Am. J. Epidemiol. 2007 Croteau 951 65

American Journal of Epidemiology
The Author 2007. Published by the Johns Hopkins Bloomberg School of Public Health.
All rights reserved. For permissions, please e-mail: journals.permissions@oxfordjournals.org.
Vol. 166, No. 8

DOI: 10.1093/aje/kwm171
Advance Access publication July 25, 2007
Original Contribution
Work Activity in Pregnancy, Preventive Measures, and the Risk of Preterm
Delivery
Agathe Croteau1,2, Sylvie Marcoux3, and Chantal Brisson3

1
National Institute of Public Health of Quebec, Quebec, QC, Canada.

CHUQ Research Center, Laval University, Quebec, QC, Canada.
3
Department of Social and Preventive Medicine, Faculty of Medicine, Laval University, Quebec, QC, Canada.
2
The objective of this case-control study was to evaluate whether occupational conditions during pregnancy are
associated with preterm delivery (PTD). Women whose work conditions changed following the use of a legally
justied preventive measure (withdrawal from work or job reassignment) were also compared with those whose
work conditions did not change. Cases (n 1,242) and controls (n 4,513) were selected from 43,898 women who
had single livebirths between January 1997 and March 1999 in Quebec, Canada. They were interviewed by
telephone after delivery. Results showed association of PTD with demanding posture for at least 3 hours per
day, whole-body vibrations, high job strain combined with low or moderate social support, and a cumulative index
composed of nine occupational conditions. The adjusted odds ratio increased from 1.0 to 2.0 for PTD (ptrend <
0.0001) and from 1.0 to 2.7 for very PTD (<34 weeks; ptrend 0.0015) as the number of conditions increased from
zero to four or more. The associations for PTD and very PTD with most of the above-mentioned work conditions
were weaker when exposures were eliminated following recourse to a legally justied preventive measure. This
study provides relevant information on the possible inuence of preventive measures on the risk of PTD in pregnant
workers.
maternal exposure; occupational exposure; posture; pregnancy outcome; premature birth; stress, psychological;
vibration; work schedule tolerance
Abbreviation: PTD, preterm delivery.
In industrialized countries, preterm delivery (PTD) is usually considered to be the most important cause of infant
mortality (14) and a major determinant of morbidity (3
5) and neurodevelopmental deficits (13). Maternal risk factors for this condition include medical conditions (1, 39),
smoking (1, 3, 6, 8), and social (1, 3, 8) and occupational (7)
factors.
Previous prospective cohorts (1024), retrospective studies (2542), and case-control studies (4350) have investigated the association between PTD and long working hours
(12, 13, 18, 19, 21, 25, 26, 28, 31, 3339, 47, 49), inconvenient schedule (17, 19, 23, 25, 28, 33, 34, 37, 40, 41, 46, 47,
49), prolonged standing (1217, 1921, 2628, 33, 34, 36

38, 43, 4649), lifting loads (10, 11, 16, 17, 19, 21, 25, 27,
28, 30, 3234, 37, 38, 43, 49), high psychosocial stress (10,
11, 13, 22, 25, 26, 29, 33, 4448, 50), or a cumulative index
of occupational fatigue (13, 15, 24, 26, 30, 33, 34, 37, 38,
42, 47, 48, 50). Results were inconsistent. Increased risk of
PTD was documented in 67 percent of studies on long working hours (12, 13, 18, 21, 25, 26, 31, 33, 34, 36, 47, 49), 62
percent on inconvenient schedule (17, 19, 25, 28, 33, 34, 41,
47), 68 percent on prolonged standing (12, 1417, 20, 26,
27, 33, 3638, 46, 47, 49), 53 percent on lifting loads (10,
17, 19, 21, 27, 30, 34, 37, 38), 71 percent on high
Correspondence to Dr. Agathe Croteau, National Institute of Public Health of Quebec, 945 Wolfe Avenue, Second Floor, Quebec, QC,
Canada G1V 5B3 (e-mail: agathe.croteau@inspq.qc.ca).
951
Am J Epidemiol 2007;166:951965
Downloaded from http://aje.oxfordjournals.org/ by guest on October 26, 2016
Received for publication November 3, 2006; accepted for publication May 9, 2007.
952 Croteau et al.
dation after completion with the pregnant worker of a standardized evaluation of her working conditions. The
pertinence of the recommendation is finally determined by
the Commission de la sante et de la securite au travail, the
governmental agency for health and safety at work.
Our objective was to estimate the association between
some occupational conditions, both individually and cumulatively, and the risk of PTD. We also assessed whether the
risk of PTD was lower when a womans work conditions
changed following the use of a legally justified preventive
measure (withdrawal from work or job reassignment) than
when they did not change.
MATERIALS AND METHODS
Study design and population
For this case-control study, the source population consisted of women living in seven regions (Saguenay-Lac-StJean, Quebec, Mauricie, Centre-du-Quebec, Estrie, Laval,
Chaudie`re-Appalaches) of the province of Quebec who gave
birth to a live singleton between January 25, 1997, and
March 7, 1999. Those regions are representative of urban,
semiurban, and rural populations of the province. We were
authorized by the Commission dacce`s a` lInformation du
Quebec to obtain the mothers name and telephone number;
type of birth (single or multiple); and birth date and length of
pregnancy, all recorded on birth certificates.
The length of pregnancy is usually estimated by comparing the actual date of delivery with the expected birth date,
the latter determined by the physician from the date of the
last menses and clinical and ultrasonic evaluations (28).
A total of 43,898 singleton livebirths were reported to us;
according to government data, this number represented 94
percent of singleton livebirths in the seven participating
regions during the study period. We classified births as
PTD cases (infants born before 37 complete weeks of gestation) (n 2,626 (6 percent)) or noncases (n 41,272).
A random sample of 20 percent of the noncases (n 8,365)
constituted the potential control group.
Data collection
After receipt of the birth certificate information, the interviewers contacted the women by telephone. The median
interval between childbirth and interview was 32 days for
cases and 30 days for controls. The interviewers explained
the study, requested the womans participation, and verified
her eligibility. Among the case mothers, 192 (7.3 percent)
could not be contacted, and 56 (2.1 percent) refused to
participate; among the control mothers, these numbers were
470 (5.6 percent) and 128 (1.5 percent), respectively. Of the
10,145 women (2,378 cases and 7,767 controls) who agreed
to participate, women who did not work (n 3,294), those
who worked less than 4 weeks from the first month of pregnancy (n 403) or less than 20 hours per week (n 457),
and those with more than one job (n 236) were excluded.
This left 5,755 eligible women (1,273 cases and 4,482 controls) who completed a computer-assisted telephone interview of 2030 minutes.
psychosocial stress (11, 26, 29, 33, 4448, 50), and 77 percent on cumulative fatigue score (15, 24, 30, 33, 34, 37, 42,
47, 48, 50). For individual exposures (1023, 2541, 43
50), most associations were of small magnitude (median
value of the 84 measures of associations: 1.29). The few
statistically significant associations either were greater than
1.50 (15, 16, 20, 25, 30, 31, 33, 36, 41, 47, 48), came from
large sample size studies (14, 34, 49), or were related to
fatigue score (30, 33, 34, 37, 42, 47, 48, 50). Results did
not vary consistently with study design. Increased risks of
PTD with long working hours, inconvenient schedule, and
lifting loads were more frequently observed in prospective
studies (10/14, 71 percent) than in retrospective and casecontrol studies (19/34, 56 percent), whereas the contrary
was observed for psychosocial stress and fatigue score
(3/7, 43 percent and 17/20, 85 percent, respectively). Several
other methodological issues could explain inconsistency in
study results. Previous studies with small sample size (10
13, 1522, 24, 25, 2733, 37, 38, 40, 41, 43, 44, 46, 47) were
more likely to show associations between PTD and standing
than were other studies. The same was true for low (under 80
percent) participation rate (1113, 16, 19, 20, 23, 24, 27, 29,
32, 33, 3537, 39, 42, 44, 45, 4749) with standing, psychosocial stress, or fatigue score and for potential confounding (1113, 15, 17, 18, 2022, 2427, 3033, 3538, 40, 42,
43, 4650) with long hours or prolonged standing.
In two meta-analyses, statistically significant pooled estimates of 1.21.3 were found for shift or night work (51,
52), long working hours (51, 52) (in a subset of high-quality
studies), prolonged standing (51, 52), and physically demanding work (52). A pooled estimate of 1.63 was reported
for a cumulative work fatigue score (52). For standing and
physically demanding work, pooled associations were consistent across study designs (52). Except for long working
hours, pooled estimates did not vary according to study
quality score (51, 52).
Studies where authors failed to take into account changes
in work conditions during pregnancy, usually experienced
by the most heavily exposed workers (1012, 14, 16, 23, 30,
34, 3638, 46, 49), are less likely to observe results associating PTD with long working hours, inconvenient schedule,
and psychosocial stress than are other studies. A possible
underestimation of PTD risk in women still working in late
pregnancy (19, 53, 54) can result from a healthy pregnant
worker effect if healthier women without pregnancy complications select themselves to continue working late in
pregnancy.
In the province of Quebec, Canada, a favorable context
exists to examine how occupational condition changes occurring during pregnancy relate to pregnancy outcome. According to the Loi sur la sante et securite du travail (Health
and Work Security Act), pregnant workers have a legal right
to be assigned to other tasks or to withdraw from work
without prejudice if working conditions present a danger
to themselves or the fetus (55). Recourse to preventive
measures is frequent (nearly half of pregnant workers)
(56). It does not depend on the womans health or on the
unions or employers willingness. The request is initiated
by the womans physician who consults a public health
occupational physician. The latter formulates a recommen-
Work Activity, Preventive Measures, and Preterm Delivery
Analysis
Odds ratios and 95 percent confidence intervals were estimated from beta coefficients and their standard errors produced by logistic regression.
All variables statistically associated (p < 0.05) with PTD
or with the use of preventive measures as those with an odds
ratio greater than or equal to 1.2 (tables 1 and 2) were
considered as potential confounders and were initially entered as covariates in all regression models. We first obtained odds ratios relating occupational conditions to PTD
by multiple logistic regression with adjustment for the
whole set of covariates. Covariates were withdrawn one

by one as long as the odds ratio was not modified by more
than 10 percent compared with the full model. In the final
model, odds ratios were also adjusted for other occupational
conditions present at the beginning of pregnancy (refer to
footnotes of tables 3, 4, and 5).
To assess the influence of changes in occupational conditions by legally justified preventive measures, we divided
workers exposed at the beginning of pregnancy to a given
working condition into three groups according to whether
the condition was eliminated by preventive measures early
(before 24 weeks) or late (at or after 24 weeks) in pregnancy
or not eliminated. The PTD risks of these three groups were
compared with the risk for unexposed workers at the beginning of pregnancy.
A secondary analysis based on job title was performed to
assess the potential for a recall bias related to self-reported
exposure. Using a 20 percent random sample of the source
population, we calculated the percentage of workers reporting exposure to a given condition for each of 60 job titles.
Women were classified according to the probability of exposure (percent) corresponding to their job title.
We calculated a cumulative index of occupational conditions for which the adjusted odds ratio was at least 1.2 when
the condition was not eliminated by a preventive measure.
The association of this index with PTD risk was adjusted for
relevant covariates. We used a v2 test to evaluate a doseresponse relation (67). We also evaluated the influence of
eliminating the indexed conditions by preventive measures
during the pregnancy using the method described previously
for single occupational conditions. For individual conditions
and for cumulative index, adjusted odds ratios were obtained for very PTD (<34 weeks).
RESULTS
Among the potential confounders, the strongest associations (odds ratio > 1.5) with PTD were observed for congenital anomalies, chronic hypertension, mothers age, and
prior adverse pregnancies. Physical activity, ethnicity, caffeine and alcohol consumption, and illicit drugs during pregnancy were not associated with PTD risk (data not shown).
Nearly half (47.3 percent) of the workers used legally
justified preventive measures: preventive withdrawal from
work (32.3 percent), job reassignment (22.0 percent), or both
(7.0 percent). Frequency use of preventive measures was
similar for the cases and the controls. The number of occupational conditions was strongly related to recourse to preventive measures and was by far the most important factor
explaining their use (table 2). The association was even
stronger for early than for late preventive measures (data
not shown). After adjustment for occupational conditions,
preventive measures remained associated with low education, family income, low maternal age, chronic hypertension, diabetes, and over-the-counter drug use. All variables
shown in table 1 were considered as potential confounders
and were entered in the initial multivariate models. This set
of covariates includes the variables associated with PTD, as
well as those associated with recourse to preventive measures in table 2.
The questionnaire documented the following working

conditions: schedule (hours and consecutive days worked
per week, day, evening, or night work, schedule regularity);
posture (sitting, standing, other demanding postures); physical effort (lifting (weight and frequency), pushing, and pulling objects); work organization (breaks, piecework or
assembly line work, psychosocial factors); and environmental occupational conditions (e.g., whole-body vibration,
temperature, exposure to environmental tobacco smoke).
We developed the questions after examining previous
questionnaires (33, 34) and existing literature (5763) and
after consulting ergonomists. Psychosocial factors were
evaluated by use of the job demand-control support model
of Karasek and Theorell (64) and a validated French version
of related scales (65, 66). Psychological demand and decision latitude scales were dichotomized at the median value.
Social support by colleagues and supervisors was categorized in two levels on the basis of the natural distribution of
the score. Four levels of job strain were obtained by crossstratifying psychological demand and decision latitude
(high, active, passive, low). The three highest levels of job
strain were also subdivided by social support level.
For the first step, we documented working conditions at
the beginning of pregnancy. If conditions were modified
during pregnancy, we asked when and documented the
new working conditions related to work schedule, posture,
and effort. Mothers also indicated when they stopped working and why (e.g., legally justified preventive withdrawal,
health problems, coming close to expected date of delivery).
The final section of the questionnaire documented obstetric history, mothers medical profile (before and during pregnancy), newborns characteristics (gender, weight, birth
date, expected date of delivery according to the physician,
congenital anomalies), mothers lifestyle (drug consumption, physical activity, and consumption in third trimester
of caffeine, tobacco, and alcohol), and sociodemographic
characteristics.
For 154 (2.7 percent) of the 5,755 women, the interview
data (date of birth and expected date of delivery) indicated
a case or control status different from that determined on the
basis of birth certificate data. Of these women, 115 (74.7
percent) gave us access to their hospital records to verify the
information. Archivists responded to 114 (99.1 percent) of
the requests. The information received resulted in an amendment of case or control status for 57 (50 percent) of these
114 subjects, resulting in 1,242 cases and 4,513 controls in
the study population.
953
954 Croteau et al.
TABLE 1. Odds ratios and 95% condence intervals for having a preterm delivery, by potential
confounding variables, among workers giving birth in Quebec, Canada, between January 1997 and March
1999
Case mothers
(n 1,242)
Odds
ratio
95%
condence
interval
No.*
No.*
23
1,200
19
1.9
96.6
1.5
50
4,415
48
1.1
97.1
1.1
1.7
1.0
1.5
765
445
32
61.6
35.8
2.6
2,254
2,157
102
49.9
47.8
2.3
1.0
0.6
0.9
0.5, 0.7
0.6, 1.4
870
255
117
70.0
20.5
9.4
3,420
809
284
75.8
17.9
6.3
1.0
1.2
1.6
1.1, 1.5
1.3, 2.0
1,185
55
95.6
4.4
4,361
152
96.6
3.4
1.0
1.3
1.0, 1.8
717
516
58.2
41.8
2,852
1,655
63.3
36.7
1.0
1.2
1.1, 1.4
1,078
127
41
86.8
10.2
3.0
4,062
351
100
90.0
7.8
2.2
1.0
1.4
1.4
1.1, 1.7
1.0, 2.1
457
388
391
37.0
31.4
31.6
1,980
1,407
1,120
43.9
31.2
24.9
1.0
1.2
1.5
1.0, 1.4
1.2, 1.9
109
242
371
400
114
8.8
19.6
30.0
32.4
9.2
364
871
1,438
1,502
325
8.1
19.4
31.9
33.4
7.2
1.2
1.1
1.0
1.0
1.4
970
102
130
36
78.4
8.2
10.5
2.9
3,684
306
407
112
81.7
6.8
9.0
2.5
1.0
1.3
1.2
1.2
1.0, 1.6
1.0, 1.5
0.8, 1.8
1,215
21
98.3
1.7
4,455
52
98.8
1.2
1.0
1.5
0.9, 2.5
1,220
20
98.4
1.6
4,473
40
99.1
0.9
1.0
1.8
1.1, 3.2
1,125
117
90.6
9.4
4,220
293
93.5
6.5
1.0
1.5
1.2, 1.9
1,186
56
95.5
4.5
4,357
156
96.5
3.5
1.0
1.3
1.0, 1.8
1,198
44
96.5
3.5
4,454
59
98.7
1.3
1.0
2.8
1.9, 4.1
1.0, 2.8
0.9, 2.5
0.9, 1.5
0.9, 1.3
0.9, 1.2
1.1, 1.7
* Totals vary because of missing data.

y Reference category.
z Examples of other chronic diseases are asthma, thyroid disorder, bowel inammatory disease, hypoglycemia,
kidney disease, heart diseases, epilepsy, and hypercholesterolemia.
Mothers age (years)

<20
2039y
40
Parity
1y
23
>3
Previous adverse pregnancy
outcomes (no.)
0y
1
2
Civil status
Married or living with a partnery
Single
Mothers education (years)
14y
<14
Family income (Can$/year)
20,000y
<20,000
Unknown
Mothers height (cm)
165188y
158164
105157
Mothers body mass index, before
pregnancy (kg/m2)
3061
2529
2224y
1921
1418
Maternal smoking, third trimester
(cigarettes/day)
0y
19
1019
20
Over-the-counter drug use
Noy
Yes
Chronic hypertension
Noy
Yes
Diabetes
Noy
Yes
Other chronic diseasez
Noy
Yes
Congenital anomaly
Noy
Yes
Control mothers
(n 4,513)
DISCUSSION
We found that an increased risk of PTD was significantly

associated with demanding posture for at least 3 hours per
day, whole-body vibrations, and high job strain combined
with low or moderate social support. Increased risks of very
PTD were associated with the last two conditions. Risks of
PTD and very PTD increased with a cumulative index composed of nine work conditions including the three above.
Most of these associations were higher in women whose
work conditions did not change compared with women
whose work conditions changed in the course of pregnancy
following a legally justified job withdrawal or reassignment.
Some earlier studies found increased PTD risk for demanding posture (21, 35, 50, 68) and vibrations (34). Ten
(11, 26, 29, 33, 4448, 50) of 14 earlier studies on psychosocial stress at work (10, 11, 13, 22, 25, 26, 29, 33, 4448,
50) showed increased PTD risk. The association with a cumulative index of exposure is consistent with a metaanalysis (52) and with most (15, 24, 30, 33, 34, 37, 42, 47,
48, 50) previous studies that evaluated the relation between
the accumulation of strenuous working conditions and PTD.
Only three studies (13, 26, 38) did not find a positive trend,
and the authors of one study report that preventive measures
(increased antenatal leave, modifications in working conditions) could be one reason for the lack of relation (38).
Given that interviews took place after delivery, a recall
bias could overestimate the associations if women who had
PTD were more likely to report adverse working conditions.
There is no clear evidence that recall bias was present in
previous retrospective studies or case-control studies of
PTD. Although retrospective (2542) and case-control
(4350) studies were more likely than prospective ones
(1024) to result in positive associations with psychosocial
stress and cumulative index of work fatigue, the contrary
was observed for long working hours, inconvenient schedule, and lifting loads, and proportions were similar for prolonged standing. Nevertheless, we conducted a secondary
analysis based on job title, which is less prone to recall bias
than analysis based on self-reported exposure. In spite of the
nondifferential misclassification induced by job title analysis and the fact that changes in work exposures during pregnancy were not taken into account, this approach confirmed
the main findings of tables 3 and 4. Indeed, the three exposures with significantly elevated odds ratios for PTD (demanding posture, vibrations, and high strain) and the three
with odds ratios greater than or equal to 1.5 for very PTD
(>5 consecutive working days, vibrations, and high strain)
remained associated with PTD or very PTD. Several occupational conditions that women may have suspected to be
detrimental to their pregnancies (e.g., night work, lifting
loads) were not associated with PTD risk. Finally, in order
to prevent differential report, the interviewers were unaware
of the mothers case or control status when they questioned
them about working conditions. Altogether, these arguments, we believe, are not in favor of a recall bias although
the latter cannot totally be excluded.
The analysis by recourse to legally justified preventive
measures is an interesting contribution of this study. Our
results are consistent with those of studies where PTD risks
Of the occupational conditions present at the beginning of

pregnancy, demanding posture (bending, squatting, arms
raised above shoulder level) for at least 3 hours per day,
whole-body vibrations, and high job strain combined with
low or moderate social support were significantly associated
with PTD (table 3). The associations with demanding posture and job strain were higher when not eliminated by preventive measures than when they were eliminated early
during pregnancy. Odds ratios of at least 1.2 were observed
for a few other conditions when they were not eliminated by
preventive measures.
Of the three work conditions associated with PTD, wholebody vibrations and high job strain combined with low or
moderate social support were related (odds ratio 1.5) to
very PTD (table 4). Workers exposed to more than 5 consecutive working days were also at increased risk of very PTD.
The odds ratios were 2.8, 2.1, and 1.7, respectively, when the
conditions were not eliminated by preventive measures, and
the first two are statistically significant (data not shown).
There was no association between PTD risk and the following conditions, whether or not the worker took recourse
to preventive measures: maximum number of hours worked
per week, possibility to sit when standing, pushing or pulling objects, having to climb stairs, absence of breaks, piece
work or assembly line, noise, long commuting time to work,
and exposure to environmental tobacco smoke at work (data
not shown).
In the job title analysis, working more than 5 consecutive
days, demanding posture, whole-body vibration, and high
strain combined with low or moderate social support remained associated with PTD or very PTD, with odds ratios
ranging from 1.3 to 2.6 (table 5).
We calculated a cumulative index, composed of the nine
occupational conditions showing an odds ratio of 1.2 or
greater for PTD when not eliminated by a preventive measure. These conditions are more than 5 consecutive working
days, irregular or shift-work schedule, standing at least 7
hours per day mostly in one spot, sitting posture for at least 3
hours per day with rare or no possibility to stand, demanding
posture for at least 3 hours per day, whole-body vibrations,
very hot or very cold temperatures, and moderate-active or
high job strain combined with low or moderate social support. Odds ratios increased from 1.0 to 2.0 for PTD and to
2.7 for very PTD, with significant trends, as the number of
indexed conditions increased from zero to 46 (table 6).
Trends were stronger when indexed conditions were not
eliminated by preventive measures. Except for workers exposed to four or more occupational conditions (2.9 percent
of workers in the source population), the PTD and very PTD
risks were lower when occupational conditions were eliminated by preventive measures. Odds ratios were higher for
very PTD than for PTD, whether the cumulative index
calculated at the beginning of pregnancy was considered
or the number of conditions not eliminated by preventive
measures.
The exclusion of newborns with congenital anomalies
had no influence on the associations between PTD risk
and individual occupational conditions or cumulative index
of conditions present at the beginning of pregnancy (data not
shown).
955
956 Croteau et al.
TABLE 2. Odds ratios and 95% condence intervals for using legally justied preventive measures to
eliminate occupational conditions that could increase risk of preterm delivery, by sociodemographic,
lifestyle, medical, and occupational variables, among workers giving birth in Quebec, Canada, between
January 1997 and March 1999
Random sample
of eligible workers
of the source
population
(n 4,750)*
% using
legally
justied
preventive
measures
Model including
occupational
conditions
Model excluding
occupational
conditions
Odds
ratioy
95%
condence
interval
1.1, 1.6
Odds
ratioz
95%
condence
interval
1.2
1.0, 1.5
Mothers age (years)

<25
2534
35
873
39.0
1.3
3,351
45.4
1.0
526
59.7
0.8
1.0
0.7, 1.0
0.8
0.7, 1.0
Parity
2,405
49.6
1.0
23
2,239
44.7
0.9
0.8, 1.0
0.9
0.8, 1.0
106
48.1
1.0
0.7, 1.5
0.9
0.6, 1.4
>3
1.0
Previous adverse pregnancy

outcomes (no.)
0
3,579
46.9
1.0
870
48.9
1.1
1.0, 1.3
1.1
1.0
0.9, 1.3
2
301
47.8
1.1
0.9, 1.4
1.1
0.9, 1.5
4,593
47.1
1.0
157
52.2
0.9
14
2,992
40.9
1.0
<14
1,758
58.1
1.6
Civil status
Married or living with a partner
Single
1.0
0.6, 1.2
0.8
0.6, 1.2
Mothers education (years)

1.0
1.4, 1.9
1.4
1.2, 1.6
Family income (Can$/year)

50,000
1,809
37.7
1.0
20,00049,999
2,468
53.3
1.5
1.3, 1.7
1.4
1.2, 1.6
<20,000
373
56.0
1.4
1.1, 1.8
1.2
0.9, 1.6
Unknown
100
40.0
0.9
0.6, 1.4
0.9
0.6, 1.3
No
3,870
45.9
1.0
Yes
880
53.3
1.0
4,260
46.7
1.0
490
52.0
1.1
3,397
48.9
1.0
1
1,353
43.3
0.9
1.0
Maternal smoking, third trimester

1.0
0.9, 1.2
0.9
0.8, 1.1
Maternal caffeine consumption, third

trimester (portions/week){
<15
15
1.0
0.9, 1.4
1.1
0.8, 1.0
0.9
0.9, 1.3
Maternal alcohol consumption, third

trimester (drinks/week)
1.0
0.8, 1.1
Table continues
were lower when pregnant women had modification of their

working conditions (69) or had access to more frequent or
longer antenatal leave (26, 69) and with those where higher
PTD risks were present among women who continued to
work later during pregnancy despite arduous working conditions (11, 28, 45, 70) or high strain (12, 45). In a study
comparing European countries, significant associations of

PTD with working conditions were observed only in countries where long prenatal leaves were infrequent (49).
As health conscious women are more likely to use preventive measures, one might question whether results are
confounded by the womans characteristics linked to
957
TABLE 2. Continued
Model excluding
occupational
conditions
Model including
occupational
conditions
Random sample
of eligible workers
of the source
population
(n 4,750)*
% using
legally
justied
preventive
measures
Odds
ratioy
No
4,695
47.2
1.0
Yes
55
52.7
1.4
No
4,706
47.2
1.0
Yes
44
54.6
1.3
No
4,433
46.9
1.0
Yes
317
52.4
1.2
No
4,586
47.2
1.0
Yes
164
49.4
1.1
No
2,613
45.7
1.0
Yes
2,137
49.2
1.1
1,624
30.7
1,720
43.0
1.7
1.5, 2.0
860
67.7
4.3
3.6, 5.2
410
78.3
7.1
5.5, 9.2
4
136
76.5
6.0
4.0, 9.1
95%
condence
interval
Odds
ratioz
95%
condence
interval
Over-the-counter drug use

1.0
0.8, 2.5
1.5
0.8, 2.7
Chronic hypertension
1.0
0.7, 2.5
1.4
0.7, 2.6
Diabetes
1.0
0.9, 1.5
1.3
1.0, 1.6
1.0
0.8, 1.5
1.0
0.7, 1.4
Complication during pregnancy**

1.0
1.0, 1.2
1.0
0.9, 1.2
Occupational conditions at beginning

of pregnancy (no.)yy
1.0
* The random sample of eligible workers of the source population (n 4,750) corresponds to 20% of the
noncases (the control group) plus a 20% random sample of women in the case group.
y Adjusted for all variables in the table except for the number of occupational conditions.
z Adjusted for all variables in the table, including number of occupational conditions.
Reference category.
{ One cup of coffee 1 portion; 1 cup of tea 0.76 portion; 1 glass of cola 0.44 portion.
# Examples of other chronic diseases are asthma, thyroid disorder, bowel inammatory disease, hypoglycemia,
kidney disease, heart diseases, epilepsy, and hypercholesterolemia.
** Examples of complications are bleeding, gestational diabetes, contractions, intrauterine growth retardation,
preeclampsia, and gestational hypertension.
yy Occupational conditions were dened as follows: more than 5 consecutive working days, irregular or shift-work
schedule, standing at least 7 hours/day mostly in one spot, sitting posture at least 3 hours/day with rare or no
possibility to stand, demanding posture at least 3 hours/day, whole-body vibrations, very hot or very cold
temperatures, and moderate-active or high job strain combined with low or moderate social support.
selective implementation of preventive measures (job withdrawal or reassignment). However, the associations shown
in table 2 do not support this argument. First, the number of
occupational exposures is the factor most strongly related to
the use of preventive measures. Second, after adjustment for
work conditions, recourse to preventive measures remains
more frequent in the youngest and less educated women and
in those with moderate or low family income, as well as in
those suffering from chronic hypertension or diabetes, and is
not affected by lifestyle. Therefore, it seems unlikely, although still possible, that personal characteristics linked to
a better pregnancy outcome increase the likelihood of reAm J Epidemiol 2007;166:951965
questing preventive measures and explain the suggested

benefit of preventive measures.
All variables linked to PTD or to recourse to preventive
measures were taken into account in the analyses leading to
tables 36. Pregnancy complications (e.g., bleeding, gestational hypertension) were not included as covariates because
they can be intermediate factors in the causal pathway linking occupational conditions to PTD. Nevertheless, the observed associations are weak, and the possibility of residual
confounding cannot be eliminated.
To minimize misclassification of outcomes, we crosschecked the case or control status established on the basis
Other chronic disease#
Elimination of condition by legally justied preventive measures during pregnancy
Condition present at beginning of

pregnancy
Cases
(no.)y
Controls
(no.)y
Odds
ratio
Early
95%
condence
interval
Cases
(no.)
Controls
(no.)
Odds
ratio
Late
95%
condence
interval
Cases
(no.)
Controls
(no.)
Odds
ratio
Not eliminated
95%
condence
interval
Cases
(no.)
Controls
(no.)
Odds
ratio
95%
condence
interval
Hours worked/weekz
2034
325
1,335
1.0
3540
775
2,729
1.2
1.0, 1.4
180
624
1.1
0.9, 1.4
81
328
1.0
0.7, 1.3
514
1,777
1.2*
1.0, 1.5
>40
135
433
1.2
1.0, 1.6
48
110
1.6*
1.1, 2.4
17
57
1.2
0.7, 2.0
70
266
1.1
0.8, 1.4
1,150
4,228
1.0
68
200
1.2
0.9, 1.6
11
40
0.9
0.5, 1.8
19
1.1
0.4, 2.9
51
141
1.3
1.0, 1.9
Consecutive days workedz,{

15
>5
Work schedulez
Day only
757
2,772
1.0
Evening# but no night** hours
345
1,218
1.0
0.8, 1.2
125
386
1.1
0.9, 1.4
52
203
0.9
0.6, 1.2
168
629
1.0
0.8, 1.2
Night** hours
126
483
0.9
0.7, 1.2
72
306
0.8
0.6, 1.1
20
70
1.0
0.6, 1.7
34
107
1.1
0.8, 1.7
24
1,045
3,803
1.0
190
694
1.0
0.9, 1.3
79
307
1.0
0.7, 1.3
26
118
0.8
0.5, 1.3
85
269
1.2
0.9, 1.6
546
1,973
1.0
644
2,422
1.1
0.9, 1.3
40
105
1.4
0.9, 2.0
10
24
1.5
0.7, 3.3
17
0.9
0.3, 2.6
24
64
1.6
0.9, 2.6
<2
296
1,165
1.0
23
299
1,010
1.2
1.0, 1.4
35
99
1.3
0.9, 2.0
17
64
1.0
0.6, 1.7
247
847
1.2
0.9, 1.4zz
46
356
1,399
0.9
0.8, 1.2
131
524
0.9
0.7, 1.2
51
225
0.8
0.6, 1.2
174
650
1.0
0.8, 1.3
7
284
923
1.1
0.9, 1.3
158
499
1.1
0.8, 1.4
52
181
1.0
0.7, 1.5
74
243
1.0
0.8, 1.4
Standing <2 hours/day
296
1,165
1.0
Standing 23 hours/day
299
1,010
1.1
Unknown
Schedule regularityz
Regular
Irregular or shift work
Sitting, by possibility to stand{,yy
Sitting <3 hours/day
Sitting 3 hours/day
Possibility to stand: anytime, often
Possibility to stand: never, rarely
Standing (hours/day)z
0.9, 1.4
In one spot: <50%
206
833
0.9
0.7, 1.1
82
336
0.8
0.6, 1.1
28
126
0.8
0.5, 1.2
96
371
1.0
0.7, 1.3
In one spot: 50%
149
565
1.0
0.7, 1.2
63
242
0.9
0.7, 1.3
22
99
0.8
0.5, 1.3
64
224
1.0
0.7, 1.4
In one spot: <50%
137
483
1.0
0.7, 1.2
86
272
1.1
0.8, 1.4
24
95
0.9
0.5, 1.4
27
116
0.8
0.5, 1.2
In one spot: 50%
145
433
1.1
0.9, 1.5
74
231
1.0
0.8, 1.4
28
79
1.2
0.8, 2.0
43
123
1.2
0.8, 1.8
Standing, by % of time remaining

in one spot
958 Croteau et al.
TABLE 3. Adjusted odds ratios and 95% condence intervals for preterm delivery, by occupational condition at beginning of pregnancy and early (<24 weeks), late (24
weeks), or no elimination of condition by legally justied preventive measures during pregnancy, among workers giving birth in Quebec, Canada, between January 1997 and
March 1999
498
2,023
>02.9
425
1,557
1.0
3
312
917
623
2,341
1.0
16
256
881
1.0
7
346
1,247
0.9
10
28
No
1,125
4,191
Yes
110
306
Never, rarely
1,082
3,977
1.0
Often, always
153
520
1.0
Never, rarely
1,178
4,297
1.0
Often, always
57
200
0.9
246
986
1.0
1.2
1.0, 1.4
125
461
1.2
0.9, 1.5
57
198
1.2
0.9, 1.7
243
898
1.1
0.9, 1.3
1.4*
1.2, 1.7
125
404
1.3*
1.0, 1.7
49
157
1.3
0.9, 1.8
138
356
1.7*
1.3, 2.1
0.9, 1.2
79
269
1.0
0.7, 1.3
34
128
0.9
0.6, 1.4
143
484
1.1
0.9, 1.3
0.8, 1.1
165
652
0.8
0.7, 1.1
52
181
1.0
0.7, 1.4
129
414
1.0
0.8, 1.3
1.0, 1.6
33
87
1.3
0.8, 1.9
46
0.6
0.3, 1.3
68
173
1.4*
1.1, 1.9
0.8, 1.2
57
219
0.9
0.6, 1.2
20
78
0.8
0.5, 1.4
76
223
1.2
0.9, 1.6
0.7, 1.3
22
93
0.8
0.5, 1.2
28
0.6
0.2, 1.5
30
79
1.3
0.8, 2.0
Lifting (kg)z
Unknown
Whole-body vibrationsz,{
1.0
1.3*
Very hot temperaturez,{,##
Very cold temperaturez,{,***
Job strain, by social support{,yyy,zzz

Low strain
Moderate-passive strain
High support
117
1.0
0.6, 1.5
1,007
1.0
0.8, 1.3
High support
116
459
1.0
0.8, 1.3
16
61
0.9
0.5, 1.7
10
35
1.0
0.5, 2.1
90
363
1.0
0.7, 1.3
Low or moderate support
207
705
1.2
0.9, 1.5
39
130
1.2
0.8, 1.7
13
54
0.9
0.5, 1.6
155
521
1.2
1.0, 1.5
Moderate-active strain
High strain
High support
33
165
0.8
0.5, 1.1
58
0.4
0.2, 1.0
18
0.8
0.3, 2.3
22
89
0.9
0.6, 1.6
304
942
1.3*
1.0, 1.5
86
315
1.0
0.7, 1.3
40
123
1.2
0.8, 1.8
178
504
1.4*
1.1, 1.8
High or moderate strain, support not

applicable
26
75
1.3
0.8, 2.2
Strain or support unknown
11
41
959
* p < 0.05.
y Totals vary because of missing data; 1,235 cases and 4,497 controls are included in the analysis.
z Adjusted for all other occupational conditions listed at the beginning of pregnancy.
Reference category.
{ Considered eliminated if withdrawal from work occurred.
# Evening hours are from 6:00 p.m. to 10:59 p.m.
** Night hours are from 11:00 p.m. to 5:59 a.m.
yy Adjusted for all other occupational conditions listed at the beginning of pregnancy, education (14 years, <14 years), and parity (1, 23, >3).
zz After withdrawal of women who were sitting 3 hours or more with rare or no possibility to stand (n 145), the odds ratio became 1.1.
Adjusted for all other occupational conditions listed at the beginning of pregnancy and parity (1, 23, >3).
{{ Bending, squatting, arms raised above shoulder level, or other demanding posture.
## Dened as very hot temperature or enough hot to cause transpiration.
*** Dened as very cold temperature or enough cold to wear coat and gloves.
yyy Adjusted for all other occupational conditions listed at the beginning of pregnancy, prior adverse pregnancy outcome (0, 1, 2), and parity (1, 23, >3).
zzz Low job strain high decision latitude and low psychological demand; moderate-passive job strain low decision latitude and low psychological demand; moderate-active job strain high decision latitude and
high psychological demand; high job strain low decision latitude and high psychological demand.
Social support was not applicable for workers without coworkers and supervisors.
28
264
Demanding posture (hours/day),{{
960 Croteau et al.
TABLE 4. Adjustedy odds ratios and 95% condence

intervals for very preterm delivery (<34 weeks), by
occupational condition at beginning of pregnancy, among
workers giving birth in Quebec, Canada, between January 1997
and March 1999
TABLE 4. Continued
Condition present at
beginning of pregnancy
95%
condence
interval
Cases
(no.)z
Controls
(no.)z
Odds
ratio
117
2,340
1.0
16
47
878
1.0
0.7, 1.4
7
64
1,243
0.9
0.6, 1.3
28
No
205
4,184
Yes
27
305
Never, rarely
200
3,969
1.0
Often, always
32
520
1.1
Never, rarely
221
4,289
1.0
Often, always
11
200
0.9
42
985
1.0
117
1.0
0.4, 2.5
50
1,005
1.1
0.7, 1.7
High support
19
459
0.9
0.5, 1.6
Low or moderate
support
36
702
1.2
0.8, 1.9
164
1.3
0.6, 2.7
65
941
1.7*
1.1, 2.5
1.2
0.4, 3.5
Lifting (kg)
Condition present at
Cases
(no.)z
Controls
(no.)z
Odds
ratio
95%
condence
interval
Hours worked/week
2034
3540
>40
64
1,334
Unknown
1.0
145
2,724
1.0
0.8, 1.5
23
431
1.0
0.6, 1.7
214
4,220
1.0
15
200
1.5
69
Consecutive days worked{

15
>5
Unknown
0.8, 2.6
2,769
1.0
1,216
0.9
0.6, 1.2
Night** hours
20
480
0.6
0.4, 1.1
24
193
3,797
1.0
39
692
1.2
High support
0.8, 1.8
Low or moderate
support
Moderate-active strain
106
1,978
1.0
Sitting 3 hours/day
Possibility to stand:
anytime, often
119
2,423
0.9
0.7, 1.3
Possibility to stand:
never, rarely
105
1.1
0.5, 2.5
High strain
High support
Standing (hours/day)
Low or moderate
support
<2
60
1,165
1.0
23
48
1,010
0.9
0.6, 1.4
High or moderate strain,

support not
applicable##
75
46
74
1,399
1.1
0.7, 1.6
Strain or support unknown
41
7
50
923
1.0
0.7, 1.7
Standing, by % of time
remaining in
one spot
Standing <2 hours/day
60
1,165
1.0
48
1,010
0.9
0.6, 1.4
In one spot: <50%
42
833
1.0
0.6, 1.6
In one spot: 50%
32
565
1.1
0.7, 1.8
In one spot: <50%
22
483
0.9
0.5, 1.6
In one spot: 50%
28
433
1.2
0.7, 2.0
Demanding posture
(hours/day)yy
0
0.4, 1.7
Moderate-passive strain
Sitting, by possibility to
stand{
Sitting <3 hours/day
0.7, 1.7
Job strain, by social

support{,{{
Low strain
Schedule regularity
Irregular or shift work
1.1, 2.7
102
2,020
1.0
>02.9
78
1,555
1.0
0.7, 1.4
3
52
914
1.0
0.7, 1.5
Table continues
* p < 0.05.
y Adjusted for other occupational conditions at the beginning of
pregnancy and parity (1, 23, >3).
z Totals vary because of missing data.
Reference category.
{ Considered eliminated if preventive withdrawal from work
occurred.
# Evening hours are from 6:00 p.m. to 10:59 p.m.
** Night hours are from 11:00 p.m. to 5:59 a.m.
yy Bending, squatting, arms raised above shoulder level, or other
demanding posture.
zz Dened as very hot temperature or enough hot to cause
transpiration.
Dened as very cold temperature or enough cold to wear coat
and gloves.
{{ Low job strain high decision latitude and low psychological
demand; moderate-passive job strain low decision latitude and low
psychological demand; moderate-active job strain high decision
latitude and high psychological demand; high job strain low
decision latitude and high psychological demand.
## Social support was not applicable for workers without coworkers
and supervisors.
146
66
Regular
1.8*
Very cold temperature{,
Evening# but no
night** hours
Unknown
1.0
Very hot temperature{,zz
Work schedule
Day only
Whole-body vibrations{
961
TABLE 5. Adjusted odds ratios and 95% condence intervals for preterm delivery and very preterm delivery, by the proportion of
workers exposed to the condition based on job title,y among workers giving birth in Quebec, Canada, between January 1997 and
March 1999
Preterm delivery (<37 weeks)
Occupational conditionz
and proportion of
workers exposed (%)
Cases
(total 1,242)
(no.){
Odds
ratio
Very preterm delivery (<34 weeks)
95%
condence
interval
Cases
(total 233)
(no.){
Odds
ratio
95%
condence
interval
3,485
971
57
972
250
20
1.0
0.9
1.1
0.8, 1.1
0.7, 2.0
175
51
7
1.0
1.1
2.6*
0.8, 1.5
1.1, 6.1
1,124
3,220
169
297
896
49
1.0
1.0
1.1
0.9, 1.2
0.8, 1.6
50
175
8
1.0
1.2
1.0
0.8, 1.7
0.5, 2.4
4,048
291
168
1,086
79
68
1.0
0.9
1.2
0.7, 1.2
0.9, 1.7
200
19
13
1.0
1.3
1.1
0.8, 2.1
0.6, 2.2
2,310
2,136
67
634
593
15
1.0
0.9
0.6
0.8, 1.0
0.3, 1.1
116
115
2
1.0
1.0
0.5
0.7, 1.3
0.1, 1.9
1,263
2,678
285
287
347
708
86
101
1.0
0.9
0.9
1.3
0.8, 1.1
0.7, 1.3
0.9, 1.7
60
141
15
17
1.0
1.1
0.8
1.3
0.8, 1.7
0.4, 1.5
0.7, 2.4
2,975
1,365
144
29
799
377
51
15
1.0
1.0
1.2
2.0*
0.9, 1.2
0.8, 1.6
1.0, 3.7
148
69
13
3
1.0
1.1
1.7
1.9
0.8, 1.4
0.9, 3.2
0.6, 6.6
1,630
2,579
304
457
708
77
1.0
0.9
0.8
0.8, 1.1
0.6, 1.1
92
127
14
1.0
0.8
0.7
0.6, 1.1
0.4, 1.2
486
3,375
603
49
112
939
182
9
1.0
1.2
1.4*
0.8
0.9, 1.5
1.1, 1.9
0.4, 1.7
22
173
35
3
1.0
1.1
1.3
1.3
0.7, 1.7
0.8, 2.4
0.4, 4.7
* p < 0.05.
y The population was categorized in 60 job titles.
z Exposure that showed odds ratios of 1.2 (table 3) if not eliminated by preventive measure (when applicable, the highest exposure level was
retained).
The proportion of workers exposed in each job title was established with subjects in the control group and a 20% random sample of the cases
to be representative of the source population.
{ Totals vary because of missing data.
# Adjusted for other individual occupational conditions at the beginning of pregnancy (hours worked/week, work schedule, schedule regularity,
standing, demanding posture, lifting, vibrations, very hot temperature, very cold temperature, and job strain by social support).
** Reference category.
yy Adjusted for other individual occupational conditions at the beginning of pregnancy (hours worked/week, work schedule, schedule regularity,
standing, demanding posture, lifting, vibrations, very hot temperature, very cold temperature, and job strain by social support), education (14 years,
<14 years), and parity (1, 23, >3).
zz Adjusted for other individual occupational conditions at the beginning of pregnancy (hours worked/week, work schedule, schedule regularity,
standing, demanding posture, lifting, vibrations, very hot temperature, very cold temperature, and job strain by social support) and parity (1, 23, >3).
Bending, squatting, arms raised above shoulder level, or other demanding posture.
{{ Very hot and very cold temperature are two distinct occupational conditions in table 3 but are combined in the job title analysis.
## Adjusted for other individual occupational conditions at the beginning of pregnancy (hours worked/week, work schedule, schedule regularity,
standing, demanding posture, lifting, vibrations, very hot temperature, very cold temperature, and job strain by social support), prior adverse
pregnancy outcome (0, 1, 2), and parity (1, 23, >3).
>5 consecutive days worked#

04**
544
4574
Irregular or shift work#
04**
544
4554
Sitting 3 hours/day and never or rare
possibility to standyy
04**
514
1524
Standing 7 hours/day 50% of time in
one spotzz
04**
544
4574
3 hours/day in demanding posturezz,
04**
544
4564
6584
Whole-body vibrations#
04**
534
3544
4554
Often or always in very hot or
very cold temperature#,{{
04**
544
4574
High strain with low or moderate support##
04**
534
3544
4564
Controls
(total 4,513)
(no.){
962 Croteau et al.
TABLE 6. Odds ratios and 95% condence intervals for preterm delivery and very preterm delivery, by cumulative index of
occupational conditionsy at the beginning of pregnancy and early (<24 weeks), late (24 weeks), or no elimination of conditions by
legally justied preventive measures during pregnancy, among workers giving birth in Quebec, Canada, between January 1997 and
March 1999
Preterm delivery (<37 weeks)
Control mothers
(total 4,513)
Case mothers
(total 1,242)
Odds
ratio
Very preterm delivery (<34 weeks)
95%
condence
interval
Case mothers
(total 233)
No.z
59
25.4
Odds
ratio
95%
condence
interval
No.z
No.z
0{
1,549
34.4
357
29.0
1.0
1,637
36.3
449
36.4
1.2*
1.0, 1.4
89
38.4
1.4*
1.0, 2.0
814
18.1
238
19.3
1.2*
1.0, 1.5
46
19.8
1.5
1.0, 2.2
Index at beginning of pregnancy
390
8.7
131
10.6
1.4*
1.1, 1.8
26
11.2
1.7*
1.1, 2.8
46
117
2.6
58
4.7
2.0*
1.4, 2.8
12
5.2
2.7*
1.4, 5.1
Trend for no. of conditions at

2
18.32;
vtrend
p < 0.0001
2
vtrend
10.08;
p 0.0015
0{
1,548
34.4
357
29.0
1.0
59
25.4
1
Eliminated early
352
7.8
88
7.1
1.0
0.8, 1.3
19
8.2
1.4
0.8, 2.3
Uncertain or late elimination
338
7.5
93
7.5
1.1
0.9, 1.5
11
4.7
0.8
0.4, 1.6
Not eliminated
947
21.0
268
21.7
1.2*
1.0, 1.5
59
25.4
1.7*
1.1, 2.4
2
Eliminated early
306
6.8
65
5.3
0.8
0.6, 1.1
14
6.0
1.1
0.6, 2.1
Uncertain, partial, or total late elimination
257
6.2
76
6.2
1.2
0.9, 1.6
11
4.7
1.0
0.5, 2.0
Not eliminated
251
7.9
97
7.9
1.6*
1.2, 2.1
21
9.1
2.2*
1.3, 3.6
0.5, 2.3
3
Eliminated early
164
4.5
55
4.5
1.3
1.0, 1.9
3.0
1.0
147
3.2
39
3.2
1.1
0.8, 1.6
11
4.7
2.0*
1.0, 3.9
79
3.0
37
3.0
1.9*
1.3, 2.9
3.5
2.6*
1.2, 5.7
0.6, 5.3
Not eliminated
46
Eliminated early
54
1.2
26
2.1
1.9*
1.1, 3.1
1.7
1.8
39
0.9
22
1.8
2.3*
1.3, 4.0
2.6
4.0*
1.6, 9.8
Not eliminated
24
0.5
10
0.8
1.6
0.8, 3.4
0.9
1.9
0.4, 8.3
Trend for no. of conditions eliminated early

Trend for no. of conditions with uncertain,
partial, or total late elimination
Trend for no. of conditions not eliminated
2
2.93;
vtrend
p 0.0872
2
vtrend
0.52;
p 0.4719
2
5.34;
vtrend
p 0.0209
2
vtrend
6.95;
p 0.0084
2
vtrend
19.31;
p < 0.0001
2
vtrend
12.11;
p 0.0005
* p < 0.05.
y No. of the following occupational conditions present at the beginning of pregnancy: more than 5 consecutive working days, irregular or shiftwork schedule, standing at least 7 hours/day mostly in one spot, sitting posture at least 3 hours/day with rare or no possibility to stand, demanding
posture at least 3 hours/day, whole-body vibrations, very hot or very cold temperatures, and moderate-active or high job strain combined with low
or moderate social support.
z Totals vary because of missing data.
Adjusted for education (14 years, <14 years).
{ Reference category.
# Adjusted for education (<14 years, 14 years), prior adverse pregnancy outcome (0, 1, 2), parity (1, 23, >3), mothers height (<158 cm,
158164 cm, >164 cm), and diabetes (yes, no).
Index by recourse to legally justied

preventive measures to
eliminate indexed conditions#
ACKNOWLEDGMENTS
Funding for this study was provided by Health Canada

under the National Health Research and Development
Program. The study was conducted with the support of
the Quebec Regional Public Health Direction. C. Brisson
holds a research scientist award from the Canadian
Institutes of Health Research.
The authors thank Dr. S. Montreuil, Dr. L. Punnett, and
Dr. L. Patry, ergonomists, for their scientific input in
questionnaire development; G. Bergeron for assistance in
data processing; and M. Desgagne, S. Mercier, and C.
Pelletier for conducting telephone interviews. The authors
would also like to thank the Regional Public Health
Directions.
Conflict of interest: none declared.
REFERENCES
1. Luke B, Papiernik E. The effects of lifestyle on prematurity. In:
Elder MG, Laumont RS, Romero R, eds. Preterm labor. New
York, NY: Churchill Livingstone, 1997:12752.
2. Forslund M, Bjerre I. Follow-up of preterm children. II.
Growth and development at four years of age. Early Hum Dev
1990;24:10718.
3. Berkowitz GS, Papiernik E. Epidemiology of preterm birth.
Epidemiol Rev 1993;15:41443.
4. Creasy RK, Resnik R. Preterm labor and delivery. In: Maternalfetal medicine. 4th ed. Philadelphia, PA: WB Saunders Co,
1999:498531.
5. Kramer MS. Intrauterine growth and gestational duration determinants. Pediatrics 1987;80:50211.
6. Meis PJ, Michielutte R, Peters TJ, et al. Factors associated with
preterm birth in Cardiff, Wales. II. Indicated and spontaneous
preterm birth. Am J Obstet Gynecol 1995;173:597602.
7. Moutquin JM. A retrospective assessment of Canadian preterm

birth prevention efforts: 19791999. Matern Child Health J
1999;3:937.
8. Olsen P, Laara E, Rantakallio P, et al. Epidemiology of preterm
delivery in two birth cohorts with an interval of 20 years. Am J
Epidemiol 1995;142:118493.
9. Stubblefield PG. Causes and prevention of premature birth: an
overview. In: Fuchs AR, Fuchs F, Stubblefield PG, eds. Preterm birth: causes, prevention, and management. New York,
NY: McGraw-Hill, 1993:339.
10. Ahlborg G Jr, Bodin L, Hogstedt C. Heavy lifting during
pregnancya hazard to the fetus? A prospective study. Int J
Epidemiol 1990;19:907.
11. Henriksen TB, Hedegaard M, Secher NJ. The relation between
psychosocial job strain, and preterm delivery and low birthweight for gestational age. Int J Epidemiol 1994;23:76474.
12. Henriksen TB, Hedegaard M, Secher NJ, et al. Standing at
work and preterm delivery. Br J Obstet Gynaecol 1995;
102:198206.
13. Hickey CA, Cliver SP, Mulvihill FX, et al. Employmentrelated stress and preterm delivery: a contextual examination.
Public Health Rep 1995;110:41018.
14. Klebanoff MA, Shiono PH, Carey JC. The effect of physical
activity during pregnancy on preterm delivery and birth
weight. Am J Obstet Gynecol 1990;163:14506.
15. Launer LJ, Villar J, Kestler E, et al. The effect of maternal
work on fetal growth and duration of pregnancy: a prospective
study. Br J Obstet Gynaecol 1990;97:6270.
16. Magann EF, Evans SF, Chauhan SP, et al. The effects of
standing, lifting and noise exposure on preterm birth, growth
restriction, and perinatal death in healthy low-risk working
military women. J Matern Fetal Neonatal Med 2005;18:
15562.
17. Misra DP, Strobino DM, Stashinko EE, et al. Effects of
physical activity on preterm birth. Am J Epidemiol 1998;
147:62835.
18. Nguyen N, Savitz DA, Thorp JM. Risk factors for preterm
birth in Vietnam. Int J Gynaecol Obstet 2004;86:708.
19. Pompeii LA, Savitz DA, Evenson KR, et al. Physical exertion
at work and the risk of preterm delivery and small-forgestational-age birth. Obstet Gynecol 2005;106:127988.
20. Teitelman AM, Welch LS, Hellenbrand KG, et al. Effect of
maternal work activity on preterm birth and low birth weight.
Am J Epidemiol 1990;131:10413.
21. Tuntiseranee P, Geater A, Chongsuvivatwong V, et al. The
effect of heavy maternal workload on fetal growth retardation
and preterm delivery. A study among southern Thai women.
J Occup Environ Med 1998;40:101321.
22. Tuntiseranee P, Olsen J, Chongsuvivatwong V, et al. Socioeconomic and work related determinants of pregnancy outcome in southern Thailand. J Epidemiol Community Health
1999;53:6249.
23. Zhu JL, Hjollund NH, Olsen J. Shift work, duration of pregnancy, and birth weight: the National Birth Cohort in Denmark. Am J Obstet Gynecol 2004;191:28591.
24. Newman RB, Goldenberg RL, Moawad AH, et al. Occupational fatigue and preterm premature rupture of membranes.
National Institute of Child Health and Human Development
Maternal-Fetal Medicine, Units Network. Am J Obstet Gynecol 2001;184:43846.
25. Bodin L, Axelsson G, Ahlborg G Jr. The association of shift
work and nitrous oxide exposure in pregnancy with birth
weight and gestational age. Epidemiology 1999;10:42936.
26. Ceron-Mireles P, Harlow SD, Sanchez-Carrillo CI. The risk of
prematurity and small-for-gestational-age birth in Mexico
of birth certificates with the status derived from interview.

Discordance was rare, and information could be checked
against hospital records for 74 percent of these newborns.
The high participation rate and the large number of potential
confounders considered in the analyses are additional characteristics that enhance the validity of our results.
In conclusion, our results show higher PTD risk for pregnant working women exposed to demanding posture for at
least 3 hours per day, whole-body vibrations, high job strain
combined with low or moderate social support, and a cumulative index composed of nine work conditions. Women who
experienced a change in these work conditions following
recourse to preventive job withdrawal or reassignment had
lower risks of PTD and very PTD than those who did not.
Although the associations are small, most of them are consistent for PTD and very PTD. A secondary analysis based
on job title supports the view that recall bias does not explain the results. As experimental studies to assess changes
in work conditions during pregnancy are not likely to be
carried out for ethical and feasibility considerations, our
observational study despite its limitations provides relevant
information on this question.
963
964 Croteau et al.
27.
28.
29.
30.
31.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47. Luke B, Mamelle N, Keith L, et al. The association between

occupational factors and preterm birth: a United States nurses
study. Research Committee of the Association of Womens
Health, Obstetric, and Neonatal Nurses. Am J Obstet Gynecol
1995;173:84962.
48. Mamelle N, Munoz F. Occupational working conditions and
preterm birth: a reliable scoring system. Am J Epidemiol
1987;126:1502.
49. Saurel-Cubizolles MJ, Zeitlin J, Lelong N, et al. Employment,
working conditions, and preterm birth: results from the Europop case-control survey. J Epidemiol Community Health
2004;58:395401.
50. Escriba-Aguir V, Perez-Hoyos S, Saurel-Cubizolles MJ.
Physical load and psychological demand at work during
pregnancy and preterm birth. Int Arch Occup Environ Health
2001;74:5838.
51. Bonzini M, Coggon D, Palmer KT. Risk of prematurity, low
birthweight, and pre-eclampsia in relation to working hours
and physical activities: a systematic review. Occup Environ
Med 2007;64:22843.
52. Mozurkewich EL, Luke B, Avni M, et al. Working conditions
and adverse pregnancy outcome: a meta-analysis. Obstet Gynecol 2000;95:62335.
53. Mamelle N, Munoz F, Collin D, et al. Fatigue professionnelle
et prematurite. (In French). Arch Fr Mal Prof 1981;42:
21116.
54. Marbury MC, Linn S, Monson RR, et al. Work and pregnancy.
J Occup Med 1984;26:41521.
55. Gouvernement du Quebec. Loi sur la sante et la securite du
travail. Chap 3. Droits et obligations. (In French). Quebec,
France: Gouvernement du Quebec, 1990.
56. Croteau A. Effet des agresseurs ergonomiques et de leur
elimination durant la grossesse sur les risques daccouchement
avant terme et dinsuffisance de poids pour lage gestationnel.
(In French). (Dissertation). Quebec, Canada: Department of
Social and Preventive Medicine, Laval University, 2002.
57. Ahlborg GA Jr. Validity of exposure data obtained by
questionnaire. Two examples from occupational reproductive studies. Scand J Work Environ Health 1990;16:
2848.
58. Baty D, Buckle PW, Stubbs DA. Posture recording by direct
observation, questionnaire assessment and instrumentation:
a comparison based on a recent field study. In: Corlett N,
Wilson J, eds. The ergonomics of working postures. London,
United Kingdom: Taylor and Francis, 1986.
59. Burdorf A, Laan J. Comparison of methods for the assessment
of postural load on the back. Scand J Work Environ Health
1991;17:4259.
60. Eskenazi B, Pearson K. Validation of a self-administered
questionnaire for assessing occupational and environmental
exposures of pregnant women. Am J Epidemiol 1988;128:
111729.
61. Pope DP, Silman AJ, Cherry NM, et al. Validity of a selfcompleted questionnaire measuring the physical demands
of work. Scand J Work Environ Health 1998;24:37685.
62. Rossignol M, Baetz J. Task-related risk factors for spinal injury: validation of a self-administered questionnaire on hospital employees. Ergonomics 1987;30:153140.
63. Wiktorin C, Karlqvist L, Winkel J. Validity of self-reported
exposures to work postures and manual materials handling.
Stockholm MUSIC I Study Group. Scand J Work Environ
Health 1993;19:20814.
64. Karasek R, Theorell T. Healthy work: stress, productivity and
the reconstruction of working life. New York, NY: Basic
Books, 1990.
32.
City: the effects of working conditions and antenatal leave.

Am J Public Health 1996;86:82531.
Estryn M, Kaminski M, Franc M, et al. Grossesse et conditions
de travail en milieu hospitalier. (In French). Rev Fr Gynecol
Obstet 1978;73:62531.
Fortier I, Marcoux S, Brisson J. Maternal work during pregnancy and the risks of delivering a small-for-gestational-age or
preterm infant. Scand J Work Environ Health 1995;21:41218.
Homer CJ, James SA, Siegel E. Work-related psychosocial
stress and risk of preterm, low birthweight delivery. Am J
Public Health 1990;80:1737.
Kaminski M, Saurel-Cubizolles MJ. Les femmes enceintes
travaillant a` lhopital: projet de recherche pour levaluation
dune politique de modification des conditions de travail. In:
Papiernik E, Breart G, Spira N, eds. Prevention de la naissance
prematuree. (In French). Vol 138. Paris, France: Colloque
INSERM, 1986:23960.
Klebanoff MA, Shiono PH, Rhoads GG. Outcomes of pregnancy in a national sample of resident physicians. N Engl J
Med 1990;323:10405.
Lerman Y, Jacubovich R, Green MS. Pregnancy outcome
following exposure to shortwaves among female physiotherapists in Israel. Am J Ind Med 2001;39:499504.
Mamelle N, Laumon B, Lazar P. Prematurity and occupational
activity during pregnancy. Am J Epidemiol 1984;119:30922.
McDonald AD, McDonald JC, Armstrong B, et al. Prematurity
and work in pregnancy. Br J Ind Med 1988;45:5662.
Peoples-Sheps MD, Siegel E, Suchindran CM, et al. Characteristics of maternal employment during pregnancy: effects on
low birthweight. Am J Public Health 1991;81:100712.
Saurel-Cubizolles MJ, Kaminski M, Rumeau-Rouquette C.
Activite professionnelle des femmes enceintes, surveillance
prenatale et issue de la grossesse. (In French). J Gynecol
Obstet Biol Reprod (Paris) 1982;11:95967.
Saurel-Cubizolles MJ, Kaminski M. Pregnant womens
working conditions and their changes during pregnancy: a national study in France. Br J Ind Med 1987;44:23643.
Saurel-Cubizolles MJ, Subtil D, Kaminski M. Is preterm delivery still related to physical working conditions in pregnancy? J Epidemiol Community Health 1991;45:2934.
Savitz DA, Olshan AF, Gallagher K. Maternal occupation and
pregnancy outcome. Epidemiology 1996;7:26974.
Uehata T, Sasakawa N. The fatigue and maternity disturbances
of night workwomen. J Hum Ergol (Tokyo) 1982;11(suppl):
46574.
Xu X, Ding M, Li B, et al. Association of rotating shiftwork
with preterm births and low birth weight among never smoking
women textile workers in China. Occup Environ Med 1994;
51:4704.
Saurel-Cubizolles MJ, Kaminski M, Llado-Arkhipoff J, et al.
Pregnancy and its outcome among hospital personnel according to occupation and working conditions. J Epidemiol Community Health 1985;39:12934.
Berkowitz GS, Kelsey JL, Holford TR, et al. Physical activity
and the risk of spontaneous preterm delivery. J Reprod Med
1983;28:5818.
Brandt LP, Nielsen CV. Job stress and adverse outcome of
pregnancy: a causal link or recall bias? Am J Epidemiol
1992;135:30211.
Brett KM, Strogatz DS, Savitz DA. Employment, job strain,
and preterm delivery among women in North Carolina. Am J
Public Health 1997;87:199204.
Hartikainen-Sorri AL, Sorri M. Occupational and sociomedical factors in preterm birth. Obstet Gynecol 1989;74:
1316.
65. Brisson C, Blanchette C, Guimont C, et al. Reliability and

validity of the French version of the 18-item Karasek
Job Content Questionnaire. Work Stress 1998;12:32236.
66. Karasek R, Brisson C, Kawakami N, et al. The Job Content
Questionnaire (JCQ): an instrument for internationally comparative assessments of psychosocial job characteristics.
J Occup Health Psychol 1998;3:32255.
67. Rothman KJ, Greenland S. Modern epidemiology. 2nd ed.
Philadelphia, PA: Lippincott Williams & Wilkins, 1998.
68. Florack EI, Pellegrino AE, Zielhuis GA, et al. Influence of
occupational physical activity on pregnancy duration and
965
birthweight. Scand J Work Environ Health 1995;21:

199207.
69. Mamelle N, Bertucat I, Munoz F. Pregnant women at work:
rest periods to prevent preterm birth? Paediatr Perinat Epidemiol 1989;3:1928.
70. Croteau A, Vezina M, Plante R. Issues de grossesses des
travailleuses sur le territoire du DSC du CHULevaluation de
limpact du retrait preventif de la travailleuse enceinte. (In
French). Presented at the 23rd International Congress of the
International Commission on Occupational Health, Montreal,
Canada, September 1990.

Am. J. Epidemiol. 2007 Croteau 951 65

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Am. J. Epidemiol. 2007 Croteau 951 65

Uploaded by

Copyright:

Available Formats

American Journal of Epidemiology

Vol. 166, No. 8

Agathe Croteau1,2, Sylvie Marcoux3, and Chantal Brisson3

National Institute of Public Health of Quebec, Quebec, QC, Canada.

Abbreviation: PTD, preterm delivery.

49), prolonged standing (1217, 1921, 2628, 33, 34, 36

Downloaded from http://aje.oxfordjournals.org/ by guest on October 26, 2016

952 Croteau et al.

Downloaded from http://aje.oxfordjournals.org/ by guest on October 26, 2016

Work Activity, Preventive Measures, and Preterm Delivery

whole set of covariates. Covariates were withdrawn one

Downloaded from http://aje.oxfordjournals.org/ by guest on October 26, 2016

The questionnaire documented the following working

954 Croteau et al.

* Totals vary because of missing data.

Downloaded from http://aje.oxfordjournals.org/ by guest on October 26, 2016

Mothers age (years)

Work Activity, Preventive Measures, and Preterm Delivery

We found that an increased risk of PTD was significantly

Downloaded from http://aje.oxfordjournals.org/ by guest on October 26, 2016

Of the occupational conditions present at the beginning of

956 Croteau et al.

Mothers age (years)

Previous adverse pregnancy

Mothers education (years)

Family income (Can$/year)

Maternal smoking, third trimester

Maternal caffeine consumption, third

Maternal alcohol consumption, third

were lower when pregnant women had modification of their

comparing European countries, significant associations of

Downloaded from http://aje.oxfordjournals.org/ by guest on October 26, 2016

Work Activity, Preventive Measures, and Preterm Delivery

Over-the-counter drug use

Complication during pregnancy**

Occupational conditions at beginning

questing preventive measures and explain the suggested

Downloaded from http://aje.oxfordjournals.org/ by guest on October 26, 2016

Other chronic disease#

Elimination of condition by legally justied preventive measures during pregnancy

Condition present at beginning of

Consecutive days workedz,{

Evening# but no night** hours

Standing <2 hours/day

In one spot: 50%

In one spot: 50%

Downloaded from http://aje.oxfordjournals.org/ by guest on October 26, 2016

Standing, by % of time remaining

958 Croteau et al.

Very hot temperaturez,{,##

Very cold temperaturez,{,***

Job strain, by social support{,yyy,zzz

Low or moderate support

High or moderate strain, support not

Strain or support unknown

Low or moderate support

Work Activity, Preventive Measures, and Preterm Delivery

Low or moderate support

Downloaded from http://aje.oxfordjournals.org/ by guest on October 26, 2016

Demanding posture (hours/day),{{

960 Croteau et al.

TABLE 4. Adjustedy odds ratios and 95% condence

Consecutive days worked{

High or moderate strain,