Drinking and Driving: Where does the Real Danger Lie?

At this writing, 36 states and the District of Columbia define "driving while intoxicated" (DWI) as having a blood alcohol concentration (BAC) of 0.10%. Fourteen states measure DWI at 0.08%. A growing number of states are adopting "zero-tolerance laws" for under-21 drivers, in conformity with U.S. Code Chapter 169, to qualify for federal highway funding. And proposed legislation sponsored by Rep. Lowey and Sen. Lautenberg will force any state that does not lower its over-21 drunk driving BAC limit from 0.10% to 0.08% to lose federal dollars; companion legislation would withhold funding for states that fail to adopt a series of mandatory minimum penalties for alcohol-related crimes.

The lobbying efforts of Mothers Against Drunk Driving (MADD) and other anti-alcohol groups on the 0.08% bandwagon haven't always been successful. This year in Minnesota, for example, moderationist voices prevailed and the state's legislature declined to pass a 0.08% law. A recent study by the Impaired Driving Division of the National Highway Traffic Safety Association (Progress in Alcohol-Impaired Driving, February 1997) strongly suggests that the statistical distinction between drivers at 0.08% and 0.10% BAC in terms of serious car crashes is vanishingly small, and thus a lower legal BAC threshold would not mean a measurable difference in prosecuting drunk drivers.

How well are the states' current laws penned? In New York, which now has six different alcohol-related driving statutes, four of the six are based on BAC alone and do not require proof of intoxication, impairment by alcohol, "under the influence", or even the slightest loss of ability to drive safely. Writing in the January 1997 issue of the New York State Bar Association Journal, retired Erie County Supreme Court Justice William Ostrowski wrote about this "patchwork of internal inconsistencies", lamenting the "failure of the legislature to define the offenses of DWI and DWAI coupled with the failure to explain the difference between them." Complicating the picture is the obfuscatory reworking of New York's Criminal Jury Instructions in 1995, and the New York Court of Appeals' approval of three contrasting formulas for defining drunk driving. "Every time you cross over the East River," as Ostrowski puts it, "the charge to the jury may be different."

Lies, damned lies, and statistics

It also seems that every "expert" has a different BAC yardstick. Hypothesizing a 160-pound man having three beers in an hour, we used six different calculators (available on the internet) to estimate BAC. Our everyman's results ranged from 0.0404% to 0.14%! One on-line resource, the Blood Alcohol Page (see sidebar), gives an apparently straightforward formula:

# fluid oz of pure alcohol = body weight x BAC x rho, where rho = for men a factor of 0.13 and for women 0.115,

stating as examples that a 100 pound woman measured at 0.08% would have consumed 1.53 cans of beer in an hour, or that a 200-pound man at the same 0.08% would have consumed 3.47 cans of beer over the same hour, to achieve the BAC level now considered as driving while intoxicated in 14 states.

A judicially approved formula (State v. Tanner, 15 Ohio St. 3d 1) shows our 160-pound man at 0.7% after his three beers.

Why this disparity? Some of our on-line calculators included fairly sensitive parameters about the type of alcohol consumed (5% "beer" vs. unspecified ABV "malt liquor") and serving size (12 oz vs. pint, presumably a 16-oz one). But some disallowed even specifying whether the subject was male or female, thus taking the rho factor out of the equation. Then there's the "burnoff" or elimination factor (elapsed time since one's first drink) to consider: is it -.01 or -.02 for each hour, or somewhere in between?

Fallible tests, gullible subjects

Levels of ingested alcohol can be measured, with varying degrees of accuracy, by testing breath, blood, or urine. The most common method, and by far the most reliable, is blood analysis. The level of alcohol in the bloodstream is calculated using the weight of alcohol (milligrams) and the volume of blood (deciliters), yielding a blood alcohol concentration that is generally expressed as a percentage such as 0.08%. Breathalyzers are significantly less accurate measuring devices, and may improperly record such substances as mouthwash.

Blood analysis itself is a superbly exact science, say the experts, producing unadulterated results. The problem is that the numbers are too frequently misinterpreted, or simply not interpreted at all. According to Morris Chafetz of the Health Education Foundation, founder of the "TIPS" training program for bar and restaurant staff and author of The Tyranny of Experts, BAC calculations are simply a "pseudoscience", because they fail to account for individual body chemistry and crucial situational factors.

There's plenty of anecdotal evidence and caselaw to suggest, chillingly, that both law enforcement and the judiciary can manipulate both the numbers and the statutes.

* Officer Friendly stops Joe Average two blocks from the pub where he's just had a couple pints. A slightly nervous but law-abiding Joe submits willingly to a breathalyzer test, but is told his puff didn't register, or he didn't puff hard enough, so he must puff into the tube again. Voila, an inflated result!

* Joe's driving home after his two pints when a rogue driver slams into his car. The smoke from his airbag gags and alarms Joe. Officer Friendly observes Joe's irritated eyes and agitated state. He testifies, in a lawsuit brought by Joe's insurer against Ron Rogue-Driver, to having completed a "Drug Influence Evaluation" course that certifies his cursory assessment of Joe as "under the influence and possibly intoxicated" (although no testing whatsoever was performed). The suit is dismissed, and Joe's left with medical bills, no car, a suspended license, and higher insurance premiums.

* Take the scene above, but Joe's been sent to hospital for observation. A blood test three hours after the crash indicates a probability of 0.08% BAC at the time Joe's car was struck. The law says Joe's consent to the test was implied; and because he consented, the statutory 2-hour period for testing is irrelevant. Moreover, the hospital's testing equipment is not state-certified to produce admissible evidence of intoxication. Nonetheless, the test results are admitted as evidence against Joe, despite objections by Joe's lawyer.

MADD neoprohibitionism

While its best-known cause is lowering BAC limits from 0.10% to 0.08% nationwide, MADD makes no bones about its ultimate intent not to have anyone drink and drive at all. Some ultraconservative groups who fund MADD and its offshoots have even noised off about a return to full-fledged Prohibition, as if history's lessons were silent!

The crusade of MADD and like-minded groups is aided by some woefully bad number-crunching from the scientific community, like the finding (Liu, et al., JAMA 1997:227(2)) that "2.5% [of 102,263 adult phone-survey participants] reported an estimated 123 million episodes of alcohol-impaired driving in 1993. This corresponds to 655 episodes of alcohol-impaired driving for each 1,000 adults." Can any qualified statistician out there defend Liu's sloppy numbers and fatally flawed inferential logic?

Kurt Dubowski, a forensic toxicologist at the University of Oklahoma's Health Sciences Center, decries the rampant misuse of numbers by anti-alcohol activists. When the first states to adopt a 0.08% standard reported a 16-18% decrease in traffic fatalities, the new laws were hailed as lifesavers and those statistics became gospel. But a few years after enactment of 0.08% laws, the fatality numbers dropped right back to where they'd been before the new law. Dubowski calls this a "halo effect" and says, furthermore, that from a scientific perspective there's absolutely nothing sacred about the number .08. It's simply the number we've been stuck with as a sort of battle standard, a maddeningly simple rallying cry for the people at MADD.

Numbers sound impressive, but docudramatics are even more effective in getting attention. Like their forebears, our generation's temperance unionists paint a grotesquely distorted picture of the evils of alcohol. Soap-opera stories of high school cheerleaders struck in the blossom of youth by a guy like our Joe may sell newspapers but, as shown below, they simply don't mirror reality.

The right target?

National Institute on Alcohol Abuse and Alcoholism (NIAAA) data indicate that a driver at 0.02-0.04% BAC is 1.4 times more likely than a driver at 0% BAC to be involved in a fatal auto crash; the risk increases to 11.1 times at 0.05-0.09%, 48 at 0.10-0.14%, and a staggering 380 times at 0.15% and above. This research is consistent with crash data for 1995, showing that fully 65% of drivers involved in fatal collisions had a BAC level of 0.15% or more; 43% of drivers in fatal crashes had over 0.20%, twice the highest legal limit in the land.

Jim Ellingson, writing in a past Minnesota Brewing News in this paper (Jan. 1997), reports that "privately, state troopers will tell you the carnage on our highways isn't caused by drivers with blood alcohol levels of 0.10% or even 0.12%. It's the 0.20, 0.30 and even 0.50% blood alcohol level drivers, many of them repeat offenders whose licenses have been revoked, who are responsible for the bloodshed." Jim Klisch, a former police officer and now the owner of Lakefront Brewery in Milwaukee, says that in all his time as a police officer, he never made any arrests or saw any accidents of anyone whose BAC was in the 0.08% to 0.10% range. So why, instead of working towards solving the problems caused by the extremely intoxicated--and often chronic--offenders who cause the vast majority of serious auto accidents where alcohol is involved, do MADD and their followers condemn all drinkers, even those who drink in moderation?

And drinking isn't the only cause of traffic accidents. The NIAAA cites driving inexperience and immaturity as significant factors in auto crashes among novice drivers, even when alcohol is not involved; changing conditions, unaccustomed routes, and unexpected occurrences can negate acquired alcohol tolerances, rendering nil the "practice effect" of mature drivers. Sleepiness, fatigue, overwork, drug interactions, and a host of other factors may affect a driver's motor skills and ability to divide his attention between several sources of visual information to at least the same degree as a driver who's had a drink or two. State and federal highway statistics show that speeders and reckless drivers cause a greater number of fatal accidents than drivers "under the influence".

Better solutions

Legislating a lower threshold for drunk-driving convictions may be politically correct, but targeting the moderate drinker won't solve the very real problem of keeping killers off the road. Already strained law enforcement, judicial, correctional, and rehabilitation resources will be needlessly dissipated by creating this new class of offenders. Enlarging the number of potential convictions will simply dilute the pool of offenders, the worst of whom may actually serve less jail time than they might have under the old law.

Several states are considering more appropriate solutions to the problem of drunk driving, including interlock ignition devices, staggered penalties based on severity of the offense, instituting or increasing mandatory jail time for repeat offenders, and education/rehabilitation programs for convicted drunk drivers. Such measures would undoubtedly do far more good than 0.08% legislation in terms of both punishment and correction.

Finally, prevention of drunk driving isn't getting nearly the attention it deserves. The simple slogan "friends don't let friends drive drunk" hits the mark, because it emphasizes recognition of warning signs, acknowledges fallibility without moralizing, and has the potential to stop accidents before they occur. The TIPS program is also highly successful in preventing tragic accidents because of its one-on-one approach.

The one-size-fits-all BAC standard, whether it's 0.08% or 0.10%, fails to account for the myriad variables involved in each particular instance of each particular person's drinking. One's outlook on life, and, by extension, one's competence to drive, after two or three beers enjoyed over a meal with friends is vastly different from having pounded two or three beers while plotting how to murder one's spouse. The bare numbers don't tell that.

The fact is that education and self-awareness may be more meaningful than BAC in identifying and dealing with impairment (see, e.g., B. Hale, Knowing Your Limit: Drinking by the Numbers Can Underestimate Impairment, Penn State Department of Public Information). Perhaps the soundest advice we can offer to readers of this paper is something they probably already understand: know what you're drinking, know how much you're drinking, know why you're drinking, know who you are when you're drinking, know where you'll be driving after drinking ... and know when to stop.

By the Numbers

How many 12-ounce servings of 5% alcohol beer would a man have consumed to reach a given blood alcohol concentration?

Man's Blood Alcohol Concentation (percentage)

Weight 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10 0.11 0.12

140 0.91 1.21 1.52 1.82 2.12 2.43 2.73 3.03 3.34 3.64

160 1.04 1.39 1.73 2.08 2.43 2.77 3.12 3.47 3.81 4.16

180 1.17 1.56 1.95 2.34 2.73 3.12 3.51 3.90 4.29 4.68

200 1.30 1.73 2.17 2.60 3.03 3.47 3.90 4.33 4.77 5.20

220 1.43 1.91 2.38 2.86 3.34 3.81 4.29 4.77 5.24 5.72

240 1.56 2.08 2.60 3.12 3.64 4.16 4.68 5.20 5.72 6.24

260 1.69 2.25 2.82 3.38 3.94 4.51 5.07 5.63 6.20 6.76

280 1.82 2.43 3.03 3.64 4.25 4.85 5.46 6.07 6.67 7.28

300 1.95 2.60 3.25 3.90 4.55 5.20 5.85 6.50 7.15 7.80

This chart calculates that a 160-pound man consumed 2.77 12 ounce 5% alcohol beers to reach 0.08% BAC; 3.47 of the same beers would put him at 0.10%. Adjusting for the "burnoff" or "elimination" factor, according to this source, an average of .02% could be subtracted from the BAC percentage shown on the chart for each hour elapsed since the last drink to arrive at a rough estimate of his current BAC.

How many "Doses" in your glass?

12 oz bottle of 2.5% low-alcohol beer = 0.03 oz of absolute alcohol

12 oz bottle of 4% beer = 0.48 oz of absolute alcohol

5 oz glass of 10% wine = 0.50 oz of absolute alcohol

1.25 oz shot of 40% vodka (80 proof) = 0.50 oz of absolute alcohol

1.25 oz shot of 151-proof rum = 0.94 oz of absolute alcohol

12.5 oz bottle 20% fortified wine cooler = 2.50 oz of absolute alcohol

40 oz bottle of 8% malt liquor = 3.20 oz of absolute alcohol

The "dose" of alcohol is calculated by multiplying the volume of an alcoholic beverage by the percentage of alcohol by volume (ABV). So, a 12-ounce beer at 5% ABV contains a dose of 0.48 ounces of absolute alcohol, i.e. alcohol without any molecules of water. Copyright 1997, Great Lakes Brewing News. No material herein may be reprinted without permission of the Great Lakes Brewing News Distributed On the W3 For personal, non-commercial enjoyment and use only. Cheers!