EPA 560/6-76-003
              EPIDEMIOLOGY STUDIES

                     TASK I - PHASE I
          PILOT STUDY OF CANCER MORTALITY
   NEAR AN ARSENICAL PESTICIDE PLANT IN BALTIMORE
                        MAY 1976
                       RNAL REPORT
                                 \
               U.S. ENVIRONMENTAL PROTECTION AGENCY
                    Office of Toxic Substances
                     4th and M Streets, S.W.
                     Washington, D.C. 20460

-------
EPA 560/6-76-003
              EPIDEMIOLOGY STUDIES

                     TASK I • PHASE I
          PILOT STUDY OF CANCER MORTALITY
   NEAR AN ARSENICAL PESTICIDE PLANT IN BALTIMORE
                        MAY 1976
                       FINAL REPORT
                U.S. ENVIRONMENTAL PROTECTION AGENCY
                     Office of Toxic Substances
                     4th and M Streets, S.W.
                     Washington, D.C.  20460

-------
EPA 560/6-76-003
                    EPIDEMIOLOGY STUDIES

                      TASK I - PHASE I

              PILOT STUDY OF CANCER MORTALITY
      NEAR AN ARSENICAL PESTICIDE PLANT IN BALTIMORE
                             BY
            Genevieve Matanoski, M.D., Dr. P. H.
                   Emanuel Landau, Ph. D.
                  Elizabeth Elliott, B. A.
                 EPA Contract No. 68-01-2490
       EPA Project Officer:  Robert J. Carton, Ph. D.
                             For
            U. S. Environmental Protection Agency
                  Office of Toxic Substances
                   4th and M Streets, S. W.
                   Washington, D. C.  20460

                          May 1976

-------
                           NOTICE
     This report has been reviewed by the Office of Toxic
Substances, Environmental Protection Agency, and approved
for publication.  Approval does not signify that the contents
necessarily reflect the views and policies of the Environmental
Protection Agency.  Mention of tradenames or commercial products
is for purposes of clarity only and does not constitute endorse-
ment or recommendation for use.

-------
                           Acknowledgements
     The authors wish to thank Drs. Baetjer, Levin, and Lilienfeld
for their suggestions and support as well as for the provision of
relevant information which permitted the completion of certain
portions of this study.  They would also like to thank Dr. C. W. Kruse'
for providing information on local wind directions.
                                   111

-------
          The measurement of  long-term effects from  low-level exposure  to
 carcinogens  in air has been a difficult problem because of  the presence of
 multiple  agents,  the changing population bases exposed, and the involvement
 of  personal  risk  factors during  the  long latent period before cancer  is
 manifest-.  Possibly because of these confounding variables,  the studies re-
 lating  levels of  air pollution to localized cancer mortality by census  tracts
 have been few in  number and frequently unrewarding.  Winkelstein and  co-
 workers (1)  in Buffalo and Zeidberg, Horton and Landau (2)  in Nashville found
 no  correlation between the level of pollution and the risk  of lung cancer.
 However,  Menck, Casagrande and Henderson (3) did find higher rates in contamin-
 ated areas of Los Angeles which  they associated with the presence of  benzo(a)pyrene
 in  air.

          Studies aimed specifically at determining  the health effects of arsenic
 in  the general environment, especially around smelters, are also limited.  Milham
 and Strong (4) have demonstrated excessive levels of arsenic in urine and hair
 of children exposed to the emissions from a copper smelter but no detrimental
 health effects were noted.  Blot and Fraumeni (5) have reported excess lung
 cancer mortality  for both males and females in counties with copper,  lead or
 zinc smelting industries.  The latter data agree with the reported risks of
 lung cancer  in populations occupationally exposed to arsenic (6, 7).  A direct
 association  of cancer and environmental arsenic has not been demonstrated.

          The current pilot project tests the feasibility of determining the
 carcinogenic effects of arsenic by examining the mortality  of populations
 living near  an insecticide-producing industry.  Cancer death rates in
 populations  from  census tracts near the plant are compared  to all the tracts
 in  the city which match on several variables.  The advantage of using this
matching  study design is that data from other projects with similar designs
might be  combined despite differences in the characteristics of the populations
 exposed to the arsenic in various cities.

Background

          For many years a chemical .plant in south Baltimore produced arsenicals
as well as other  chemicals for agricultural use.  Arsenic acid was manufactured
in  the early 1900*s but this operation was discontinued over 20 years ago and
 the plant reconstructed in 1952.  The insecticide portion of the plant produced
several arsenates but these chemical procedures have also been terminated.
Lead arsenate production was discontinued in 1967, calcium  arsenate in 1973 and
sodium arsenate in 1974.   The plant continues to package many of these arsen-
 icals for distribution.   The  insecticide plant has also produced chlorinated
hydrocarbons and organic phosphates.

          The area around the plant is heavily industrialized.  Adjacent to the
chemical plant was a Gas and Electric Company operation.   Rubber production also
took place nearby.  The significance of these possible sources of contamination
must be investigated.

-------
 Methods

           Index census  tracts with possible environmental exposure  to arsenic
 were defined  empirically  as  those tracts  for which at  least  fifty percent  of
 the area was  included within a  3/4 mile radius of the  plant  (Figure 1).  This
 distance was  selected because any smaller radii would  have included only the
 single  tract  in which the plant was  located and larger radii would  have
 included tracts across  the river in  an area with heavy industry and few
 private  dwellings within  one mile of the  chemical plant.  Four census tracts
 which fell within these boundaries are 2303, 2302, 2404 and  2301.   The
 populations of  the  first  three  tracts were similar in  demographic and socio-
 economic characteristics  (See tables 1 and 2, Figure 2).  The 1970  census
 data indicated  that the  residents were an older, stable, white population
 with median incomes from  $8400-9200  per annum.  The population of tract  2301
 had a lower median  income, a higher  proportion of families at poverty level,
 and a higher  percentage of blacks than the other three tracts.  Separate
 matching control tracts were selected for the first three index tracts, Match
 I,  and  for 2301, Match  II.   All census tracts in Baltimore City which matched
 the index  tracts on six variables were chosen as controls.   The variables  and
 the range  of  differences  which  were  acceptable for matching  were:

            Age distribution      — 10% for each age group
            Race                  _ 15% expressed as % black
            Sex                   — 5%  expressed as  % male
            Median  income         ± $1000
            % below poverty  level  ± 10%
            % head  of household   ± 20%
              over  65 years

            Several of  the matching  tracts were adjacent to  the index tracts and
 have been  analyzed  separately in the event that the population of those tracts
 may also have had minimal exposure.   One  possible control tract was  dropped
 because  of an excess  of white females in  the older age group resulting from
 the presence of a nursing home  in that tract.

            For some  of the  analyses, the control tracts have been  subdivided
 into groups according to  geographic  location since the populations  in various
 areas of -the city differ  in  their exposure to heavy industry or in mobility.
 The  southern  tracts have  several areas of industrialization.  The central
 area includes inner city  tracts with  both problems of  industrial exposure
 and  mobility.   The northern  tracts are found in primarily residential areas.
No Match II controls were found in the south.

            All death certificates in Baltimore City for the three years 1970-
 1972 have been  examined for cancer listed anywhere among the causes  of death.
All  cancers have been classified according to the 1965 ISCD  code.  If no
cancer code appeared on the certificate,  the' cancer was classified by the
 staff; otherwise the coding as recorded by city vital  statistics personnel
was  accepted.

-------
            The occupation and place of employment as recorded on the certificate
were  listed in order to detect deaths of chemical plant  workers.  Such infor-
mation iJ.ght also identify other dangerous occupations.

            The census tract codes noted on the certificates were used for
classifying residence.  The validity of the codes was checked by selecting a
seven and  ten percent systematic sample of the cancer deaths in the years 1971
and 1972 respectively.  The census tracts incorrectly coded were eight percent
in the first sample and fifteen percent in the second sample.  To avoid
erroneously adding cancer deaths to index tract mortality from misclassification,
the residence was listed for all cancer deaths in the index tracts and recoded
correctly.  This procedure allowed elimination of deaths from the index tracts
but no additions from misclassifications in other tracts, thus providing a
minimum estimate of cancer mortality in these tracts.

Results

            The crude cancer death rates for selected sites and total cancers
in each sex are presented in tables 3 and  4 for the combined tracts.  The
data-suggest that the risk of lung cancer in males is 1.4 times higher in
the index  as compared to the non-adjacent control tracts.  The risk in the
mixed racial index tract is even higher in comparison to its matched controls.
The lung cancer mortality for females is not higher in the index as compared
to the adjacent control tracts.

            The rates in these tables include cancer as listed anywhere on the
death certificate but the relationship is not changed if only underlying cause
is used.  The inclusion of lung cancers classified as secondary or unspecified
does  not change these observations.

            Although the tracts have been matched by age, race and sex so that
crude rates should be comparable in the populations, the sex-specific rates
of the white population have been adjusted for age in tables  5  and 6  in
order to determine whether this variable had any influence on lung cancer
mortality.  The 1970 Baltimore City population has been used as the standard
for a direct adjustment.  In these tables, the matching tracts have been
subdivided by their location in northern, central and southern geographic
areas in order to detect any possible internal differences in controls.
None  of these adjustments have altered the conclusions based on the crude
rates although the magnitude of the excess risk of lung cancer mortality
varies depending on which geographic area control is used.

            Race-specific rates adjusted for age and sex in tables 7 and 3
indicate that the blacks in the one index tract 2301 may not have a higher
rate  than inner city populations after adjustment although the white population
still has a higher rate.  The number of deaths in blacks is small.

            The crude death rates for individual index tracts are presented in
tables 9   and  10 •   The data in table  9  indicate that the differences
between lung cancer death rates in white males in tract 2303 where the plant
is located and the northern and southern controls are significant for both
comparisons with probabilities of .0002 and .0003 respectively (b, y).  The

-------
relative risk weighted by age of lung cancer deaths in males in this tract
is  3.3  and 4.0  times higher than in the southern and northern controls
respectively.   Although  the relative risk of lung cancer is higher in males
in  tracts 2301  and 2302, suggesting a possible gradient in death rates for
this  disease, the differences between index and control rates are significant
for 2301 only.  The death rates for several other cancer sites (oral, stomach,
pancreas, and prostate)  appear high in one or more of the census tracts near
the plant but the numbers of deaths are so small that no importance can be
attached to  the observation.

             It  was suspected that employees of the plant who lived in the local
area  might be contributing to the excess mortality in the index tracts.   An
examination  of  the industries of employment recorded on the death certificates
revealed one man with lung cancer who worked in a chemical plant.  Employment
records from the insecticide plant were checked to determine whether any of
the cancer deaths had occurred in workers.  The same case of lung cancer noted
above and a  prostatic cancer death were matched to plant employees.  Removing
the case of  lung cancer  reduces the rate for tract 2303 from 395 to 355 per
100,000 white males.  The significant difference in lung cancer deaths in
tract 2303 as compared to controls remains.

             The occupation as listed on the death certificate indicated no
other work exposure which might explain the excess deaths.  It is recognized
how inaccurate  such data are in determining an individual1s occupational history.

             Out of the 96 cancer deaths in the index tracts 28 had no reported
occupation and  an additional 19 were listed as housewives.  Among the remaining
49  cancers,  5 occurred in laborers, 4 in machinists, 4 in truck drivers, 3 in
salesmen with all other occupations being noted less frequently.  The industries
recorded on  the death certificates also did not indicate any remarkable excesses.
Fifty cancer deaths had no industry noted and the 19 housewives are included
here.  The only industry noted three times was "shipbuilding and repair" which
may relate to the presence of a large shipyard in the south-eastern corner of
the city.

             The data were suggestive of a gradient in lung cancer mortality
in  a  northerly  direction with intermediate rates in index tracts 2302 and 2301
as  compared  to  the tract with the chemical plant and lower rates in the
adjacent control tracts.  A spot map of the residences of individuals dying
of  lung and  other cancers during the three year period  1970-72  are noted in
Figure 3.  These data also suggest a clustering of lung cancer deaths.  An
examination  of  the direction of wind currents in Baltimore   indicate the winds
with highest velocity flow from the wast and northwest 32 percent of the time
in  the winter.  These winds should dump most of the contamination into relatively
unpopulated  areas.  The south and southwest winds which could be carrying
contamination into the northern tracts occur 22 percent of the time in the
summer and only 13 percent of the time in the winter and .have a lower velocity.
Figure 4 demonstrates wind roses for Dundalk Terminal four decades ago; they
are not believed to have changed significantly since that time.

-------
Discussion

            A matched control design similar to that utilized in the present
pilot study might be effective in determining th« cancer risks associated with
environmental pollution.  Frequently the population exposed to any single
source of pollution nay be too small and the cancers too infrequent to be
sure ȣ significant results.  By using the matched census tract design to
study several exposure sites even in different cities, one may be able to
add the data or at least find consistent results which will support an
association between a specific contaminant and a specific cancer even though
there are basic differences in the characteristics of the populations exposed
at each of the various sites.

            In the current pilot project there is a significant increase in
lung cancer mortality in males in a census tract adjoining an industrialized
area containing an arsenical-producing chemical plant.  The data also suggest
a decreasing northerly gradient in rates for this cancer which could be related
to distance from the plant.  Although cancer rates for several other sites
may appear high, the numbers are too small to attach any  weight to the
observation.

            The lack of any excess lung cancer mortality in females raises
questions about whether the differences should be attributed to environmental
or occupational exposure.  An examination of plant records indicates that
employment in the chemical plant cannot explain the male-female discrepancy
or the excess lung cancer mortality.  From death certificate data no other
industrial exposure predominated in these men but this occupational informa-
tion is incomplete and unreliable.  It is possible, however, that environmental
exposure to agents such as arsenic may require the synergistic action of
cigarette smoking before lung cancers are manifest.  This hypothesis would
be compatible with the data on occupational lung cancer in asbestos and uranium
mine workers who experienced an excess of lung cancers only in association
with cigarette smoking.  Exposure to arsenic in females may not produce lung
cancer because of infrequent cigarette smoking in the past, although it may
produce cancer at a site other than the lung.  Moss (10) reported that female
woolen textile workers had an excess risk of oral cancers which he attributed
to arsenic exposure from sheep-dip; he did not report an excess risk of lung
cancers.  Thus, health effects from environmental exposure to an agent may
be different depending on sex, age, and simultaneous exposure to other personal
risk factors.

            The data from this study are still preliminary.  The information
should be expanded with deaths from additional years, validation of pathology
records and further information on the personal characteristics of the deceased.
Sampling of soil and dust in-the area to detect the presence of arsenic or
other contaminants should be accomplished.

-------
Summary

        The present study was a preliminary attempt to determine the feasibility
of using cancer mortality data to indicate the carcinogenic effects of possible
arsenic exposure in the environment.  The four census tracts for which fifty
percent of the areas lay within a 3/4 mile radius of an arsenical-producing
insecticide plant  were matched by such demographic characteristics as age, race,
sex, and socioeconomic status with all other similar tracts in Baltimore City.
The matching tracts were then subdivided according to suspected differences
in exposure to environmental pollution.  The lung cancer mortality was higher
for the combined tracts adjacent to the plant as compared to the control tracts.
The lung cancer mortality for the tract in which the plant was located was three
to four times higher than in controls tracts, a difference which was significant.
The data also support a northerly gradient in rates with increasing distance
from the plant.

        The difference in lung cancer mortality which was present only in males
could not be explained by employment in the insecticide plant or in other high
risk occupations.  It is possible that differences in smoking habits in pre-
vious time periods might explain the discrepancy in rates by sex.  Additional
deaths as well as further information on occupation and personal habits are
needed to validate the findings.

-------
               Figure 1



Map of area surrounding chemical plant

-------

                     1306.02
                        X**"*.
                        •   •
1203  *'v
                                   03  '--g;.'

                                   %:  .•**"
                                    ?•:•:•: '•'•**
            2503.03
INDEX TRACTS

CONTROL TRACTS ••

MATCH  II  £:v:  .




PLANT  LOCATION *
                                            PATAPSCO  RIVER
^12504.0!  ':">*
  •*d     • • • • ^ •
                                 Figure 2               ^


                                                             ^--ts
     Map of Baltimore City showing location of index and control census

     tracts.                          0

                                     o

-------
                                                   Lung
                                                   Other
                                                   Census
                 cancer death
                  cancer death
                   tract boundary

N
           Figure 3

Spot map showing cancer deaths
for 1970-72 by residence at
death, excluding chemical plant
employees.

-------
                                           FIGURE 4
  NW
w
     SW
                                      NW
                                  W
              ANNUAL
             LEGEND
tZZZdVELOCITIES 4  THRU  15  MPH
••^VELOCITIES 16 MPH  AND OVER
 VELOCITIES  0 THRU  3  MPH RECORDED AS CALM.
            NUMBERS ARE  PER CENT OF
 TIME WINDS  BLEW FROM DIRECTION INDICATED

 NOTE: WIND ROSES  BASED ON ANALYSIS OF MUNICIPAL
 AIRPORT WBO HOURLY WIND RECORDS (1933 THRU 1938)
                                                                        NW

                                                                               SW
                                                                                    WINTER
                                                                            PER CENT OF TIME
                                                                            5   0   5  10
                                                                                WIND DATA
                                                                              BALTIMORE, MD
                                                                     Dace: March  1950     Drawn by:  THG

-------
                          Table 1.  Designation of census tracts


Match I





Match II

Index
Tracts
2302
2303'
2404



2301

Adjacent
Controls
2401
2402
2403



2101
2201
South
Controls
2005
2502.05
2503.03
2504.01
2504.02
2505


Central
Controls
101
104
105
601
602
2608
703

North
Controls
1308.02
2604.01
2702



905
1203
•Plant located in this tract.

-------
                        Table 2.  Characteristics of Index and control census tracts from 1970 census for selected Baltimore areas.




                                                                                   TRACTS
Characteristics
Total Population
Range between tracts
% 
-------
                        Table .3.  Crude death rates for specific types of cancer for total population
                            In Index tracts and matched controls for both Match I and Match II,
                                          average rates per 100,000 per year,
                                               total deaths for 1970-72,
                                                     males only.
                                                  Match I
Match II
Type of cancer and
8th revision (SCO code
Oral
(140-149)
Stomach
(151)
Colon
(153)
Rectum
(154)
Pancreas
(157)
Lung
(162)
Prostate
(185)
Bladder
(188)
Lymphomas
(200-203)
Leukemia
(204-207)
All cancer
(140-209)
#
rate
#
rate
#
rate
#
rate
#
rate
#
rate
#
rate
#
rate
#
rate
#
rate
#
rate
Index
2
15.46
1
7.73
2
15.46
2
15.46
2
15.46
19 h
146.88
3
23.19
0
-
1
7.73
0
-
38
293.75
Adjacent
Controls
3
21.99
1
7.33
3
21.99
3
21.99
0
-
10
73.31
3
21.99
1
7.33
0
0
-
40
293.23
Non-adjacent
Controls
6
6.47
13
14.02
15
16.18
7
7.55
10
10.79
97
104.61
13
14.02
9
9.71
11
11.86
6
6.47
241
259.92
Index
1
19.11
0
-
0
-
0
-
1
19.11
11 a
210.24
3
57.34
0
-
1
19.11
1
19.11
24 a
458.72
Adjacent
Controls
2
23.68
1
11.84
3
35.52
0
-
1
11.84
10
118.41
3
35.52
0
-
2
23.68
0
-
31
367.08
Non-adjacent
Controls
0
-
1-
6.20
4
24.80
0
-
0
-
12 a
74.40
3
18.60
1
6.20
0
3
18.60
40 a
248.02
a Index significantly different from non-adjacent controls by Woolf-Haldane method (8, 9).
b Index not significantly different from non-adjacent controls.
                                                       13

-------
Table 4.  Crude death rates for specific types of cancer for total population
    In Index tracts and matched controls for both Match I and Match II,
                  average rates per 100,000 per year,
                       total deaths for 1970-72,
                            females only.
                                Match I
Match II
Type of cancer and
8th revision ISCD code
Oral
(140-149)
• Stomach
(151)
Colon
(153)
Rectum
(154)
Pancreas
(157)
Lung
(162)
Breast
(174)
Cervix
(180)
Bladder
(188)
Lymphomas
(200-203)
Leukemia
(204-207)
All Cancer
(140-209)
#
rate
#
rate
#
rate
#
rate
#
rate
#
rate
#
rate
#
rate
#
rate
#
rate
#
rate
#
rate
Index
2
14.87
0
1
7.44
0
3
22.31
3
22.31
1
7.44
1
7.44
0
0
2
14.87
18
133.84
Adjacent
Controls
3
20.83
3
20.83
2
13.89
1
6.94
1
6.94
1
27.78
1
6.94
2
13.89
0
1
6.94
0
28
194.44
Non-adjacent
Controls
3
2.98
8
7.93
21
20.83
8
7.93
14
13.89
15
14.88
27
26.78
5
4.96
3
2.98
10
9.92
5
4.96
177
175.55
Index
0
1
18.36
1
iase
0
0
1
18.36
5
91.78
1
18.36
0
0
1
18.36
16
293.69
Adjacent
Controls
1
11.94
1
11.94
3
35.83
0
2
23.89
3
35.83
3
35.83
4
47.77
1
11.94
0
0
23
274.69
Non-adjacent
Controls
0
3
16.83
6
33.65
1
5.61
2
11.22
2
11.22
6
33.65
1
5.61
2
11.22
1
5.61
0
33
185.09
                                14

-------
               Table 5.  Death rates for selected cancer In white male populations of
                    Match I and white Match II Index and control census tracts,
                              rates per 100,000 adjusted for age to
                                1970 Baltimore City white males.
Type of cancer and
8th revision ISCD code
Oral (140-149)
Colon (153)
Pancreas (157)
Lung (162)
Prostate (185)
Lymphomas (200-203)
Leukemias (204-207)
All Cancer (140-209)
a This rate includes ISCD
b This rate includes ISCO
Baltimore
City
12.06
36.34a
11.47
95.62
24.13
20.89b
282.75
codes 153 and
codes 200-209.
Index
13.11
18.71
13.11
181.52
37.38
15.89
608.91
154.
Adjacent
Controls
22.91
27.85
-
90.12
17.43
5.47
308.76

South
Controls
2.53
24.82
14.48
144.07
23.21
16.66
8.15
328.91

Central
Controls
10.06
13.24
12.71
105.13
16.29
12.53
9.67
301.03

North
Controls
9.75
20.21
5.03
103.01
10,11
4.72
9.40
274.55

Table 6. Death rates for selected cancers In white female populations of
Match 1 and white Match II index and control census tracts,
rates per 100,000 adjusted for age to
1970 Baltimore City white females.
Type of cancer and
8th revision ISCO code
Oral (140-149)
Colon (153)
Pancreas (157)
Lung (162)
Breast (174)
Cervix (180)
Lymphomas (200-203)
LeuKemia (204-207)
All Cancer (140-209)
Baltimore
City
4.08
32.51°
9.35
18.43
35.67
7.90
16.06b
192.96
Index
15.60
8.88
26.69
26.77
44.03
8.88
6.16
144.64
Adjacent
Controls
17.57
23.57
6.00
29.18
18.01
29.22
• 6.00
240.28
South
Controls
5.98
24.83
21.61
11.95
35.40
7.81
6.89
11.04
212.06
Central
Controls
-
28.27
23.89
16.27
30.84
3.28
15.04
2.72
212.20
North
Controls
5.01
39.25
5.01
26.81
47.66
3.55
14.22
5.01
277.08
a This rate includes ISCD codes 153 and 154.
b This rate includes ISCD codes 200-209.
                                             15

-------
                   Table 7.  Death rates for selected cancers In populations of
                    Match I and whites In Match II Index and control census
                      tracts, rates per 100,000 adjusted for age and sex to
                             white 1970 Baltimore City population.
Type of cancer and
8th revision ISCD code
Oral (140-149)
Colon (153)
Pancreas (157)
Lung (162)
Lymphomas (200-203)
Leukemia (204-207)
All Cancer (140-209)
Baltimore
City
7.85
34.32a
10.35
54.88
18.34b
235.36
Index
14.42
13.53
20.28
99.85
7.50
3.25
363.87
Adjacent
Controls
20.09
25.59
3.17
57.96
5.75
272.61
South
Controls
4.35
24.82
18.24
74.34
11.50
9.68
267.24
Central
Controls
4.75
21.17
18.61
58.23
13.86
6.00
254.15
North
Controls
7.25
30.26
5.01
62.79
9.73
7.09
275.89
a This rate includes ISCD codes 153 and 154.
b This rate includes (SCO codes 200-209.
                                              16

-------
                Table 8.  Death rates for selected cancers In black populations of
                   Match II Index and control census tracts, rates per 100,000
                adjusted tor age and sex to black 1970 Baltimore City population.
Type of cancer and
8th revision ISCD code
Oral (140-149)
Colon (153)
Pancreas (157)
Lung (162)
Lymphomas (200-203)
Leukemia (204-207)
All Cancer (140-209)
Baltimore
City
4.28
17.778
8.01
39.98
11.03b
167.93
Index
9.07
9.97
9.07
79.77
31.37
236.33
'Adjacent
Controls
17.21
17.02
25.63
49.26
8.59
215.82
Central
Controls
-
60.11
-
118.46
.
353.44
North
Controls
-
35.51
-
35.98
17.56
206.31
a This rate includes ISCD codes 153 and 154.
b This rate includes ISCD codes 200-209.

-------
                                           Table 9.  Crude cancer death rates for selected sites In males
                                                     by Individual tract and matched controls,
                                                      average annual death rate per 100,000.


                                                                   MATCH I
oo
Cancer Site 2303
Oral
Pancreas
Lung
Prostate
39.5
39.5
394.9s
39.5
Index
2302
19.8
-
138.9b
19.8
2404
-
18.6
37.3b
18.6
Adjacent
Controls
22.0
-
73.3
22.0
South
Controls
2.2
11.1
109.2
15.6
Central
Controls
10.3
21.9
102.8
17.1
North
Controls
10.7
29.7
96.4
5.4
MATCH II
Index
Cancer Site WM
Oral
Pancreas
Lung
Prostate
-
-
146.0°
36.5
2301
BM
40.1
40.1
280.8°
80.2
Adjacent Controls
WM BM
24.0
-
144.2
-
23.3
23.3
93.4
70.0
Central
WM
-
-
114.8
-
Controls
BM
-
-
153.1
.
North Controls
WM BM
-
-
56.2 48.2
18.8 32.1
                    a Significant difference compared to each control.
                    b Not significant compared to each control.
                    c Lung cancer rates combined for races significantly different from comparable rates of north and adjacent controls.

-------
Table 10.  Crude cancer death rates for selected sites In females
           by Individual tract and matched controls,
            average annual death rate per 100,000.
                    MATCH I
Index Adjacent
Cancer Site 2303 2302 2404 Controls
Oral
Pancreas
Lung
Breast
Cervix
77.6
19.4 3
58.3
19.4
19.4

20.8
4.9 6.9
27.8
6.9
13.9
MATCH II
Index 2301 Adjacent Controls
Cancer Site WF BF WF BF
Oral
Pancreas
Lung
Breast
Cervix
-
-
39.3 - 2E
196.5 - 5C
34.4 7£
22.9
45.8
>.0 45.8
).0 22.9
>.0 22.9
South Central
Controls Controls
4.1
14.4 21.9
8.2 18.8
24.7 31.3
6.2 3.1

Central Controls
WF BF

37.7
43.2
-
-
North
Controls
5.0
'-
24.7
24.7
5.0

North Controls
WF BF
-
17.4
17.4
87.1 14.0
14.0

-------
                           REFERENCES
 1.  W. Winkelstein, S. Kantor, et al, "The relationship of
     air pollution and economic status to total mortality and
     selected respiratory system mortality in men,1' Arch
     Environ Health 14; 1620 (1967).

 2.  L. Zeidberg, R.J.M. Horton/ and E. Landau, "The Nashville
     air pollution study.  V. Mortality from diseases of the
     respiratory system in relation to air pollution," Arch
     Environ- Health 15: 214  (1967)

 3.  H.R. Menck, J.T. Casagrande and B.E. Henderson, "Industrial
     air pollution:  Possible effect on lung cancer," Science
     183: 210 (1974).

 4.  S. Milham,  Jr., and T. Strong, "Human arsenic exposure in
     relation to a copper smelter," Environ Res 7: 176 (1974).

 5.  W.J. Blot,  and J.F. Fraumeni, Jr., "Arsenical air pollution
     and lung cancer," Lancet,  142  (1975).

 6.  M.G. Ott, B.B. Holder, and H.L. Gordon, "Respiratory cancer
     and occupational exposure to arsenicals," Arch Environ Health
     29: 250  (1974).

 7.  A.M. Lee, and J.F. Fraumeni, Jr., "Arsenic and respiratory
     cancer in man, an occupational study," JNCI 42, 6-, 1045  (1969)

 8.  B. Woolf, "On estimating the relation between blood group and
     disease," Ann Eugen  19: 251  (1955).

 9.  J.B.S. Haldane, "The estimation and significance of the
     logarithm of a ratio of frequencies," Ann Human Genetics
     20: 309  (1956).

10.  E. Moss, "Oral cancer in textile workers," Conference on
     Occupational Carcinogenesis, New York Academy of Sciences,
     (Radiation and Particulate Matter) New York, New York (1975);

-------
                             TECHNICAL REPORT DATA
                       (Please read Instructions on the reverse before completing)
1. REPORT NO.

    560/6-76-003
                         2.
                                                  3. RECIPIENT'S ACCESSION1 NO.
4. TITLE AND SUBTITLE
    EPDIEMIOLOGY STUDIES/ TASK I - PHASE I
    Pilot Study of Cancer Mortality Near an
    Arsenical Pesticide  Plant in Baltimore
           5. REPORT DATE
             Mav  1976
           6. PERFORMING ORGANIZATION CODE
7. AUTHOFUS)
    Genevieve Matanoski,  MD,  Dr. PH, Johns Hopkins
    Emanuel Landau/ PhD,  APHA E. Elliott/ BA;  JE
                                                  8. PERFORMING ORGANIZATION REPORT NO,
9. PERFORMING ORGANIZATION NAME AND ADDRESS
   American Public Health  Association
   1015 Eighteenth Street  N.  W.
   Washington, D. C.  20036
                                                  10. PROGRAM ELEMENT NO.
           11. CONTRACT/GRANT NOT
                                                     68-01-2490
 12. SPONSORING AGENCY NAME AND ADDRESS
    Office of Toxic Substances
    Environmental Protection Agency
    401 M St. S. W.  Washington, D. C.
                                                  13. TYPE OF REPORT AND PERIOD COVERED
                                                     Final Report	
           14. SPONSORING AGENCY CODE
   20460
15. SUPPLEMENTARY NOTES
16. ABSTRACT
     The population of the  area surrounding  a  chemical plant which pro-
 duced arsenical pesticides had possibly been exposed to arsenic  in the
 environment.  In order  to determine whether  there had been a carcino
 genie effect from the exposure, a pilot study of the cancer mortality
 of  the four census tracts for which at least half of the tract area lay
 within three quarters of  a mile of the plant was conducted for the year
 1970-1972.  The rates were compared to those of the other 23 tracts in
 the city which were similar to the index tracts for five variables mea-
 suring age, race, and sex distribution and socioeconomic characteristic
     The results indicate that lung cancer mortality in males is signifi-
 cantly higher for the tract in which the plant is located as compared
 to  control areas.  Two  of the other three  tracts also had higher lung
 cancer mortality than controls but the difference was not significant
 and the values were lower than those of the  tract in which the plant
 was located.  The four-fold excess in lung cancer deaths did not appear
 to  be related to occupational exposure.
     Total cancers in females and those at other sites in males were too
 few in number to place  any reliance on death rates.  The high mortality
 from oral cancer, especially in females, deserve further study.
17.
                           KEY WORDS AND DOCUMENT ANALYSIS
               DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS  C.  COSATI Field/Group
  Mortality analysis
  Lung Cancer
  Insecticide/arsenic  exposure
  Air pollution
  Analysis by census tracts
  Arsenic
  Insecticides
  Cancer
  Census tracts
13. DISTRIBUTION STATEMENT
  Release Unlimited
                                       19. SECURITY CLASS (ThisReport)
                                       Unclassified
                       21. NO. OF PAGES
                          26
20. SECURITY CLASS (Thispage)

 Unclassified
22. PRICE
EPA Form 2220-1 (9-73)

-------
                                                       INSTRUCTIONS

   1.   REPORT NUMBER
        Insert the EPA report number as it appears on the cover of the publication.

   2.   LEAVE BLANK

   3.   RECIPIENTS ACCESSION NUMBER
        Reserved for use by each report recipient.
   •»•   i • i kc *wnw awo 111 uc
        Title should indicate clearly and briefly the subject coverage of the report, and be displayed prominently. Set subtitle, if used, in smaller
        type or otherwise subordinate it to main title. When a report is prepared in more than one volume, repeat the primary title, add volume
        number and include subtitle for the specific title.

   6.   REPORT DATE
        Each report shall carry a date indicating at least month and year. Indicate the basis on which it was selected (e.g., date of issue, date of
        approval, date of preparation, etc.).

   6.   PERFORMING ORGANIZATION CODE
        Leave blank.

   7.   AUTHOR(S)
        Give name(s) in conventional order (John R. Doe, J. Robert Doe, etc.).  List author's affiliation if it differs from the performing organi-
        zation.

   8.   PERFORMING ORGANIZATION REPORT NUMBER                             ••-•>-	           --
        Insert if performing organization wishes to assign this number.

   9.   PERFORMING ORGANIZATION NAME AND ADDRESS
        Give name, street, city, state, and ZIP code. List no more than two levels of an organizational hirearchy.

   10.  PROGRAM ELEMENT NUMBER
        Use the program element number under which the report was prepared. Subordinate numbers may be included in parentheses.

   11.  CONTRACT/GRANT NUMBER
        Insert contract or grant number under which report was prepared.

   12.  SPONSORING AGENCY NAME AND ADDRESS
        Include ZIP code.

   13.  TYPE OF REPORT AND PERIOD COVERED
        Indicate interim final, etc., and if applicable, dates covered.

   14.  SPONSORING AGENCY CODE
        Leave blank.

   IS.  SUPPLEMENTARY NOTES
        Enter information not included elsewhere but useful, such as: Prepared in cooperation with, Translation of, Presented at conference of,
        To be published in. Supersedes, Supplements, etc.

   16.  ABSTRACT
        Include a brief (200 words or less) factual summary of the most significant information contained in the report. If the report contains a
        significant bibliography or literature survey, mention it here.

   17.  KEY WORDS AND DOCUMENT ANALYSIS
        (a) DESCRIPTORS - Select from the Thesaurus of Engineering and Scientific Terms the proper authorized terms that identify the major
        concept of the research and are sufficiently specific and precise to be used as index entries for cataloging.

        (b) IDENTIFIERS AND OPEN-ENDED TERMS - Use identifiers for project names, code names, equipment designators, etc. Use open-
        ended terms written in descriptor form for those subjects for which no descriptor exists.

        (c) COS ATI FIELD GROUP - Field and group assignments are to be taken from the 1965 COS ATI Subject Category List. Since the ma-
        jority of documents are multidisciplinary in nature, the Primary Field/Group assignments) will be specific discipline, area of human
        endeavor, or type of physical object. The application(s) will be cross-referenced with secondary Field/Group assignments that will follow
        the primary posting(s).

   18.  DISTRIBUTION STATEMENT
        Denote releasability to the public or limitation for reasons other than security for example "Release Unlimited." Cite any availability to
        the public, with address and price. /

   19. & 20. SECURITY CLASSIFICATION
        DO NOT submit classified reports to the National Technical Information service.

   21.  NUMBER OF PAGES
        Insert the total number of pages, including this one and unnumbered pages, but exclude distribution list, if any.

   22.  PRICE
        Insert the price set by the National Technical Information Service or the Government Printing Office, if known.
EPA Form 2220-1 (9-73) (Reverse)

-------