Last updated on January 6th, 2024 at 08:34 am

Are you really unsafe to drive at a blood alcohol level of 0.08 BAC?

History of the problem and the conundrum about professionals who were alcoholics yet were considered by peers to be quite competent in their professions.

In the 1960s and 1970s early in my career as a clinical psychologist I knew half a dozen physicians and attorneys both socially and during my professional contacts whom I knew to be alcoholics or at least drunks who practiced their professions while intoxicated.  I was puzzled by the fact that these professionals, without exception, were considered by their peers to have a high level of competency in practicing their profession.  I appeared on a radio talk show with a psychiatrist and we were being interviewed about whether or not the serious illegal drug problem on both coasts would be coming to Wichita?  I noticed the interviewer was puzzled by the psychiatrist’s answers but I could not alert the interviewer to the fact that the psychiatrist was drunk!  The psychiatrist was also a private pilot who flew a complex airplane, often referred to as a doctor killer, and as far as I know he never had an accident while I am sure flying drunk.  Flying a complex airplane competently, as well as practicing a high level profession skillfully, suggests intact higher level mental abilities including the very important frontal lobe functions, which reflect the more executive abilities such as complex decision making, judgment and the ability to profit from feedback from one’s actions.

It is well known that chronic alcoholism causes brain damage but these professionals were in the early stages of their habit and had not as yet experienced significant structural damage to their brain and their higher level abilities.  Thirty years or more later I think I have a greater understanding of this conundrum than earlier.

Field Sobriety testing (SFST) for Driving While Intoxicated (DWI).

The SFST includes a battery of 3 tests measuring horizontal gaze nystagmus (HGN), the walk and turn, and the one leg stand test among others.  These tests assess balance, coordination, and the ability of the driver to divide his attention to more than one task at a time.  The SFST seems very heavily loaded in motor, balance and coordination skills but there is no assessment of higher order mental abilities such as planning, judgment, insight and the more complex abilities. In development of the SFST I suspect the researchers and physician consultants just assumed that with such extensive deterioration of the motor system due to the effects of alcohol that the whole brain was impaired but this assumption may not be true according to the results of my pilot study blow.

One might ask are the motor and related skills diagnostic or critical to impaired driving while under the influence of alcohol or is it the impaired higher level abilities that are the major source of the often tragic consequences of driving while intoxicated?  Those who swear by sobriety testing may think that the substantial motor impairment may be highly correlated with impaired higher level abilities as well but this is really an experimental question and the following pilot study I conducted may shed some light on such questions.

The Ruthven Impairment Assessment (RIA).

The RIA is a brief, computer delivered, performance measure of normal and non-specific impairment of mental or cognitive functions.  The five tasks are performed on a computer screen with a standard 12 hour clock face on the right of the screen and task instructions appear on the right half of the computer screen.

Task 1 is a measure of simple reaction time (press the space bar as quickly as you can to any lighted number), task 2 is a measure of complex reaction time (press the space bar as quickly as possible to only even lighted clock numbers), and task 3 is a measure of conditional RT (press the space bar to the number following two even numbers but not if that number is a six or a ten, the two numbers changing after each administration of the test).  Task 4 is a measure of attention/memory.  Tasks 4 and 5 are performed by the subjects operating a computer mouse and left clicking on the lighted clock numbers as explained in the task instructions on the left of the screen.  Task 4 involves two series of 3,4,5, and 6 lighted numbers and the S, when given the signal “Go” on the screen, must replicate with the computer mouse and left click on the number series just presented.  The highest possible score on task 4 is 8 points.  Task 5, the only task that is highly correlated with general intelligence and higher level mental abilities, is composed of 13 items, each of which has a series of 5 lighted numbers in sequence such as 1, 2 ,3 ,4 , 5.  The S is instructed to try to discover a theme or pattern within each 5 number series and to left click with the mouse on two numbers that would seem to continue or complete the pattern or theme.  The writer has called this skill anticipatory thinking or complex problem solving that is likely to involve higher level or more frontal lobe functions.  The item above in task 5 (1,2,3,4,5) is 1 of the 3 simple items (the correct answer of course is 6 and 7) to better explain what is involved in solving the remaining 10 items.   

The author/creator of the RIA has RIA norms on 55 college students tested 3 times over a 10 day period, the re-testing showing little or no practice effects on serial testing (Ref. #1).

Participants in the pilot study

The sample included 10 young (21 to 35 years) college graduates (7 females and 3 males), employees and friends from a health club, 6 of whom are personal trainers.  Comprehensive health screenings were not conducted but inquiry did not reveal any health or psychiatric or alcohol use history that would preclude participation in the study.  Non-drinkers and the other end of the alcohol use spectrum were of course rightly excluded from the pilot study.   Experienced social drinkers were defined as those who had 3 to 5 drinks at one session in the past 90 days, the attempt being to select those who have had some experience with the intoxicating effects of alcohol but none of any long term adverse effects.   All participants signed disclosure and consent forms and were paid $125.00 in cash for participation in the study. 

Study protocol

Participants came two at a time to the study site knowing in advance the procedures to take place, having read, signed and witnessed a full disclosure document which were brought by each subject to the study site.  The participants were requested (and later signed) forms which included abstention from alcohol or recreational drugs within 48 hours of the study appointment.  The participants were requested to forego breakfast on the morning of the study (the participants arriving at 8AM), which allowed less alcohol to reach peak Blood Alcohol Content (BAC).  Soon after arrival each participant took a Breathalyzer test (the ALCO-SENSOR FST, which is used by many law enforcement agencies) to insure a 0.00% alcohol BAC reading before alcohol ingestion.  Time involvement for each participant was approximately 4.5 hours and each had to have a licensed driver transport each subject from the study site since some alcohol remained in their system after the initial (and final) alcohol dosing.

Absorption of alcohol in the bloodstream depends greatly on gender and weight, with males of equal weight with females requiring greater amounts of alcohol than females to reach comparable BAC levels.  Female participants in the study required approximately 8 oz. of 80 proof Vodka or Bourbon in a carbonated mix to reach a targeted BAC of 0.08% alcohol or above vs. 10 to 11 oz. of 80 proof alcohol for males.

Participants took the RIA for a baseline prior to the alcohol dosing (consuming the dosing within 10 minutes after the initial RIA testing), and a final dosing to bring the subjects’ BAC to a low of 0.08 to a high of 0.145 BAC in one female subject. At this point participants re-took the RIA for a second and final time.  Following this the subjects were released from the study site under the care of a licensed driver.

Results:

There are 9 important measures that make up RIA performance, which include the number of correct RTs to the 3 RT tests, Mean RT on tasks 1, 2 and 3 (along with any errors on the 2 number exception presentations), number correct on task 4 (out of a possible 8) and number correct on the 13 items of task 5.  The first 3 items on task 5 are very concrete and are given to better explain the concepts involved in the task.  However, one male health professional (not in this pilot study) gave a substantially impaired RIA performance and missed the 1,2,3,4,5 item and later on  comprehensive neuropsychological testing was diagnosed with a progressive frontal lobe dementia.  This patient’s greatest impairment on the RIA was on task 5 as expected and his only mild impairment on task 4 (memory) was considered incompatible with AZ.

Results of the pilot study

At a gross level the 5 task measures were subjected to an analysis of variance of the pre and post RIA data on the 5 tasks; the only significant difference was that alcohol (at the BAC levels studied) impaired the group Mean for simple reaction time (task 1) but there were no other impairing effects of the alcohol on the remaining measures on the other 4 tasks.  The impaired group performance on simple RT is diagnostic of reduced processing speed.

Looking at the individual subject performance an impairment score is defined as an impaired score of at least one standard deviation in the impaired direction compared to the college norms.

On the pre-dosing RIA testing only 3 of the 10 subjects (2 females and 1 male) had only one of their performances in the impaired range.  2 of the 3 subjects were impaired on task 5 and one on complex RT (task 2).  One error is not considered diagnostic of impairment and as a group these 10 subjects on the RIA performed very much like the normative group of college students.

After final alcohol dosing 3 subjects showed no impairment in any of their 9 RIA performances while 7 of the 10 subjects had at least 1 of the 9 performances in the impaired range!  However, there was no consistent pattern to the impairment of these subjects. 3 subjects had impairment on task 5 (anticipatory thinking), 2 subjects on task 4 (memory), and 1 subject was impaired on complex RT (task 2).

 A female with the highest BAC in the group (0.145) had 3 performances in the impaired range, one each on task 2 (only 7 correct RTs on this task), task 4 (memory) and task 5 (anticipatory thinking). This female subject’s impairment on the RIA, especially tasks four and five, demonstrates the effects of the alcohol are beginning to compromise the higher level cognitive abilities, which are likely to be critical and possibly mandatory for defining impaired and reckless driving.

When BAC is 0.12 and above the data looks as if higher level and perhaps frontal lobe abilities are beginning to show serious impairment.  Many drunken and reckless drivers are found to have BACs at 0.45 and above and the present pilot study sampled only those with mild to low moderate BAC levels. 

6 of 8 subjects with BACs of 0.08 to 0.11 had either none or 1 impaired performance, which at such a BAC level in a driver would, according to the data, be wrongfully accused of driving while intoxicated!  On the RIA these subjects did not demonstrate any measurable cognitive impairment accept for a mild decrease in mental processing speed (task 1 simple RT), which is not in itself considered a critical measure of impaired driving.

On the basis of the data in this pilot study the current BAC of 0.08 is not sufficient evidence for a DWI arrest and valid conviction. The legal limit for drunken driving might be raised significantly to possibly a BAC of 0.150 and above. If one could establish a more valid threshold for DWI than the present one the court could legitimately accept as proof of impaired driving on the basis of the BAC alone as long as the BAC is above the impaired driving threshold, i.e., 0.150 and above.

If a driver is arrested for a DWI on the basis of the BAC alone before a court trial the driver might be offered a cognitive assessment at the driver’s BAC at the time of the driver’s arrest.  Such a cognitive post-assessment should occur at the time of the day or night when the driver was arrested for the offence.  Facing such a cognitive test these subjects under their BAC level would certainly have high motivation to give their best performance.

The pilot study suggests that the field sobriety testing, relying very heavily on motor, coordination and balance difficulties, does not assess the critical sources of the reckless behavior of many drunken drivers, which are likely due to impaired judgment, limited insight into one’s impairment and other higher cognitive deficits.

Further research needed in this area

A major limitation of the pilot study was of course the too restricted range of the BAC in assessing cognitive impairment.  Another pilot study assessing a BAC of 0.150 to perhaps 0.400 on impaired cognition would be most appropriate.

Caveats on the above pilot study

I should certainly not be falsely accused of advocating that people should drive after drinking alcohol.  However, the current SFST and a BAC of 0.08 and above are insufficient evidence to charge a driver with a DWI.  The performance data of the pilot study suggests that impairment of motor functions are insufficient to establish impaired driving and that impaired higher level and judgmental abilities must also be present to sustain a true conviction.  Setting a cut off BAC between “unimpaired” and impaired driving is arbitrary but the pilot study data suggests that the BAC should be at least 0.15 and above to establish a charge of DWI.  

Reference:

 “Introduction to the Ruthven Impairment Assessment (RIA):  a Stability Study”.  Gerald Goldstein, Jibo He, Leslie Ruthven, Jon Walker.  Applied Neuropsychology, Published March 2017.