You are here

Article Text

Article menu
Workplace
Original article
Physically demanding work, fetal growth and the risk of adverse birth outcomes. The Generation R Study
  1. Claudia A Snijder1,2,
  2. Teus Brand3,
  3. Vincent Jaddoe1,4,5,
  4. Albert Hofman1,4,
  5. Johan P Mackenbach2,
  6. Eric A P Steegers6,
  7. Alex Burdorf2
  1. 1The Generation R Study Group, Erasmus Medical Center, Rotterdam, The Netherlands
  2. 2Department of Public Health, Erasmus Medical Center, Rotterdam, The Netherlands
  3. 3Netherlands Centre for Occupational Diseases, Coronel Institute of Occupational Health, Academic Medical Centre, University of Amsterdam, Amsterdam, The Netherlands
  4. 4Department of Epidemiology, Erasmus Medical Center, Rotterdam, The Netherlands
  5. 5Department of Paediatrics, Erasmus Medical Center, Rotterdam, The Netherlands
  6. 6Department of Obstetrics and Gynaecology, Erasmus Medical Center, Rotterdam, The Netherlands
  1. Correspondence to Professor Alex Burdorf, Department of Public Health, Erasmus University Medical Centre, PO Box 2040, 3000 CA Rotterdam, The Netherlands; a.burdorf{at}erasmusmc.nl

Abstract

Objectives Work-related risk factors, such as long work hours, and physically demanding work have been suggested to adversely influence pregnancy outcome. The authors aimed to examine associations between various aspects of physically demanding work with fetal growth in different trimesters during pregnancy and the risks of adverse birth outcomes.

Methods Associations between physically demanding work and fetal growth were studied in 4680 pregnant women participating in a population-based prospective cohort study from early pregnancy onwards in the Netherlands (2002–2006). Mothers who filled out a questionnaire during mid-pregnancy (response 77% of enrolment) were included if they conducted paid employment and had a spontaneously conceived singleton live born pregnancy. Questions on physical workload were obtained from the Dutch Musculoskeletal Questionnaire and concerned questions on lifting, long periods of standing or walking, night shifts and working hours. Fetal growth characteristics were repeatedly measured by ultrasound and were used in combination with measurements at birth.

Results There were no consistent significant associations between physically demanding work nor working hours in relation to small for gestational age, low birth weight or preterm delivery. Women exposed to long periods of standing had lower growth rates for fetal head circumference (HC), resulting in a reduction of approximately 1 cm (3%) of the average HC at birth. Compared with women working <25 h/week, women working 25–39 h/week and >40 h/week had lower growth rates for both fetal weight and HC, resulting in a difference of approximately 1 cm in HC at birth and a difference of 148–198 g in birth weight.

Conclusion Long periods of standing and long working hours per week during pregnancy seem to negatively influence intrauterine growth.

  • Physically demanding work
  • fetal growth
  • birth outcomes
  • occupation

https://doi.org/10.1136/oemed-2011-100615

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    What this paper adds

    • With the increased labour force participation, the likelihood that pregnant women may be exposed to chemical, physical and psychological risks at work increased.

    • Work-related risk factors, such as long work hours, and physically demanding work have been suggested to adversely influence pregnancy outcome.

    • This study shows that long periods of standing and long working hours per week during pregnancy may negatively influence intrauterine growth, such as fetal weight and HC.

    • Reducing occupational risk factors such as physical demands in the workplace during pregnancy may be beneficial for the health in the offspring.

    Introduction

    Developmental diseases, such as structural birth defects, functional alterations, growth restrictions and preterm delivery account for more than 25% of infant mortality and morbidity.1 ,2 Environmental exposures and lifestyle behaviours act at different stages of fetal development and may result in adverse birth outcomes, such as preterm birth, low birth weight, small for gestational age, certain congenital defects and fetal death.3–5 Although women in paid employment seem to have better pregnancy outcomes than those without paid jobs,6–8 certain work-related risk factors, such as exposure to chemicals,9 long working hours,7 ,10 high physical workload, prolonged standing11 and psychological job strain,12 ,13 have been suggested to adversely influence pregnancy outcome.

    Two reviews have summarised the literature on physical workload and adverse pregnancy outcomes. Mozurkewich et al14 performed a meta-analyses on 29 studies and concluded that physically demanding work may significantly increase risks of preterm delivery (pooled OR 1.22; 95% CI 1.16 to 1.29), small for gestational age (pooled OR 1.37; 95% CI 1.30 to 1.44) and hypertension or pre-eclampsia (pooled OR 1.60; 95% CI 1.30 to 1.96). A review by Bonzini et al15 on 49 studies described the relationship between five common occupational exposures (prolonged working hours, shift work, lifting, standing and heavy physical workload) and three major adverse outcomes, namely preterm delivery, low birth weight and pre-eclampsia/gestational hypertension. Due to the small effects, low population attributable fractions (PAFs), and conflicting results, mandatory restrictions were not justified.

    Variations in study findings may be due to differences in exposure assessment, definitions of physical workload and components of indices that were used to score physical workload and timing during pregnancy, the same activity may carry different risks if it occurred late in pregnancy compared with only a few weeks after conception. Despite the substantial body of evidence on physically demanding work and birth outcomes, it is still unclear how occupational activities of pregnant women should be managed. Furthermore, studies on physically demanding work have primarily focused on adverse birth outcomes, which are important from an obstetric point of view but are rather crude measures of fetal growth. To gain more insight in how physically demanding work influences birth outcomes, studies on fetal growth characteristics during pregnancy are needed in order to identify critical periods in which exposure is deleterious for fetal growth and development.

    The aims of this study were to examine associations between various aspects of physically demanding work with fetal growth in different trimesters during pregnancy and the risks of adverse birth outcomes.

    Materials and methods

    Study design

    This study was embedded in the Generation R Study, a population-based prospective cohort study from fetal life until young adulthood in Rotterdam, the Netherlands.16 ,17 Briefly, all pregnant women who had an expected delivery date between April 2002 and January 2006 and lived in the study area of Rotterdam were invited to participate. In total, 9778 pregnant women (response 61%) were enrolled in the study of which 8880 women were enrolled during pregnancy and another 898 at birth of their child. Extensive assessments were carried out in each trimester, including physical examinations, questionnaires, interviews and biological samples. The study was approved by the Medical Ethics Committee at Erasmus University Medical Centre Rotterdam, The Netherlands (MEC 198.782/2001/31).

    The occupational information required for this study was collected in the questionnaire completed during mid-pregnancy (send out at 30 weeks of gestation), which was filled out by 6830 women (77% of enrolment). The mean gestational age for completing the questionnaire was 30.8 weeks (SD 2.4 weeks). For this study, we selected women who were prenatally enrolled, with paid employment at the time of the questionnaire (5154 women), and we excluded women with twin pregnancies (76 women), with pregnancies of non-spontaneous origin (84 women) and with fetal death (4 women). For each woman, we included the first pregnancy within the Generation R cohort in our study, excluding women who participated more than once (310 women). The population for analysis consisted of 4680 women; the flowchart of the study population is depicted in supplementary figure 1.

    Fetal ultrasounds

    For this study, we used the ultrasound measures of fetal head circumference (HC) and estimated fetal weight (EFW) since these measures are essential characteristics to describe fetal growth. In our research facility, we measured fetal HC, abdominal circumference (AC) and femur length to the nearest millimetre using standardised ultrasound procedures in the second (median 20.5, minimum–maximum 18–25 weeks) and third (median 30.4, minimum–maximum 25.8–37 weeks) trimesters. Since use of the last menstrual period for pregnancy dating has several limitations,18 and a large number of women in our study population did not know the exact date of their last menstrual period (76%), we used crown–rump length for pregnancy dating until a gestational age of 12 weeks (2308 women) and biparietal diameter for pregnancy dating thereafter (2372 women) in all women.19 ,20 First trimester measurements (3459 women) were primarily used to establish gestational age and therefore not included in the growth analyses. EFW was calculated using the formula by Hadlock et al.21 The intraclass correlation coefficient of fetal growth measurements was 0.95, tested on 21 subjects.22

    Verburg et al showed that fetal growth reference curves for fetal weight and fetal HC during pregnancy typically have a parabolic pattern. Based on these reference curves, SD scores for all growth characteristics were constructed,18 reflecting the commonly used z-scores for child growth as proposed by WHO.23 The SD score indicates the relative position of the fetus on the observed distribution, for example, a SD score of 1 for fetal HC indicates for that particular child his HC measurement is larger than approximately 84% of all children. This approach enables linear analyses of the fetal growth characteristics since the reference curve is a curve with a mean SD score of 0.

    Birth outcomes

    Information about gender at birth, gestational age, weight, length and HC at birth was obtained from medical records and hospital registries. Low birth weight was defined as birth weight <2500 g Small for gestational age at birth was defined as a gestational age-adjusted birth weight below the 5th percentile in the whole study cohort (n=8880) (<−1.71 SD), and preterm birth was defined as a gestational age at <37 weeks at birth.

    Occupation and working conditions

    The mid-pregnancy questionnaire (send out at 30 weeks of gestation) contained questions about work status, occupation and working conditions and focused on the periconception and pregnancy period. The question on current work status, with seven categories (paid labour, self-employed, unemployed, disabled, homemaker, student or other), was used to select women with paid employment. The question on starting date of the current occupation provided information if women started working before pregnancy or somewhere during the first trimester of pregnancy. For the current study, we selected both women with paid employment who started before pregnancy and women who started working during the first trimester.

    The number of weekly working hours of the mothers with paid employment was assessed by means of an open question, ‘How many hours per week do you work?’ Working hours were categorised into ‘1–24’, ‘25–39’ and ‘40 or more hours a week’.7 The questions on physical workload were obtained from the Dutch Musculoskeletal Questionnaire and concerned questions on manually handling loads of 25 kg or more, long periods of walking, long periods of standing and night shifts. A four-point scale was used with ratings ‘seldom or never’, ‘occasionally’, ‘often’ and ‘very often’ during a regular workday. These factors were all considered as separate variables in the analyses since multicollinearity was not present (Spearman's correlation coefficient ρ=−0.14 to 0.23) except for long periods of standing and walking (ρ=0.56). We reclassified long periods of walking and standing, manual handling of load of 25 kg or more and night shifts into three categories, namely ‘seldom or never’ (reference group), ‘occasionally’ and ‘often/very often’.24 ,25

    In a postnatal questionnaire, we collected information in which pregnancy week women had stopped working and whether this was due to pregnancy complaints.

    Potential confounders

    The following variables were considered as possible confounders: maternal age, pre-pregnancy weight, height, educational level, ethnicity, parity, smoking, alcohol use and folic acid supplement use. Information about maternal age, educational level, ethnicity, parity and folic acid supplement use was obtained by questionnaire at enrolment in the study. Maternal smoking habits and alcohol use were assessed on the basis of three questionnaires in each trimester and classified as no, until pregnancy was known or during pregnancy.26 ,27 Maternal height was measured at intake in the study.

    Statistical analyses

    The associations between occupational risk factors and the risk of preterm delivery, small for gestational age and low birth weight were analysed with multiple logistic regression analyses. In all analyses, the reference group consisted of women who were not exposed to that particular physical risk factor. Second, cross-sectional analyses were performed using linear regression analysis to demonstrate the influence of physically demanding work on HC, AC and EFW in the second and third trimesters of pregnancy. Third, occupational risk factors associated with birth outcomes were selected for the longitudinal analyses of HC and weight (second and third trimester EFW and birth weight) using unbalanced repeated measurement analysis, which enables optimal use of the available data, taking into account correlations within subjects and assessing both time dependant and independent associations. In these linear longitudinal models, we used SD scores as parameter of fetal growth (dependent variable). The final model can be written as follows (eg, for fetal weight): SD score of fetal weight = β0 + β1×gawks + β2×exposuregroup + β3×gawks×exposuregroup (gawks = gestational age in weeks). In this model, β0 reflects the intercept and β2×exposuregroup tests the difference in intercept between exposed and non-exposed groups. The coefficient β3 reflects the slope (interaction of exposure with gestational age) and tests whether the groups of exposed and non-exposed grow at the same rate over time. The latter coefficient is the main interest for this article since this β represents the average decline or increase in SD for fetal weight per gestational week for exposed women versus non-exposed women. The regression models were adjusted for lifestyle and socioeconomic confounders used in previous studies on maternal occupational exposure6 and known determinants of fetal growth: maternal age, educational level, ethnicity, parity, pre-pregnancy weight, height at intake, smoking during pregnancy, alcohol use during pregnancy, folic acid supplement use and fetal gender.

    Missing values in confounders were handled by multiple imputations (fully conditional specification, Markov Chain Monte Carlo method) by generating five independent data sets for all analyses using SPSS V.17.0 for windows. Variables included in the imputation procedure (these variables were both imputed and used as predictors of missing data) were maternal age, educational level, ethnicity, parity, pre-pregnancy weight, height at intake, smoking during pregnancy, alcohol use during pregnancy, folic acid supplement use, fetal gender and gestational age at birth. Table 1 presents the proportion of missing values for each variable that was imputed. All multivariate analyses were performed with the multiple imputation data sets, and pooled estimates were calculated across these five independent data sets. The maximal allowed threshold for imputations was set on a maximum of missing values of 30%. However, missing values for parameters of physically demanding work were not imputed and, thus, the analysis on each exposure of interest was based on slightly different number of subjects due to some missing values.

    View this table:
    Table 1

    Baseline characteristics of the study population (n=4680)

    In total, three sensitivity analyses were performed: the first to evaluate whether women who started working before conception differed from women who started working during the first trimester, the second to analyse whether women with a certain last menstrual period and regular cycle differed from women whose pregnancy was dated by means of an ultrasound and the third to study the influence of the subgroup of women who stopped working before 34 weeks of gestation because of pregnancy complaints. Results from the logistic regression analyses on birth outcomes were used to estimate PAFs, expressing the proportion of the adverse health outcomes in the general population that is attributable to the risk factors of interest.28 The repeated measurement analyses were conducted with the Proc Mixed module of the Statistical Analysis System (V.9.2; SAS Institute Inc.).

    Results

    Table 1 shows the baseline characteristics of the study population. In total, 38.6% of the women were exposed to long periods of standing at work, 45.5% to long periods of walking at work, and 6.1% to heavy lifting at work. About 4.2% of the women regularly worked night shifts. Part-time jobs were common among women since 47.5% worked 25–39 h/week, 25.5% worked less than 25 h/week and 23.2% worked more than 40 h/week.

    Table 2 shows the associations between maternal occupational exposure to physically demanding work and adverse birth outcomes. There were no consistent associations between physically demanding work, long working hours and adverse birth outcomes. Furthermore, there was no clear dose–response relationship, and women often exposed to a certain occupational risk factor were not consistently at higher risk for adverse birth outcomes. In these multivariate models with adverse birth outcomes, the following confounders significantly influenced the outcome (in descending order of magnitude): maternal age, pre-pregnancy weight, height at intake, parity, ethnicity, smoking and folic acid use. Joint effects of several physically demanding work risk factors and working hours were investigated; however, we did not found any statistically significant joint effect on adverse birth outcomes (data not shown).

    View this table:
    Table 2

    Associations between maternal occupational exposure to physically demanding work during pregnancy with the risk of adverse birth outcomes in the Generation R Study

    Table 3 shows the cross-sectional analyses between long periods of standing, lifting >25 kg and working hours with HC, AC and EFW during the second (∼20 weeks of gestation) and third (∼30 weeks of gestation) trimesters in pregnancy. After adjustment for potential confounders, no associations of physically demanding work or working hours with fetal growth characteristics during the second trimester were found. In the third trimester of pregnancy, after adjustments for potential confounders, long periods of standing were significantly associated with a decreased fetal HC. For working hours, we observed effects on AC and EFW; however, after adjustment for potential confounders, these effects did not remain statistically significant.

    View this table:
    Table 3

    Associations between maternal occupational exposure to physically demanding work during pregnancy and fetal growth measured by ultrasound second and third trimester analyses

    Figures 1 and 2 show the association between long periods of standing and working hours on longitudinally measured growth (fetal weight and fetal HC). Long periods of standing at work were associated with slower growth rates in HC (−0.32 SD and −0.33 SD at birth), which corresponds to approximately 1 cm difference (3%) compared with the average HC of 33.9 cm at birth. Women working >25 h/week showed reduced fetal growth rates in both domains of fetal growth, namely fetal weight and HC. In these models, educational level and ethnicity significantly influenced fetal growth but did not influence the relationship between physically demanding work and fetal growth, resulting in comparable effect estimates.

    Figure 1
    Figure 1

    Relative differences in SD scores for women occupationally exposed to long periods of standing at work compared with non-exposed women for fetal weight and fetal head circumference. Values are based on repeated linear regression models and reflect the difference in SD score of fetal weight (1261 measurements) and fetal head circumference (10 411 measurements) in the offspring of mothers who work 25–39 or >40 h/week compared with the offspring of mothers who work <25 h/week. The reference value is a SD score of 0. *p value <0.05. Models were adjusted for maternal age, height at intake, weight before pregnancy, educational level, ethnicity, parity, smoking, alcohol use, folic acid supplement use, fetal gender and self-perceived health.

    Figure 2
    Figure 2

    Relative differences in SD scores for women who work 25–39 or >40 h/week compared with women who work <25 h/week for fetal weight and fetal head circumference. Values are based on repeated linear regression models and reflect the difference in SD score of fetal weight (1261 measurements) and fetal head circumference (10 411 measurements) in the offspring of mothers who work 25–39 or >40 h/week compared with the offspring of mothers who work <25 h/week. The reference value is a SD score of 0. *p value <0.05. Models were adjusted for maternal age, height at intake, weight before pregnancy, educational level, ethnicity, parity, smoking, alcohol use, folic acid supplement use, fetal gender and self-perceived health.

    In total, 4177 (89.3%) women filled out the question concerning the starting date of their current occupation, 4068 women (97.4%) started working before conception, whereas 109 (2.6%) women started working somewhere during their first trimester of pregnancy. In the sensitivity analyses, starting work before or during conception and having a regular menstrual cycle or not did not change the results. Furthermore, women who stopped working earlier than planned (before 34 weeks of gestation) often had medical reasons (71.4%), and these women had a higher risk of preterm delivery and low birth weight. When excluding these women from the analysis, duration of work during pregnancy was not associated with fetal growth, and we found no negative effect of working till 34–36 weeks of pregnancy on any of the birth outcomes in this study population. Stopping working before 34 weeks of gestation was not associated with physically demanding work or working hours.

    Discussion

    This population-based prospective cohort study suggests that long periods of standing at work and working >25 h/week were associated with lower fetal growth rates for fetal weight and HC in pregnancy. These findings were not reflected in adverse birth outcomes. Additional cross-sectional analyses showed that the differences are demonstrable from the third trimester onwards.

    Several mechanisms have been suggested to explain the possible adverse influence of physically demanding work during pregnancy on the fetus. Heavy physical work is thought to reduce the blood flow to the uterus and placenta, thereby reducing the availability of oxygen and nutrients for the fetus.29 ,30 Furthermore, lifting and trunk bending may increase intra-abdominal pressure, which in turn may lead to preterm delivery, especially in the last trimester when space in the abdominal cavity is maximally constrained.10 Also an increased release of catecholamines, through mediation of the sympathetic nerve system, has been hypothesised to play a role.31 Occupational risk factors, such as working in a specific occupation,8 ,32 shift work,33 ,34 job stress,12 ,13 ,35 standing, lifting36 and work hours,37–39 have been related to adverse birth outcomes. Two reviews have suggested an influence of physically demanding work on pregnancy outcomes.14 ,15 In addition to previous studies, which looked at adverse birth outcomes, we have looked at fetal growth measured in the second and third trimester of pregnancy. Although birth outcomes are important from an obstetric perspective, they are rather crude measures of fetal growth during pregnancy. We could not demonstrate an effect of working hours >25 h/week on adverse birth outcomes; however, effects on fetal growth rates during pregnancy could be demonstrated, suggesting that the latter analyses are more sensitive for picking up more subtle differences in fetal weight and HC. The PAFs for the occupational risk factors in this study were small, for small for gestational age at birth with the highest contribution of lifting >25 kg, PAF 4.2%, for preterm delivery with the highest contribution of working >40 h/week, PAF 1.5% and for low birth weight with the highest contribution of lifting >25 kg/week, PAF 3.6%. In this community-based study, physically demanding work had little influence on the prevalence of adverse birth outcomes, but in specific occupations with a high prevalence of physically demanding work, this contribution could be higher.

    Long working hours were associated with impaired fetal weight, resulting in a decrease in SD at birth varying between −0.27 and −0.36 SD at birth. This corresponds to approximately 150–200 g difference in birth weight. This effect seems to be of similar magnitude than the effects of other well-known lifestyle factors, such as smoking and caffeine intake with reported reductions of −0.3 SD and −0.1 to −0.3 SD.27 ,40 However, we must note that the PAFs of specific categories of physically demanding work were very low, and the effects of fetal growth were subtle since these effects were not reflected in adverse birth outcomes. The results of the current study hampers sound advice for pregnant women exposed to these risk factors.

    Women working as nurse, childcare giver or saleswoman most often reported lifting heavy loads (together accounting for 44.6% of all working women). For standing, several occupations were reported, most notably saleswoman, working with toddlers, schoolteachers and administrative employees (21.5%). Night shifts were most frequently reported by stewardesses, physicians and nurses (60.3%).

    In this study, we used ultrasound measurements for pregnancy dating, which seems superior to dating based on the last menstrual period.18 A disadvantage is that growth variations in early pregnancy are assumed to be zero, impairing analyses on first trimester growth. The repeated measurements based on gestational age-adjusted SD scores, comparable to standardised z-scores, enable us to identify pathological smallness instead of constitutional smallness. The advantage of SD scores as relative measure of difference is that the SD scores can be used in linear regression models, whereas absolute differences in fetal growth were highly skewed since growth curves during pregnancy have a typical parabolic shape that must be described by fractional polynomials instead of normal distributions.

    The strength of this study is the population-based approach with recruitment during the prenatal period and the availability of a large number of potential confounders. A limitation of this study is lower selective participation among mothers from ethnic minorities and with lower socioeconomic status.41 The non-response would lead to biased effect estimates if the association between physically demanding work and fetal development would be different between those included and those not included in the analyses. However, this seems unlikely since biased estimates in large cohort studies mainly arise from loss to follow-up rather than from non-response at baseline.42 Information on psychosocial stress or general fatigue, which could correlate with working hours and fetal growth, was not available in this study. Furthermore, this study did not take into account other sources of physically demanding activities outside employment, such as exercise, housework and volunteer work. However, it is unlikely that these activities are strongly related to physically demanding work risk factors, but they may lead, in some extent, to residual confounding. Women working in physically demanding jobs could have a more unhealthy lifestyle that was not fully adjusted for in the analysis by including smoking and alcohol use as confounders. Education is an important determinant of health behaviour, but adjustment for educational level did not affect the relationship between physically demanding work and birth outcomes or fetal growth. This suggests that life style-related risk factors most likely do not bias the relationship between long working hours and fetal growth.

    A limitation of this study is the semi-quantitative nature of the exposure information in four self-reported categories. This did not allow us to investigate duration of standing and walking per week or frequency of lifting heavy weights. Recall bias is unlikely since the information obtained was not biased by the outcome since the questionnaire was completed in mid-pregnancy. In this study, we used multiple imputations for missing values in covariates. This reduces selection bias due to non-random missing in the covariates.

    In the current study, we selected women with paid employment around week 30 of pregnancy, and this might have resulted in a more healthy and affluent study population since these women generally have better pregnancy outcomes than women without paid employment. Women in paid employment might have stopped working earlier during pregnancy due to pregnancy complaints, and technically, these women would be on sick leave. The sensitivity analyses on women who reported stopping working before 34 weeks because of pregnancy complaints showed that these women were at higher risk of preterm delivery and low birth weight. However, this was not associated with physically demanding work and, thus, will not have influenced the reported associations. When excluding these women from the analysis, duration of work during pregnancy was not associated with fetal growth. We were unable to find a clear negative effect of working till 34–36 weeks of pregnancy on any of the birth outcomes in this study population. When we corrected the longitudinal models for the duration of work during pregnancy (thus pregnancy week when women stopped working), it did not change the effect estimates, suggesting that the relationship between physically demanding work and fetal growth is independent of work duration.

    In the study, we found that physically demanding work during pregnancy was associated with lower fetal growth rates. We believe that optimising the work environment is important since participation of women in the reproductive age in the workforce continues to increase. Preventive measures reducing certain occupational conditions, such as shift work, night hours, standing, lifting and noise, have proven to reduce the risks of adverse birth outcomes.33 ,43 In the current study, we were unable to pinpoint the effects of physically demanding work in specific trimesters or of cumulative exposure over pregnancy since occupational activities were only measured once during pregnancy. We were able to demonstrate differences in fetal growth during the third trimester, and we hypothesised that differences might already originate earlier during pregnancy but were too small to be noticed. Preventive measures therefore may be most beneficial when focusing on the weeks before the third trimester. However, this study does not present concrete information on the required reduction in duration and level of work demands, which hampers sound advice. The results of this study need to be confirmed by future research.

    Acknowledgments

    The Generation R Study is conducted by the Erasmus Medical Centre in close collaboration with the School of Law and Faculty of Social Sciences of the Erasmus University Rotterdam, the Municipal Health Service Rotterdam area, Rotterdam, the Rotterdam Homecare Foundation, Rotterdam, and the Stichting Trombosedienst & Artsenlaboratorium Rijnmond (STAR-MDC), Rotterdam. We gratefully acknowledge the contribution of children and parents, general practitioners, hospitals, midwives and pharmacies in Rotterdam.

    References

    1. ↵
      1. Liu X,
      2. Roth J
      . Development and validation of an infant morbidity index using latent variable models. Stat Med 2008;27:971–89.
      OpenUrlCrossRefPubMed
    2. ↵
      1. Stillerman KP,
      2. Mattison DR,
      3. Giudice LC,
      4. et al
      . Environmental exposures and adverse pregnancy outcomes: a review of the science. Reprod Sci 2008;15:631–50.
      OpenUrlAbstract/FREE Full Text
    3. ↵
      1. Wigle DT,
      2. Arbuckle TE,
      3. Turner MC,
      4. et al
      . Epidemiologic evidence of relationships between reproductive and child health outcomes and environmental chemical contaminants. J Toxicol Environ Health B Crit Rev 2008;11:373–517.
      OpenUrlCrossRefPubMedWeb of Science
    4. ↵
      1. Wigle DT,
      2. Arbuckle TE,
      3. Walker M,
      4. et al
      . Environmental hazards: evidence for effects on child health. J Toxicol Environ Health B Crit Rev 2007;10:3–39.
      OpenUrlPubMedWeb of Science
    5. ↵
      1. Windham G,
      2. Fenster L
      . Environmental contaminants and pregnancy outcomes. Fertil Steril 2008;89:111–17.
      OpenUrlCrossRefPubMedWeb of Science
    6. ↵
      1. Burdorf A,
      2. Brand T,
      3. Jaddoe VW,
      4. et al
      . The effects of work-related maternal risk factors on time to pregnancy, preterm birth and birth weight: the Generation R Study. Occup Environ Med 2011;68:197–204.
      OpenUrlAbstract/FREE Full Text
    7. ↵
      1. Jansen PW,
      2. Tiemeier H,
      3. Verhulst FC,
      4. et al
      . Employment status and the risk of pregnancy complications: the Generation R Study. Occup Environ Med 2010;67:387–94.
      OpenUrlAbstract/FREE Full Text
    8. ↵
      1. Savitz DA,
      2. Olshan AF,
      3. Gallagher K
      . Maternal occupation and pregnancy outcome. Epidemiology 1996;7:269–74.
      OpenUrlPubMedWeb of Science
    9. ↵
      1. Mattison DR
      . Environmental exposures and development. Curr Opin Pediatr 2010;22:208–18.
      OpenUrlCrossRefPubMedWeb of Science
    10. ↵
      1. Bonzini M,
      2. Coggon D,
      3. Godfrey K,
      4. et al
      . Occupational physical activities, working hours and outcome of pregnancy: findings from the Southampton Women's Survey. Occup Environ Med 2009;66:685–90.
      OpenUrlAbstract/FREE Full Text
    11. ↵
      1. McCulloch J
      . Health risks associated with prolonged standing. Work 2002;19:201–5.
      OpenUrlPubMed
    12. ↵
      1. Vrijkotte TG,
      2. van der Wal MF,
      3. van Eijsden M,
      4. et al
      . First-trimester working conditions and birthweight: a prospective cohort study. Am J Public Health 2009;99:1409–16.
      OpenUrlCrossRefPubMedWeb of Science
    13. ↵
      1. Henrich W,
      2. Schmider A,
      3. Fuchs I,
      4. et al
      . The effects of working conditions and antenatal leave for the risk of premature birth in Berlin. Arch Gynecol Obstet 2003;269:37–9.
      OpenUrlCrossRefPubMed
    14. ↵
      1. Mozurkewich EL,
      2. Luke B,
      3. Avni M,
      4. et al
      . Working conditions and adverse pregnancy outcome: a meta-analysis. Obstet Gynecol 2000;95:623–35.
      OpenUrlCrossRefPubMedWeb of Science
    15. ↵
      1. Bonzini M,
      2. Coggon D,
      3. 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;64:228–43.
      OpenUrlAbstract/FREE Full Text
    16. ↵
      1. Jaddoe VW,
      2. van Duijn CM,
      3. van der Heijden AJ,
      4. et al
      . The Generation R Study: design and cohort update until the age of 4 years. Eur J Epidemiol 2008;23:801–11.
      OpenUrlCrossRefPubMedWeb of Science
    17. ↵
      1. Jaddoe VW,
      2. van Duijn CM,
      3. van der Heijden AJ,
      4. et al
      . The Generation R Study: design and cohort update 2010. Eur J Epidemiol 2010;25:823–41.
      OpenUrlCrossRefPubMedWeb of Science
    18. ↵
      1. Verburg BO,
      2. Steegers EA,
      3. De Ridder M,
      4. et al
      . New charts for ultrasound dating of pregnancy and assessment of fetal growth: longitudinal data from a population-based cohort study. Ultrasound Obstet Gynecol 2008;31:388–96.
      OpenUrlCrossRefPubMedWeb of Science
    19. ↵
      1. Altman DG,
      2. Chitty LS
      . New charts for ultrasound dating of pregnancy. Ultrasound Obstet Gynecol 1997;10:174–91.
      OpenUrlCrossRefPubMedWeb of Science
    20. ↵
      1. Robinson HP,
      2. Sweet EM,
      3. Adam AH
      . The accuracy of radiological estimates of gestational age using early fetal crown-rump length measurements by ultrasound as a basis for comparison. BJOG 1979;86:525–8.
      OpenUrlCrossRef
    21. ↵
      1. Hadlock FP,
      2. Harrist RB,
      3. Sharman RS,
      4. et al
      . Estimation of fetal weight with the use of head, body, and femur measurements—a prospective study. Am J Obstet Gynecol 1985;151:333–7.
      OpenUrlCrossRefPubMedWeb of Science
    22. ↵
      1. Verburg BO,
      2. Mulder PG,
      3. Hofman A,
      4. et al
      . Intra- and interobserver reproducibility study of early fetal growth parameters. Prenat Diagn 2008;28:323–31.
      OpenUrlCrossRefPubMedWeb of Science
    23. ↵
      World Health Organisation. Global database on Child Growth and Malnutrition. Chapter 5: The Z-Score or Standard Deviation Classification System. http://www.who.int/nutgrowthdb/about/introduction/en/index4.html (accessed 1 Nov 2010).
    24. ↵
      1. Elders LA,
      2. Burdorf A
      . Interrelations of risk factors and low back pain in scaffolders. Occup Environ Med 2001;58:597–603.
      OpenUrlAbstract/FREE Full Text
    25. ↵
      1. Hildebrandt VH,
      2. Bongers PM,
      3. van Dijk FJ,
      4. et al
      . Dutch Musculoskeletal Questionnaire: description and basic qualities. Ergonomics 2001;44:1038–55.
      OpenUrlCrossRefPubMedWeb of Science
    26. ↵
      1. Jaddoe VW,
      2. Troe EJ,
      3. Hofman A,
      4. et al
      . Active and passive maternal smoking during pregnancy and the risks of low birthweight and preterm birth: the Generation R Study. Paediatr Perinat Epidemiol 2008;22:162–71.
      OpenUrlCrossRefPubMedWeb of Science
    27. ↵
      1. Jaddoe VW,
      2. Verburg BO,
      3. de Ridder MA,
      4. et al
      . Maternal smoking and fetal growth characteristics in different periods of pregnancy: the generation R study. Am J Epidemiol 2007;165:1207–15.
      OpenUrlAbstract/FREE Full Text
    28. ↵
      1. Landis JR,
      2. Koch GG
      . The measurement of observer agreement for categorical data. Biometrics 1977;33:159–74.
      OpenUrlCrossRefPubMedWeb of Science
    29. ↵
      1. Hart A,
      2. Morris N,
      3. Osborn SB,
      4. et al
      . Effective uterine bloodflow during exercise in normal and pre-eclamptic pregnancies. Lancet 1956;271:481–4.
      OpenUrlPubMedWeb of Science
    30. ↵
      1. Stein ZA,
      2. Susser MW,
      3. Hatch MC
      . Working during pregnancy: physical and psychosocial strain. Occup Med 1986;1:405–9.
      OpenUrlPubMed
    31. ↵
      1. Katz VL,
      2. Jenkins T,
      3. Haley L,
      4. et al
      . Catecholamine levels in pregnant physicians and nurses: a pilot study of stress and pregnancy. Obstet Gynecol 1991;77:338–42.
      OpenUrlPubMedWeb of Science
    32. ↵
      1. Naeye RL,
      2. Peters EC
      . Working during pregnancy: effects on the fetus. Pediatrics 1982;69:724–7.
      OpenUrlAbstract/FREE Full Text
    33. ↵
      1. Croteau A,
      2. Marcoux S,
      3. Brisson C
      . Work activity in pregnancy, preventive measures, and the risk of delivering a small-for-gestational-age infant. Am J Public Health 2006;96:846–55.
      OpenUrlCrossRefPubMedWeb of Science
    34. ↵
      1. Fortier I,
      2. Marcoux S,
      3. 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;21:412–18.
      OpenUrlPubMedWeb of Science
    35. ↵
      1. Brandt LP,
      2. Nielsen CV
      . Job stress and adverse outcome of pregnancy: a causal link or recall bias? Am J Epidemiol 1992;135:302–11.
      OpenUrlAbstract/FREE Full Text
    36. ↵
      1. Wergeland E,
      2. Strand K,
      3. Bordahl PE
      . Strenuous working conditions and birthweight, Norway 1989. Acta Obstet Gynecol Scand 1998;77:263–71.
      OpenUrlCrossRefPubMedWeb of Science
    37. ↵
      1. Hatch M,
      2. Ji BT,
      3. Shu XO,
      4. et al
      . Do standing, lifting, climbing, or long hours of work during pregnancy have an effect on fetal growth? Epidemiology 1997;8:530–6.
      OpenUrlCrossRefPubMedWeb of Science
    38. ↵
      1. Peoples-Sheps MD,
      2. Siegel E,
      3. Suchindran CM,
      4. et al
      . Characteristics of maternal employment during pregnancy: effects on low birthweight. Am J Public Health 1991;81:1007–12.
      OpenUrlCrossRefPubMedWeb of Science
    39. ↵
      1. Tuntiseranee P,
      2. Geater A,
      3. Chongsuvivatwong V,
      4. 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;40:1013–21.
      OpenUrlCrossRefPubMedWeb of Science
    40. ↵
      1. Bakker R,
      2. Steegers EA,
      3. Obradov A,
      4. et al
      . Maternal caffeine intake from coffee and tea, fetal growth, and the risks of adverse birth outcomes: the Generation R Study. Am J Clin Nutr 2010;91:1691–8.
      OpenUrlAbstract/FREE Full Text
    41. ↵
      1. Jaddoe VW,
      2. Mackenbach JP,
      3. Moll HA,
      4. et al
      . The Generation R Study: design and cohort profile. Eur J Epidemiol 2006;21:475–84.
      OpenUrlCrossRefPubMedWeb of Science
    42. ↵
      1. Nohr EA,
      2. Frydenberg M,
      3. Henriksen TB,
      4. et al
      . Does low participation in cohort studies induce bias? Epidemiology 2006;17:413–18.
      OpenUrlCrossRefPubMedWeb of Science
    43. ↵
      1. Croteau A,
      2. Marcoux S,
      3. Brisson C
      . Work activity in pregnancy, preventive measures, and the risk of preterm delivery. Am J Epidemiol 2007;166:951–65.
      OpenUrlAbstract/FREE Full Text

    Supplementary materials

    • Supplementary Data

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

      Files in this Data Supplement:

    • Press release

      Press release

      Files in this Data Supplement:

    Footnotes

    • Funding This work was supported by financial support from the Erasmus Medical Centre, Rotterdam, the Erasmus University Rotterdam, the Netherlands Organization for Health Research and Development (ZonMw), the Netherlands Organisation for Scientific Research (NWO), the Ministry of Health, Welfare and Sport and the Ministry of Youth and Families. The present study was conducted in commission of the European project CONTAMED with funding from the European Union's Seventh Framework Programme (FP7) for Research and Technology Development [EU grant agreement 212502].

    • Competing interests None.

    • Patient consent Obtained.

    • Ethics approval Medical Ethics Committee at Erasmus University Medical Centre Rotterdam, The Netherlands (MEC 198.782/2001/31).

    • Provenance and peer review Not commissioned; externally peer reviewed.