|
Items in cart [0] |
Questions? Call 888-624-8373 |
| Copyright © 2009. National Academy of Sciences. All rights reserved. Terms of Use and Privacy Statement |
Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 336
Reducing the Costs of
Dnnking and Driving
DAVIl) S. REED
INTRODUCTION
Public concern over the dangers of drunken driving is almost as old as
the automobile. Indeed, few authors on the subject can resist citing the
"motor wagons" editorial in the Quarterly Journal of Inebriety in 1904.
Despite the long history of concern and the many attempts to control
the problem, drunken driving is still perceived as a major highway safety
problem.
Paradoxically, the widespread familiarity in our society with drinking,
driving, and their combination may have hindered the development of
effective countermeasures. An individual who is very familiar with the
elements of a problem may let preconceived notions interfere with the
gathering, assimilating, and applying of information. The first section
of this paper examines the costs generated by the drinking-driving prob-
lem, what part of these costs are potentially preventable, and how the
magnitude and distribution of costs might be viewed in assessing the
priority of the problem. The next section taps the extensive experience
worldwide with programs to reduce drunken driving. We find some
David S. Reed, a doctoral candidate in the program in public policy at the John F.
Kennedy School of Government, Harvard University, is currently an intern at the Federal
Communications Commission.
This work was completed during my enrollment at the John F. Kennedy School of
Government, Harvard University. I would particularly like to thank Mark Moore and
Dean Gerstein, who supervised its preparation.
336
OCR for page 337
Reducing the Costs of Drinking and Driving
337
promising avenues for future action and equally important some un-
promising avenues that still have vocal advocates. The following section
examines efforts to reduce the risk associated with a given amount of
drunken driving. Because there have been few such efforts, this section
offers few conclusions; it does raise questions that seem to warrant
further investigation.
-I- The final section examines the manner in which the federal govern-
ment has designed and managed programs of drinking-driving counter-
measures. It concludes that changes are necessary if we are to learn
from experience and improve our ability to reduce the costs of drunken
. · .
arlvlng.
COSTS AND THEIR IMPLICATIONS FOR THE ROLE
OF GOVERNMENT
THE NEED FOR EVALUATION
Any attempt to ameliorate dr~nking-driving problems will consume
scarce government resources and may impose monetary and nonmo-
netary costs (such as some restriction of civil liberties) on individuals.
We must therefore determine the costs generated by the problem in
order to compare them with the costs imposed by possible solutions.
This section examines the magnitude and distribution of costs resulting
from drunken driving. It also provides information to help determine
the priority of reducing this problem by governmental efforts.
This question of costs is of more than academic interest. The comp-
troller general of the U.S. (1979, p. 3) estimates that "Federal, State,
and local governments spent over $100 million in 1976 for their drinking-
driver countermeasure activities." This level of resources, and, as we
discuss below, the large human and economic costs potentially at stake
suggest the importance of determining the appropriate role of govern-
ment in efforts to reduce the costs of drunken driving.
PREVENTABLE ACCIDENTS
Information presented to policy makers about drunken driving (U.S.
Department of Transportation 1968, Noble 1978a, Comptroller General
of the U.S. 1979) has typically expressed the importance of the problem
in terms of the costs associated with it. For example, Noble (1978a, p.
61) reports that "approximately one-third of the . . . injuries and one-
half of the fatalities [from traffic accidents] are alcohol related." The
term "alcohol-related" refers to any accident in which a driver, or some-
OCR for page 338
338
REED
TABLE 1 Percentage of Accident-Involved Drivers and Control
Group Drivers Found to Be Within Various BAC Ranges, by Worst
Consequence of Accident and Place and Time of Study
Fatal,a Vermont, Huntsville, Ala.,
1967-1968, 197~1975,
BAC (%) Ar~itlent/~natro1 A~nirlent/(~nntrn1
injury, Grand
Rapids, Mich.,
1962-1963,
Accident/Control
Property
Damage, Grand
Rapids, Mich.,
1962-1963,
Accident/Control
<0.01 64~o 84%
0.01~.049 5% 9%
0.05 0~999)
0.10 0.149~31%
~0.015 ~
73.98~o 88.18%
4.86% 4.29~o
7.97% 4.6B~o
4.84~o 2.11%
8.35% 0.74% )
81.83~o 89.01%
6.55% 7.76%
4.11~o 2.46%
83.87% 89.01%
6.89% 7.76%
3.54% 2.46%
0.76% 5.70% 0.76%
a Vermont data adjusted per Appendix B.
Source: Appendix A.
Note: BAC levels less than 0.01% are effectively equal to zero. The minimum BAC at
which it is illegal to drive in most states is 0.10%.
times any person involved in an accident, had a positive blood alcohol
content (BAC).i
Clearly, a better indication of the importance of solving a problem
than calculating the associated costs would be to calculate the costs that
would be eliminated if it were solved. To determine the maximum pre-
ventable costs of drunken driving, I first examine the BAC levels of
drivers involved in various types of accidents and those of control groups
of drivers selected at random from times and places similar to those at
which the accidents being controlled for occurred. Table 1 shows these
data, which I have selected from several studies (also see Appendix A).
As Table 1 shows, the BAC levels of drivers involved in more serious
accidents are generally higher than those of drivers involved in less
serious accidents.
Once the distribution of BAC levels among accident-involved drivers
and the control groups is known, it is possible to compute how many
accidents would be avoided if all driving was done at the risk level
associated with the lowest BAC level; that is, accidents that would be
prevented by a "perfect" drinking-driving countermeasure. Appendix
C demonstrates these computations, and the results are shown in Table
2.
Since the figures in Table 2 are based on the association between
' Blood alcohol content (BAC) and the equivalent term blood alcohol concentration refer
to the standard measure of the concentration of alcohol present in a person's body at a
given time (not including any alcohol that has been drunk but not yet absorbed).
OCR for page 339
Reducing the Costs of Drinking and Driving 339
TABLE 2 Expected Reduction in Motor Vehicle Traffic Accidents
If All Drivers Had a Zero BAC
Type of Accident,
Place and Time
Expected Reduction
Fatal,
Vermont, 1967-1968
Injury,
Huntsville, 197~1975
Injury,
Grand Rapids, 1962-1963
Property damage,
Grand Rapids, 1962-1963
23.7%
15.8%
8.2~o
s.75rO
Source: Appendix C.
BAC and accident involvement, we must ask whether any other factors,
correlated with both BAC while driving and accident risk, are con-
founding our analysis. There appear to be two confounding factors bias-
ing the results in opposite directions. First, there is evidence that greater
frequency of drinking is positively associated with more frequent drun-
ken driving and negatively associated with accident risk at any given
BAC (Borkenstein et al. 1974, Hurst 1973~. As Appendix C shows, the
risk of accident as computed from the data on Grand Rapids and Ver-
mont is actually lower for the 0.10-0.049 BAC range than for the <0.01
range. The correlations with drinking frequency bias Table 2 toward
underestimating the accident reduction, since those who currently drive
drunk, if sober, would have a lower accident risk than those who cur-
rently drive sober.
The second bias results from a disputed but probable positive asso-
ciation between "problem drinking" or "alcoholism" and both drunken
driving and accident risk while sober (Smart 1969, Noble 1978b, pp.
238-240~. This bias would result in drunken drivers, as a group, having
higher accident risk than the general driving population even if they
were prevented from driving with positive BAC levels.
Available information is not sufficient to quantify either of these
biases, therefore I present the expected reductions in Table 2 as they
are and hope that the biases are small or cancel each other out.
A final caution is that my figures for maximum achievable accident
reduction are based on samples at specific times and places. Their va-
lidity for the nation as a whole is not ensured. It is heartening that the
Huntsville and Grand Rapids studies of injury accidents, despite their
spatial and temporal separation, yield reduction estimates within 8 per-
centage points of each other. The Vermont data on distribution across
OCR for page 340
340
REED
BAC ranges of drivers killed is in agreement with several other studies
of driver fatalities (Jones and Joscelyn 1978, p. 12, Noble 1978b, pp.
233-240~.2 I have encountered no studies of property-damage-only
crashes other than the Grand Rapids study.
MAGNITUDE OF PREVENTABLE COSTS
The figures on accident reduction can be roughly converted to savings
in terms of lives, injuries, and dollars of property damage. In 1977,
motor vehicle traffic accidents resulted in 49,500 deaths, 1.9 million
disabling injuries, arid $15.5 billion of property damage in the United
States (National Safety Council 1976~. (The $15.5 billion figure is based
on $520 per accident-involved vehicle [Jones and Joscelyn 19783.) Mul-
tiplying these figures by the percentages of accident reductions in Table
2 yields maximum achievable savings in 1977 of 11,700 deaths, 156,000
to 300,000 disabling injuries, and $963 million in property damage (based
on the total reduction in accident-involved vehicles).
Of course, like any estimates based on sketchy information, these
figures must be interpreted with some care. The figures in Table 2 refer
to reductions in the number of accidents, not directly to reductions in
the consequences of accidents. For instance, alcohol-related accidents
more frequently involve only a single car and driver than do accidents
in general (Borkenstein et al. 1974, p. 105, Noble 1978b, p. 235~. There-
fore, the reduction in the number of people killed, the number of people
injured, and the number of cars damaged will be less than the reduction
in number of accidents. It is also likely, however, that the average
severity of injury and damage in the accidents prevented will be greater
than the severity of injury and damage in accidents in general; therefore
only in the case of fatalities do the above figures clearly overstate po-
tential savings from drinking-driving countermeasures.
It is beyond the scope of this paper to develop estimates of preventable
costs for all other problems that may compete with drinking-driving for
government resources. Table 3, however, provides some perspective.
For example, we see from Table 3 that 22 percent of all accidental deaths
resulted from alcohol-related traffic accidents and that 10 percent of
accidental deaths would have been prevented if all drivers had zero
BAC.3
2 I could not produce accident reduction estimates from these other studies because they
lacked control groups.
3 Table assumes that 50 percent of all traffic accident deaths are alcohol related (Comp-
troller General of the U.S. 1979, Noble 1978b), and that 23.7 percent of all traffic accident
deaths would be prevented if all drivers had zero BAC (see Table 2~.
OCR for page 341
Reducing the Costs of Drinking and Driving
TABLE 3 Deaths from Alcohol-Related Motor Vehicle Traffic
Accidents, and Deaths Prevented If All Drivers Had Zero BAC as
Percentages of Various Categories of Deaths, 1977
341
Cause of Death
Related to
Drinking-Driving
Prevented If All
Drivers Had Zero
BAC
Alcohol-related motor vehicle traffic
accidents
All motor vehicle traffic accidents
All accidents
All causes except cardiovascular diseases
and malignancies
All causes
100%
50%
22%
4%
1%
47%
24~o
10%
Do
0%
Source: Bureau of the Census (1978' pp. 75, 78~.
In considering the data in Table 3, one should be aware that deaths
caused by traffic accidents, particularly alcohol-related traffic accidents'
occur typically among a younger age-group than deaths from other major
causes. For example, Perrine et al. (1971, p. 46) report that of drivers
killed in accidents, 47 percent were under 30 years of age, and of those
with positive BAC levels, 49 percent were under 30. Many think that
deaths of younger persons should be counted more heavily in making
policy decisions, because each such death results in more years of life
lost.
Another problem of weighting involves deaths of drinking drivers
themselves versus deaths of innocent victims of drinking-driving acci-
dents. I estimate that 61 percent to 78 percent of persons killed in
drinking-driving traffic accidents are drivers with positive BAC levels
(see Appendix B). One may argue that since people can choose how
much to drink before driving, it may not be the proper role of govern-
ment to protect them from the consequences of a decision freely made.
Alternatively, one might argue that people may drive after excessive
drinking due to a momentary lapse of judgment, an episode of severe
stress, or a habitual behavior pattern that is difficult to control (i.e., an
"addictions. One study reports that 75 percent of drivers admit to
driving after drinking at least occasionally (U.S. Department of Trans-
portation 1968, p. 614.4 Viewed in this way, drunken driving looks less
4 This is not to say that 75 percent of drivers drive with the high BAC levels associated
with greatly increased accident risk (see Table C-1 in Appendix C). However, there is
evidence that less frequent drinkers, and less frequent drinking drivers, show elevated
accident risk at lower BAC levels than does the drinking-driving population in general
(Hurst 1973~.
OCR for page 342
342
REED
like a risk some choose to take and more like a random risk to which
many people will be exposed whether or not they would choose to in
a moment of calm reflection. Furthermore, the death or injury of some-
one who is driving drunk may impose costs on friends, loved ones, those
dependent for financial support, and the economy in general.
In setting policy, the proper weight to give to deaths of drunk drivers
cannot be determined by empirical or logical analysis, although these
may inform the decision. I feel that such deaths should be weighted
equally to others for purposes of setting policy, but that we should be
wary of providing perverse incentives to potential drunk drivers- a topic
I address later in this paper.
In conclusion, it seems that drinking-driving countermeasures can be
legitimate and useful government actions, but that even if such coun-
termeasures were perfectly successful, the savings in lives, injuries, and
property loss would be less than widely quoted figures would lead one
to believe. Which countermeasures show the best prospect of success
is the topic of the rest of this paper.
EXPOSURE REDUCTION
OVE RVIEW
The strategy of drinking-driving countermeasure that occurs first to most
people is exposure reduction: reducing the amount of drunken driving
that takes place and thereby reducing accident costs. This section ex-
amines several potential ways of achieving exposure reduction:
· General deterrence: countermeasures that seek to prevent drivers
in general from combining driving with drinking in excess of legally
prescribed limits (0.10 percent BAC in most states).
· Recidivism reduction (specific deterrence): countermeasures that
seek to specifically compel those people who have already been arrested
for driving while intoxicated (DWI) not to drive drunk again.
· Third-party intervention: countermeasures that seek to influence
those around potential drunk drivers (servers of alcohol, fellow party
guests, or bar patrons, etc.) to prevent them from driving while intox-
icated.
· Altering the legal minimum drinking age.
· Screening the driving population for those most likely to drive
drunk.
· Installing devices in vehicles to automatically detect drunk drivers.
· Providing alternative transportation for potential drunk drivers.
OCR for page 343
Reducing the Costs of Drinking and Driving
GENERAL DETERRENCE
Risk of Punishment
343
The most effective programs of general deterrence seem to have been
those that raised drivers' perceived risk of arrest and punishment for
drunk driving.
The classic program of this type is the British Road Safety Act of
1967. The act "defined driving with a BAC of .08% or higher per se as
an offense," and "gave police the power to require pre-arrest roadside
tests [for breath alcohol] from drivers who had been involved in traffic
accidents or moving violations or where there was reasonable cause to
suspect them of drinking and driving. Drivers who refused to provide
a breath test were considered guilty" (Cameron 1978, p. 22~. Passage
of the act was accompanied by a great deal of publicity and public
awareness of its provisions (Jones and Joscelyn 1978, pp. 66-67~.
The immediate impact of the act was positive and dramatic. For the
3 months following passage of the act, casualties from traffic accidents
were reduced 16 percent from the same period the previous year, and
fatalities were reduced 23 percent (Ross 1973, p. 20~. The percentage
of fatally injured drivers with BAC levels of greater than or equal to
0.08 percent dropped from 27 percent before passage to 17 percent the
following year (Comptroller General of the U.S. 1979, p. 27~. A careful
quasi-experimental study by Ross (1973) attributed these reductions
substantially to the act and the wide publicity it received. He also noted
(p. 75~:
Unfortunately. there are many signs that the initial effect of the legislation
is diminishing. Although there are problems in speculating on what would have
happened in the absence of the legislation, the significant change in the slope
of the casualty rate curves . . . suggests that the savings achieved ought to be
regarded as temporary. This conclusion is bolstered by the fact that blood
analysis of fatalities shows that the initial drop in the percentage with an illegal
alcohol concentration. from 25 percent in 1967 to 15 percent in 1968, was
progressively diminished and the percentage has now returned to its former
level.
The trend noted by Ross appears to have continued. By 1975 the
percentage of drivers killed in England and Wales in road accidents with
a BAC level of 0.08 percent or more had reached 36 percent, substan-
tially above its pre-1967 level (Comptroller General of the U.S. 1979,
p. 27~.
The explanation offered by Ross ( 1973, pp. 75-78), which has achieved
wide acceptance, is that the well-publicized passage of the act convinced
OCR for page 344
344
REED
many drivers that the risk of arrest when driving drunk had become
much higher than it used to be be. Potential drunk drivers were deterred
by the threat of arrest and punishment. As time went on, however,
drivers realized that enforcement was rather slack and that risk of arrest
was not all that high. This realization caused an "evaporation" of the
act's deterrent effect.5
In September 1975, the Cheshire County Police seemed to recapture
the Road Safety Act's deterrent effect through a publicized policy of
administering breath tests to all drivers pulled over for violations or
involved in accidents during "drinking hours." The resulting accident
reduction "evaporated" a month after the policy came to a publicized
end (Ross 1977~.
How applicable is Britain's experience to that of the United States?
Ross (1973, p. 1) asserts that "in broad culture and narrower legal
structure Britain is the closest of European countries to the United
States." Comparative statistics indicate that Britain is very similar to the
United States in number of traffic fatalities per vehicle mile (Borkenstein
et al. 1974, p. 20) and has a slightly smaller representation of alcohol-
influenced drivers among driver fatalities (Organisation for Economic
Co-operation and Development 1978, p. 25~. More convincing is the
fact that the Road Safety Act experience has been partially replicated
in countries other than Britain. As Robertson (1977, p. 6) describes:
A law providing for breath testing and penalties for blood alcohol above .08~o
by weight or refusing the test came into force in Canada in 1969 but with less
widely publicized predictions of increased chances of arrest than in Britain.
Death rates were about 8 percent less than expected in the subsequent two years
but, again, the effect of the law was temporary. After 2 years of the law, death
rates in Canada returned to levels that would have been expected without the
law.
In 1978, France amended its drunk-driving law to allow police to
5 There are alternatives to the "evaporation" theory regarding the British Road Safety
Act. In support of alternative or supplementary explanations. we should note that since
the date of Ross's study, incidence of illegally high BAC levels among drivers killed in
Britain has continued to climb, until it now substantially exceeds the incidence before
passage of the act. Mere evaporation of the deterrent effect cannot account for this. Other
factories) must be strengthening the association between BAC level and driver fatalities
over time. It is not certain how much of what Ross observed was due to "evaporation"
and how much was the result of the aforementioned unidentified factories).
The "evaporation" theory, however, is still attractive as a partial explanation. It is well
grounded in theory and predicts the outcomes of several other programs patterned after
the British Road Safety Act.
OCR for page 345
Reducing the Costs of Drinking and Driving
345
administer breath tests to drivers even in the absence of an accident or
traffic violation. Dorozynski and Volnay (1978, p. 44) describe the initial
effects (presented here in translation):
A certain number of [breath testing] operations were organized immediately,
[and] announced in the press, in several French provinces and in Paris. Their
results were entirely unexpected: less than O.S~o of the "alcooltests" were pos-
itive, and in Paris, zero.
Taking account of what is known of the accuracy of the "alcooltest," and of
its sensitivity, this roughly indicates that the "French at the wheel" had stopped
drinking.
While the information from France so far is too sketchy to tell us
anything, it does not appear inconsistent with Ross' observations in
Britain.
Unfortunately, the only agreed-upon success in the United States
similar to the British Road Safety Act was a 1-year countermeasure
program at Lackland Air Force Base, Texas, in 1959. While the program
achieved "a statistically significant decrease of 50% to 60% in the num-
ber and rate of accidents, driver injuries, and other injuries during the
operational period" (Cameron 1978, p. 23), it is not clear how applicable
this experience would be to a civilian environment.
Despite the lack of documented successes in the United States for
countermeasures to deter drunken driving by increasing the risk of arrest
and punishment, this approach appears to have won favor among state
highway safety officials. A recent survey by the U.S. General Account-
ing Office (Comptroller General of the U.S. 1979) asked officials from
the 50 states, the District of Columbia, and Puerto Rico to choose and
rank the three most important current or past efforts in their state to
combat drunken driving. "Instituting or increasing the use of special
police patrols for the drinking driver" was ranked number one by 25
percent of respondents, more than any of the other nine responses.
Another 27 percent of respondents ranked it second.
Of six states examined in depth in the GAO report (California, Geor-
gia, Louisiana, Minnesota, New York, Washington state), five were
operating some police patrols targeted at drinking drivers, four reported
increased numbers of driving while intoxicated (DWI) arrests, and two
reported evaluations of the patrol's effect on accident rates. In King
County, Washington, a "drinking-driver emphasis patrol" was operated
as part of the federally funded Alcohol Safety Action Project (ASAP).
Evaluation failed to find a change in accident fatalities or injuries re-
sulting from the patrol, although DWI arrests had increased. In Hen-
OCR for page 346
346
REED
nepin County, Minnesota, emphasis patrols were also operated as part
of ASAP. "Alcohol involvement in fatal crashes was reduced from 63
percent in 1972 to 38 percent in 1976" (Comptroller General of the U.S.
1979, p. 22~. Even in the absence of a control group, this finding is very
suggestive of success.
From present evidence, then, it appears that when drivers' perceived
risk of arrest and punishment for drunk driving is sufficiently increased,
drunk driving is deterred and accidents are reduced. In Britain, fatalities
from traffic accidents decreased by 23 percent in response to the Road
Safety Act, and similar legislation in Canada brought about a temporary
reduction of 8 percent. In order to sustain a high perceived risk of arrest
and punishment, the risk must be made high and kept high.
Targeting patrols by day of week, time, and geographic location;
legislative and technical progress toward making breath tests for alcohol
easier to administer; and simplifying the process of making a DWI arrest
and providing police with motivation to make such arrests are all ways
to increase the risk of arrest. Using such methods, ASAPs were able
to double and triple the number of DWI arrests, although it is unclear
how much of this increase resulted merely from charging drivers with
DWI rather than a specific moving violation (Zimring 1978 pp. 151-
152~.
What remains unknown is just what levels of risk are necessary to
achieve various degrees of deterrence and what it would cost to bring
about such increases in risk. These questions appear to require empirical
study.
Severity of Punishment
If increasing risk of punishment can deter drunken driving, then what
about increasing severity of punishment? It seems at first glance easier
and less expensive to hand out stiffer penalties to convicted drinking
drivers than to beef up enforcement. The archetypes of the severe pun-
ishment approach are the Scandinavian countries. Imprisonment or fines
exceeding one-tenth of the convicted person's after-tax income, com-
bined with license suspensions exceeding 1 year, are common punish-
ments for first DWI offense in these countries.
The Scandinavian drunk-driving laws are widely reputed to be effec-
tive deterrents, and there is anecdotal evidence in accord with this rep-
utation, but during a 3-month study in Scandinavia, Ross (1975) was
unable to find any scientific evidence that the laws had deterrent effects.
He then performed time series analyses of drunk driving and traffic
OCR for page 377
Reducing the Costs of Drinking and Driving
3~17
Control group was six cars going in the same direction at same place,
time of day, and day of week as each accident in case group. "All site
visits were made in 1960, within a few weeks of the calender week of
occurrence" (p. 812~.
Of the control group 0.39 percent refused the BAC test. BAC test
was performed by the municipal medical examiner on each member of
case group.
TABLE A-7 New York: Survey Results and Relative Risk
BAC # Control No Control # Accident % Accident Ratio Risk
0 195 77.38 14 41.18 0.53 1.00
<0.02 14 5.56 0 0.00 0.00 0.00
0.02~.099 34 13.49 3 8.82 0.65 ~ 1.23
0.1~0.249 9 3.57 2 5.88 1.65 3.12
0.25 0.399 0 0.00 15 44.12
I have interpolated these results into my standard BAC categories by
assuming uniform density of observations within each of the study?s
BAC categories.
<0.01 202.00 80.31 14.00 41.12 0.51 1.00
0.01-0.049 19.48 7.75 1.10 3.23 0.42 0.82
0.05-0.099 21.09 8.39 1.86 5.46 0.65 1.27
0.10-0.149 2.96 1.18 0.66 1.94 1.64 3.21
~0.15 5.99 2.38 16.43 48.25 20.27 39.73
The predominant flaw of this study is the small sample size, which allows
us to put little confidence in the results.
VERMONT
M. W. Perrine, Wailer, J. A., and Harris, L. S. (1971) Alcohol and
Highway Safety: Behavioral and Medical Aspects. National Highway
Traffic Safety Administration technical report DOT-HS-800-599. Wash-
ington, D.C.: U.S. Department of Transportation.
Case group was fatally injured drivers in Vermont from July 1, 1967,
to April 30, 1968.
Control group was pooled data of: (1) six drivers at same place, time
of day, and day of week as each observation in case group, either within
a few weeks or one year after accident being matched; and (2) six drivers
at same place, time of day, and day of week as each observation from
OCR for page 378
378
REED
a group of crashes occurring in Vermont in 1966 and resulting in an
injury warranting hospitalization, each observation of which was chosen
as a close match to one observation in the case group.
The report's account of control group selection is unclear and self-
contradicting. The report justifies pooling the two samples by showing
that there is a significant difference in the distribution within either
sample of only 3 of 22 variables. I suspect that the control group shows
fewer high BAC levels than were actually present among drivers at times
and places similar to those of case group crashes.
Of control group 1.3 percent refused the BAC test. BAC test was
performed on all members of the case group.
The lowest BAC category used in the report is <0.02. I have inter-
polated into my standard BAC categories by assuming that the density
of observations in the interval 0.01 SAC <0.02 is the same as the
density in the interval 0.02 ~ BAC ~0.04. All other categories in the
report are comparable to mine.
TABLE A-8 Vermont: Survey Results and Relative Risk
BAC # Control % Control # Accident Do Accident Ratio Risk
<0.01 942.10 83.74 47.62 44.92 0.54 1.00
0.01~.049 104.90 9.32 5.38 5.08 0.55 1.02
0.05-0.099 54 4.80 9 8.49 1.77 3.27
0.1~0.149 13 1.16 14 13.21 11.39 21.07
~0.15 11 0.98 30 28.30 28.88 53.43
The "risk" figures for the higher BAC categories are probably biased
upward somewhat due to flaws in the control group.
APPENDIX B: ADJUSTING VERMONT DATA TO
REFLECT ALL FATAL ACCIDENTS
The Vermont Study (see Appendix A) reports the BAC levels only for
fatally injured drivers, but for purposes of computing the possible savings
from drinking-driving countermeasures we need to know the BAC dis-
tribution of all drivers involved in accidents in which anyone was killed.
To adjust the Vermont data to reflect all fatal accidents, I use data
presented by Sterling-Smith (1976, p. 135~. Sterling-Smith studies the
drivers judged to have been "most responsible" for each of 267 fatal
traffic accidents in the Boston area from 1971 to 1974. He presents a
breakdown of this group of drivers by BAC range and whether the
fatality was the driver or another person. Table B-1 shows these results.
OCR for page 379
Reducing the Costs of Drinking and Driving
TABLE B-1 Number of Drivers
Responsible for Fatal Accidents, Boston,
1971-1974
BAC (5to) Driver Killed Other Killed
<0.01 29 1 16
0.01-0.04 6 13
~0.05 68 35
TOTAL 103 164
Source: Sterling-Smith (1976, p. 135~.
379
Table B-1 cannot be directly applied to the Vermont data. Since
Boston is much more urbanized than Vermont, there is a greater density
of vehicles and pedestrians for a driver to collide with, and we would
expect a lower ratio of driver fatalities to all fatalities than in Vermont.
The data bear this out. According to Table B-1 driver fatalities comprise
39 percent of fatal accidents in the Boston area, whereas Perrine et al.
(1971, p. 42) report that the ratio in Vermont in the early 1970s was 59
percent.
To adjust the Boston data to those of Vermont, I reduced the number
of nondriver fatalities ("other killed") until driver fatalities comprised
59 percent of total fatalities. I multiply the number of nondriver fatalities
in each BAC range by 72/164 to arrive at the figures in Table B-2.
TABLE B-2 Adjusted Number of Drivers
Responsible for Fatal Accidents
BAC (%) Driver Killed Other Killed Total Killed
<0.01 29 5 1 80
0.01-0.04 6- 6 12
~0.05 68 15 83
TOTAL 103 72 175
By dividing the "total killed" figure by the "driver killed" figure for
each BAC range in Table B-2, I obtain for each range an estimate of
the ratio of all fatal accidents to driver-fatal accidents in Vermont. Let
us call the ratio for the ith BAC range Ri. The percentage of driver
fatalities in each BAC range, as listed in Appendix A, is D'. Then it is
clear that the percentage of all fatalities in the ith BAC range is
Ti = ~;ij; lOO~o.
OCR for page 380
380
REED
Table B-3 shows Ti for each BAC range. The control group figures need
no adjustment, so they are merely reproduced in Table B-3 as they
appeared in Appendix A (with the three highest BAC ranges summed).
TABLE B-3 Percentage of Drivers
Involved in Fatal Accidents, and of
Control Group Drivers, Estimated to Be
Within Three BAC Ranges
BAC (DO)
Accident (5to) Control (%)
84
<0.01 64
0.01-0.04 5
30.05
31
APPENDIX C: RELATIVE RISK AND ACCIDENT
REDUCTION
CONTROLLED EPIDEMIOLOGICAL STUDIES
Among the most valuable studies of drinking-driving problems are con-
trolled epidemiological studies, such as those reviewed in Appendix A
of this paper. In these studies, a random (or exhaustive) sample of
drivers involved in accidents in a set geographic area and time period
is made. The BAC of each driver in the sample is measured at the time
of the accident. These drivers constitute the "accident group."
In addition, researchers match a "control site" to each observation
in the accident group. The control site is ideally the same place, at the
same time of day and day of the week, under similar weather conditions,
and otherwise nearly identical to the circumstances under which the
accident being matched occurred. Several drivers passing each control
site are randomly selected and their BAC levels measured. These ob-
servations constitute the "control group."
Drivers in both the accident group and control group are classified
into BAC ranges, and the results are presented in Table C-1.
I use the results of a controlled epidemiological study, as presented
in Table C-1, to estimate the results of an imaginary experiment that
is impractical to actually perform. The imaginary experiment is to ran-
domly sample "units of accident exposure" and record driver BAC and
whether an accident occurs for each unit of exposure selected. A unit
of accident exposure would be a period of driving during which a "stand-
ard" driver would have some particular expected number of accidents.
OCR for page 381
Reducing the Costs of Drinking and Driving
TABLE C-1 Results of a Controlled
Epidemiological Study
Number of
Accident
BAC Range Group Drivers
Number of
Control
Group Drivers
O a`, c<,
1 al Cl
n
a,,
en
381
For example, driving for 1 hour (or 1 mile) at night in the rain on a
narrow twisting road would constitute more units of exposure to accident
than would 1 hour (or 1 mile) of driving on a sunny day on a well-
designed expressway.
If we could perform the imaginary experiment, then we could answer
two important questions: (1) What are the probabilities of an accident
occurring when one unit of accident exposure is driven by a driver at
various BAC levels? (2) What would be the reduction in number of
accidents if all driving was done by drivers in the lowest BAC range?
ESTIMATING RELATIVE RISKS
What is the probability that a random unit of exposure results in an
accident, given that the driver is in some particular BAC range? To
answer this question, I invoke Bayes's theorem:
P`A~i~ - (A n i)
= .
p (i) ~
where
(1)
P(A~i) is the probability that a given unit of exposure results
in an accident, given that the driver is in the ith BAC
range.
UP n id is the probability that a randomly chosen unit of driving
results in an accident, and that the driver is in the ith
BAC range.
P(i) is the probability that a randomly selected unit of ex-
posure is driven by a driver in the ith BAC range.
'The Bayesian analysis used in this and the following section has been applied to drunk
driving before, notably by Hurst (1970, 1973~.
OCR for page 382
382
Also from Bayes's theorem:
Substituting (2) into (1~:
REED
P(Ani) =P(i~A)P(A).
P`A~i' P(i~A) P(A)
.
(2)
(3)
Referring back to Table C-1, we see that Pupil) = at/~.ai, since the
accident group is a random sample of units of exposure resulting in
accidents. We can use c,/Ici as a proxy for P(i), because the probability
of a driving trip being included in the control group is primarily deter-
mined by the likelihood that a similar trip resulted in an accident. Thus,
the control group approximates a random sample of units of exposures
We can thus rewrite (3) as:
2 The probability of any given trip being included in the control group can be expressed
as:
P(T ~ C) = N E D (SIN - R),
where
P(T ~ C) is the probability that trip T becomes an element of the control group.
N is the number of trips (including trip T) to which trip T could be matched
as a control (i.e., the number of trips passing sites for which the site of
T might be a control site).
is the mean exposure to accident of each of the N trips.
D is the risk multiplier due to the fraction of drivers among the N trips who
have elevated BAC levels.
S is the number of trips sampled at each control site.
R is the probability that trip T ends in an accident before the car reaches the
site where it would have been sampled for the control group.
I make the simplifying assumptions that L) approaches one and R approaches zero. We
are left with:
P(T ~ c) = ES.
S is a constant, so the probability that trip T is included in the control group is ap-
proximately equal to the expected value of the exposure to accident of trip T.
