EXPOSURE OF HUMANS TO CARBON MONOXIDE COMBINED
WITH INGESTION OF ETHYL ALCOHOL
AND THE COMPARISON OF HUMAN PERFORMANCE
WHEN EXPOSED FOR VARYING PERIODS OF TIME TO CARBON MONOXIDE
Report No.: CRC APRAC CAPM-3-68 MCOW-ENVM-CO-74-2
From the Department of Environmental Medicine, The Medical College of
Wisconsin, Milwaukee, Wisconsin 53226
Supported by Contract CRC-APRAC, Project No. CAPM-3-68, From the
Coordinating Research Council, Inc., and The Environmental Protection Agency
-------
The automobile driver who is being exposed to carbon monoxide (CO)
while driving home after a social cocktail plus an analgesic to relieve his
tension headache is a common figure in our society. Consequently,
when trying to assess the potential hazard of CO, the possibility of a
synergistic effect with commonly used drugs and ethyl alcohol must
be considered. CO exposure along with ingestion of phenacetin or
diphenhydramine hydrochloride was the topic of a. previous report. This
report presents the findings of two sets of experiments, referred to as
Study A and Study B designed to assess the effects of CO and ethyl alcohol
on behavioral test performance and to assess the effects of a 24-hour
exposure as compared .to a 5-1/2 hour exposure to identical concentrations
of CO.
STUDY A
/
In the fall of 1970, four experiments were conducted to study the
effect of the following agents on man: carbon monoxide, ethyl alcohol,
carbon monoxide plus ethyl alcohol and ambient air. Because it was not
desirous to have inebriated subjects incapable of performing the tests,
a low dose level of 1. 6 ml - 100 proof alcohol/kg body weight was chosen.
The alcohol experiments were conducted immediately after completion
of experiments defining the effect of CO on time perception, reported
earlier. Consequently, a CO exposure level of 200 ppm was retained.
-------
-2-
METHODS
Protocol
Thirty minutes prior to each exposure the subjects were given an
interim physical examination at which time they were questioned about
subjective feelings including nausea, headache, dizziness, chest pain,
abdominal pain and eye, nose and throat irritation. Pre-exposure blood
samples were also obtained at this time for determinations of percent
carboxyhemoglobin concentration (COHb).
During the first 150 minutes of each exposure, time tests were
performed and correlated with the corresponding COHb saturations.
(1)
The results of these tests were reported earlier. Between 150 and 195
minutes of exposure time each subject was given 1.6 ml alcohol (Smirnoff's
Vodka) per kilogram body weight mixed with orange juice and equally divided
between three drinks. The subjects were given 15 minutes to finish each
drink. After completion of the drinks, a blood sample was obtained
for blood alcohol and % COHb determinations. Immediately following
the blood sampling, the behavioral tests were administered over the next
30 minutes and the exposure was terminated at 240 minutes. A post
exposure blood sample was then obtained for blood alcohol and % COHb
determinations.
-------
-3-
Subjects
The subjects were five male and one female sophomore medical
students, ranging in age from. 22 to 26 years, selected from a larger group
of volunteers on the basis of being in good health and because of their
expressed willingness to expose themselves to CO. Prior to their selection,
subjects were completely informed as to the purpose of the experiments
and the potential hazard to themselves. Subjects were paid $1. 50 per hour
for each hour in the chamber, with a bonus of $1. 00 per hour conditional
upon satisfactory completion of the study. Baseline % COHb saturations
attested that all six subjects were nonsmokers. Prior to and after com-
pletion of the study, each subject was given a comprehensive medical
examination which included a complete history and physical examination,
a 12-lead EKG and an EEG using the standard 10-20 electrode configuration.
Exposure Chamber
All CO exposures were carried out in the controlled environment
chamber located in the Department of Environmental Medicine, The Medical
(1)
College of Wisconsin. The environmental system (Carrier) provided
accurate control of temperature (72 +_ 2°F) and relative humidity (40 4^5% RH)
within the chamber, a room measuring 20 x 20 x 8 feet. The chamber was
operated at a slight negative pressure with an air circulation rate of 1, 200 cfm.
Exhaust and makeup air capacities of 1,000 cfm provided a rapid chamber
flushing capability when desired. The chamber featured pleasant lighting,
-------
-4-
comfortable chairs, individual testing carrels and a restroom facility.
The subjects were under continuous visual surveillance by medical personnel
while in the chamber, and in addition, their activities were visually
monitored and periodically video taped by closed circuit TV (Sony).
Exposure Chamber Atmosphere
On non-exposure days, outside ambient air was flushed through the
chamber. On exposure days, CO was continuously metered into the
chamber's environmental system from a compressed gas cylinder in the
adjacent command laboratory. Chemically pure CO with a minimum purity
of 99. 5% was used. The concentration of CO in the chamber atmosphere
was continuously recorded from an infrared spectrophotometer (Beckman
IR-10) equipped with a 10-meter path-length gas cell which was continuously
flushed with air drawn from the chamber through 1/4" diameter polyethylene
tubing. An MSA CO Meter and Alarm, Model 701, also independently
sampled chamber air to provide a second independent means of continuously
monitoring chamber concentration. The chamber atmosphere was also
periodically monitored by a gas chromatograph (GC) (Varian Aerograph)
equipped with a sequential sampler.
All three methods of monitoring CO concentration during an exposure
were calibrated from within the chamber with standards prepared in saran bags,
-------
-5-
Analysis of Carboxyhemoglobin Saturation
Five-ml. aliquots of venous blood were collected in Vacutainer
tubes containing edetic acid and immediately analyzed for the hemoglobin
concentration and the percent carboxyhemoglobin saturation using a
CO-Oximeter (Instrumentation Laboratories, Inc.).
Behavioral Tests
The Flanagan Coordination Test (Science Research Associates, Inc.,
259 East Erie Street, Chicago, Illinois) measured the subject's ability to
rapidly and accurately follow a spiral pathway with a pencil. The subject,
while sitting comfortably at a desk in the individual carrel, was allowed
40 seconds to complete each of 6 spirals. The first two were considered
practice and the last four were scored and totaled. The score was determined
by the distance covered in each spiral minus the number of times the
sides of the spiral were touched with the pencil and had a maximum of 100.
The random number inspection test measured the speed with which
a subject could detect the number "3" in rows of random numbers on an
8-1/2 x 11 inch page. The subject was asked to inspect each row of
numbers beginning at the top of the page and mark as many 3's as possible
in two minutes. The subject's score was the total number of 3's struck
with an unachievable maximum score of 203. Ten different pages of random
numbers were used sequentially to prevent memorization of test answers.
-------
-6-
The arithmetic test, which measured the subject's ability to work
with numbers, was divided into two parts. The first part, timed at five
minutes, consisted of simple addition and subtraction problems while the
second part, timed at three minutes, consisted of multiplication and
division. The maximum score available, if all answers were correct,
was 125; however, no subject completed the test in the allotted time. In
order to prevent memorization of answers, ten different tests generated
from random number tables were used sequentially.
The Crawford collar-and-pin test, (produced by The Psychological
Corp. . 304 E. 45th Street, New York, New York 10017) measured a
subject's manual dexterity. The test required the subject to use forceps
to pick up a pin (0. 657" long, 0. 063" diameter), place the pin in a hole
(0.08" diameter) and place a loosely fitting collar over the pin. The test
score was the number of collar and pin combinations placed in three minutes
and had a maximum of 42.
Hand Steadiness Test
In the AAA hand steadiness test, a subject passed a hand-held
metal wand down through a gradually narrowing V-shaped vertical slot.
Contact with either side of the slot completed an electrical circuit which
flashed a light. The slot was numerically graduated for recording wand
-------
-7-
position, and the score was tabulated by an observer who totaled the results
of five trials. The test had a maximum score of 80 and took 1 minute to
perform.
Data Analysis
Data analysis of the behavioral test scores used an analysis of
variance for two factors. The within treatments mean square was used
o
as the estimate of uncontrolled variability in calculating F ratios.
RESULTS
The blood alcohol levels and COHb saturations determined for
each subject at the conclusion of each of the four experiments are presented
in Table 1. Subject K. H. (native Chinese) could not tolerate this level
of alcohol and was not given any alcohol after the first drink on the first
day. His data were, not included in the final analysis. The remaining
subjects displayed the usual post alcohol ingestion periods of gaiety followed
by drowsiness. They remained in good health throughout the study.
Table 2 presents the analysis of variance (AOV) F ratios and their
significance. A significant F value appears that CO had an effect on the
performance of the collar and pin test. However, this effect was an increase
in test score, or an improvement in test performance. The Coordination
Test performance was affected by alcohol but not by CO. The test data
reveals this effect was a decrease in test performance. Performance of
the Arithmetic Test and Hand Steadiness Test were not affected by either
alcohol or CO at these levels.
-------
-8-
STUDY B
The second series of experiments was conducted in the fall of 1972,
and had several distinct differences from the first series: exposure for
24-hours were included, CO exposure was lowered from 200 to 70 ppm and
30 ppm but the steady state % COHb saturation was reached by an initial
30 minutes at 1, 200 ppm, and behavioral testing was conducted under the
direct supervision of an experimental psychologist. Simultaneous exposure
to ethanol and CO was conducted as the last experiment in this series.
METHODS
Protocol
To obtain information on the prolonged'effects of CO exposure,
four subjects were selected to remain in the chamber overnight for 24 hours,
being exposed continuously to CO concentrations of 0, 30, or 70 ppm. A
total of six 24-hour sessions were run--three at 0 ppm, one at 30 ppm,
and two at 70 ppm.
For the 30 ppm session, there was an initial 20-minute exposure
to 1,000 ppm CO, which caused a rapid rise of COHb saturation to ap-
proximately 5%. For one of the 70 ppm sessions, an initial 30 minutes
at 1,000 ppm caused a rapid rise of COHb saturation to approximately 10%.
-------
-9-
CO concentration levels were presented in a random, order and
every attempt was made to prevent subjects from knowing the CO concen-
tration present during a given session. Questionnaire data obtained at
the end of each session confirmed subject's naivete in this regard. The
four 24-hour subjects, together with four other subjects who left the chamber
overnight, were given behavioral tasks to perform during the first 5 hours
and during the last 5 hours of this 24-hour period. A comparison of
performance of the 24-hour subjects between the first and the last 5-hour
test periods was made to determine the effects of prolonged CO exposure.
In addition, all eight subjects were exposed to the three concentrations
of CO during regular 5-hour sessions which were scheduled the day
before the 24-hour sessions. Four, 5-hour sessions were held, two
sessions at 0 ppm, one at 30 ppm, and one at 70 ppm. As with the 24-hour
sessions, exposure concentrations were selected randomly for the 5-hour
sessions. The purpose of the 5-hour sessions was to obtain additional
information on the effects of CO during short exposures.
To determine whether alcohol potentiated the effects of CO, all
subjects were studied during two separate 5-hour sessions. These sessions
occurred as the final two sessions of the experiment. During the first
of these sessions, subjects were given 1.5 ml 100% proof alcohol/kg body
weight and exposed to a CO concentration of 70 ppm. During the second
-------
-10-
ses sion, which occurred the next day, subjects were given the same amount of
i
alcohol but no CO was present in the chamber. As always, subjects were unaware
of actual CO concentrations.
Subjects
The subjects were eight male college students, ranging in age from 20
to 26 years. Of the eight subjects, six were nonsmokers and two were light to
moderate smokers who were requested to refrain from smoking for the duration
of the study.
Behavioral Tests
Subjects were tested simultaneously while seated at two tables in the
environmental chamber. Partitions on the table tops extended several inches
beyond the edge of the tables thus reducing visual interaction among subjects
and providing a somewhat isolated testing area for each subject. All behavioral
tests were administered under the direct supervision of an experimental psychol-
ogist and were the same as in the first set of experiments (A) except for the
deletion of the Crawford Collar and Pin Test and the Hand Steadiness Test, and,
the addition of the Time Estimation Tests and a Vigilance Test which are described
below.
The vigilance test required subjects to detect and report a complex
visual signal presented intermittently over an extended period of time. The
test apparatus consisted of eight General Motors readiness-to-drive consoles,
one for each subject. During the test, the subject sat facing his console and
random four-digit numbers were briefly flashed on a front window of the
console after which he pressed the number back into the consol via the
console keyboard. Flash duration was 0.7 seconds and the time interval
between numbers averaged 18 seconds during a brief, 3-minute alert phase,
and 90 seconds during a subsequent 45-minute phase. Subjects remained
-------
-11-
at the task for 48 minutes followed by a 12-minute rest period. This
procedure continued until 4 hours had elapsed with the beginning and end
of each 48-minute session indicated by the offset and onset, respectively,
of a light on the console. Over the 4-hour period, 160 four-digit numbers
were presented to each subject. The number of four-digit numbers correctly
reported within 5 seconds after the number was flashed defined the vigilance
performance of a given subject. Subjects reported the numbers by pressing
appropriate buttons on the front of the console, labeled 0 to 9, in the same
sequence as the number which had been presented. The 5-second period
allowed for reporting the number was identified by a white light on the console.
Immediate feedback was provided to each subject in the form of a green
or red light which flashed briefly at the end of the 5-second interval, depending
a
on whether the number was reported correctly or incorrectly. Unlike the
other behavioral measures, subjects were not given extensive training on
the vigilance test prior to the experiment because performance on this
sort of task which does not test reaction time does not appear to improve
with practice. A single 4-hour practice session was conducted to insure
that subjects understood the procedure and to familiarize them with the
apparatus. During this practice session, and at various times throughout
the experiment, subjects were instructed not to talk or interact in any way
with other subjects so as to minimize distraction during the test.
-------
-12-
The Marquette Time Test consisted of a series of nine tone stimuli
followed by a series of nine light stimuli. Each series contained stimuli of
approximately 1.0, 1.5, . . . 5.0 sec. duration presented in random order.
At termination of each stimulus, the subject depressed a push button switch
for that interval of time he estimated to be equal in duration to that of the
original auditory or light stimulus. This provided a measure of his ability
to estimate the duration of the stimulus; the interval between stimulus
termination and onset of the response -was a measure of reaction time.
The ten-and thirty-second time estimation tests consisted of each ',',
subject depressing the push button described above for an interval equal
to that desired (10 or 30 seconds). This -was repeated an additional two
times for each test.
!)
RESULTS
!!
ii
The blood alcohol levels and % COHb saturations obtained from Study B
are presented in Table 3. The subjects displayed the usual post alcohol j
ii
11
ingestion periods of gaiety and then drowsiness. They remained in good health
.11
throughout the experiment. jj
' ii
Comparison of data obtained from the behavioral testing during the
experiments designed to bring the % COHb saturation to a plateau very
I j|
quickly to that data obtained during experiments when this plateau was '.'
ii
reached in a 5 or 8 hour period revealed no significant difference in performance.
Therefore, all data from the 30 ppm or 70 ppm exposures were averaged
i
for a comparison of the performance at 1 hour of exposure versus 24 hours of
Fl
exposure. j
-------
-13-
Coordina.tion Test
Figure 1 presents performances of the four 24-hour subjects and
1 and 20 hour of exposure to 0, 30 and 70 ppm of CO. Comparisons of
performances during the 1st and 20th hour indicates the degree to which
subjects were affected by CO after prolonged exposure. It is apparent that
there were no consistent changes in coordination for subjects exposed to
0 or 30 ppm CO for 20 hours. While all four subjects showed slight
improvements in coordination performance after 20 hours of exposure to
70 ppm, the small magnitude of the mean difference and the large amount of
variability, as expressed by the standard errors, provide little confidence
in the reliability of this finding.
Figure 2 shows coordination performances during the last two
sessions of the experiment when subjects were given alcohol plus CO
(70 ppm) or alcohol alone. Clearly, there was no systematic tendency
for alcohol to potentiate the effects of CO relative to alcohol-along baselines--
four subjects showed decreased coordination with alcohol plus CO while the
remaining four showed increased coordination.
Finally, Figure 3 summarizes overall coordination performances
under each condition of the experiment. Again it is apparent that subjects
did not show systematic changes in coordination performance in response to
CO; however, all subjects did show deteriorations in coordination following
ingestion of alcohol and as demonstrated in the AOV (Table 4), this was found
to be highly significant.
-------
-14-
Arithmetic Test
Figure 4 presents arithmetic scores of the 24-hour subjects under
each concentration of CO and allows a comparison of performances obtained
during the 1st and 20th hour of the 24-hour sessions. As may be seen,
arithmetic performance did not change with prolonged exposure to CO,
regardless of concentration level.
Figure 5 shows, in addition, that alcohol did not systematically
potentiate the effect of CO. Three subjects showed a decrease in arithmetic
scores but the remainder showed an increase in arithmetic scores when
alcohol was combined with CO.
From Figure 6 and from the AOV (Table 4), alcohol as a treatment
variable had little effect on ability to solve arithmetic problems, at least
at this blood alcohol level.
Inspection Test
Mean and individual inspection scores of the 24-hour subjects after
the 1st and after the 20th hour of CO exposure are shown in Figure 7. Only
one slight but systematic trend is apparent. Under 70 ppm CO, all four
subjects showed decreases in scores after 20 hours of exposure to CO.
This finding is in contrast to the slight increase that was noted in coordination
after prolonged exposure to 70 ppm CO. As such, and until this finding
can be reliably reproduced in future research, it is best viewed as an
artifact of the small number of subjects employed in the present study.
-------
-15-
Figure 8 shows the inspection scores of each subject in response
to alcohol plus CO (70 ppm) and to alcohol alone. Clearly, there is no
consistent tendency for alcohol to potentiate the effect of CO on performance
of this test since some subjects showed decreased scores while others
showed an increase.
Looking at the overall data obtained on the inspection test under
each condition of the experiment, summarized in Figure 9 and in the AOV
Table, Table 4, it is apparent that the inspection test was not a sensitive
measure of responses to either CO or alcohol at these levels.
Time Tests
The AOV F values in Table 4 show that neither time estimations
nor reaction time was affected by either CO or alcohol for either sound or
light stimuli.
Vigilance Performance
Interpretation of the data obtained from the series of vigilance
tests must take into account the adequacy of the test procedures followed.
Unfortunately, for the reasons given below, the circumstances under which
the vigilance data were collected create serious questions about the validity
of the data.
A vigilance task has usually been defined as one in which an individual
is required to respond to unpredictable signals in fairly isolated and non-eventful
surroundings over long periods of time. For this reason, group-testing
-------
-16-
procedures are not conveniently used in the study of vigilance. When
subjects work together in groups, especially over extended periods of time,
there is the distinct possibility of uncontrolled social interactions which
will distort performances. Furthermore, such distortions generally
will be in the direction of heightened efficiency, thus complicating efforts
to detect behavioral effects of CO or any other toxic substances.
The design of the present study was predicated on the assumption
that it would be possible to prohibit interactions among subjects during
prolonged group testing of vigilance performance so that valid data could
be obtained. Unfortunately, social interaction could not be prevented.
Despite efforts to enlist the cooperation of the subjects, including repeated
admonishments, a variety of, interactions were observed to occur among
the subjects including talking, singing, whistling, eye-to-eye contact
and even on one occasion, the playing of a game of chess. For these reasons,
the data presented below can hardly be viewed as adequately answering the
question of whether low levels of CO can influence vigilance performance.
Their main value is to indicate the inadequacy of group procedures as a
method for studying vigilance performance in reaction to CO.
Figure 10 shows the mean and individual vigilance scores of the
24-hour subjects as each concentration of CO during the first and last 4 hours
-------
-17-
of the Z4-hour sessions. The only apparent trend is a minor increase in
vigilance scores during the last four hours of the 70 ppm CO sessions. Com-
t
parisons of the vigilance performance of each subject under alcohol plus
CO (70 ppm) versus alcohol alone, shown in Figure 11, do not indicate
any trend for alcohol to potentiate the effects of CO. Of the eight subjects,
four showed increased and four showed decreased vigilance performance
under the CO-plus-alcohol condition.
Figure 12 summarizes performance under each treatment condition
of the experiment. As can be seen, there were no differences across CO
levels. Figure 12 does indicate substantial decreases in vigilance performance
under alcohol, an effect present in every subject. That alcohol will
result in disruptions of performance requiring sustained attention is a
well-established finding, and in the context of the present study shows
that the group testing procedure, despite its inadequacies, was not totally
insensitive to the effects of alcohol.
An important characteristic of vigilance performance is that it
deteriorates progressively over extended test sessions. Figure 13, which
presents scores during each of the four hours shows that this did not occur
in the present experiment. If there is any trend, it is that scores were
lowest during the second hour and gradually increased to a maximum during
the fourth hour. (In the case of the alcohol condition this pattern may
have been caused by the decrease in blood alcohol level with time. ) The
-------
-18-
general finding that vigilance performance did not deteriorate during the
tests bears out the aforementioned conclusions concerning the inadequate
sensitivity of the procedures. It is likely that interactions ensuing
between the subjects during testing permitted them to maintain a high level
of alertness during the 4-hour course of the tests. Any subtle behavioral
effects of CO on vigilance would be effectively masked under these conditions.
DISCUSSION
The purpose of these two studies was to test the hypotheses that a
24-hour exposure to low levels of CO would have deleterious effects on behavior
and that alcohol would potentiate such effects. The data obtained from tests
of coordination, arithmetic ability, inspection, manual dexterity, time
estimations and vigilance did not provide evidence to substantiate either
of these hypotheses.
The lack of CO effect was not unexpected as previous studies in this
(1)
laboratory for acute exposure have shown no effect of CO on time estimations,
(2)
arithmetic, inspection or coordination.
The experimental design did not include replicates of each exposure
combination and thus CO-alcohol interactions could not be calculated. From
the mean scores, however, there was no evident synergistic effect of the two.
Negative findings place a special burden on the researcher obtaining such
-------
-19-
findings. They may reflect the fact that the variable in question indeed
has no influence. But it also is possible that negative findings reflect
inadequacies of the experimental method, such as sample size, insensitive
procedures, and the lack of control of extraneous variables.
Investigators working in another laboratory have reported per-
(3,4)
formance decrements as a consequence of exposure to low levels of CO.
The present results, on the other hand, as well as those of several other
(1,5,6,7)
prior investigations have not indicated such effects. The issue is whether
these differences in findings should be interpreted to mean that CO has no
behavioral effects, or whether some procedures are insensitive to real
effects.
There are at least three aspects of the behavioral testing procedures
of this work which may have contributed to negative results. An obvious
factor concerns the procedure of testing groups of subjects rather than
individual subjects. It may be more than coincidence that Beard, who has
obtained positive results, has characteristically studied subjects one at
a time. If low concentrations of CO do have behavioral effects, those
effects seem most likely to appear under conditions of isolation. Testing
in groups undoubtedly created states of alertness and arousal which may
have effectively masked subtle cognitive and perceptual-motor effects
of CO. However, a test of that hypothesis conducted using individual
-------
-20-
subjects in an experimental design identical to Beard's showed no deleterious
(1)
effects of CO. Group interaction then does not appear to be an applicable
explanation in this case.
A second factor possibly contributing to error and insensitivity
involves social relationships which may develop between subjects and
the research staff. In some respects the procedures in this laboratory
are informal and may be characterized as treating subjects more as
collaborators in the research endeavor than as subjects. Even though
approach may not be inappropriate for strictly physiological or medical
work, it is a potential source of serious experimental error in behavioral
work. It is advisable, if not essential, to develop more formal procedures
which will reduce as much as possible social interactions between subjects
and the research staff. These procedures must begin at the point that
the subject is enlisted into the experiment, and should be followed strictly
at every point along the way. Only by imposing this sort of rigor will
it be possible to observe behavioral reactions to CO uncontaminated by other
variables.
Finally, and quite apart from the question of experimental control,
is the question of test sensitivity. There are no simple rules that can
be applied to determine which behavioral measure stands the best chance
of revealing changes in response to CO, and there 'was justification for using
the tests of the present experiment. There is reason to believe that highly
-------
-21-
skilled performances will be resistant to disruption by toxic agents, and
for this reason training before testing may have worked against the possibility
of detecting CO effects. In any case, to show subtle behavioral effects,
tests at least potentially sensitive to many variables must be used. The
present study showed no decrements on tests of inspection, arithmetic
ability, or time estimations under the influence of alcohol. A test which
is insensitive to alcohol could probably not be expected to be sensitive
to CO. However, these are the tests which other investigators have shown
to be affected by CO and, therefore, they were included in this research.
SUMMARY
The purpose of these two studies was to test the hypothesis that
a 24-hour exposure to low levels of CO would have deleterious effects on
behavior and that alcohol would potentiate such effects. The data obtained ^
from both studies for tests of coordination, arithmetic, inspection, manual
dexterity, time estimation, and vigilance did not provide evidence to substantiate
either of these hypotheses. The only positive result was alcohol adversely
affects eye-hand coordination.
ACKNOWLEDGEMENTS
The authors gratefully acknowledge the technical assistance of
S. Graff, K. Donohoo, D. Fleischfresser and K. Kujawski. They also
thank Miss Susan. Kamke for her help in preparation of this manuscript.
-------
-22-
REFERENCES
1. Stewart, R. D. , Newton, P.E., Hosko, M. J. and Peterson, J. E. ,
"The Effect of Carbon Monoxide on Time Perception", (a) CRC
Report No. CRC APRAC CAPM-3-68, MCOW-ENVM-CO-72-1,
(b) Arch. Environ. Health: 27:155-160, 1973.
2. Stewart, R.. D. , Newton, P. E. , "The Effect of Carbon Monoxide
on Eye-Hand Coordination, Manual Dexterity and Mathematical
Abilities", unpublished report.
3. Beard, R. R. , Wertheim, G.A. , "Behavioral Impairment Associated
with Small Doses of Carbon Monoxide", Am. J. Public Health 57:2012-2022,
1967.
4. Beard, R. R. , Wertheim, G. A. , "Behavioral Manifestations of
Carbon Monoxide Absorption", read before the 16th International
Congress on Occupational Health, Tokyo, 1969.
5. Stewart, R. D. , et al, "Experimental Human Exposure to Carbon
Monoxide", Arch. Environ. Health, 21:154-160, 1970.
6. O'Donnell, R. D. , et al, "Low Level Carbon Monoxide Exposure
and Human Psychomotor Performance", Toxicol. Appl. Pharmacol. ,
18:593-602, 1971.
7. O'Donnell, R. D. , Cherkos, P., Theodore, J. , "Effect of Carbon
Monoxide Exposure on Human Sleep and Psychomotor Performance",
J. Appl. Physiol. , 31:513-518, 1971.
-------
TABLE 1
SUBJECT BLOOD ALCOHOL LEVEL AND PERCENT CARBOXYHEMOGLOBIN SATURATION
IMMEDIATELY FOLLOWING TESTING SESSION - STUDY A
Exp. 1
Exp. 2
Exp. 3
Exp. 4
SUBJECT
R.S.
F.G.
D. T.
T.K.
.K.H.
T.M.
Blood
Alcohol
mg. %
--
--
--
--
--
--.
COHb
0.4
0.3
1.0
1.4
1.0
0.9
Blood
Alcohol
mg. %
--
--
--
--
--
COHb
15.9
14.5
N.A.
15.2
N.A.
14.9
i
Blood
Alcohol
mg. %
! 48.2
36.6
69.9
45.6
. <-5
46.8
COHb
0.6
0.7
1.6
0.6
0.5
0.5
Blood
Alcohol
mg. %
50.7
52.7
74.0
82.0
< .5
42.0
COHb
15.2
13.6
15.4
13.8
13.9
13.9
CO
-------
-24-
TABLE 2
ANALYSIS OF VARIANCE FOR THE EFFECT OF CARBON MONOXIDE
AND ETHYL ALCOHOL ON BEHAVIORAL TESTS - STUDY A
TEST EFFECT df F Ratio
Collar and Pin
Coordination
Arithmetic
Hand Steadiness
Significant at p <. 01
Significant at p < .05
Alcohol
CO
Alcohol
CO
Alcohol
CO
Alcohol
CO
1
1
1
1
1
1
1
1
15.75^
#:
121.00
*
60.49
1.49
1.67
0.69
1.47
0.01
-------
-25-
TABLE 3
SUBJECT BLOOD ALCOHOL LEVELS AND
PERCENT CARBOXYHEMOGLOBIN SATURATION IMMEDIATELY
PRIOR TO TESTING SESSION - STUDY B
SUBJECT
J.H.
J.R.
R.S.
R. C.
B. F.
E.G.
S.P.
J.W.
Exp.
Blood
Alcohol
mg. %
45.3
58.8
58.2
59.2
54.3
58.8
59.9
60.5
1
COHb
8.7
7.8
9.1
9.3
9.5
8.7
8.4
9.1
Exp.
Blood
Alcohol
mg. %
27.9
56.3
42.3
37.4
54.2
48.4
53.7
51.1
2
COHb
1.5
0.9
0.5
0.21
0.5
0.4
0.4
1.0
-------
-26-
TABLE 4
ANALYSIS OF VARIANCE FOR THE EFFECT OF CARBON MONOXIDE
AND ETHYL ALCOHOL ON BEHAVIORAL TESTS - STUDY B
TEST
Marquette:
Sound Stimulus:
Es timate /Stimulus
I Estimate-Stimulus
Reaction Time
Light Stimulus:
Estimate/Stimulus
I Estimate-Stimulus
Reaction Time
10 Second Estimations
30 Second Estimations
Arithmetic:
Coordination:
Inspection:
EFFECT
df
F RATIO
Alcohol
CO
Alcohol
CO
Alcohol
CO
Alcohol
CO
Alcohol
CO
Alcohol
CO
Alcohol
CO
Alcohol
CO
Alcohol
CO
Alcohol
CO
Alcohol
CO
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
3.23
0.36
4.25
0.34
0.04
0.13
15.98
' 35.96
26.10
31.31
1.39
1.00
0.58
0.00
0.01
0.70
0.72
0.01
718.67*
8.53
0.87
0.08
^Significant at p< .001
-------
70
si
I
40
O
Co
x
X
X
' i
IZII JIELJLi ."
70 pp
13 \^\ koue
S3
. i - i ; i < . \ < ' i . i i
, . . :__;- i i i
---- p- -rp --+-.
C-J
CM
O
CJ
rv '.o in
fT) C'l
CO
. 5TAAiT-> A«.T» -e(?(?r>D.
«\S
-------
-1-70
S"b
V _ -J..
-2.0
:
-~3o
1:40
-.!.:::.:::
d~n"
'I I""''"1 L_V
i S \
TT
C\J
CM
CM
o
PJ
. . | .
_ . l_
Cl
Ti
TT
n
Q
-70 f>p/m
S8
JO
15
. -.1 .L
I ..J.,.:T
~.:_p_l
"T~
o
co
co
^>\us CO
I
N5
00
*
-------
£ »
g ^o
9
° 10
3?
£
o
~io
-ao
.;:.<3i^ ....,S5!
* CO OJ « O
cu r»j CM r>i c\j
. . . . I . . . ..:..:..!:'.:
Oi K>
S"3*" S^
^
to m
i.
ro rg
JL
-si-
___i_..-.:._i :J_I__L_
tf! UT V >1 !N! «-«
FiG,3.
soe>jecT
eond»Vfo\o
FOR
-------
too
70
foe-
ppnm
CO
X
Co
X
x
x
p p
Q
YlOv*L
I
o
c_
s» sa. . 53
...Li
I
f-i Cl
i :. i'i ! j !"!'
Oi to r^ vo a-j
i !
i . i
1 : }__}
;X* r^ xi
ot CO
THE.
!) AS
-------
-.70
five
.: ..! ; ; j .!._:._.
7O
ALCOHOL
Cl
I
LO
MSI , :::
S5"
O JO
0
r^. \O to
SCOPES
n oj
O CO
(-70
AS
-------
u
Ho
-70
": : ::::':.:
' i
. i
: . .
" i i ; : :
'
i
' . 1 ' : :
-1 j
' 1
1
~ r
...
1 : .
: |...
. : ; :
|
. . j
! : : : 1
- : !
oOt>i£CV ^>
''".' i ! | '
-- ri:L_-- L-""L
-'I- .... , ;
'
i i i ' ' ! ' ; i
1-
T
r.
O
Ol
IT) IN
FiEei CENTAL
-------
Co
""" K
110
H
. _
..:..-!..
: rrH
i
I
!__._; I . .
I b:i :.{_.
! .1
CO
::::_: .1'
i
u>
i
i
''\"-''1
i T i ;
i
i i
,
i ' ' : !
. . : . | . : , .
; :.,. ,
...... . | . . ;, I . - - - i
' i ! i
! . i i
' 1 ' 1 I ........ 1 .
-. 'I :[-::L[_I"! :';i '
SI
o
en
i<&. "7,
Scopes e>\
O.O
to.
± 4^
-------
y
c/60
-/&>
^
~f\i'
:e/)-r/*»
^ --/&
^
x /^)
K
_vj'
. , -//o
^r^ '
! . r\*
-1 _ ' \^<"/00
'" rj-:::VA
^^;-:
-------
-------
ISo
V
I
ho
I
O
1 ' I .'
CO
..
1
:_ i_. 1 r:
1-
-51
-SSL ,
x
x
X*
X"
CO
i
LO
WEAN
S6
: i ' ; ;
S3
m
rj
r-j CM
O
C4
i _
r^.
%?
,, »o.
^4-KooQ
eAcu ceyjeL
CO
-------
rrHf^ :
&i
~IO
CO
l~^- L _T1
L".:mrz:T -_LJ!: mi .an.:'.
T~
-H-
'
\ ' '
If - !
C|
? \
1 ' 1
.'. I. '. .:
> < i
sa
.
< , i
..:<
,
S3
!
. . . . j . . . .
^ t }-***
^ vl K>s
'(.
::.:...:
*
i
u>
n c-j
CO
f
Aee
AS
-------
r_DPCTR
24--W.
.._ _,_.. ^_ rr:~'_ i
~~""~'"' ~ ~~ ~
-------
i
i
!
1
- - t
.
....
"~ .'. ."
i .
.
'
.
. , 1 '
III.
i
' '
1 '
! '
L' ' '
i ';
i ! : '.
; . 1 !
' i i
i ' ' '
. i .'
' ' '
! : : :
! i
!iii) i >
:;;(::
~\(P
..;.,..,..
1 ' . .
' '
; : ' i . ' :
I.]
!
j
' '-i. '.' * '
... ......... . ..........
;- ; | ' ' t "'"' '">' ; ' ' ' ' ' ' ' : "~
it 1 ' '
1 1
: . . . . - t j - - - ...... . . . - ' | -
- ! - ! - : .!-.-. .... | ...:.. . . . . j
I ! -! -! : = i . -1
01
' .
ef\cV\
d.
VO
I
------- |