in predicting Blood Alcohol Concentration (BAC) above 0.08%. Re-analysis suggests SFSTs are less accurate than previously reported, especially at lower BAC levels.
Area of Science:
- Forensic Science
- Toxicology
- Traffic Safety
Background:
- Standardized Field Sobriety Tests (SFSTs) are commonly used to assess alcohol impairment.
- Previous reports suggested high accuracy (91%) in predicting Blood Alcohol Concentration (BAC) ≥ 0.08%.
Purpose of the Study:
- To re-evaluate the accuracy of SFSTs using a more comprehensive statistical analysis.
- To determine the reliability of SFSTs across different BAC levels.
Main Methods:
- Re-analysis of existing data from a previous SFST accuracy study.
- Application of advanced statistical evaluation methods.
Main Results:
- SFST accuracy is significantly dependent on the individual's BAC level.
- The overall accuracy is considerably lower than previously reported by Stuster and Burns.
- SFSTs are most effective in identifying individuals with BAC substantially above 0.08%.
Conclusions:
- Current SFST evaluation methods require modification to account for BAC variability.
- Further development, including diverse BAC sampling and scoring adjustments, could improve SFST predictive capabilities.
- SFSTs show potential for improved DUI screening with further refinement.