PB82-247180
The University of Akron Study on Air
Pollution and Human Health Effects II
Effects on Acute Respiratory Illness
Akron Univ., OH
Prepared for
Health Effects Research Lab.
Research Triangle Park, NC
1981
U.S. Departaott of Commerce
Nstisnsl Technical Wormatkm Service
-------
t TECHNICAL REPORT DATA
' . • If lease rtad lnttruct:cni on the rcurte before completing! j
£ '- REPORT NO. 2-
! EPA-rtMn/l-M-fpn JOURNAL ARTICLE
tl. TITLE AND SUBTITLE
,' *
^ The Univeristy of Akron Study on Air Pollution and Human
;;i Health Effects II. Effects on Acute Respiratory Illness
: ,'. AUTHORIS)
R. A. Mostardi, N.R. Woebkenberg, D.L. Ely, M. Conlon,
G. Atwooa
'. PERFORMING ORGANIZATION NAME AND ADDRESS •
Oeparments of Biology and Chemical Engineering
The University of Akron
^ Akron1, Ohio 44325
; 12. SPONSORING AGENCY NAME ANO ADDRESS
jr: Office of Research and Development
1 Health Effects Research Laboratory
US Environmental Protection Acency
Research Triangle Park, NC 27711
3-f Pbti2-2l71ttO
9. REPORT DATE ' .
6. PERFORMING ORGANIZATION COOE
8. PERFORMING ORGANIZATION REPORT NO.
C9XA1C
11. CONTRACT/CiRANT NO.
Grant No: R804256
13. TYPE OF REPORT ANO PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA-600/11 -'.'•
IS. SUPPLEMENTARY NOTES
, •
'' Reference: Archieves of Environmental Health, 36(5) : 250-255, September/October 1981
6. AfiTRACT
The purpose of this study was to determine the effects of air pollution on aetite
respiratory illness (ARI). Levels of air pollutants were monitored on a daily 24-hour
basis at two schools in Akron, Ohio. The children at each school completed daily
diaries which served as a screening mechanism for detecting ARI. Once an ARI was
isolated, pulmonary function tests (PFT) were run during the symptomatic phase; once
the child became asymptomatic, tests were continued for 2 wk.
17. KEY WORDS ANO DOCUMENT ANALYSIS . - •'
t. ' DESCRIPTORS
•
8. DISTRIBUTION STATEMENT
Release to Public
b.lOENTIFIERS/OPEN ENDED TERMS
•
19. SECURITY CLASS (Ttiii Report)
Unclassified
20. .SECURITY CLASS I This panel
c. COSATI Field/Group
• . '
21. NO. OF PAGES
t
22. PRICE " 4
-------
EPA-600/J-82-020
JOURNAL ARTICLE
The University of Akron Study on Air Pollution
and Human Health Effects II.
Effects on Acute Respiratory Illness
RICHARD A. MOSTARDI, Ph.D.
NANCY R. WOEBKENBERG, M.S.
DANIEL L. ELY, Ph.D.
MARY CONLON, M.S.
Department of Biology
GLENN ATVVOOD. Ph.D.
Department of Chemical Engineering
The University of Akron
Akron, Ohio 44325
ABSTRACT. The purpose of this study was to determine
the effects of air pollution on acute respiratory illness (ARI).
Levels of air pollutants were monitored on a daily 24-hour
basis at two schools in Akron, Ohio. The children at each
school completed daily diaries which served as a screening
mechanism for detecting ARI. Once an ARI was isolated,
pulmonary function tests (PFT) were run during the symp-
tomatic phase; once the child became asymptomatic, tests
were continued for 2 wk.
The results of t!iis study indicate that SO2 and NO2
levels are higher vhat is usually
reported in the United States. Few studies, however, have
examincd'the prevalence and severity of acute respiratory
illness (ARI) and daily symptomology in an urban area
where air pollution is a potential health factor. Those stud-
ies that have been conducted have not carefully monitored
the air quality to which the population was exposed. For
this reason, as well as others, there is a paucity of data avail-
able regarding cause and effect of air pollutants on ARI.
in young children.
An earlier report5 established that the levels of sulfur
dioxide (SO2) and nitrogen dioxide (NO2) were significant-
ly higher in one school area located adjacent to industry
than another that was 4 km away. Sulfur dioxide appears
250
Archives of Environmental Health
-------
to be absorbed very rapidly in the upper respiratory tract,
while NO2 reaches the small peripheral airways where it
exerts its effect. When the annual arithmetic mean of these
gases exceeds 100 pg/m3, increased risk of acute and
chronic respiratory illness and possible changes in pulmon-
ary function6"8 may result. Even though SO2 is rapidly
absorbed in the membranes of the upper respiratory sys-
tem, there are still peripheral airways effects via the vagus.
therefore, it is concluded that both SO2 and NO2 can
affect the dynamics of the respiratory airways.
In 1973 Lebowitz* used multivariate statistical tech-
r iqucs to demonstrate air pollution's effect on human
rcalth. However, the relationships were causal, but defini-
tive cause and effect was not shown. Lcbowitz suggested
that an intensive research design would be required to com-
pletely answer the questions concerning air pollution and
effects on human health, which would include question-
naires, daily diaries, frequent PFT during illness episodes,
frequent personal communication with the volunteers
involved, and daily monitoring of air pollutants.
Goldsmith and Friberg10 have stated: "The control of
air pollution, and the extent that it is based upon health
effects, should be based on the most sensitive groups of
persons." Because school children appear to be very sensi-
tive and specifically reactive to the effects of air pollution,
this would be a most suitable group to study.
The authors, therefore, chose to study the cause and
effect relationship between gaseous air pollutants (i.e.,
SO; and NO2) and ARI in a stable population of children
who attended schools in Akron, Ohio, where air pollu-
tant levels were widely divergent
METHODS AND MATERIALS
Two schools locate'.) in Akron, Ohio were selected for
this study. The Seiber ing school is located adjacent to an
industrial complex and the Betty jane school is located 4
km east. These schools were selected on the basis of prelim-
inary observations and data from the Akron Region Air Pol-
lution Control Board (ARAPC). The highest levels of SO2
in Akron have been recorded in the Seiber ling area, making
it an excellent area for study.
Air pollution measurements. Methods of acrometric
measurements used during this study were identical to those
described earlier.5 Total suspended particulatcs (TSP) sus-
pended sulfatc, suspended nitrate, SO2, and NO2 were
measured daily.
Hca'th data. To determine if a child was contracting an
ARI, a daily dairy was initiated. The dairy, which was dis-
tributed each day during homeroom by the teacher, was a
simple chart; the child checked cither yes or no to cough,
runny nose, sore throat, chest congestion, cold, and eye irri-
tation. The diaries were then collected by the homeroom
teacher for use by the research staff.'Prior lo initiating the
study, all participating homeroom teachers were given a
description of the diary and how it was to be used.
On Monday, Wednesday, and Friday of each week, a
team of two technicians visited each school. Initially, the
diaries were examined for a potential ARI. If runny nose
and several other symptoms were checked by a student,
they were requested to leave class and go to the clinic,
where the student was given a verbal and visual examina-
tion by the two technicians to determine the nature and/or
severity of the ARI. If an ARI was not found, the symp-
toms were re-explained to the student and instructions were
provided regarding conditio'ts under which a given symp-
tom should be checked. The student then returned to class.
If the individual had an ARI, pulmonary function tests
were conducted. The techniques for conducting PFT and
measuring the spirograms included5: forced vital capacity
(FVC), forced expiratory volume at 1 sec (FEV, 0), and
maximal mid-expiratory flow (MMF).
To cross validate the assessment of ARl among the
school children, the technical staff exchanged schools and
evaluated the existing ARI in both symptomatic and
asymptomatic phases. Each symptom was graded from 0 to
4 with increasing severity, enabling oojective comparative
evaluations. Statistical comparisons were net made on the
graded symptoms, but in all cases there was excellent agree-
ment among the technicians concerning presence and xver-'
ity of individual symptoms.
During the ARI and recovery period (i.e., symptomatic
phases A and B, respectively) of an individual who had an
ARI, two PFT were conducted every other day until he or
she was completely asymptomatic. The individual was then
monitored for 2 wk during the asymptomatic phase, which
included PFT every other day. The entire process usually
included five different scries of PF T.
Accuracy of measurement was crucial to the outcome
of this study, therefore, brief comments on these tech-
niques follow. Ruled t; iangles were aligned with the hori-
zontal chart paper lines, which enabled near-perfect verti-
cal lines to be drawn. The l-?ec time line was measured,
drawn, and rechecked for accuracy; the 25% and 75% lines
for MMF were drawn, measured, and rechecked against the
calculated values; and all linos were drawn with very thin,
hard lead. These techniques provided a 99% level of accur-
ac/ for a given tcst/re-test calc:.'ation for FEVj 0 and FVC,
and 98% for MMF.
The techniques used to adjust data were similar to those
described previously.5 Briefly, age, height, and weight were
used as independent variables to adjust each pulmonary
function variable in a multiple linear regression model with
forward order of inclusion. This analysis compared all of a
given student's symptomatic and asymptomatic PFT vari-
able and effectively used each child as his o^n control.
Other statistical methods employed included the student's
t test.
RESULTS
The air pollutiut levels found in .he school areas in Akron
have been detailed earlier.5 Brie-fly during the school year,
mean levels of SO2 were 77.5 pg/r.3 and 21.4 ^g/m3 at
Seibcrlingand Betty Jane schools, respectively, and NO2
mean levels were 54.5 jug/m3 and 36.9Aig/m3, respectively.
These data confirm the d.ita from the ARAPC and verify
that there arc large differences in air quality to which the
children are being exposed. Figures 1 and 2 show all diary
data collected during the school year. E-ach week's data were
totaled for a given symptom at each school and expressed
as a percentage of the '->tal reporting students. Although
September/October 1981 [Vol. 36 (No. 5)|
251
-------
o COUCH
• RUNNY NOSE
o SORE THROAT
» CHEST CONGESTION
• EYE ISRITATION
• COLD
234
OCT
234
NOV
234
DEC
234
JAN
234
FEB
2 3 4 5
MARCH
234
APRIL
234
MAY
Fig. 1. Daily diary data for children attending Seiberling school.
40- .
» COUCH
• RUNNY NOSE
o' SORE THROAT
» CHEST CONGESTION
• EYE IRRITATION
« COLO
1234
OCT
234
NOV
1234
DEC
1234
JAN
2341
FEB
23451234
MARCH APRIL
234
MAY
Fig. 1. Daily diary data for children attending Betty )ane school.
these data were not analyzed statistically, empirical obser-
vation indicates that the Seiberling students reported any
daily symptom twice as often as the Betty Jane students.
Histograms summarizing the results of the PFT during
the symptomatic and asymptomatic phases of the ARI arc
presented in Figures 3-5. The data were obtained from sub--:
trading the specific symptomatic and asymptomatic PFT *
from the baseline values after adjustment for age, height,
and weight. Adjusted data, which provide examples of
adjusted means, as well as sex, and the sample sizes involved
in the ARI testing, are presented in Tables 1 and 2. The num-
ber of students participating in the symptomatic B phase
was less than the number who participated in the sympto-
matic A phase at the Betty Jane school because several
became asymptomatic after only 3 days. This might indicate
that the Bctty'Janc itudents.were contracting ARI but
that their symptoms were not as prolonged as those at the
Seiberling school. Another possibility could be that the sub-
jective evaluations by the technicians at the two schools var-
ied. This is unlikely, however, as ARI assessment by tcch-
252
Archives of Environmental Health
-------
ISO
1*0
*
D
s«ib«rhtt|
4l
*•-
*!,• I-S »ll»
Fig. 3. Comparative differences for forced vital capacity (FVC) dur-
ing symptomatic and asymptomatic phases of acute respiratory ill-
ness (mean i standard error).
190
ISO'
so
Sim
i
D<
sie»e
,« i-s At, in 4-6
Fig. 4. Comparative differences for forceJ expiratory volume at 1 sec
(FEVi 0) during symptomatic and asymptomatic phases of acute res-
piratory illneo (mean t standard error).
Table 1.-Acute Respiratory Illness and Recovery Period PFT Results for Children
Reporting at Betty Jane School
Baseline
ARI
Symptomatic A
phase
FVC (L)
FEVj.o/FVC
MMF (L/sec)
Symptomatic B
phase
FVC (L)
FEVj.ofL)
FEVj.o/FVC
MMF (L/sec)
2.66 1 .03*
2.16 t. 03
0.81 t .01
2.39 ± .09
2.69 1.04
2.171.04
0.81 ± .01
2.34 ±.11
2.531.03
2.04 i .03
0.81 t .01
2.32 i .09
2.55 1.04
2.07 1.04
0.81 i .01
2.27 1.11
.131.02 46
.121.02 46
.00 1.01 46
.07 1.04 46
.141.024 33
.101.020 33
.00 i .00$ 33
.01
.01
ns
ns
.01
.01
ns
.071.047 33 ns.
NOTE: The symptomatic A phase occurred during acute respiratory illness period;
the symptomatic B phase occurred during the recovery period. For complete
description, see METHODS.
•Values expressed as mean i standard error.
nical staff was cross validated (sec METHODS).
DISCUSSION
During this study, children suffering from chronic med-
ical symptoms were eliminated from the study, as well as
children with physician-confirmed asthma or other chronic
respiratory illnesses. The lung function determinations con-
ducted during this study were done with numerous inter-
nal quality controls. The students had participated in the
study during 2 consecutive years, and were thoroughly
familiar with thv .-quipment, technicians, and the FVC man-
euver. At this time in the study, nearly all the children
were efficient at the FVC maneuver and 3 or 4 efforts were
very reproducible. The technicians were per-vunally familiar
with each of the students and were able to accurately assess
the magnitude of each symptom, and also determine
between single symptom allergic reactions and a true ARI.
This work was to have been a 2-yr study of ARI, but dur-
ing the first 5 months (i.e., January-May), the study team
was not able to resolve all the technical problems and pro-
duce reliable data. Consequently, data collection during
this half year established good ARI surveillance techniques
and eliminated other problems.-
The effects of siblings were carefully considered during
the study. At the Scibcrling school there were 9 siblings in
the study group. During the 1977-1978 school year, there
was only one instance when both had ARI. There were
September/October 1981 (Vol. 36 (No. S)|
253
-------
j »•»» r •»•••
U.,>„!,.!
*»• A Sf*i • *•!• I-} Atrm 4-6 At,in r-9
Fig. 5. Comparative differences for maximal mid-expiratory flow
(MMF) during symptomatic and asymptomatic phases of acute
respiratory illness.
five siblings at the Betty jane school; during this same
school year there was not any case when both were suffer-
ing from ARI.
Absenteeism was carefully monitored by examining the
daily attendance records for both schools. The fifth and
sixth grade populations were approximately the same size
and weekly attendance records were similar for both
schools. Occasional spikes in absenteeism were seen at both
schools, but did not occur at the same time, nor were the
spikes related to increased incidence of reported ARI.
The data indicate several interesting observations. (1) The
students at Seibcrling school are reporting more daily symp-
toms than those at Betty Jane. This is also exemplified
because there arc more ARI being reported at the Seibcrling
school. (2) The PFT data recorded during the symptomatic
and asymptomatic phases of an ARI show more airway or
airflow obstruction in the Seiberling school, i.e., the PFT
variables thought to be most sensitive to airway obstruction
arc describing the greatest difference, namely FEVJ o and
MMF. (3) There is more severe airway obstruction in the
Seiberling children and the involved airways require a longer
time for recovery. We interpret this to indicate ihat the air-
ways are affected by the ARI and then further compromised
by the effects of SO2 and NO2. In effect, one condition was
superimposed upon the other.
The symptoms that were reported most frequently by
the students were prmarily upper respiratory in nature.
Therefore, the question arises: "Are such upper respiratory
symptoms capable of producing changes in peripheral air-
ways?"
Several studies have shown spiromctric changes assoc-
iated with upper respiratory infection indicating peripheral
airway involvement. In some cases this was a spontaneous
disease process,' '"'3 but in others the disease was initiated
by an influenza vaccine.-14'ls
Having established that upper respiratory illness can pro-
duce spirometric changes, it must be questioned why the
Seibcrling children have a greater reduction in FEV1O and
MMF than the Betty Jane children. The populations were
carefully matched with respect to sociocconomic status,
race, and smoking habits, thus ruling out or minimizing such
effects. That the levels of SO2 and NO2 are significantly dif-
ferent between the schools could be important factors. We
maintain that SO2 and NO2 are exerting subtle effects on
Table 2.-Acute Respiratory Illness and Recovery Period PFT Results for Children
Reporting at Seiberling School /
Baseline
ARI
Symptomatic A
phase
FVC(L)
FEVi.o(L)
FEV,.o/FVC
MMF (L/sec)
Symptomatic B
phase
FVC (L)
FEV1-0(L)
2.52 ±.036* 2.40 t.037
2.0S ± .043 1.91 ± .042
0.81 ± .009 0.80 ± .009
2.39 ±.068 2.16 ±.076
.12 ±.012 76 .01
.14 ±.020 76 .01
.01 ± .006 76 .10
.23 ± .048 76 .01
MMF (L/sec)
2.52 ± .038
2.04 ± .046
0.81 ±.009
2.40 ± .072
2.39 ± .039
1.88 ±.045
0.79 ±.011
2.10 ±.072
.13±.016
.16 ±.021
.02 ± .008
70
70
70
.30 ± .040 70
.01
.01
.05
.01
NOTE: The symptomatic A phase occurred during acute respiratory illness period;
the symptomatic B phase occurred during the recovery period. For complete
description, see METHODS.
•Values expressed as mean ± standard error.
254
Archives of Environmental Health
-------
the respiratory symptoms of these children, which manifest
themselves in baseline PFT data, questionnaire response
data, daily diaries, and ARI.
In other studies the supcrimposition of ARI on existing
small airway diseases which were attributed to smoking16
and chronic obstructive respiratory disease17 have pro-
duced more pronounced lung function changes. To our
knowledge, the supcrimposition of irritating atmospheric
gases, such as SO2 and NOz, cither singly or in combination
with the ARI, has not been demonstrated.
All PFT variables were affected by the ARI. In the clean
area, reduction in FVC and FEVj 0 were the best indicators
of dysfunction, while in the polluted area FEVt 0 and MMF
were more sensitive. At the Seiberling school the differ-
ences from baseline for MMF were extremely large when
Compared to Betty jane. Hogg et al.17 suggest that the
small caliber of peripheral airways in children may enhance
changes in PFT when the children have an ARI, or as in this
study, exposed to irritating airborne gases. Both of the
above factors would seem to affect the more peripheral air-
ways which would be most sensitive to changes in MMF.
In conclusion, it was thought that the degree of data
reliability would be excellent within the framework of this
study design, where repeated lung function tests were con-
ducted during a 9-month period of time, as well as daily
symptoms and acromctrics. Within this study, the frequency
of daily symptoms, as well as syndromes reported by the
children, were higher at the polluted school. Repeated lung
function tests, which were carefully and systematically
carried out during the 9-month school year, further
enhanced what we perceive is an air pollution effect, by
characterizing reduced airway flow values at the polluted
school. These data provide evidence that gaseous air pollu-
tants may exacerbate ARi in fifth and sixth grade children
at the Seiberling schook
Follow-up studies of these children are currently under-
way to determine long-term consequences on lung growth
of these levels of SO2 and NO2. Many students will also
begin smoking; the relative contribution of this will be
studied with respect to accumulated effect from the pollu-
tant gases. Other factors, e.g., family history of respiratory
disease and a prospective history of respiratory diseases,
need to be considered as factors affecting the development
of chronic obstructive pulmonary disease.
The authors wish to thank the Akron Board of Education, espec-
ially Mrs. Dann and Mrs. Schole, principals of Betty lane arid Seiber-
ling schools. The parents and participating students were most patient
and helpful during the 2 years of this work.
Supported by U.S. EPA Grants 804256-01,02.
Submitted for publication February 16,1981; revised;
accepted for publication |unc 1, 1981.
Requests for reprints should be sent to: uichard A. Mostardi,
Department of Biology, The University of Akron, Akron, OH 44325.
REFERENCES
1. Douglas, I.W.B., and Waller, R.E. 1966. Air pollution and respir-
atory infection in children. Br / Prev Sac Med 20:1 -8.
2. Lunn, I.E.; Knowclden, |.;and Handyside, A.J. 1967. fv.ltcrns
of respiratory illness in Sheffield infant school children. Br I Prcv
Med 21:7-16.
3. Lunn, I.E.; Knoweldcn, (.; and Roe, J.W. 1970. Patterns of res-
piratory illness in Sheffield junior school children,A follow-up
study. Br I PrevSoc Med 24:223-28.
4. Coilcy, J.R.T., and Reid, D.D. 1970. Urt jn and social origin of
childhood bronchitis in England and Wales. Br Med I 2:213-17.
5.' Mostardi, R.A.; Ely, D.L.; Wocbkcnbjrg, N.R.; Richardson. B.;
and (arrctt, M. 1981. The University of Akron study on air pol-
lution and human health effects. I. Methodology, baseline data,
and acromctrics. Arch Environ r ealth 36: 243-49.
6. World Health Organization.. 19" 2. Air Quality Criteria and Guides
for Urban Air Pollutants. Report or a WHO expert committee,
WHO Technical Report Serir; No. 506 (Geneva).
7. United States Environmental Protection Agency. 1971./1/V Qual-
ity Criteria lor Nitrogen Oiidcs, Air Pollution Control Office
Publication No. AP-84. Washington, O.C.: U. S. Government
Printing Office.
8. U. S. Department of Health, Education, and Welfare. 1969./1/V
Quality for Sulfur Ox,dcs. National Air Pollution Control Publi-
cation No. AP-50. Washington, D.C.: U.S. Government Printing
Office.
9. Lebowitz.M.D. 19/2. A comparison of the relationships of mor-
tality and morbidity with air pollution-weather and the impli-
cations for further research. Sci Total Environ 2:191-95.
10. Goldsmith, |.R.,and Friberg, L.T. 7977. Effects of Air Pollution
on Human Health in Air Pollution, A. C. Stern, ed., Vol. 2,
pp. 458-610. New York: Academic Press.
11. Collier, A.M.; Pimmcl, R.L.; Hassclblad. V.; Clyde, W.A., |r.;
Knelson, J.H.; and Brooks, J.G. 1978. Spiromctric changes in
normal children with upper respiratory infections. Am Rev
KespirDii 117:47-53.
12. Cate, T.R.; Roberts, |.S.; Russ, M.A.;and Pierce, J.A. 1973.
Effects of common colds on pulmonary function. Am Rev
RespirDis 108:858-65:
13. Hall, W.J.; Douglas, R.G., |r.; Hyde, R.W.; Roth, F.N.; Cross,
A.S.;and Spccrs, D.W. 1976. Pulmonary mechanics after uncom-
plicated influenza A infection. Am Rev Respir Dis 113:141-47.
14. Rosenzweig, D.Y.; Dwycr, D.|.; Fcrstcnfcld, I.E.;and Rytcl,
M.W. 1975. Changes in small airway function after live atten-
uated vaccination. Am Rev Respir Dis 111:399-404.
15. Zeck, R.; Solliday, N.;' Kehoc, T.; and Berlin, B. 1976. Respir-
atory effects of live influenza virus vaccine: Healthy older sub-
jects aid patients with chronic respiratory disease. Am Rev
Respir Dis 114:1061-68.
16. Fridy, W.W., )r.; Ingrarr, R.H., |r.; Hicrholzcr, J.C.; and Cole-
man, M.T. 1974. Airwj/s function during mild viral respiratory
disease: The effects of fhinovirus infection in cigarette smokers.
Ann Intern Med 80:": 0-55.
17. Hogg, |.C.; Williams, |.; Richardson, J.G.; MacKlcm, P.T.; and
Thurlbcck, W.M. 1970. Age as a factor in the distribution of
lower-airway conductance and in (he pathologic anatomy of
obstructive lung disease. N Engl I Med 282:1283-87.
September/October 1981 (Vol. 36 (No. 5)|
255
-------
------- |