Evidence for increased risks of preterm delivery in a population residing near a Freeway in Taiwan

Evidence for increased risks of preterm delivery in a population residing near a Freeway in Taiwan

Chun-Yuh Yang

IN MANY COUNTRIES, road traffic is the major source of ambient air pollution in urban areas where most individuals live. Air pollution from motor vehicles is a complex mixture of volatile hydrocarbons, airborne particles, nitrogen oxides, and carbon monoxide. (1,2) Previous research on the health effects of air pollution has been focused on respiratory health and has been concerned primarily with workers exposed occupationally to high concentrations of automotive emissions. (3-13)

Although the contribution of traffic to community air pollution levels is recognized by many, surprisingly few researchers have investigated the association between traffic-related air pollution and respiratory illnesses or symptoms in children and adults. (14-28) In most previous studies, authors (14-24) have reported significantly positive associations between respiratory illnesses and respiratory symptoms in children and adults and traffic-related air pollution. However, in other studies no such association has been found. (25-28)

Although air pollution is not typically considered a possible determinant of adverse pregnancy outcome(s), (29) 2there is growing concern regarding the negative health effects of air pollution on the reproductive system. (30) Recently, several investigators have reported associations between air pollution levels and preterm deliveries. (31-34) However, Landgren (35) conducted a study in southern Sweden and found no evidence connecting air pollution levels with the incidence of preterm delivery.

Despite the fact that an association has been found between air pollution levels and preterm delivery in several studies, the data do not support or refute a causal inference. (29) In addition, it is not known whether the associations between air pollution and preterm delivery found in some populations can be replicated in others. (32)

In Taiwan, outdoor air pollution is caused primarily by industrial sources–specifically the petrochemical and petroleum industries. (34,36-42) Previous studies have examined the relationship between preterm deliveries and industrial pollutants (34,42); however, the relationship between preterm delivery and residential proximity to traffic-related air pollution is currently unknown. In the present study, we explored the potential association between traffic-related air pollution levels and preterm deliveries among women who resided near a freeway in Taiwan. This study is one of a series in which potential hazards posed by exposure to traffic-related air pollution in Taiwan were investigated.

Method

Study location. The Zhong-Shan Freeway, built in 1978, is the only highway connecting Keelung City (in northern Taiwan) with Kaohsiung City (in southern Taiwan). Kaohsiung–the 2nd largest city in Taiwan–has a population of 1.5 million (October 2000 census) that occupies an area of 153.6 [km.sup.2]. The Zhong-Shan Freeway passes through the San-Ming district (a residential area in East Kaohsiung), and most traffic uses this freeway to access that district. The traffic density on the East Kaohsiung section of the freeway is approximately 93,000 vehicles/day on weekdays. There were no industrial sources of air pollution near the study areas. Additional details regarding the study location have been presented elsewhere. (27) In summary, these factors afforded us a unique opportunity to investigate the issue of exposure to traffic pollution levels and its health effects on preterm deliveries.

The study area comprised a zone along the Zhong-Shah Freeway that included the 1,500-m areas on both sides of the freeway within the San-Ming Ward, Nan-Tzu Ward, Chienchen Ward, and Linya Ward–all of which are residential areas in East Kaohsiung. Given that the freeway was occupied by high-density traffic, we thought it unlikely that air pollution exposure at distances equal to or greater than 1,500 m from the freeway was measurably influenced by traffic from the freeway. We divided the study areas into 2 districts: (1) an area located within 500 m of the freeway, and (2) an area located 500-1,500 m from the freeway.

Subjects and outcomes. Data on pregnancy outcomes were obtained from the Taiwan Birth Registry. Registration of all births is required by law in Taiwan. It is the responsibility of the parents or the family concerned to register infant births at a local Household Registration Office within 15 days of birth. Computerized data on live births are collected from the Household Registration System, which is managed by the Taiwan Ministry of the Interior. The registration form–which contains questions about maternal age, education, parity, gestational age, date of delivery, and the infant’s gender and birth weight–is completed by the attending physician. Given that (a) most deliveries in Taiwan occur in either a hospital or clinic, (43) (b) birth certificates are completed by physicians who attend the delivery, and (c) it is mandatory that all live births be registered at local household registration offices, the birth registration data are considered complete and accurate. The aforementioned data have been used in our previous studies. (34,42,44) Twins or multiple pregnancies were not included in our analysis. In addition, gestational ages for live births that occurred outside the range of 20-50 wk were not considered in our study. (45)

There were 14,395 singleton deliveries within 1,500 m of the freeway between January 1, 1992, and December 31, 1997 (i.e., dates of study). Of the 14,389 births for which information on parity was available, first-parity births comprised 44.51% of the study subjects. Of 6,404 first-parity singleton live births, 29 subjects for whom there was invalid or missing information on gestational age were excluded. Among the remaining 6,375 subjects, birth weight or maternal age data were missing for 66. Of the 6,309 first-parity births for which there was complete information on the above variables, 58 were excluded because data were missing for at least one other variable (e.g., maternal education, marital status, infant gender, month of birth, birthplace). As a result of these exclusions, there remained 6,251 first-parity singleton live births for study. The number of births considered for final analysis in the district located closest to the freeway (i.e., within 500 m) was 3,211; there were 3,040 births in the outer band (i.e., between 500 m and 1,500 m from the highway).

Statistics. The outcome of interest in this study was preterm delivery, which was defined as a gestational period (on the basis of dates of each mother’s last menstrual period) of less than 37 complete weeks. We further restricted this study to each mother’s first-parity singleton live birth. Therefore, prior pregnancy history would not be a confounder. We used an unconditional multiple logistic-regression model to estimate the effect of traffic air pollution on the risk of preterm births. (46) Odds ratios (ORs) and 95% confidence intervals (CIs) were computed. Several potential confounding factors could have influenced the association between traffic air pollution and preterm delivery and were, therefore, examined in this study. We included all of the confounders: maternal age ([less than or equal to] 20 yr, 21-34 yr, [greater than or equal to] 35 yr); season (spring, summer, autumn, winter); marital status (married, unmarried); maternal education (< 12 yr, [greater than or equal to] 12 yr); and infant gender (male, female) in the final model to calculate the adjusted ORs. The analyses were performed using the SAS statistical package. (47) All statistical tests were 2-sided. We considered p values of less than 0.05 to be significant.

Results

Collectively, 6,251 first-parity singleton live births, with complete information, were included in the analysis (Table 1). The prevalences of preterm deliveries within 500 m, and between 500 m and 1,500 m, of the freeway were significant at 5.29% and 4.08%, respectively.

The ORs for preterm deliveries, on the basis of comparisons (logistic regression) between mothers who lived within 500 m and mothers who lived between 500 m and 1,500 m of the freeway are shown in Table 2. After we controlled for possible confounders (including maternal age, season, marital status, maternal education, and infant gender), the adjusted OR was 1.30 (95% CI = 1.03, 1.65) for delivery of preterm infants to mothers who lived close to the freeway (i.e., 500 m) vs. mothers who lived between 500 m and 1,500 m from the freeway.

Discussion

The results of the present study indicated that residence in a geographical area characterized by higher air pollution (specifically, traffic emissions) than prevailed in other areas resulted in a significant excess of preterm births. No industrial sources of air pollution were present near the study areas and, on the basis of one author’s (CY) professional observations, there was no long-range drift of industrial pollution to the study areas. We assumed, therefore, that automobile exhaust was the single major contributor to local air pollution, and that the effects observed in the study were not caused by another external source of air pollution.

The major difficulty in studying adverse health effects associated with pollution is the assessment of exposure. In previous studies, investigators have attempted to quantify the concentration of air pollutants (e.g., carbon monoxide, sulfur dioxide, total suspended particulates, particulate matter with diameters less than 10 [micro]m) and assign exposure values to study subjects, (31-33) or to use area-based exposure to approximate individual exposures. (35) In our study, we used an “extreme point contrast” to investigate the effects of traffic air pollution on delivery of preterm birth infants, in order to maximize the inherent power of the study design. (48-49) In summary, if an extreme point contrast is to be established, a specific traffic-exposed population must be compared with what may reasonably be considered a corresponding “placebo” group (i.e., nonexposed population), which is used as an appropriate reference background for the determinant. (37,50,51) In our study, we assumed that women who lived within 500 m of the freeway, on average, experienced a higher exposure to air pollutants than women who lived between 500 m and 1,500 m of the freeway. Information on the outcomes and covariates, however, were collected from individual birth records. Therefore, our birth design was semi-individual, which is considered a valid design, compared with traditional ecologic studies. (52)

Despite the fact that we used extreme point contrast to assess exposure, the potential for misclassification of exposure remained. Mobility during pregnancy was likely a problem in this study. In 2 U.S. studies, investigators reported that approximately 25% (53) and 37% (54) of women moved during pregnancy. No data were available concerning the proportion of women in Taiwan who moved during their pregnancies. Therefore, “true” personal exposures of mothers were inevitably misclassified. However, the potential exposure errors would likely be random, and would, therefore, be more likely to lead to an underestimation, rather than an overestimation, of the effect(s) of air pollution.

There are additional risk factors, not included in our study, that are thought to affect the probability of delivery of preterm birth infants, such as prepregnancy weight, weight gain, nutritional status, cigarette smoking, occupational exposures, and intrauterine infections. (55,56) Unfortunately, no information was available regarding these variables for the individual study subjects, and the variables could not be adjusted for directly in our analysis. Individuals who live closest to a freeway (i.e., within 500 m) are likely to be poorer than those who live in the outer band (i.e., 500-1,500 m), and they likely work in more-hazardous industrial jobs. Inasmuch as no information was available on income levels for our study subjects, we used years of maternal education as a proxy for socioeconomic status. This inclusion of the maternal education variable in our model may have helped to control for potential confounding from variables such as maternal nutrition status or occupational exposure. Also, the adjusted effect estimate was nearly identical to the crude one–a result that may suggest that the observed effects were not attributable to regional variations in this confounder.

We understand that some study subjects who suffered from chronic hypertension; chronic renal, cardiac, or lung diseases; or diabetes during pregnancy should have been excluded because the effects of air pollution on such pregnancies may be far outweighed by the influence of the mother’s medical condition. (33) We were unable to do this because we lacked the medical records for the study subjects. However, there is no reason to believe that exposure to traffic-related pollutants would be associated with some of these medical conditions.

The prevalence of smoking among Taiwan’s female population is extremely low. According to the 1993 national smoking survey, only 3-4% of women aged 1 6 yr or older were smokers (57); therefore, maternal smoking status should not have been a significant confounder in our study. In addition, we adjusted for maternal age and maternal education factors that our Bureau of Tobacco and Alcohol Monopoly has identified as influencing the smoking behavior of some women. We, therefore, may have adjusted indirectly for the confounding that resulted from smoking. Also, fears of preterm deliveries are unlikely to deter women from living near a freeway because the possible role of air pollution as a risk factor for preterm delivery has not been highly publicized.

Motor traffic can produce a mixture of volatile organic compounds, airborne particles, nitrogen oxides, and carbon monoxide. (1,2) We lacked the data necessary to estimate the levels of these pollutants. The most important indoor source of these pollutants is active or passive smoking in study areas. We do not know whether the contribution of indoor air pollution would differ between homes occupied by mothers located closest to the freeway and mothers located in the outer band (i.e., 500-1,500 m). Therefore, on the basis of the results of our study, we were unable to rule out the possible effect of indoor air pollution on the risk of preterm delivery. Nevertheless, given the ubiquitous occurrence and distribution of car exhaust, misclassification of exposure is likely; however, this result would probably reduce the magnitude of associations, rather than introduce a positive bias in the associations.

The biological mechanisms whereby air pollution might cause preterm delivery remain unknown. (31) Although a range of social and behavioral determinants of preterm births have been identified, the biological mechanisms that lead to preterm births are not well understood. (56,58) Maternal infection in pregnancy has been reported as a cause of preterm delivery. (55,59) Perhaps repeated respiratory, genitourinary, or systemic infection during pregnancy–possibly related to elevated air pollution levels–may play a role.

In summary, the results of our study indicate that traffic air pollution plays a role in adverse pregnancy outcomes (e.g., preterm delivery). Our results appear to confirm recent findings in China, (31) the United States, (33) and the Czech Republic, (32) and add to the growing body of evidence that reproductive health hazards are caused, in part, by air pollution. (30)

Table 1.–Maternal Characteristics and Prevalence of

Preterm Delivery, by Distance from a Freeway in Kaohsiung,

Taiwan, Republic of China

Distance from freeway

500-1,500m<500m

Variable n % n %

No. of singleton live births 3,040 3,211

Gestational age (wk)

<371244.081705.29*

[greater than or equal to] 37 2,916 95.92 3,041 94.71

Maternal age (yr)

[less than or equal to] 20 117 3.85 146 4.55

21-34 2,688 88.42 2,797 87.11

[greater than or equal to] 35 235 7.73 268 8.34

Season

Spring 732 24.08 761 23.70

Summer 708 23.29 690 21.49

Autumn 780 25.66 843 26.25

Winter 820 26.97 917 28.56

Marital status

Married 2,936 96.58 3,094 96.36

Unmarried 104 3.42 117 3.64

Maternal education (yr)

<122,41379.382,53979.07

[greater than or equal to] 12 627 20.62 672 20.93

Infant gender

Male 1,592 52.37 1,659 51.67

Female 1,448 47.63 1,552 48.33

Place of birth

Hospital/clinic 3,040 100 3,211 100

*Significantlydifferentfromcontrol(p<0.05).

Table 2.–Estimated Odds Ratios (ORs) and 95% Confidence Intervals

(Cls) for Preterm Delivery in First-Parity Singleton Live Births

Associated with Traffic Air Pollution

Variable Crude 0.95 CI Adjusted

OR OR

Distance from freeway[dagger] (m)

<5001.321.04,1.671.30

500-1,500 1.00 1.00

Maternal age (yr)

[greater than or equal to] 35 0.85 0.47, 1.53 0.86

21-34 0.56 0.39, 0.80 0.61

[less than or equal to] 20 1.00 1.00

Season

Spring 0.89 0.64, 1.25 0.91

Summer 0.94 0.67, 1.32 0.95

Autumn 1.09 0.80, 1.49 1.08

Winter 1.00 1.00

Marital status

Unmarried 2.34 1.48, 3.70 2.07

Married 1.00 1.00

Maternal education (yr)

<120.980.74,1.310.89

[greater than or equal to] 12 1.00 1.00

Infant gender

Female 0.90 0.71, 1.14 0.90

Male 1.00 1.00

Variable 0.95 CI *

Distance from freeway[dagger] (m)

<5001.03,1.65

500-1,500

Maternal age (yr)

[greater than or equal to] 35 0.47, 1.57

21-34 0.42, 0.89

[less than or equal to] 20

Season

Spring 0.65, 1.27

Summer 0.68, 1.33

Autumn 0.79, 1.48

Winter

Marital status

Unmarried 1.29, 3.31

Married

Maternal education (yr)

<120.66,1.19

[greater than or equal to] 12

Infant gender

Female 0.71, 1.14

Male

* Data were based on multiple logistic-regression models in which all

6 variables were included.

([dagger]) Kaohsiung, Taiwan, ROC.

References

(1.) Watson AY, Bates RR, Kennedy D (Eds). Air Pollution, the Automobile, and Public Health. Washington, DC: National Academies Press, 1988; pp 39-75.

(2.) Godlee F. Health and the environment: air pollution. II. Road traffic and modern industry. Br Med J 1991; 303:1539-43.

(3.) Speizer FE, Ferris BG. The prevalence of chronic nonspecific respiratory disease in road tunnel employees. Am Rev Respir Dis 1963; 88:205-12.

(4.) Speizer FE, Ferris BG. Exposure to automobile exhaust. I. Prevalence of respiratory symptoms and disease. Arch Environ Health 1973; 26:313-18.

(5.) Speizer FE, Ferris BG. Exposure to automobile exhaust. II. Pulmonary function measurements. Arch Environ Health 1973; 26:319-24.

(6.) Tollerud DJ, Weiss T, Elting E, et al. The health effects of automobile exhaust. VI. relationship of respiratory symptoms and pulmonary function in tunnel and turnpike workers. Arch Environ Health 1983; 38:334-40.

(7.) Ayres SM, Evans R, Licht D, et al. Health effects of exposure to high concentrations of automotive emissions. Studies in bridge and tunnel workers in New York City. Arch Environ Health 1973; 27:168-78.

(8.) Evans RG, Webb K, Homan S, et al. Cross-sectional and longitudinal changes in pulmonary function associated with automobile pollution among bridge and tunnel officers. Am J Ind Med 1988; 14:25-36.

(9.) Gamble J, Jones W, Minshall S. Epidemiological-environmental study of diesel bus garage workers: chronic effects of diesel exhaust on the respiratory system. Environ Res 1987; 44:6-17.

(10.) Ulfvarson U, Alexandersson R, Aringer L, et al. Effects of exposure to vehicle exhaust on health. Scand J Work Environ Health 1987; 13:505-12.

(11.) Nielsen OR, Nielsen ML, Gehl J. Traffic-related air pollution: exposure and health effects in Copenhagen street cleaners and cemetery workers. Arch Environ Health 1995; 50:207-13.

(12.) Bener A, Galadari I, Mutawa JK, et al. Respiratory symptoms and lung function in garage workers and taxi drivers. J R Soc Health 1998; 118:346-53.

(13.) Yang CY, Chen YF, Chuang HY, et al. Respiratory and irritant health effects in tollbooth collectors in Taiwan. J Toxicol Environ Health 2002; 65:237-43.

(14.) Wjst M, Reitmeir P, Dold S, et al. Road traffic and adverse effects on respiratory health in children. Br Med J 1993; 307:596-600.

(15.) Nitta H, Sato T, Nakai S, et al. 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.

(16.) Edwards J, Waiters S, Griffiths RK. Hospital admissions for asthma in preschool children: relationship to major roads in Birmingham, United Kingdom. Arch Environ Health 1994; 49:223-27.

(17.) Weiland SK, Mundt KA, Ruckmann A, et al. Self-reported wheezing and allergic rhinitis in children and traffic density on street of residence. Ann Epidemiol 1994; 4: 243-47.

(18.) Nakai S, Nitta H, Maeda K. Respiratory health associated with exposure to automobile exhaust. II. Personal N[O.sub.2] exposure levels according to distance from the roadside. J Expo Anal Environ Epidemiol 1995; 5:125-36.

(19.) Nakai S, Nitta H, Maeda K. Respiratory health associated with exposure to automobile exhaust. III. Results of a cross-sectional study in 1987, and repeated pulmonary function tests from 1987 to 1990. Arch Environ Health 1999; 54:26-33.

(20.) Oosterlee A, Drijver M, Lebret E, et al. Chronic respiratory symptoms in children and adults living along streets with high traffic density. Occup Environ Med 1996; 53: 241-47.

(21.) Brunekreef B, Janssen N, de Hartog J, et al. Air pollution from truck traffic and lung function in children living near motorways. Epidemiology 1997; 8:298-303.

(22.) Studnicka M, Hackl E, Pischinger J, et al. Traffic-related N[O.sub.2] and the prevalence of asthma and respiratory symptoms in seven year olds. Eur Respir J 1997; 10:2275-78.

(23.) van Vliet P, Knape M, de Hartog J, et al. Motor vehicle exhaust and chronic respiratory symptoms in children living near freeways. Environ Res 1997; 74:122-32.

(24.) Ciccone G, Forastiere F, Agabiti N, et al. The SIDRIA collaborative group. Road traffic and adverse respiratory effects in children. Occup Environ Med 1998; 55:771-78.

(25.) Waldron G, Pottle B, Dod J. Asthma and the motorways: one district’s experience. J Public Health Rep 1995; 17: 85-89.

(26.) Livingstone AE, Shaddick G, Grundy C, et al. Do people living near inner city main roads have more asthma needing treatment? A case-control study. Br Med J 1996; 312: 676-77.

(27.) Yang CY, Yu ST, Chang CC. Respiratory symptoms in primary school children living near a freeway in Taiwan. J Toxicol Environ Health 2002; 65:747-55.

(28.) Venn A, Lewis S, Cooper M, et al. Local road traffic activity and the prevalence, severity, and persistence of wheeze in school children: combined cross sectional and longitudinal study. Occup Environ Med 2000; 57: 152-58.

(29.) Wang X, Ding H, Ryan L, et al. Association between air pollution and low birth weight: a community-based study. Environ Health Perspect 1997; 105:514-20.

(30.) Sram RJ. Impact of air pollution on reproductive health. Environ Health Perspect 1999; 107:542-43.

(31.) Xu X, Ding H, Wang X. Acute effects of total suspended particles and sulfur dioxides on preterm delivery: a community-based cohort study. Arch Environ Health 1995; 50:407-15.

(32.) Bobak M. Outdoor air pollution, low birth weight, and prematurity. Environ Health Perspect 2000; 108:173-76.

(33.) Ritz B, Yu F, Chapa G, et al. Effect of air pollution on preterm birth among children born in Southern California between 1989 and 1993. Epidemiology 2000; 11: 502-11.

(34.) Lin MC, Chiu HE, Yu HS, et al. Increased risk of preterm delivery in areas with air pollution from a petroleum refinery plant in Taiwan. J Toxicol Environ Health 2001; 64: 637-44.

(35.) Landgren O. Environmental pollution and delivery outcome in southern Sweden: a study with central registries. Acta Paediatr 1996; 85:1361-64.

(36.) Yang CY, Wang JD, Chan CC, et al. Respiratory and irritant health effects of a population living in a petrochemical polluted area in Taiwan. Environ Res 1997; 74: 145-49.

(37.) Yang CY, Chiu HF, Chiu JF, et al. Cancer mortality and residence near petrochemical industries in Taiwan. J Toxicol Environ Health 1997; 50:265-73.

(38.) Yang CY, Wang JD, Chan CC, et al. Respiratory symptoms of primary school children living in a petrochemical polluted area in Taiwan. Pediatr Pulmonol 1998; 25: 299-303.6

(39.) Yang CY, Cheng MF, Chiu JF, et al. Female lung cancer risk and petrochemical air pollution in Taiwan. Arch Environ Health 1999; 54:180-85.

(40.) Yang CY, Tsai SS, Cheng BH, et al. Sex ratio at birth associated with petrochemical air pollution in Taiwan. Bull Environ Contain Toxicol 2000; 65:126-31.

(41.) Yang CY, Cheng BH, Hsu TY, et al. Female lung cancer mortality and sex ratios at birth near a petroleum refinery plant. Environ Res 2000; 83:33-40.

(42.) Lin MC, Yu HS, Tsai SS, et al. Adverse pregnancy outcome in a petrochemical polluted area in Taiwan. J Toxicol Environ Health 2001; 63:565-74.

(43.) Wu SC, Young CL. Study of the birth reporting system. J Natl Public Health Assoc (ROC) 1986; 6:15-27.

(44.) Yang CY, Cheng BH, Tsai SS, et al. Association between chlorination of drinking water and adverse pregnancy outcome in Taiwan. Environ Health Perspect 2000; 108: 765-68.

(45.) Alexander GR, Tompkins ME, Comely DA. Gestational age reporting and preterm delivery. Public Health Rep 1990; 105:267-75.

(46.) Hosmer DW, Lemeshow S. Applied Logistic Regression. New York: John Wiley & Sons, 1989; pp 25-36.

(47.) SAS/STAT software (changes and enhancements through release 6.12). Cary, NC: SAS Institute, Inc., 1997.

(48.) Miettinen OS. Theoretical Epidemiology. New York: John Wiley & Sons, 1985; pp 29-36.

(49.) Rothman KJ, Greenland S. Modern Epidemiology, 2nd ed. New York: Lippincott-Raven, 1998; pp 96-100.

(50.) Yang CY, Chiu HF, Cheng MF, et al. Chlorination of drinking water and cancer mortality in Taiwan. Environ Res 1998; 78:1-6.

(51.) Yang CY, Cheng MF, Tsai SS, et al. Fluoride in drinking water and cancer mortality in Taiwan. Environ Res 2000; 82:189-93.

(52.) Kunzli N, Tager IB. The semi-individual study in air pollution epidemiology: a valid design as compared to ecologic studies. Environ Health Perspect 1997; 105:1078-83.

(53.) Shaw GM, Malcoe LH. Residential mobility during pregnancy for mothers of infants with or without congenital cardiac anomalies: a reprint. Arch Environ Health 1992; 47:236-38.

(54.) Khoury M, Stewart W, Weinstein A, et al. Residential mobility during pregnancy: implications for environmental teratogenesis. J Clin Epidemiol 1988; 41:15-20.

(55.) Kline J, Stein Z, Susser M. Conception to Birth: Epidemiology of Prenatal Development. New York: Oxford University Press, 1989; pp 174-207.

(56.) Berkowitz GS, Papiernik E. Epidemiology of preterm birth. Epidemiol Rev 1993; 15:414-43.

(57.) Bureau of Tobacco and Alcohol Monopoly (BTAM), Taiwan Provincial Government. Annual Report of Tobacco, Alcohol Consumption Investigation in the Taiwan Area. Taipei, ROC: BTAM, Taiwan Provincial Government, 1993.

(58.) Bobak M, Leon DA. Pregnancy outcomes and outdoor air pollution: an ecologic study in districts of the Czech Republic 1986-1988. Occup Environ Med 1999; 56: 539-43.

(59.) Gibbs RS, Romero R, Hillier SL, et al. A review of premature birth and subclinical infection. Am J Obstet Gynecol 1992; 166:1515-28.

CHUN-YUH YANG

CHIH-CHING CHANG

HUNG-YI CHUANG

Institute of Public Health

Kaohsiung Medical University

Kaohsiung, Taiwan

CHI-KUNG HO

TRONG-NENG WU

Institute of Occupational Safety and Health

Kaohsiung Medical University

Kaohsiung, Taiwan

SHANG-SHYUE TSAI

Department of Healthcare Administration

I-Shou University

Kaohsiung, Taiwan

This study was supported by a grant from the National Science Council, Executive Yuan, Taiwan (NSC-90-2320B-037037).

Submitted for publication April 30, 2002; revised; accepted for publication June 19, 2003.

Requests for reprints should be sent to Chun-Yuh Yang, Institute of Public Health, Kaohsiung Medical University, 100 Shih-Chuan 1st Rd, Kaohsiung, Taiwan 800708, Republic of China.

E-mail: chunyuh@kmu.edu.tw

COPYRIGHT 2003 Heldref Publications

COPYRIGHT 2004 Gale Group