ISEE-471 The Effect of the WTC Tragedy on Pregnancy Outcomes Sally Ann Lederman, PhD1, Lisa Weiss, MPH1, Deliang Tang, MD, DrPH1, Howard Andrews, PhD1, Virginia Rauh, ScD1, Mark Becker, MA2, Richard Wang DO3, Frederica Perera,

1
ISEE-471 The Effect of the WTC Tragedy on
Pregnancy Outcomes Sally Ann
Lederman, PhD1, Lisa Weiss, MPH1, Deliang Tang,
MD, DrPH1, Howard Andrews, PhD1, Virginia
Rauh, ScD1, Mark Becker, MA2, Richard Wang DO3,
Frederica Perera, DrPH1. 1. Columbia Center
for Children's Environmental Health, Mailman
School of Public Health, Columbia University, NY,
NY 2. CIESIN, Columbia University, Palisades,
NY 3. Division of Laboratory Sciences, CDC,
Atlanta, GA
Figure 1. Perceived Air Quality Scale
Purpose of study To examine the utility of
womens report of air quality as a way of
assessing detailed, local exposures. Population
Between Dec. 2001 and June 2002, project
interviewers enrolled 329 pregnant, non-smoking
women when they came for labor and delivery at 3
hospitals close to the WTC site, Beth Israel, St.
Vincents, and NYU Downtown. All willing women
between 18 and 39 years old, reporting no
diabetes, hypertension, HIV/AIDS or use of
illegal drugs in the last year and pregnant with
a single infant were enrolled. Methods
Birth outcome data and characteristics of the
pregnancy and delivery were abstracted from the
medical records. The mother was interviewed
shortly after delivery (in English, Spanish or
Chinese) to determine demographics, reproductive
history, and lifestyle sources of toxicant
exposures. In addition, she reported her home and
work locations in the 4 weeks after 9-11-01. All
home and work locations were geocoded and their
linear distances from the WTC site were computed.

Sample question (for home site, first week)
How would you describe the condition of the
air outside the place where you were living
most during the first week, Sept 11th to 17th?
Would you say it was Not visibly hazy
or smoky, and with only occasional or no bad
smell. 1 Rarely visibly hazy or
smoky, with a bad smell during some of the
time... 2 Very hazy or smoky only
some of the time, but with a bad smell during
most of that time. 3
Dense with visible haze or smoke and with a
bad smell during a lot of that time.. 4
As an immediate measure of the womens
exposure at home and work, women were asked to
report, for each week at each home and work
location, their perception of the average air
quality on a 4-point Perceived Air Quality Scale
that we developed for this purpose (Figure 1).
Average reported values declined over the four
weeks. Womens Perceived Air Quality values
were used to develop an areal Model of Air
Quality for each of the 4 weeks after 9-11-01.
The intent was to develop a Model of Air Quality
not highly influenced by individual sensitivity
to poor air quality and to determine whether
reported air quality can be used to identify
populations with high exposure. The Model of Air
Quality integrated information on reported air
quality for all work and home locations by week,
using nearly 1860 perceived air quality values
for the 4 weeks, from the womens 1190 home site
reports and 669 work site reports of air quality
for those women who were working. The Model was
then used to assign to each home and work
location for each of 4 weeks, a new, modeled, air
quality value. Correlations were computed
between the modeled air quality values and the
original, reported air quality values and between
the reported and modeled values and distance to
the WTC site.
Figure 2
Figure 3
Results The Perceived Air Quality Scale is
shown in Figure 1. The geographical
distribution of the raw air quality values for
week 1 reported by the women in the study using
this scale is shown in Figure 2. The Model
of Air Quality was used to generate the map in
Figure 3. This shows the areal distributions of
values from the Model of Air Quality for the
first week after 9-11-01. Table 1 shows the
correlations between the Perceived Air Quality
Scale values and Modeled Air Quality values for
each of the 4 weeks after 9-11-01 for home and
work locations. As expected, modeled and reported
air quality are highly correlated. The
correlations of reported and modeled values with
distance from the WTC site are also shown in
Table 1.
Table 1. Correlations of modeled and reported air
quality values with each other and with distance
from the WTC site for 4 weeks after 9-11-01.
Modeled Modeled Modeled Modeled Reported
Reported Reported Reported air
air air air air air
air air wk1 wk2 wk3
wk4 wk1 wk2 wk3
wk4 HOME Reported air
wk10.654 Reported air
wk20.637 Reported air
wk3...0.538 Reported air wk4
... 0.565 WORK
Reported air wk1 0.738 Rep
orted air wk20.761 Reported air
wk3 0.765 Reported air
wk4... 0.740 HOME
Distance fr WTC wk10.893
0.592 Distance fr WTC
wk20.896 ...
0.587 Distance fr WTC wk3...
0.848 0.489 Distance fr WTC
wk4... 0.789 .
0.492 WORK Distance fr WTC
wk10.883 0.670 Distance
fr WTC wk20.896...
0.684 Distance fr WTC wk3 .
0.901.. 0.704 Distance fr
WTC wk4...0.889
. 0.710
The data show that that Modeled Air Quality
values correlate more strongly with distance from
the WTC site than do Perceived Air Quality
values. At the time covered by these reports,
the WTC collapses, fires and rescue and cleanup
operations were a major source of air pollution
in the New York area. The high correlation
between Modeled Air Quality and distance from the
WTC site suggests that the modeling procedure was
successful in eliminating variation among
reported air quality values due to individual
hyper/hypo-sensitivity to air quality.
Conclusion These results indicate that
reported perception of air quality based on a
4-point scale may be useful for generating models
of air quality in other settings where there are
insufficient data from standard air pollution
detectors.
Upcoming Work A report of the relationship
of pregnancy outcomes to the women's work and
home locations after 9-11-01 is under peer
review. We expect the results to be published
shortly.