Comparison of residential geocoding methods in population-based study of air quality and birth defects

Our population-based case-control study of air quality and birth defects in Texas relied on the geocoding of maternal residence from vital records for the assignment of air pollution exposures during early pregnancy. We attempted to geocode the maternal addresses for 5,338 birth defect cases and 4,574 frequency-matched controls using an automated procedure with standard matching criteria in ArcGIS 8.2 and 8.3. Initially, we matched 7,266 observations (73%). To increase the proportion of successful matches, we used an interactive procedure for the 2,646 addresses that were initially not geocoded by the software. This yielded an additional 985 matches (37%). Using the same 2,646 initially unmatched addresses, we compared the results of this interactive procedure to those of an automated procedure using lower standards. The automated procedure with lower standards yielded more matches (n=1,559, 59%) but with questionable accuracy. We included the interactively geocoded observations in our final data set. Their inclusion did not affect the estimates of air pollution exposure but increased our statistical power to detect associations between air quality and risk of selected birth defects. The geocoded and not geocoded populations differed in the distribution of Latino ethnicity (51% vs 59%) and ethnicity was independently associated with air pollution exposures (P<0.05). Geocoding status also appeared to modify the association between ethnicity and risk of birth defects; Latina women appeared to have a slightly lower risk of birth defects than non-Latina women in the geocoded population and to have a slightly higher risk in the not geocoded population. Incomplete geocoding may have resulted in a selection bias because of the under-representation of Latinas in our study population.

Community influenza activity and risk of acute influenza-like illness episodes among healthy unvaccinated pregnant and postpartum women

This study investigated the relation between weekly levels of influenza activity and the risk of acute influenza-like illness episodes among 8,323 healthy pregnant and postpartum women enrolled in a Puget Sound region, Washington, health maintenance organization, Group Health Cooperative, between June 1991 and December 1997. The authors classified weeks between October and May for isolate activity level based on surveillance data for influenza, respiratory syncytial virus, parainfluenza, and adenovirus infection. Influenza-like illness episodes were identified from medical encounters assigned a diagnostic code consistent with a symptomatic influenza infection. The authors compared the occurrence of influenza-like illness episodes within each pregnancy stage for periods with varying levels of influenza isolate detection in the community. Repeated-measures logistic regression methods accounted for time-dependent factors. The adjusted strength of association between influenza exposure and influenza-like illness episodes increased as the pregnancy stage progressed (first trimester odds ratio = 1.12, 95% confidence interval: 0.79, 1.59; second trimester odds ratio = 1.30, 95% confidence interval: 0.97, 1.73; third trimester odds ratio = 1.84, 95% confidence interval: 1.31, 2.59; postpartum period odds ratio = 2.28, 95% confidence interval: 1.42, 3.68). Pregnancy stage modified the association between influenza activity and influenza-like illness episodes. Findings estimate that 20-43 pregnant/postpartum women would need to be vaccinated with an 80% effective vaccine to prevent one influenza-like illness episode.

Attitudes toward participation in a pregnancy and child cohort study

While epidemiological studies aim for high participation rates, it is becoming increasingly difficult to recruit and retain participants in lengthy observational studies. We surveyed women who recently participated in the Pregnancy, Infection, and Nutrition Study during their pregnancy to learn more about what initially motivated them to participate in the study, their attitudes about the study protocol, and whether they would allow their child to participate in future studies. Most women were motivated by their interest in science and learning about their pregnancy. In general, women felt quite comfortable with most aspects of the study. Consent forms, telephone interviews and self-administered questionnaires were the most acceptable components of the study, but even specimen collection was well tolerated by this cohort. Women were less comfortable with the possibility of their child participating in future research. This survey confirmed that once women are enrolled, they tend to be willing to complete most components of an intensive study, suggesting that initial efforts for recruitment are most important.

Pesticide exposure and timing of menopause: the Agricultural Health Study

Age at menopause has implications for fertility and risk of hormonally related chronic diseases. Some pesticides disrupt reproductive hormones or are toxic to the ovary, but little is known about the association between pesticide exposure and timing of menopause. Cox proportional hazards modeling was used to examine the association between use of pesticides and age at menopause among 8,038 women living and working on farms in Iowa and North Carolina. Premenopausal women aged 35-55 years were followed from enrollment (1993-1997) to the date of their last menstrual period, or their follow-up interview (1999-2003) if still premenopausal. Women who experienced surgical menopause were censored at the date of surgery. Approximately 62% of the women reported ever mixing or applying pesticides; women who had never used pesticides were the comparison group for all analyses. After control for age, smoking status, and past use of oral contraceptives, the median time to menopause increased by approximately 3 months for women who used pesticides (hazard ratio = 0.87, 95% confidence interval: 0.78, 0.97) and by approximately 5 months for women who used hormonally active pesticides (hazard ratio = 0.77, 95% confidence interval: 0.65, 0.92). Pesticide use may be associated with a later age at menopause.

Direct observation of neighborhood attributes in an urban area of the US south: characterizing the social context of pregnancy

BACKGROUND

Neighborhood characteristics have been associated with poor maternal and child health outcomes, yet conceptualization of potential mechanisms is still needed. Census data have long served as proxies for area level socioeconomic influences. Unique information captured by neighborhood inventories, mostly conducted in northern US and Canadian urban areas, has shown important aspects of the community environment that are not captured by the socioeconomic and demographic aggregated individual statistics of census data. In this paper, we describe a neighborhood data collection effort tailored to a southern urban area.

METHODS

This study used data from the Pregnancy, Nutrition and Infection (PIN) prospective cohort study to describe neighborhoods where low- and moderate-income pregnant women reside. Women who participated in the PIN study and who resided in Raleigh, NC and its surrounding suburbs were included (n = 703). Neighborhood attributes captured by the inventory included litter, housing condition, road condition, and social interactions that informed theoretical constructs of physical incivility, territoriality and social spaces. US Housing and Population Census 2000 data at the block group level were also assessed to identify the unique contribution of directly observed data. We hypothesize that neighborhood environments can influence health through psychosocial mediated pathways that lead to increased stress, or through disadvantage leading to poor neighborhood resources, or by protective attributes through increased social control.

RESULTS

Findings suggest that directly observed neighborhood attributes distinguished between different types of areas in which low-income pregnant non-Hispanic white and non-Hispanic black women lived. Theoretically informed scales of physical incivilities, territoriality and social spaces were constructed and found to be internally consistent. Scales were weakly associated indicating that these constructs capture distinct information about these neighborhoods. Physical incivilities, territoriality and social spaces scales were poorly explained by traditional census variables used to proxy neighborhood environment.

CONCLUSION

If neighborhoods influence health through psychosocial mediated pathways then careful detailing of neighborhood attributes that contribute to stress or deterioration, beyond traditional socioeconomic status, are needed. We believe that measuring physical incivility, territoriality and social spaces as expressions of underlying issues of maintenance and social communication make important contributes to this field.

Physical exertion at work and the risk of preterm delivery and small-for-gestational-age birth

OBJECTIVE

To assess whether exposure to standing, lifting, night work, or long work hours during 3 periods of pregnancy are associated with an increased risk of preterm or small-for-gestational-age birth.

METHODS

The Pregnancy, Infection and Nutrition study is a prospective cohort with a nested case-control component that was conducted through clinic and hospital settings in Central North Carolina. A total of 1,908 women pregnant with a singleton gestation were recruited during prenatal visits from January 1995 through April 2000 and provided information during telephone and face-to-face interviews about physical exertion for the 2 longest-held jobs during pregnancy.

RESULTS

No significant elevations in preterm delivery were observed among women who lifted repeatedly or stood at least 30 hours per week, with no changes in risk estimates over the course of pregnancy. A 50% elevation in the risk of preterm delivery (relative risk 1.5, 95% confidence interval 1.0-2.0; first trimester) was observed among women who reported working at night (10:00 PM to 7:00 AM), whereas a 40% reduction in risk was observed among women working at least 46 hours per week (relative risk 0.6, 95% confidence interval 0.4-0.9; first trimester), regardless of period of exposure. No elevations in small-for-gestational-age birth were observed among women exposed to any of the 4 types of occupational exertion.

CONCLUSION

Physically demanding work does not seem to be associated with adverse pregnancy outcomes, whereas working at night during pregnancy may increase the risk of preterm delivery. Studies to examine the effect of shift work on uterine activity would help to clarify the possibility of a causal effect on preterm birth.

Occupational magnetic field exposure and the risk of acoustic neuroma

BACKGROUND

Acoustic neuroma is the intracranial tumor subtype showing the highest relative risk in relation to ionizing radiation but other environmental risk factors are largely unknown. This study was performed to investigate the effect of power frequency magnetic fields.

METHOD

A total of 793 cases between 1987 and 1999 were identified through the Swedish cancer registry and 101,762 controls were randomly selected from the total population. Information about occupation was obtained from censuses and linked to gender specific job-exposure matrices based on actual measurements of 50 Hz magnetic field exposure.

RESULT

We investigated time-weighted average, peak values, and rate of change of magnetic field exposure considering several different time windows in relation to cancer diagnosis. We found no increases in risks regardless of exposure metric or time window of exposure.

CONCLUSION

This study is the largest ever on acoustic neuroma and the first study to evaluate this tumor subtype specifically in relation to extremely low frequency magnetic fields. The results do not support the hypothesis that 50 Hz magnetic fields increase the risk of acoustic neuroma.

Urbanization and breast cancer incidence in North Carolina, 1995-1999

PURPOSE

Breast cancer incidence rates are reported to be higher in urban compared with rural areas in the United States. We investigated the relationship between urbanization and breast cancer in North Carolina (1995-1999), and considered hospital characteristics as an explanation.

METHODS

We calculated age-adjusted in situ and invasive female breast cancer incidence rates stratified by race, urbanization (Urban Influence Codes), and the presence of a hospital with a cancer registry and cancer program approval in a county.

RESULTS

For white women, incidence rate ratios (IRRs) comparing the most urban with the most rural counties were 1.60 for in situ and 1.18 for invasive cancer. For non-white women, IRRs were 1.27 and 0.99, respectively. IRRs for incidence in registry hospital counties versus those without were all > 1.00 and differences were greater for in situ cancer than invasive. For most strata, urban excesses were attenuated when further stratified by registry hospital status.

CONCLUSIONS

For most strata, we observed excess incidence in urban counties, but it appeared to be explained through the urban preponderance of registry hospitals. Counties with these hospitals may have higher incidence because of increased detection. Area hospital characteristics should be considered when evaluating geographic patterns of breast cancer incidence.

Considerations for improving the accuracy of exposure to disinfection by-products by ingestion in epidemiologic studies

Disinfection by-product (DBP) exposure characterization studies are often based on the analysis of a limited number of samples collected from a distribution system (DS) in which DBP levels are variable over time and space. A compositing technique was developed to simplify the sample collection procedures for integrating over temporal variations in DBPs measured in terms of trihalomethanes (THMs), haloacetic acids (HAAs), and total organic halogen (TOX). Over the course of 5 days analysis, the single composited sample was within 94-100% of the average THM concentration in all grab samples, 92-105% of HAAs, and 130% of the TOX concentration. Additionally, temporal variability factors such as timing of sample collection and the handling of tap water prior to consumption were found to influence DBP levels in consumers' drinking water. Included in our study of home water use are the effects of boiling which removed up to 98% of THMs and point of use (POU) devices which all showed DBP removal but differed depending on the device used. These factors should be taken into consideration in DBP exposure characterization for epidemiologic studies.

Characteristics that predict locating and interviewing mothers identified by a state birth defects registry and vital records

BACKGROUND

State vital records are often used to select population-based controls in record-linkage studies of birth defects. However, locating and contacting individuals based on these data sources to collect additional data can be a challenge.

METHODS

A large case-control study of air quality and birth defects was conducted in 7 Texas counties in which cases were selected from the Texas Birth Defects Registry and controls from state vital records. In 2004, data from these sources were used to trace mothers of cases and controls who delivered babies in the year 2000 (n=2477) for participation in a computer-assisted telephone interview. A number of factors that predicted whether an individual would be located and interviewed were identified.

RESULTS

Between March and August 2004, 38% of the mothers were located, and 38% of the located mothers were interviewed. Case mothers were more likely than control mothers to be located (44 vs. 30%) and, if located, to be interviewed (43 vs. 31%). We compared the characteristics of mothers who were not located (case n=760; control n=777), mothers who were located but not interviewed (case n=344; control n=236), and mothers who were interviewed (case n=256; control n=104). Among both cases and controls, older mothers (>or=30 years) were more likely than younger mothers to be located, and non-Hispanic black mothers were least likely to be located and interviewed.

CONCLUSIONS

Despite the utility of vital records as a source of population-based controls in record-linkage analyses, the poor response rate discourages the use of these data sources to contact individuals for a follow-up study 4 years after delivery.

Risk of spontaneous preterm birth is associated with common proinflammatory cytokine polymorphisms

BACKGROUND

Preliminary data suggest that common genetic variation in immune response genes can contribute to the risk for spontaneous preterm birth and possibly small-for-gestational age (SGA).

METHODS

We investigated the relationship of polymorphisms in 6 cytokine genes associated with inflammation-interleukin (IL)1alpha, IL1beta, IL2, IL6, tumor necrosis factor (TNF), and lymphotoxin alpha (LTA)-with spontaneous preterm and SGA birth in a nested case-control study drawn from a prospective pregnancy cohort. Women were recruited between 24 and 29 weeks' gestation at the Wake County and University of North Carolina, Chapel Hill obstetric clinics between February 1996 and June 2000. We inferred haplotypes using the EM algorithm and the Bayesian method, PHASE. We then compared haplotype frequency distributions and implemented semi-Bayesian hierarchical logistic regression analyses to obtain odds ratio (OR) estimates and 95% confidence intervals (CIs) for each polymorphism.

RESULTS

Two haplotypes spanning the TNF/LTA genes were associated with increased risk for spontaneous preterm birth in white subjects (for the AGG haplotype, OR = 1.5 [95% CI=0.8-2.6]; for the GAC haplotype, 1.6 [0.9-2.9]). Additionally, carriers of the GAG haplotype were found to have decreased risk of spontaneous preterm birth (0.6; 0.3-1.0). The TNF(-488)A and LTA(IVS1-82)C variants, constituents of the AGG and GAC haplotypes respectively, were also strongly associated with increased risk of spontaneous preterm birth.

CONCLUSIONS

Our results suggest that common genetic variants in proinflammatory cytokine genes could influence the risk for spontaneous preterm birth. Selected TNF/LTA haplotypes were associated with spontaneous preterm birth in both African-American and white subjects. Our data do not support an inflammatory etiology for SGA.

Risk of small-for-gestational age is associated with common anti-inflammatory cytokine polymorphisms

BACKGROUND

Anti-inflammatory cytokines play a key role in pregnancy maintenance. Genetic variation in anti-inflammatory cytokines could influence a woman's risk of adverse reproductive outcomes.

METHODS

We investigated the relationship of polymorphisms in interleukin 4 (IL4), IL5, IL10, IL13, and transforming growth factor (TGFbeta1) with spontaneous preterm birth and small-for-gestational age (SGA) in a nested case-control study of a prospective pregnancy cohort. Women were recruited between 24 and 29 weeks' gestation at the Wake County and University of North Carolina, Chapel Hill obstetric clinics between February 1996 and June 2000. We inferred haplotypes using the EM algorithm and the Bayesian method, PHASE. Semi-Bayesian hierarchical logistic regression was used to obtain odds ratio (OR) estimates and 95% confidence intervals (CIs) for each polymorphism.

RESULTS

African-American mothers who carried the IL4 GCC haplotype had greater risk of spontaneous preterm birth (OR = 2.9; 95% CI = 1.2-7.4). In white mothers, carriers of the "low-producing" IL4 CC and IL10 ATA haplotypes had markedly reduced risk of SGA (for the CC haplotype, 0.2 [0.0-1.2]; for the ATA haplotype, 0.5 [0.3-0.8]), whereas carriers of the "high-producing" IL4(-589)T variant had increased risk of SGA in both African-American and white mothers.

CONCLUSIONS

Variants related to decreased anti-inflammatory cytokine production may lower risk of SGA. Furthermore, the same mechanism that protects against SGA might increase risk of spontaneous preterm birth.

Measures of placental growth in relation to birth weight and gestational age

Fetal growth depends in part on placental growth. The authors tested placental measures derived from digital images for reliability and to evaluate their association with birth weight and gestational age. A total of 628 women recruited into the Pregnancy, Infection, and Nutrition Study, a prospective cohort study of preterm birth in central North Carolina between 2002 and 2004, delivered singleton liveborn infants after 24 completed weeks' gestation. Novel chorionic plate morphometric parameters captured off digital images of the gross placenta were analyzed as estimators of gestational age and birth weight. Without acknowledgment to placental weight, digitally obtained lateral chorionic plate growth measures accounted for 17 percent of gestational age variance and 35 percent of birth weight variance, overall. Chorionic plate measures accounted for 10 percent of birth weight variance beyond that accounted for by placental weight alone. Among preterm births, 34 percent of gestational age variance and 63 percent of birth weight variance were accounted for by lateral chorionic plate growth measures. Intraclass correlation coefficients for the novel digital measures ranged from 0.96 to 0.98. Reliable digital measures of lateral chorionic plate growth estimate birth weight variance more strongly than gestational age, project variance that is not accounted for by placental weight, and project these outcomes to a greater degree in preterm births than at term.

Neighborhood crime, deprivation, and preterm birth

PURPOSE

Significant racial disparity in preterm birth (PTB; birth at <37 weeks' gestation) exists, poorly explained by Individual-level factors. This research explores whether neighborhood crime contributes to the racial disparity in PTB.

METHODS

Geocoded Wake County, NC, birth records and crime-report data for 1999 to 2001 were merged with US Census data (2000). Race-stratified logistic and multilevel logistic models produced odds ratios (ORs) and 95% confidence intervals (CIs) for block-group violent, theft, property, and vice crime rates and singleton PTB.

RESULTS

A total of 13,960 women resided in a 114-block-group crime area. Non-Hispanic black women were more likely than non-Hispanic white women to deliver preterm (12.8% versus 6.7%), live in economically deprived block groups (42.2% versus 19.3% in the highest deprivation quartile), and experience more crime (32.0% versus 3.8% in the highest violent-crime-rate quartile). Quartiles of violent, theft, property, and vice crimes were associated with PTB in unadjusted models. Living in very high violent-crime-rate block-group quartiles was suggestive of increased odds of PTB for white and black non-Hispanic women (OR = 1.5; 95% CI, 0.9-2.6; and OR = 1.4; 95% CI, 1.0-2.1, respectively) in adjusted models. Other crime effects were attenuated after adjustment.

CONCLUSIONS

Differential neighborhood exposures may contribute to racial disparity in PTB.

A quantitative risk-benefit analysis of changes in population fish consumption

Although a rich source of n-3 polyunsaturated fatty acids (PUFAs) that may confer multiple health benefits, some fish contain methyl mercury (MeHg), which may harm the developing fetus. U.S. government recommendations for women of childbearing age are to modify consumption of high-MeHg fish, while recommendations encourage fish consumption among the general population because of nutritional benefits. To investigate the aggregate impacts of hypothetical shifts in fish consumption, the Harvard Center for Risk Analysis convened an expert panel (see acknowledgements). Effects investigated include prenatal cognitive development, coronary heart disease mortality, and stroke. Substitution of fish with high MeHg concentrations with fish containing less MeHg among women of childbearing age yields substantial developmental benefits and few negative impacts. However, if women instead decrease fish consumption, countervailing risks substantially reduce net benefits. If other adults (mistakenly and inappropriately) also reduce their fish consumption, the net public health impact is negative. Although high compliance with recommended fish consumption patterns can improve public health, unintended shifts in consumption can lead to public health losses. Risk managers should investigate and carefully consider how populations will respond to interventions, how those responses will influence nutrient intake and contaminant exposure, and how these changes will affect aggregate public health.

A quantitative analysis of fish consumption and coronary heart disease mortality

Although a rich source of n-3 polyunsaturated fatty acids (PUFAs) that may confer multiple health benefits, some fish contain methyl mercury (MeHg), which may harm the developing fetus. U.S. government recommendations for women of childbearing age are to modify consumption of high-MeHg fish to reduce MeHg exposure, while recommendations encourage fish consumption among the general population because of the nutritional benefits. The Harvard Center for Risk Analysis convened an expert panel (see acknowledgements) to quantify the net impact of resulting hypothetical changes in fish consumption across the population. This paper estimates the impact of fish consumption on coronary heart disease (CHD) mortality and nonfatal myocardial infarction (MI). Other papers quantify stroke risk and the impacts of both prenatal MeHg exposure and maternal intake of n-3 PUFAs on cognitive development. This analysis identified articles in a recent qualitative review appropriate for the development of a dose-response relationship. Studies had to satisfy quality criteria, quantify fish intake, and report the precision of the relative risk estimates. Relative risk results were averaged, weighted proportionately by precision. CHD risks associated with MeHg exposure were reviewed qualitatively because the available literature was judged inadequate for quantitative analysis. Eight studies were identified (29 exposure groups). Our analysis estimated that consuming small quantities of fish is associated with a 17% reduction in CHD mortality risk, with each additional serving per week associated with a further reduction in this risk of 3.9%. Small quantities of fish consumption were associated with risk reductions in nonfatal MI risk by 27%, but additional fish consumption conferred no incremental benefits.

A quantitative analysis of fish consumption and stroke risk

Although a rich source of n-3 polyunsaturated fatty acids (PUFAs) that may confer multiple health benefits, some fish contain methyl mercury (MeHg), which may harm the developing fetus. U.S. government recommendations for women of childbearing age are to modify consumption of high-MeHg fish to reduce MeHg exposure, while recommendations encourage fish consumption among the general population because of the nutritional benefits. The Harvard Center for Risk Analysis convened an expert panel (see acknowledgements) to quantify the net impact of resulting hypothetical changes in fish consumption across the population. This paper estimates the impact of fish consumption on stroke risk. Other papers quantify coronary heart disease mortality risk and the impacts of both prenatal MeHg exposure and maternal intake of n-3 PUFAs on cognitive development. This analysis identified articles in a recent qualitative literature review that are appropriate for the development of a dose-response relationship between fish consumption and stroke risk. Studies had to satisfy quality criteria, quantify fish intake, and report the precision of the relative risk estimates. The analysis combined the relative risk results, weighting each proportionately to its precision. Six studies were identified as appropriate for inclusion in this analysis, including five prospective cohort studies and one case-control study (total of 24 exposure groups). Our analysis indicates that any fish consumption confers substantial relative risk reduction compared to no fish consumption (12% for the linear model), with the possibility that additional consumption confers incremental benefits (central estimate of 2.0% per serving per week).

Pregnancy Intendedness, Maternal Psychosocial Factors and Preterm Birth

OBJECTIVES

This study examined associations between reported pregnancy intendedness and several maternal psychosocial factors in relation to preterm birth (<37 weeks' completed gestation).

METHODS

Women were recruited into a prospective cohort study between the 24th and 29th weeks of pregnancy in central North Carolina from 1996 to 2000. Prior to delivery, participants responded to questions about pregnancy intendedness, life events impacts, depressive symptoms, and coping style.

RESULTS

Women who reported not intending their pregnancy had increased odds of reporting low, medium and high levels of perceived stress during pregnancy (OR = 1.4 [95% CI: 1.1, 1.9], OR = 2.2 [95% CI: 1.7, 2.8], and OR = 3.4 [95% CI: 2.6, 4.5], respectively, relative to very low), medium and high levels of depressive symptoms (OR = 2.2 [95% CI: 1.8, 2.9] and OR = 3.1 [95% CI: 2.4, 3.9], respectively), and medium and high levels of several coping styles. Reporting not intending the pregnancy was not associated with increased risk of preterm birth (Risk Ratio [RR] = 1.0, 95% CI: 0.8, 1.1), but reporting the highest quartile of perceived stress (RR = 1.6, 95% CI: 1.1, 2.3) and the highest tertile of distancing coping style (compared with lowest quartile) was associated with preterm birth (RR = 1.4, 95% CI: 1.1, 1.9). Interactions between pregnancy intendedness and the psychosocial variables perceived stress, depression or coping style did not modify the psychosocial variable's associations with preterm birth.

CONCLUSIONS

Pregnancy intendedness remains an important concept in the reproductive health literature integrally tied to indicators of maternal mental health, but not necessarily to pregnancy outcomes.

Maternal urine albumin excretion and pregnancy outcome

BACKGROUND

Vascular dysfunction has been hypothesized as a causal pathway for preeclampsia, impaired fetal growth, and early parturition. The relationship between increased urine albumin excretion (albuminuria), a marker of endothelial dysfunction, and preterm birth has not been fully evaluated.

METHODS

We conducted a nested case-control study of 404 pregnancies from 1998 to 2000 within the Pregnancy, Infection and Nutrition cohort. Cases consisting of live births delivered before 37 weeks of gestation (preterm birth, n = 111) were compared with term births from the same cohort (n = 293). Albumin-creatinine ratio (in milligrams per gram) was measured in urine collected around 27 weeks of gestation. We compared risks for preterm birth in mothers with albuminuria with albumin levels of 3 to 20 mg/g and greater than 20 mg/g relative to those with albumin levels less than 3 mg/g by using logistic regression.

RESULTS

Median albuminuria was albumin level of 2.7 and 4.3 mg/g for term and preterm births, respectively. Albuminuria was strongly associated with preterm birth in a dose-response fashion, with adjusted odds ratios of 1.9 (95% confidence interval, 1.1 to 3.1) and 4.7 (95% confidence interval, 1.7 to 12.6) for albuminuria with albumin of 3 to 20 mg/g and greater than 20 mg/g, respectively. The association was present for both spontaneous and medically induced preterm births, but the effect was decreased and the dose-response relationship was eliminated by excluding high-risk groups and those with pregnancy complications.

CONCLUSION

Low levels of albuminuria are associated with preterm birth. The mechanism underlying this association warrants additional exploration.

Diagnosis of bacterial vaginosis from self-obtained vaginal swabs

OBJECTIVE: To determine the concordance between vaginal fluid Gram stains and pH obtained at speculum exam with similar stains and pH prepared from self-obtained vaginal swabs. METHODS: Using vaginal fluid Gram stain, 129 pregnant women were screened for bacterial vaginosis at 24 to 29 weeks' gestation. Two smears were collected from each woman during the same prenatal visit: the first was prepared from a self-obtained vaginal swab and the second from a physician-obtained speculum examination. Vaginal pH was recorded for each swab. Kappa coefficient was used to quantify agreement between the two sets of results. RESULTS: When compared with the physician-obtained smear, the ability of the self-obtained Gram stain to diagnose bacterial vaginosis had a sensitivity of 77%, specificity of 97%, positive predictive value of 71% and negative predictive value of 97%. There was substantial agreement (weighted kappa=0.82) between the two techniques in the ability to determine the grade of vaginal flora. CONCLUSION: When compared with physician-obtained vaginal smears, self-obtained smears have substantial agreement in the diagnosis of bacterial vaginosis.

Maternal Pre-pregnancy Overweight and Obesity and the Risk of Cesarean Delivery in Nulliparous Women

PURPOSE

To examine the effect of maternal pre-pregnancy overweight and obesity on the risk of term cesarean delivery in nulliparous women.

METHODS

The authors examined data from 641 nulliparous women with a term pregnancy that participated in the Pregnancy, Infection, and Nutrition Study from 1995 to 2002. Unadjusted and adjusted risk ratios and 95% confidence intervals (CI) were computed for normal weight (BMI 19.8-26.0 kg/m(2)), overweight (BMI 26.1-29.0 kg/m(2)), and obese (BMI>29.0 kg/m(2)) women. Normal weight women served as the referent population.

RESULTS

The unadjusted risk ratio for cesarean delivery for overweight women compared with normal weight women was 1.4 (95% CI, 0.97, 2.1) and for obese women compared with normal weight women was 1.4 (95% CI, 1.03, 2.0). After controlling for maternal height, education, weight gain during pregnancy, and labor induction, the adjusted risk ratio for cesarean delivery among overweight women was 1.2 (95% CI, 0.8, 1.8). The adjusted risk ratio for obese women was 1.5 (95% CI, 1.05, 2.0).

CONCLUSION

Our analysis confirms that there is a moderate association between maternal pre-pregnancy obesity and an unplanned term cesarean delivery. However, the risk is not as large as previously reported.

Relation between ambient air quality and selected birth defects, seven county study, Texas, 1997-2000

A population-based case-control study investigated the association between maternal exposure to air pollutants, carbon monoxide, nitrogen dioxide, ozone, sulfur dioxide, and particulate matter <10 microm in aerodynamic diameter during weeks 3-8 of pregnancy and the risk of selected cardiac birth defects and oral clefts in livebirths and fetal deaths between 1997 and 2000 in seven Texas counties. Controls were frequency matched to cases on year of birth, vital status, and maternal county of residence at delivery. Stationary monitoring data were used to estimate air pollution exposure. Logistic regression models adjusted for covariates available in the vital record. When the highest quartile of exposure was compared with the lowest, the authors observed positive associations between carbon monoxide and tetralogy of Fallot (odds ratio = 2.04, 95% confidence interval: 1.26, 3.29), particulate matter <10 microm in aerodynamic diameter and isolated atrial septal defects (odds ratio = 2.27, 95% confidence interval: 1.43, 3.60), and sulfur dioxide and isolated ventricular septal defects (odds ratio = 2.16, 95% confidence interval: 1.51, 3.09). There were inverse associations between carbon monoxide and isolated atrial septal defects and between ozone and isolated ventricular septal defects. Evidence that air pollution exposure influences the risk of oral clefts was limited. Suggestive results support a previously reported finding of an association between ozone exposure and pulmonary artery and valve defects.

Regression models for clustered binary responses: implications of ignoring the intracluster correlation in an analysis of perinatal mortality in twin gestations

PURPOSE

Dependent binary responses, such as health outcomes in twin pairs or siblings, frequently arise in perinatal epidemiologic research. This gives rise to correlated data, which must be taken into account during analysis to avoid erroneous statistical and biological inferences.

METHODS

An analysis of perinatal mortality (fetal deaths plus deaths within the first 28 days) in twins in relation to cluster-varying (those that are unique to each fetus within a twin pregnancy such as birthweight) and cluster-constant (those that are identical for both twins within a sibship such as maternal smoking status) risk factors is presented. Marginal (ordinary logistic regression [OLR] and logistic regression using generalized estimating equations [GEE]) and cluster-specific (conditional and random-intercept logistic regression models) regression models are fit and their results contrasted. The United States "matched multiple data" file of twin births (1995-1997), which includes 285,226 twins from 142,613 pregnancies, was used to examine the implications of ignoring of clustering on regression inferences.

RESULTS

The OLR models provide variance estimates for cluster constant covariates that ranged from 7% to 71% smaller than those from GEE-based models. This underestimation is even more pronounced for some cluster-varying covariates, ranging from 21% to 198%.

CONCLUSIONS

Ignoring the cluster dependency is likely to affect the precision of covariate effects and consequently interpretation of results. With widespread availability of appropriate software, statistical methods for taking the intracluster dependency into account are easily implemented and necessary.

Predictors of vaginal bleeding during the first two trimesters of pregnancy

This study evaluates maternal age, race, cigarette smoking, prior spontaneous abortion, prior induced abortion, and prior preterm birth in relation to vaginal bleeding during the first two trimesters of pregnancy. Information on vaginal bleeding and predictors came from the Pregnancy, Infection, and Nutrition Study, which enrolled 2806 pregnant women at 24-29 weeks' gestation during 1995-2000 in central North Carolina, USA. Generalised estimating equations were applied to take into account repeated episodes of vaginal bleeding during pregnancy. Women with advanced maternal age and passive smoking exposure were more likely to experience more intense vaginal bleeding during pregnancy, as were women with prior preterm birth. More intense bleeding was also more likely to be reported among women with multiple prior spontaneous abortions or multiple prior induced abortions, but not among women with a single prior spontaneous or induced abortion. The combination of prior spontaneous and induced abortion showed a dose-response association with the occurrence of vaginal bleeding during pregnancy.

Probability samples of area births versus clinic populations for reproductive epidemiology studies

Studies of pregnancy outcome are generally based either on geographically defined populations, often from birth records, or on clinic or hospital populations. We compared women recruited into a clinic-based study, the Pregnancy, Infection, and Nutrition (PIN) Study, with women who resided in the geographical area of the study (Alamance, Orange, and Wake Counties in North Carolina) and gave birth over the corresponding time period (1996-2000). Clinic participants were more likely to be Black, younger, have lower education, be unmarried, have a more frequent history of adverse pregnancy outcome, obtain prenatal care later, and smoke more cigarettes. Despite that profile, the proportion of clinic participants delivering preterm was somewhat lower than among area women overall (10.8% vs. 11.3%). Black/White risk ratios for preterm birth were markedly different for area (1.6) vs. clinic women (1.1), whereas other predictors were similar. Patterns may differ across groups for many reasons, including self-selection of clinics and varying clinical practices.

Reducing misclassification in assignment of timing of events during pregnancy

BACKGROUND

Perinatal epidemiology studies often collect only the calendar month in which an event occurs in early pregnancy because it is difficult for women to recall a specific day when queried later in pregnancy or postpartum. Lack of day information may result in incorrect assignment of completed gestational month because calendar months and pregnancy months are not aligned.

METHODS

To examine the direction and magnitude of misclassification, we compared 3 methods for assignment of completed gestational month: 1) calendar month difference, 2) conditional month difference, and 3) imputed month midpoint. We used data from the Pregnancy, Infection, and Nutrition Study for simulations.

RESULTS

Calendar month difference misclassified 54% of events as 1 month later in pregnancy compared with the actual completed month of gestation. Each of the other 2 methods misclassified approximately 12% of events to 1 month earlier and 12% to 1 month later.

CONCLUSIONS

Calendar month difference, a common method, has the greatest misclassification. Conditional month difference and imputed month midpoint, which require little effort to implement, are superior to calendar month difference for reducing misclassification.

A time-series analysis of air pollution and preterm birth in Pennsylvania, 1997-2001

Preterm delivery can lead to serious infant health outcomes, including death and lifelong disability. Small increases in preterm delivery risk in relation to spatial gradients of air pollution have been reported, but previous studies may have controlled inadequately for individual factors. Using a time-series analysis, which eliminates potential confounding by individual risk factors that do not change over short periods of time, we investigated the effect of ambient outdoor particulate matter with diameter < or = 10 microm (PM10) and sulfur dioxide on risk for preterm delivery. Daily counts of preterm births were obtained from birth records in four Pennsylvania counties from 1997 through 2001. We observed increased risk for preterm delivery with exposure to average PM10 and SO2 in the 6 weeks before birth [respectively, relative risk (RR) = 1.07; 95% confidence interval (CI), 0.98-1.18 per 50 microg/m3 increase; RR = 1.15; 95% CI, 1.00-1. 32 per 15 ppb increase], adjusting for long-term preterm delivery trends, co-pollutants, and offsetting by the number of gestations at risk. We also examined lags up to 7 days before the birth and found an acute effect of exposure to PM10 2 days and 5 days before birth (respectively, RR = 1.10; 95% CI, 1.00-1.21; RR = 1.07; 95% CI, 0.98-1.18) and SO2 3 days before birth (RR = 1.07; 95% CI, 0.99-1.15), adjusting for covariates, including temperature, dew point temperature, and day of the week. The results from this time-series analysis, which provides evidence of an increase in preterm birth risk with exposure to PM10 and SO2, are consistent with prior investigations of spatial contrasts.

The epidemiology of threatened preterm labor: a prospective cohort study

OBJECTIVE

The purpose of this study was to describe the occurrence, timing, and outcomes of hospital-based diagnoses of preterm labor.

STUDY DESIGN

Administrative records identified hospital admissions for preterm labor among 2534 women in an ongoing cohort study. Factors that were considered risks for prematurity were examined by logistic regression for an association with any preterm labor diagnosis, a preterm labor diagnosis <33 weeks of gestation, or > or =33 weeks of gestation.

RESULTS

Of 234 women (9%) who experienced hospitalization for preterm labor, 90 women (38%) were delivered in the first episode. Previous preterm birth consistently was associated with a diagnosis of preterm labor. Reporting a sexually transmitted infection (odds ratio, 1.8; 95% CI, 1.1-3.0) or bacterial vaginosis (odds ratio, 2.6; 95% CI, 1.7-4.1) early in pregnancy was associated with hospitalization for preterm labor between 24 and 32 weeks of gestation.

CONCLUSION

The incidence of first-time hospitalization for preterm labor was 9%, with most episodes not resulting in preterm birth. Previous preterm birth was associated therefore with a preterm labor diagnosis.

Should spontaneous and medically indicated preterm births be separated for studying aetiology?

An increasing proportion of preterm births result from medical interventions, and the practice of aggregating all preterm births vs. splitting into spontaneous and medically indicated preterm births is inconsistent. While mechanistic and clinical arguments can be offered for either approach, we empirically evaluated the predictiveness of a range of risk factors for preterm birth in the Pregnancy, Infection, and Nutrition Study. Most influences were shared across the two subsets - African-American ethnicity, advancing age, delivery at a university medical centre, prior preterm birth and smoking. Medically indicated preterm births appeared to be associated with intensity of medical care, higher in the university medical centre and lower for the poorest women. Body mass index was positively associated with medically indicated preterm birth and inversely with spontaneous preterm birth. Given the complexity of the aetiological pathways, both aggregation and disaggregation are well justified and should be included in studies of the causes of preterm birth.

Maternal prepregnancy overweight and obesity and the pattern of labor progression in term nulliparous women

OBJECTIVE

To examine the effect of maternal overweight and obesity on labor progression.

METHODS

We analyzed data from 612 nulliparous women with a term pregnancy that participated in the Pregnancy, Infection, and Nutrition Study from 1995 to 2002. The median duration of labor by each centimeter of cervical dilation was computed for normal-weight (body mass index [BMI] 19.8-26.0 kg/m2), overweight (BMI 26.1-29.0 kg/m2), and obese (BMI > 29.0 kg/m2) women and used as a measurement of labor progression.

RESULTS

After adjusting for maternal height, labor induction, membrane rupture, oxytocin use, epidural analgesia, net maternal weight gain, and fetal size, the median duration of labor from 4 to 10 cm was significantly longer for both overweight and obese women, compared with normal-weight women (7.5, 7.9, and 6.2 hours, respectively). For overweight women, the prolongation was concentrated around 4-6 cm, whereas for obese women, their labor was significantly slower before 7 cm.

CONCLUSION

Labor progression in overweight and obese women was significantly slower than that of normal-weight women before 6 cm of cervical dilation. Given that nearly one half of women of childbearing age are either overweight or obese, it is critical to consider differences in labor progression by maternal prepregnancy BMI before additional interventions are performed.

Pesticide use and menstrual cycle characteristics among premenopausal women in the Agricultural Health Study

Menstrual cycle characteristics may have implications for women's fecundability and risk of hormonally related diseases. Certain pesticides disrupt the estrous cycle in animals. The authors investigated the cross-sectional association between pesticide use and menstrual function among 3,103 women living on farms in Iowa and North Carolina. Women were aged 21-40 years, premenopausal, not pregnant or breastfeeding, and not taking oral contraceptives. At study enrollment (1993-1997), women completed two self-administered questionnaires on pesticide use and reproductive health. Exposures of interest were lifetime use of any pesticide and hormonally active pesticides. Menstrual cycle characteristics of interest included cycle length, missed periods, and intermenstrual bleeding. The authors used generalized estimating equations to assess the association between pesticide use and menstrual cycle characteristics, controlling for age, body mass index, and current smoking status. Women who used pesticides experienced longer menstrual cycles and increased odds of missed periods (odds ratio = 1.5, 95% confidence interval: 1.2, 1.9) compared with women who never used pesticides. Women who used probable hormonally active pesticides had a 60-100% increased odds of experiencing long cycles, missed periods, and intermenstrual bleeding compared with women who had never used pesticides. Associations remained after control for occupational physical activity.

Second trimester folate status and preterm birth

OBJECTIVE

The purpose of this study was to show that maternal folate status during pregnancy may be related to preterm birth.

STUDY DESIGN

Women were recruited at 24 to 29 weeks' gestation from 1995 to 2000 into the Pregnancy, Infection, and Nutrition Study. Those who completed an interview and a food frequency questionnaire, or provided a blood sample for radioassay of serum (n = 2026) and red blood cell (n = 1034) folate were included.

RESULTS

Mean daily dietary folate intake was 463 microg (SD +/- 248). Intake </=500 microg was associated with increased preterm delivery (RR = 1.8, 95% CI 1.4-2.6) controlling for total energy intake. Serum folate levels <16.3 ng/mL and red blood cell folate levels </=626.6 ng/mL yielded adjusted risk ratios of 1.8 (95% CI 1.3-2.5) and 1.7 (95% CI 1.1-2.6), respectively. Patterns were similar for spontaneous and overall preterm birth.

CONCLUSION

These results support the hypothesis that low folate levels during the second trimester of pregnancy are associated with an increased risk of preterm birth.

Multivitamin use and the risk of preterm birth

Previous research suggests that multivitamin use before and during pregnancy can diminish diet-related deficiencies of certain micronutrients and potentially prevent preterm birth. To assess this association, the authors performed an analysis by using data from the Pregnancy, Infection, and Nutrition Study (n = 2,010). Women were recruited at 24-29 weeks of pregnancy from four prenatal care clinics in North Carolina from August 1995 to June 2000. For women who took multivitamins prior to pregnancy, compared with nonusers, the adjusted risk ratio was 0.50 (95% confidence interval: 0.20, 1.25) for delivering preterm (<37 weeks). In contrast, prenatal and periconceptional use, compared with nonuse, were not related to preterm birth, with adjusted risk ratios of 1.1. Preconceptional multivitamin use was inversely associated with both early (<35 weeks; adjusted odds ratio = 0.59, 95% confidence interval: 0.12, 2.76) and late (35-36 weeks; adjusted odds ratio = 0.40, 95% confidence interval: 0.12, 1.40) preterm birth; findings were based on only two and three exposed cases, respectively. These results suggest that, compared with nonusers, women who take multivitamin supplements prior to conception may have a reduced risk of preterm birth, but further studies are needed with a larger sample of preconceptional users.

Leisure-time physical activity among pregnant women in the US

Few population-based data exist that describe leisure-time physical activity among pregnant women. The purpose of this study was to characterise the prevalence of leisure activity and to examine characteristics associated with participation in leisure activity during pregnancy. Using data collected from the year 2000 Behavioral Risk Factor Surveillance System, information on leisure activity was collected during telephone interviews from 1979 pregnant women and 44,657 non-pregnant women 18-44 years of age, representative of the US population. The prevalence of any leisure activity in the past month was 65.6%, 95% confidence interval [95% CI] 62.0, 69.1, among pregnant women and 73.1% [95% CI 72.4, 73.9] among non-pregnant women. The prevalence of recommended activity was also lower among pregnant women than non-pregnant women (15.8% vs. 26.1%). The most common leisure activity for pregnant women was walking, followed by activities such as swimming laps, weight lifting, gardening, and aerobics. Among pregnant women, any leisure activity was significantly higher for those with higher education, younger age, and excellent or very good health than for those with fair or poor health. Pregnant women reported less leisure activity than non-pregnant women of the same age. Studies to understand the enablers and barriers to participating in leisure activity during pregnancy are needed.

Psychosocial factors and preterm birth among African American and White women in central North Carolina

OBJECTIVES

We assessed associations between psychosocial factors and preterm birth, stratified by race in a prospective cohort study.

METHODS

We surveyed 1898 women who used university and public health prenatal clinics regarding various psychosocial factors.

RESULTS

African Americans were at higher risk of preterm birth if they used distancing from problems as a coping mechanism or reported racial discrimination. Whites were at higher risk if they had high counts of negative life events or were not living with a partner. The association of pregnancy-related anxiety with preterm birth weakened when medical comorbidities were taken into account. No association with preterm birth was found for depression, general social support, or church attendance.

CONCLUSIONS

Some associations between psychosocial variables and preterm birth differed by race.

Vaginal bleeding during pregnancy and preterm birth

This study investigated the relation between self-reported vaginal bleeding during pregnancy and preterm birth in a prospective cohort of 2,829 pregnant women enrolled from prenatal clinics between 1995 and 2000 in central North Carolina. The overall association between vaginal bleeding and preterm birth was modest (risk ratio (RR) = 1.3, 95% confidence interval (CI): 1.1, 1.6). Bleeding in the first trimester only was associated with earlier preterm birth (< or =34 weeks' gestation) (RR = 1.6, 95% CI: 1.1, 2.4) and preterm birth due to preterm premature rupture of the membranes (PPROM) (RR = 1.9, 95% CI: 1.1, 3.3). Bleeding in both trimesters was associated with preterm birth due to preterm labor (RR = 3.6, 95% CI: 1.9, 6.8). Bleeding of multiple episodes, on multiple days, and with more total blood loss was associated with an approximate twofold increased risk of earlier preterm birth, PPROM, and preterm labor. In contrast, bleeding in the second trimester only, of a single episode, on a single day, and with less total blood loss was not associated with any category of preterm birth. Vaginal bleeding was not associated with preterm birth among African Amercians (RR = 1.2, 95% CI: 0.9, 1.7). This study indicates that more intense but not less intense bleeding is associated with earlier preterm birth and spontaneous preterm birth presenting as PPROM or preterm labor, and it suggests that bleeding is less predictive of preterm birth among African-American compared with White women.

Risk factors for preterm birth in Vietnam

OBJECTIVE

To identify the risk of preterm birth and possible determinants among women in Hanoi, Vietnam.

METHOD

Prospective cohort study of 1709 women with singleton live births at Hanoi Obstetrics and Gynecology Hospital, Hanoi, Vietnam, June-October 2002. Logistic regression analysis was used to examine predictors of preterm birth (<37 weeks' gestation).

RESULT

The risk of preterm birth was 11.8%. Physically demanding work during pregnancy, two or more prior spontaneous abortions, history of preterm birth, vaginal bleeding, inadequate prenatal care during the first 20 weeks of gestation, and history of intrauterine device use with removal less than 12 months before the current pregnancy were associated with increased risk of preterm birth (adjusted odds ratios between 1.8 and 2.6).

CONCLUSION

Preterm birth is relatively frequent in this population. Beyond established risk factors, these data implicated agricultural work and physical work demands with preterm birth, as well as history of recent IUD use.

Maternal medication use and neuroblastoma in offspring

The association between a mother's use of specific medications during pregnancy and lactation and neuroblastoma in her offspring was evaluated in a case-control study. Newly diagnosed cases of neuroblastoma (n=504) in the United States and Canada were identified between 1992 and 1994 at 139 hospitals affiliated with the Pediatric Oncology Group or the Children's Cancer Group clinical trial programs. One age-matched control was sampled from the community of each case by means of random digit dialing. Exposure information was ascertained retrospectively from mothers in a structured telephone interview. Odds ratios were estimated using conditional logistic regression, with adjustment for maternal sociodemographic factors. The results did not support an association between neuroblastoma and maternal exposure to diuretic agents, antiinfective agents, estrogens, progestins, sedatives, anticonvulsant drugs, or drugs that may form N-nitroso derivatives. Mothers of cases were more likely to report using medications containing opioid agonists while they were pregnant or nursing than were mothers of controls (odds ratio=2.4, 95% confidence interval: 1.3, 4.3). Specifically, more mothers of cases reported using medications containing codeine while pregnant or nursing than did mothers of controls (odds ratio=3.4, 95% confidence interval: 1.4, 8.4). This preliminary finding may be due to bias, confounding, or chance, and additional studies are needed for confirmation.

Recruitment for a community-based study of early pregnancy: the Right From The Start study

Despite the high incidence of spontaneous abortion, little is known about its causes, in part because of the challenge of assembling a large cohort of women in early pregnancy for prospective study. We describe the effectiveness of recruitment strategies used in Right From The Start (RFTS), a prospective, community-based study of spontaneous abortion. Between December 2000 and September 2002, 803 pregnant women enrolled in RFTS, 103 of whom were recruited while trying to conceive. The mean gestational age at enrollment was 52 days, with 25% of the cohort enrolling before 6 completed weeks' gestation. Participants recruited directly from the community typically enrolled earlier in their pregnancies (mean of 44 days) and accounted for 24% of the total cohort and 83% of all participants who were recruited while trying to conceive. Posting brochures in drug stores and targeted mailings to new homeowners were the most effective community recruitment strategies. Recruitment at private and public prenatal care sites accounted for 57% and 19% of the participants respectively. Recruitment from public clinics required direct contact by RFTS staff and yielded women who enrolled at later gestational ages (mean of 58 days), but was valuable for inclusion of minorities and lower income women with less favourable health behaviours. Although intensive, diverse efforts were required, when recruitment efforts were maximised, we successfully recruited over 10% of the estimated number of pregnant women in the community.