The Rest of the Story: False Positive Rates of the Standardized Field Sobriety Tests
Counterpoint Volume 8: Issue 1 - Article 1 (March 2024)
An article in the Core Skills II Course
Joshua Ott
Caselock Inc.
Abstract:
Standardized Field Sobriety Tests (SFSTs) are often presented as extremely accurate tests to assess if a driver is impaired. A closer look shows that these tests have never been validated to determine driving impairment and have high False Positive rates.
Despite the published overall accuracy rates of 80-95%, there are high rates of False Positives. The overall False Positive rates are as high as 28% in the San Diego Study.
The studies provide the overall accuracy rates of the tests, but many of the studies never publish the actual False Positive rates. It is then left up to the reader to use the data to calculate the False Positive rates on their own.
Officers and legal professionals often learn these tests indicate impairment even though the San Diego Study clearly states they do not. Only the objective measurement of Blood Alcohol Concentration (BAC) was used to determine the accuracy of the tests.
This article is a call for a more in-depth understanding of the studies behind the SFSTs. Officers, legal professionals, and the public should be aware of the pros and cons of these tests. That includes transparency when it comes to the error rates of these tests, and that these tests do not measure driving impairment.
Validating the SFSTs
What was the purpose of the Standardized Field Sobriety Test validation studies?
The validation studies were conducted to find reliable tests for officers to use on the side of the road to assist them in making arrest decisions. By having standardized tests that all officers use, it also would assist judges and juries with determining how much weight to give the evidence from the tests. SCRI wanted to see if these tests were able to correlate to a specific BAC. These studies have never validated the SFSTs as indicators of driving impairment.
What is a False Positive?
A false positive incorrectly indicates that a condition is present. When looking at the studies, a False Positive occurs when the field sobriety tests indicate that a person was at or above the specified BAC, but a breath or blood test showed they were actually below the specified BAC.
Now that the basics are covered, the first study that will be addressed is the most recent validation study. After that, the other studies will be covered, primarily to show that there have consistently been issues with False Positives.
Validation of the Standardized Field Sobriety Test Battery at BACs Below 0.10 Percent (1998)
Validation of the Standardized Field Sobriety Test Battery at BACs Below 0.10 Percent, Stuster and Burns (San Diego, 1998).
|
This study is one of the most important to understand because this is the most recent validation study and officers are trained to testify to the accuracy of the field sobriety tests based on this study. It is commonly referred to as the San Diego Study. It was a field study and was validating the field sobriety tests for a BAC of 0.08 grams/dL or more.
What does NHTSA say?
The overall accuracy of the field sobriety tests is 91% in determining that a person is 0.08 grams/dL or more.
Additionally, NHTSA broke down each test and listed its individual accuracy:
|
What were the False Positive rates?
- Horizontal Gaze Nystagmus: Four or more clues occurred in 37% of the people under 0.08 grams/dL.
- Walk and Turn: Two or more clues occurred in 52% of the people under 0.08 grams/dL.
- One Leg Stand: Two or more clues occurred in 41% of the people under 0.08 grams/dL.
- Overall: 28% of the people arrested were under 0.08 grams/dL.
Two more interesting notes to consider:
- The average BAC of the 217 people arrested was 0.150 grams/dL.
- The average BAC of the 80 people not arrested was below 0.050 grams/dL.
This is important to note because it is easier to make a correct decision the farther away from 0.08 grams/dL as a person is. An example would be a person with a high BAC who is obviously intoxicated; It will be evident to the officer that an arrest is likely the correct decision.
Psychophysical Tests for DWI Arrest (1977)
Psychophysical Tests for DWI Arrest, Marcelline Burns and Herbert Moskowitz, (1977).
|
This was the first study and was used to determine which tests were the most accurate and easily administered by officers in the field. The BAC that these field sobriety tests indicated was 0.100 grams/dL or above. There were six tests that were involved in the study.
At the end of this study, the three most accurate field sobriety tests were determined to be Alcohol Gaze Nystagmus, Walk and Turn, and One Leg Stand.
|
The False Positive rate in this study was 27%. This means that 27% of the people having a BAC of under 0.100 grams/dL were incorrectly determined to be at or over the BAC level by the officers.
Development and Field Test of Psychophysical Tests for DWI Arrests (1981)
Development and Field Test of Psychophysical Tests for DWI Arrests, Tharp, Burns and Moskowitz (1981)
|
This study was also using a BAC of 0.100 grams/dL. A portion was conducted in a laboratory environment while the other segment was conducted in the field. For the laboratory portion, volunteers were dosed to specific BAC levels. Some of the volunteers received a placebo dose, which means their BAC would have been 0.000 grams/dL. The volunteers were brought in on two separate occasions.
According to NHTSA this study showed that the Standardized Field Sobriety Tests were 81% accurate in determining if a person was at or above 0.100 grams/dL.
Only the placebo-dosed individuals will be looked at for the False Positive rate in this study.
|
During the first testing period, 18% of the placebo-dosed individuals were incorrectly determined to be impaired by the officers. During the second testing period, 10% were incorrectly determined to be impaired. This is a high rate of false positives for people that had no alcohol in their system.
Field Evaluation of a Behavioral Test Battery for DWI (1983)
Field Evaluation of a Behavioral Test Battery for DWI, Theodore Anderson, Robert Schweitz, Monroe Snyder (1983).
|
This study was again using a BAC of 0.100 grams/dL. The study was only conducted in the field.
According to NHTSA this study showed that the Standardized Field Sobriety Tests were 83% accurate in determining if a person was at or above 0.100 grams/dL.
Of the incorrect decisions made by officers, 94% of those incorrect decisions were incorrectly arresting someone whose BAC ended up being below 0.100 grams/dL.
|
Field Evaluation Study of the Standardized Field Sobriety Test (SFST) Battery (1995)
Field Evaluation Study of the Standardized Field Sobriety Test (SFST) Battery, Burns and Anderson (Colorado, 1995).
|
This study is commonly referred to as the Colorado study and was a field study only. The BAC level involved in this study was 0.100 grams/dL, but it was not considered a false positive unless the person’s BAC was below 0.050 grams/dL.
The average BAC of those arrested was 0.152 grams/dL.
The False Positive rate was 24%.
|
Field Evaluation Study of the Standardized Field Sobriety Test (SFST) Battery (1997)
Field Evaluation Study of the Standardized Field Sobriety Test (SFST) Battery, Burns and Dioquino (Florida, 1997).
|
This study is commonly referred to as the Florida Study and it was a field study only. This was the first study dealing with a BAC of 0.080 grams/dL.
The average BAC of those arrested was 0.150 grams/dL, and the average BAC of those correctly released was 0.031 grams/dL.
The False Positive rate was 18%.
|
The Robustness of the Horizontal Gaze Nystagmus Test (2007)
The Robustness of the Horizontal Gaze Nystagmus Test, Burns (2007).
|
This was not a Validation Study. This is a laboratory study that only involved Horizontal Gaze Nystagmus. The study was conducted to determine if changes in the administration of HGN would affect the results. There were three elements of HGN tested:
|
Looking at the results from the times in which HGN was administered correctly, the False Positive rate was 67%. Additionally, 65% of the people below a BAC of 0.05 grams/dL also had four clues or more identified. There was a person with six clues identified who had a BAC of 0.029 grams/dL.
What about the times when the test was not administered correctly?
|
These numbers show that it is imperative that officers position the stimulus correctly or the False Positive Rates increase to even higher rates.
How did NHTSA address the extremely high false positive rate?
They changed the standards to lower the number of reported False Positives! In the current training material (2023 SFST Manual) and the San Diego Study, four or more clues correlated to a BAC of 0.08 grams/dL or more. In this study, four clues correlated to a BAC of 0.03 grams/dL or more. This drastically lowered the published False Positives. There was no justification given for this changed standard.
Evaluation of Field Sobriety Tests for Identifying Drivers Under the Influence of Cannabis
Evaluation of Field Sobriety Tests for Identifying Drivers Under the Influence of Cannabis, Marcotte, Umlauf, Grelotti, Sones, Mastopietro, Suhandynata, Huestis, Grant, Fitzgerald (2023).
|
This is the most recent study and not a validation study. It was published in the Journal of the American Medical Association (JAMA) in 2023. This study was examining “the classification accuracy of FST’s with respect to cannabis exposure and driving impairment (as determined via driving simulation).”
If this study was addressing marijuana DUIs, then why is it important in relation to the False Positive Rates of the SFST’s for alcohol DUIs?
It’s because there was a group of placebo dosed participants who were administered multiple field sobriety tests including the Walk and Turn and One Leg Stand. Due to this group of participants, we now have information on what the False Positive Rates are for the Walk and Turn and One Leg Stand when administered to sober people.
|
The study used Drug Recognition Expert (DRE) Instructors to administer the field sobriety tests. This is the highest level of training that officers can receive. The participants were current cannabis users who were screened prior to inclusion in the study. On the day of the testing, the participants completed a urine drug screen, a breath test for alcohol, and an oral fluid sample was collected for THC detection. The participants also completed a driving simulation prior to being dosed with an active or placebo dose of THC.
There were 63 placebo dosed individuals. Were these individuals sober or were they impaired by prior usage of THC? The study addressed this possibility. “We found no evidence of residual effects on pretreatment simulator performance.” This means that all evidence indicates that the placebo dosed participants were sober.
What were the False Positive Rates on the Walk and Turn and One Leg Stand for the placebo group?
|
This study shows that a high percentage of people will exhibit two or more clues on the Walk and Turn and One Leg Stand even when they are sober.
Final thoughts:
As can be seen, the False Positive rates remained high in all these studies. The Standardized Field Sobriety Tests, especially HGN, have been given considerable weight in court. Officers testify to the accuracy of the field sobriety tests based on the NHTSA data, but the rest of the story is never heard.
Does this mean the SFSTs are useless and should be eliminated?
No, it means that the tests should continue to be used for the purpose they were created, which is to assist officers in screening drivers and determining whether or not to arrest the driver based on the standard of Probable Cause. The False Positive rates are critical when in a courtroom and the “Beyond a Reasonable Doubt” standard applies. Then a jury needs to hear, and understand, that these tests do have False Positives and can be wrong.
|
I believe that there should be additional studies conducted on the SFSTs, involving a large-scale study involving many people from 18 years of age through 70 years of age or older, all of whom are completely sober. This will provide us with a more accurate picture of how many people in the general population exhibit two or more clues on the Walk and Turn and One Leg Stand while sober. Additionally, a study of this nature would provide additional data as to how age affects the False Positive Rates of the tests.
To use a “one size fits all” approach of two or more clues indicates a BAC of 0.08 grams/dL or more does not seem like an accurate method. It is likely that a 21 year old would perform better on the test than a 70 year old, but research is needed to definitively answer this question. After conducting this baseline study, I believe additional studies should be conducted to try to find solutions to lower the False Positive rates of the SFSTs while not increasing the False Negative rates. This may involve adding or eliminating clues or changing the number of clues to indicate that a person is 0.08 grams/dL or more.
Finally, Officers, Prosecutors, and Judges should be taught about the False Positive rates, so that they can make more informed decisions both on the road and in the courtroom.
Practice Tip: |
This concludes the Core Skills II-3 course
Coming Summer 2024 - The Core Skills III course
About the Author:
Joshua Ott operates Caselock Inc., specializing in expert testimony and reviews for cases involving alcohol and/or drug impairment. Joshua has extensive knowledge and training obtained from ten years as a police officer in Metro Atlanta. He has extensive experience training law training law enforcement officers as well as courtroom preparation and testimony. He resides in Woodstock, Georgia with his wife, daughter and Miniature Schnauzer, Gus.
|
Send me your questions or comments:
Comments and questions will be posted here with their responses:
Comments and questions will be posted here with their responses:
For further study:
- NHTSA, DUI Detection and Standardized Field Sobriety Testing Manuals, 2006, 2013, 2015, 2018, and 2023.
- Marcelline Burns and Herbert Moskowitz, Psychophysical Tests for DWI Arrest, 1977.
- Tharp, Burns and Moskowitz, Development and Field Test of Psychophysical Tests for DWI Arrests, 1981.
- Theodore Anderson, Robert Schweitz, Monroe Snyder, Field Evaluation of a Behavioral Test Battery for DWI, 1983.
- Burns and Anderson, Field Evaluation Study of the Standardized Field Sobriety Test (SFST) Battery, Colorado, 1995.
- Burns and Dioquino, Field Evaluation Study of the Standardized Field Sobriety Test (SFST) Battery, Florida, 1997.
- Stuster and Burns, Validation of the Standardized Field Sobriety Test Battery at BACs Below 0.10 Percent, San Diego, 1998.
- Burns, The Robustness of the Horizontal Gaze Nystagmus Test, 2007.
- Marcotte, Umlauf, Grelotti, Sones, Mastopietro, Suhandynata, Huestis, Grant, & Fitzgerald, Evaluation of Field Sobriety Tests for Identifying Drivers Under the Influence of Cannabis, 2023.