Secondary Logo

Journal Logo

Air Pollution: Original Article

Residential Proximity to Major Roads and Preterm Births

Yorifuji, Takashia,b; Naruse, Hirooc; Kashima, Saoria,d; Ohki, Shigerue; Murakoshi, Takeshic; Takao, Soshia; Tsuda, Toshihidef; Doi, Hiroyukia

Author Information
doi: 10.1097/EDE.0b013e3181fe759f
  • Free

Abstract

Preterm births, defined as births before 37 gestational weeks, are associated with perinatal mortality and adverse consequences in later adulthood.1,2 In the United States, the proportion of preterm births among all live births was about 12.7% in 2007,3 whereas in Japan the proportion was 5.8% in the same year.4 The proportion of preterm births is increasing in developed countries (eg, from 4.1% in 1980 to 5.8% in 2007 in Japan),4 despite increased knowledge of the risk factors.

Potential risk factors associated with preterm births have been extensively studied.5 Air pollution is potentially an important risk factor,6–8 because it can affect many people. Furthermore, some components of air pollutants that affect health are similar to those that result from maternal smoking.9,10

Reviews of air pollution and preterm births published in 2004 and 20059–12 concluded that the evidence was insufficient to confirm this relationship. Subsequently, several studies have been published. Seven studies with case-control or cohort designs, which used local air-pollution measurement difference as exposure, demonstrated a positive association between air pollution and preterm births.13–19 Only 2 of these 7 studies reported effects of air pollution on risk of preterm birth before 30 weeks13,19; the other studies did not differentiate the gestational age of preterm births.

Preterm births can be classified by gestational age: less than 28 weeks' gestation (extreme prematurity), 28–31 weeks (severe prematurity), 32–33 weeks (moderate prematurity), and 34–36 weeks (near term).5 The outcome of preterm births is different between extreme prematurity and near-term preterm births.1 Therefore, evaluating whether air pollution increases preterm births at different gestational ages could be important to consider with regard to the medical outcome and public health burden of preterm births.

Preterm births are also known as “preterm parturition syndromes,” reflecting the fact that preterm births can have multiple potential causes and can be induced by multiple mechanisms.20,21 Some authors have discussed the necessity of separating preterm births to elucidate potential causes and mechanisms.22,23 Thus, evaluating whether air pollution increases certain types of preterm births with specific potential causes could help to elucidate the potential mechanisms associated with this risk factor.24

We evaluated the associations between proximity to major roads (as an index for air pollution) and preterm births classified by gestational age. We also examined the specific clinical types of preterm births (such as preterm premature rupture of the membranes [PPROM]), pregnancy hypertension, and other conditions) associated with air pollution.

METHODS

Participants

Data were extracted from a perinatal database maintained since January 1997 at the Seirei Hamamatsu General Hospital, Shizuoka, Japan. Shizuoka is located in central Japan (Figure) and has an area of 7770 km.2 It has a southern coastline facing the Pacific Ocean, and there are mountains exceeding 3000 m in altitude to the north.25 The western part of Shizuoka recorded 12,367 births in 2008.26 The hospital is the only perinatal medical center in the western part of Shizuoka prefecture, and accepts both low- and high-risk deliveries. The hospital reported 1539 new live births in 2008; thus, about one-eighth of the babies born in the western part of Shizuoka were born in this hospital. The database includes information about all of the mothers admitted to the department of obstetrics in the hospital (n = 18,645 from 1997 to 2008). In this study, we restricted the analysis to mothers who delivered liveborn single births (22 or more weeks of gestation) between January 1997 and December 2008, and with an Apgar score at 1 minute after birth greater than one. Using these criteria, we retrieved 16,382 births from the database.

FIGURE.
FIGURE.:
Map of the study area in Shizuoka prefecture. Major roads (white roads) and other roads for which there was traffic volume information (lines) are shown.

We excluded 206 births that lacked residential information (a prerequisite for estimating air pollution exposure from proximity to major roads). Furthermore, because we used major roads information in Shizuoka prefecture as an exposure indicator, we removed 1653 births whose mothers' address was in another prefecture.

The hospital is located in the western part of Shizuoka prefecture. We also removed 257 births whose mothers lived in central or eastern parts of Shizuoka prefecture, because these mothers were likely to have returned to their parents' home during pregnancy to give birth (a Japanese tradition). We defined the western part of Shizuoka prefecture on the basis of administrative limits, as shown in the Figure. These exclusions left 14,266 births for analysis. All mothers lived within 50 km of the hospital.

Exposure Data

We used the mothers' residential proximity to major roads, defined by the amount of traffic, as an index for air pollution exposure. First, we identified the geocode (latitude and longitude coordinates) at the residential address at the time of delivery. Among 14,266 births, 13,987 births could be geocoded at the exact address level, whereas 279 births could be geocoded at a census enumeration district (this gave more detail than the Japanese 7-digit zip code). We dichotomized our exposure indicator at 200 m (≤200 m; >200 m) based on previous epidemiologic studies27,28 and studies showing exponential decay in exposure with increasing distance from major roads.29,30

Among the 14,266 births, there were 265 mothers who were known to have returned to their parents' home during pregnancy to give birth. We considered these 265 mothers, as well as the 279 mothers with less detailed address information, in the sensitivity analyses.

We defined major roads as those having more than 50,000 vehicles on a weekday.25,31,32 First, we obtained road-type and traffic-volume data for Shizuoka from the 2005 Road Traffic Census conducted by the Road Bureau of the Ministry of Land, Infrastructure, Transport and Tourism. The traffic-volume data were recorded for 1 day during the period from September to November in 2005. We excluded 253 local roads and selected 92 roads with 555 locations to measure traffic volume. We averaged traffic volumes from count locations on each road, applying the average value to the entire length of the road. The correlation between the average traffic volume on each road and the original traffic-volume information measured at 555 locations was 0.85. The roads defined as “major” in this study corresponded almost exactly with the existing expressways or primary national highways.

Data on nitrogen dioxide were available from the environmental database managed by the National Institute for Environmental Studies in Japan. The correlation between distance from major roads and concentration of nitrogen dioxide, averaged over the period from 1997 to 2007 at 67 regulatory monitoring stations in Shizuoka prefecture, was −0.43. All geographic variables were collected by the Geographic Information System (GIS) software ArcGIS (ESRI Japan Inc, version 9.3).

Outcome Data

We divided preterm births by gestational age: less than 37 weeks, less than 32 weeks (severe prematurity), and less than 28 weeks (extreme prematurity). We also divided preterm births less than 37 weeks by specific clinical manifestations: PPROM, pregnancy hypertension, and other clinical manifestations. Gestational ages were based on the last menstrual period, and mostly confirmed or corrected by ultrasound measurements at about 10 weeks of gestational age. The clinical manifestations before preterm delivery were diagnosed by trained obstetricians in the hospital. PPROM was defined as membrane rupture before the onset of labor. Pregnancy hypertension was defined as hypertension after 20 weeks of gestational age with or without proteinuria. All other preterm births were designated as “other clinical manifestations.”

Other Covariate Data

We retrieved information about potential confounding factors from the perinatal database. This information was obtained from mothers by trained obstetricians or midwives at the time of the prenatal examination when the expected due date was confirmed (at about 10 weeks of gestational age), and added or corrected at admission or delivery.

We also included area-based socioeconomic status (SES) variables in the study: the proportion of white collar workers >15 years of age and the proportion of unemployed individuals in households in the corresponding census region where mothers lived. These data were obtained from the 2000 national census. We defined the following occupations as white-collar work: professional or technical workers, managerial workers, and clerical workers. The study area had 1872 census regions and the median and mean areas of the census regions were 0.4 and 1.3 square kilometers (standard deviation, 3.6), respectively. Approval for this study was obtained from the Institutional Review Board of Seirei Hamamatsu General Hospital.

Data Analyses

We calculated the proportion of preterm births according to characteristics of subjects. We next estimated the multivariate adjusted odds ratios (ORs) for preterm births, classified by gestational age or specific clinical manifestations, according to major roads, using the logistic regression model.33 When we estimated the adjusted ORs for preterm births classified by gestational age, we treated all other births as noncases. For example, when we assessed the risk of preterm birth before 28 weeks, we used all births at >28 weeks as noncases. When estimating the adjusted ORs for preterm births with specific clinical manifestations, we used full-term births as noncases.

We first adjusted for maternal age, maternal occupation (housewives; part-time workers; self-employed workers; employees; professional workers), maternal smoking (never smoked; ex-smoker including mothers who quit smoking during pregnancy; smoker), paternal smoking (smoker; nonsmoker), and the proportion of white-collar workers as measured by the area-based SES mandatory. We further adjusted for maternal alcohol consumption (drinker; nondrinker) and maternal body mass index (BMI) and examined whether point estimates changed. BMI was defined as the mothers' body weight before pregnancy (kg) divided by her height squared (m2). Maternal age, maternal BMI, and the proportion of white-collar workers were treated as continuous variables. Maternal age was entered as a linear and quadratic term into the models because a U-shaped association was expected between maternal age and preterm births. These potential confounding factors were chosen a priori, based on previous studies.13–19 Our results did not change when we further adjusted for maternal alcohol consumption and maternal BMI, and we therefore present results from the more parsimonious model.

Traffic exposure at home is likely to be stronger among those who spend more time at home. We, therefore, compared effect estimates between housewives and those who worked outside the home (part-time workers; self-employed workers; employees; professional workers).

In sensitivity analyses, we repeated the analysis using paternal occupation (unemployment; part-time workers; self-employed workers; employees; professional workers) as an individual SES variable instead of maternal occupation. We also repeated the analysis using the proportion of unemployment in households as an area-based SES variable instead of the proportion of white-collar workers. Finally, to reduce possible exposure misclassification, we conducted the same analysis excluding births for which the mothers' residential information was based only on the census enumeration district (n = 279), as described previously. We also excluded 265 mothers who returned to their parents' home during pregnancy to give birth. All confidence intervals (CIs) were estimated at the 95% level. PASW software (SPSS Japan Inc, version 18.0J) was used for the analysis.

RESULTS

The baseline characteristics of mothers, fathers, and newborns (n = 14,266) for full-term births and preterm births are shown in Table 1. The mean maternal age was 30.2, similar to the national mean of 29.3 in 1997 and 30.7 in 2007.4 Two percent of mothers were less than 20 years of age. The mean birth weight of all births was 2886 g, lower than the national mean birth weight of 3000 g in 2007.4 The overall proportion of preterm births was 10.7%, higher than the proportion of preterm births in Japan as a whole (5.8% in 2007).4 The proportion of preterm births over four 3-year calendar periods from 1997 to 2008 was 9.1%, 10.7%, 11.5%, and 11.2%, respectively. Although mean maternal age did not differ between full-term births and preterm births, younger and older mothers had a higher risk of a preterm birth. Maternal smokers experienced higher risk of preterm birth, as expected. Furthermore, preterm births were less common among professional workers.

TABLE 1
TABLE 1:
Descriptive Characteristics of Mothers, Fathers, and Newborns

Table 2 shows the number and the adjusted ORs of preterm births classified by gestational age. We found positive associations between proximity to major roads and preterm births for each of the gestational age categories. Specifically, living within 200 m increased the risk of preterm births of less than 37 weeks by 1.5 times (95% CI = 1.2–1.8), preterm births of less than 32 weeks by 1.6 times (1.1–2.4), and preterm births of less than 28 weeks by 1.8 times (1.0–3.2).

TABLE 2
TABLE 2:
Association Between Proximity to Major Roads and Preterm Births of Less Than 37, Less Than 32, and Less Than 28 Weeks

We found positive associations between proximity to major roads within 200 m and preterm births with PPROM 1.9 (1.3–2.8) or with pregnancy hypertension 2.0 (1.2–3.3) (Table 3). The risk of preterm births from all other conditions was only slightly elevated. Among the preterm births with specific clinical manifestations, 7 preterm births had both PPROM and pregnancy hypertension.

TABLE 3
TABLE 3:
Association Between the Proximity to Major Roads and Preterm Births With Specific Clinical Manifestations Among Preterm Births of Less Than 37 Weeks

When we stratified subjects by maternal occupation (housewives vs. outside workers), there were fewer than 10 exposed cases in each stratum, except for preterm births before 37 weeks and preterm births with PPROM and other conditions. We found consistently higher ORs among housewives; for preterm births before 37 weeks the ORs were 1.6 (1.2–2.1) for housewives and 1.3 (0.9–1.8) for outside workers; for preterm births with PPROM, 2.0 (1.2–3.2) for housewives and 1.9 (1.0–3.5) for outside workers; and for preterm births with other conditions, 1.5 (1.1–2.1) for housewives and 0.8 (0.5–1.4) for outside workers.

In sensitivity analyses, the effect estimates did not change substantially. When we considered paternal occupation as the individual-level SES variable, the ORs for preterm births of less than 37, 32, and 28 weeks were 1.5 (1.2–1.9), 1.7 (1.2–2.4), and 1.7 (1.0–3.1), respectively. In addition, when we adopted the proportion of unemployment in households as the area-based SES, these ORs were 1.4 (1.1–1.8), 1.6 (1.1–2.3), and 1.7 (1.0–3.1), respectively. Furthermore, the ORs did not change after excluding 279 births with residential information only at the census level. Effect estimates did not change after excluding 265 mothers who returned to their parents' home to give birth.

DISCUSSION

We evaluated associations between proximity to major roads (as an index for air pollution) and preterm births, classified by gestational age or specific clinical manifestations. Proximity to major roads was associated with higher risks of preterm births at any gestational age. Proximity to major roads was also associated with increased risk of preterm births with PPROM and pregnancy hypertension.

Consistent with other recent studies,13–19our findings suggest that air pollution exposure increases the risk of preterm birth. Furthermore, our study adds to the preliminary findings of Brauer et al13 and of Wu et al19 in showing associations of air pollution with risk of preterm birth before 30 weeks. Because the medical outcome and public-health burden of preterm births are much higher with extreme and near-term preterm births, it is important to pursue further the possible effects of air pollution on very early preterm birth.

Potential mechanisms by which air pollution might increase preterm births include inflammation, endothelial dysfunction (hypertension), endocrine disruption, and genetic changes in germ cells.7 A recent study by Wu et al19 suggested a positive association between exposure to air pollution and development of preeclampsia during pregnancy, which is characterized by elevated blood pressure, edema, and proteinuria. Our findings provide further evidence to support this mechanism, demonstrating that air pollution could increase preterm births through the mechanism of inflammation (PPROM)34 or endothelial dysfunction (pregnancy hypertension). Indeed, some components of air pollutants that affect reproductive outcomes are considered to be similar to those that result from maternal smoking.9,10 Smoking is known to increase PPROM and pregnancy hypertension.35

Most previous air pollution studies used register-based participants. Our study participants were pairs of mothers and newborns who attended one general hospital with a perinatal center, which provided more detailed clinical information. In contrast to previous studies that used birth certificates,13–16,18 we were able to collect maternal behavior and physical attribute variables as well, such as active and passive smoking behavior, alcohol consumption, and BMI. Our study demonstrated positive associations even after adjusting for these variables. Furthermore, clinical diagnoses regarding maternal complications could be validated and standardized because of a single hospital-based sampling method.

Not all babies in the western part of Shizuoka are born in this particular hospital; however, the hospital is the only perinatal center in the area and accepts high-risk deliveries. As a consequence, the mean birth weight of our study group was less than the national mean birth weight, and the proportion of preterm births was higher than that reported nationally, as expected. This hospital-based sampling method might introduce selection bias. However, the proportion of preterm births delivered in this hospital was higher mainly among mothers residing far from the hospital. The proportion of preterm births was 7.4% among mothers residing close to the hospital (lowest 10th percentile of the distance from the hospital) and 23.4% among mothers residing far from the hospital (highest 10th percentile of the distance). The hospital is located relatively close to major roads (Figure), hence, this type of selection bias, if it exists, would underestimate the results.

We used the mothers' proximity to major roads as an index for air pollution exposure. Air-monitoring stations are sited for policy and regulatory purposes, and may not be ideally placed for epidemiologic studies.8 Furthermore, the air-monitoring stations are limited in spatial resolution. In contrast, the approach we adopted is limited in temporal resolution. Traffic information in this study was obtained from the Road Traffic Census conducted in 2005, and the traffic volume data were recorded for only 1 day, during the period from September to November 2005. Thus, our exposure indicator could not reflect year-to-year or seasonal variations in traffic exposure. Because major roads did not change, the traffic volume would not have varied substantially in the year-to-year comparison. The average change (increase) in traffic volume over a 24-hour period from 1999 to 2005 was only 117 for the whole of Shizuoka prefecture.36 However, our exposure indicator did not provide any indication of seasonal variation. Recently, seasonal factors have been discussed as potential confounding factors or effect-measure modifiers in the association between air pollution and reproductive outcomes.7 Thus, future studies should incorporate seasonal variations in air pollution using more sophisticated models, such as a temporally adjusted land use regression model.13

Exposure misclassification can occur due to a mothers' mobility during pregnancy, as determined by analysis of maternal residential information at delivery. We reduced the possibility of movement by restricting our study subjects to those who lived in the western part of Shizuoka prefecture. In addition, we had information about 265 mothers who were known to have returned to their parents' home during pregnancy to give birth; in a sensitivity analysis that excluded these subjects, results were unchanged. To reduce exposure misclassification during daytime, we divided participants according to whether they were housewives or outside workers and repeated the analyses. The point estimates in housewives were consistently higher than those in outside workers, as expected, if the influence of traffic exposure at home was stronger among those who spent more time at home during their pregnancy. Although we could not obtain other information about maternal mobility, this type of exposure misclassification would not explain our findings.

In conclusion, exposure to traffic-related air pollution was associated with an increased risk of preterm birth, perhaps even more strongly for preterm births before 30 weeks. The mechanism responsible for this association may work through PPROM and pregnancy hypertension.

ACKNOWLEDGMENTS

We appreciate the contributions of the staff at Seirei Hamamatsu General Hospital in maintaining the perinatal dataset. We also thank Tomo Hirayama for his statistical advice.

REFERENCES

1. Moster D, Lie RT, Markestad T. Long-term medical and social consequences of preterm birth. N Engl J Med. 2008;359:262–273.
2. Saigal S, Doyle LW. An overview of mortality and sequelae of preterm birth from infancy to adulthood. Lancet. 2008;371:261–269.
3. Hamilton B, Martin J, Ventura S. Births: Preliminary Data for 2007. National Vital Statistics Reports. Vol. 57. Hyattsville, MD: National Center for Health Statistics; 2009;1–23.
4. Health and Welfare Statistics Association. Statistics and Information Department in Japanese Ministry of Health Labour and Welfare. Vital statistics of Japan 2007. Tokyo:Health and Welfare Statistics Association; 2009.
5. Goldenberg RL, Culhane JF, Iams JD, Romero R. Epidemiology and causes of preterm birth. Lancet. 2008;371:75–84.
6. Ritz B, Wilhelm M. Ambient air pollution and adverse birth outcomes: methodologic issues in an emerging field. Basic Clin Pharmacol Toxicol. 2008;102:182–190.
7. Slama R, Darrow L, Parker J, et al. Meeting report: atmospheric pollution and human reproduction. Environ Health Perspect. 2008;116:791–798.
8. Woodruff TJ, Parker JD, Darrow LA, et al. Methodological issues in studies of air pollution and reproductive health. Environ Res. 2009;109:311–320.
9. Maisonet M, Correa A, Misra D, Jaakkola JJ. A review of the literature on the effects of ambient air pollution on fetal growth. Environ Res. 2004;95:106–115.
10. Sram RJ, Binkova B, Dejmek J, Bobak M. Ambient air pollution and pregnancy outcomes: a review of the literature. Environ Health Perspect. 2005;113:375–382.
11. Glinianaia SV, Rankin J, Bell R, Pless-Mulloli T, Howel D. Particulate air pollution and fetal health: a systematic review of the epidemiologic evidence. Epidemiology. 2004;15:36–45.
12. Lacasana M, Esplugues A, Ballester F. Exposure to ambient air pollution and prenatal and early childhood health effects. Eur J Epidemiol. 2005;20:183–199.
13. Brauer M, Lencar C, Tamburic L, Koehoorn M, Demers P, Karr C. A cohort study of traffic-related air pollution impacts on birth outcomes. Environ Health Perspect. 2008;116:680–686.
14. Genereux M, Auger N, Goneau M, Daniel M. Neighbourhood socioeconomic status, maternal education and adverse birth outcomes among mothers living near highways. J Epidemiol Community Health. 2008;62:695–700.
15. Leem JH, Kaplan BM, Shim YK, et al. Exposures to air pollutants during pregnancy and preterm delivery. Environ Health Perspect. 2006;114:905–910.
16. Ponce NA, Hoggatt KJ, Wilhelm M, Ritz B. Preterm birth: the interaction of traffic-related air pollution with economic hardship in Los Angeles neighborhoods. Am J Epidemiol. 2005;162:140–148.
17. Ritz B, Wilhelm M, Hoggatt KJ, Ghosh JK. Ambient air pollution and preterm birth in the environment and pregnancy outcomes study at the University of California, Los Angeles. Am J Epidemiol. 2007;166:1045–1052.
18. Wilhelm M, Ritz B. Residential proximity to traffic and adverse birth outcomes in Los Angeles county, California, 1994–1996. Environ Health Perspect. 2003;111:207–216.
19. Wu J, Ren C, Delfino RJ, Chung J, Wilhelm M, Ritz B. Association between local traffic-generated air pollution and preeclampsia and preterm delivery in the south coast air basin of California. Environ Health Perspect. 2009;117:1773–1779.
20. Haram K, Mortensen JH, Wollen AL. Preterm delivery: an overview. Acta Obstet Gynecol Scand. 2003;82:687–704.
21. Romero R, Espinoza J, Kusanovic JP, et al. The preterm parturition syndrome. BJOG. 2006;113(suppl 3):17–42.
22. Adams MM. Perinatal Epidemiology for Public Health Practice. New York:Springer; 2009.
23. Savitz DA, Dole N, Herring AH, et al. Should spontaneous and medically indicated preterm births be separated for studying aetiology? Paediatr Perinat Epidemiol. 2005;19:97–105.
24. Ananth CV, Vintzileos AM. Epidemiology of preterm birth and its clinical subtypes. J Matern Fetal Neonatal Med. 2006;19:773–782.
25. Kashima S, Yorifuji T, Tsuda T, Doi H. Application of land use regression to regulatory air quality data in Japan. Sci Total Environ. 2009;407:3055–3062.
26. Shizuoka prefecture. Vital statistics of Shizuoka 2008. Shizuoka, Japan; 2010.
27. Hart JE, Laden F, Puett RC, Costenbader KH, Karlson EW. Exposure to traffic pollution and increased risk of rheumatoid arthritis. Environ Health Perspect. 2009;117:1065–1069.
28. Jerrett M, Arain A, Kanaroglou P, et al. A review and evaluation of intraurban air pollution exposure models. J Expo Anal Environ Epidemiol. 2005;15:185–204.
29. Nitta H, Sato T, Nakai S, Maeda K, Aoki S, Ono M. Respiratory health associated with exposure to automobile exhaust. I: results of cross-sectional studies in 1979, 1982, and 1983. Arch Environ Health. 1993;48:53–58.
30. Ono M. Pollution around a road [in Japanese]. In: Koizumi A, Murakami M, eds. Introduction of Environmental Health. Karadano Kagaku. Tokyo: Nihonhyouronnsha; 1990.
31. Ross Z, English PB, Scalf R, et al. Nitrogen dioxide prediction in Southern California using land use regression modeling: potential for environmental health analyses. J Expo Sci Environ Epidemiol. 2006;16:106–114.
32. Ryan PH, LeMasters GK. A review of land-use regression models for characterizing intraurban air pollution exposure. Inhal Toxicol. 2007;19(suppl 1):127–133.
33. Kleinbaum DG, Klein M, Pryor ER. Logistic Regression: A Self-Learning Text. Statistics for Biology and Health. 2nd ed. New York: Springer; 2002.
34. Aagaard-Tillery KM, Nuthalapaty FS, Ramsey PS, Ramin KD. Preterm premature rupture of membranes: perspectives surrounding controversies in management. Am J Perinatol. 2005;22:287–297.
35. Salihu HM, Wilson RE. Epidemiology of prenatal smoking and perinatal outcomes. Early Hum Dev. 2007;83:713–720.
36. Development Bureau Road Department of Chubu Area. Road Traffic Census. Chubu, Japan: Development Bureau Road Department of Chubu Area; 2006.
© 2011 Lippincott Williams & Wilkins, Inc.