D Lincoln

Impact of ambient air pollution on birth weight in Sydney, Australia

Background: Studies in Asia, Europe, and the Americas have provided evidence that ambient air pollution may have an adverse effect on birth weight, although results are not consistent. Methods: Average exposure during pregnancy to five common air pollutants was estimated for births in metropolitan Sydney between 1998 and 2000. The effects of pollutant exposure in the first, second, and third trimesters of pregnancy on risk of “small for gestational age” (SGA), and of pollutant exposure during pregnancy on birth weight were examined. Results: There were 138 056 singleton births in Sydney between 1998 and 2000; 9.7% of babies (13 402) were classified as SGA. Air pollution levels in Sydney were found to be quite low. In linear regression models carbon monoxide and nitrogen dioxide concentrations in the second and third trimesters had a statistically significant adverse effect on birth weight. For a 1 part per million increase in mean carbon monoxide levels a reduction of 7 (95% CI −5 to 19) to 29 (95% CI 7 to 51) grams in birth weight was estimated. For a 1 part per billion increase in mean nitrogen dioxide levels a reduction of 1 (95% CI 0 to 2) to 34 (95% CI 24 to 43) grams in birth weight was estimated. Particulate matter (diameter less than ten microns) in the second trimester had a small statistically significant adverse effect on birth weight. For a 1 microgram per cubic metre increase in mean particulate matter levels a reduction of 4 grams (95% CI 3 to 6) in birth weight was estimated. Conclusion: These findings of an association between carbon monoxide, nitrogen dioxide, and particulate matter, and reduction in birth weight should be corroborated by further study.

Socioeconomic and maternal determinants of small-for-gestational age births: patterns of increasing disparity

Objective. To better characterize the relation between socioeconomic disadvantage and small‐for‐gestational age births (SGA). Design. Analysis of data from a mandatory population‐based surveillance system. Setting. Public or private hospitals and at home. Population. All 877,951 singleton births occurring in New South Wales, Australia, between 1994 and 2004. Methods. Multilevel models were developed to determine the factors associated with babies weighing less than the 3rd percentile for gestation and gender. Main outcome measures. Odds of SGA. Results. The risk of SGA increased with increasing socioeconomic disadvantage. Smoking accounted for approximately 40% of the increased risk associated with socioeconomic disadvantage, and delayed antenatal care approximately 5%. While the absolute rate of SGA remained stable over the study period, the odds of SGA in mothers living in the most disadvantaged areas compared to those in the least disadvantaged areas increased from approximately 1.7 to 2.2. This trend persisted after accounting for maternal smoking. The risk of SGA over this period also increased in mothers commencing antenatal care after the first trimester. After accounting for smoking, socioeconomic disadvantage and clinical conditions, mothers under 21 years of age were at reduced risk of SGA, but mothers over 35 were at increased risk. Conclusions. Socioeconomic disadvantage remains one of the dominant determinants of SGA, even in a developed country with universal insurance. This relation appears to be strengthening. Smoking patterns, inadequate antenatal care and clinical conditions partially account for this association and trend, however, most is mediated by other factors.

Spatio-temporal variation in trihalomethanes in New South Wales

AIM This paper describes the spatio-temporal variation of trihalomethanes in drinking water in New South Wales, Australia from 1997 to 2007 METHOD We obtained data on trihalomethanes (THMs) from two metropolitan and 13 rural water utilities and conducted a descriptive analysis of the spatial and temporal trends in THMs and the influence of season and drought. RESULTS Concetrations of monthly THMs in the two metropolitan water utilities of Sydney/Illawarra (mean 66.8 μg/L) and Hunter (mean 62.7 μg/L) were similar compared to the considerable variation between rural water utilities (range in mean THMs: 14.5-330.7 μg/L). Chloroform was the predominate THM in two-thirds of the rural water utilities. Higher concentrations of THMs were found in chlorinated water distribution systems compared to chloraminated systems, and in distribution systems sourced from surface water compared to ground water or mixed surface and ground water. Ground water sourced supplies had a greater proportion of brominated THMs than surface water sourced supplies. There was substantial variation in concentration of THMs between seasons and between periods of drought or no drought. There was a moderate correlation between heavy rainfall and elevated concentrations of THMs. CONCLUSION There is considerable spatial and temporal variation in THMs amongst New South Wales water utilities and these variations are likely related to water source, treatment processes, catchments, drought and seasonal factors.

Small area estimation of sparse disease counts using shared component models-application to birth defect registry data in New South Wales, Australia

In the field of disease mapping, little has been done to address the issue of analysing sparse health datasets. We hypothesised that by modelling two outcomes simultaneously, one would be able to better estimate the outcome with a sparse count. We tested this hypothesis utilising Bayesian models, studying both birth defects and caesarean sections using data from two large, linked birth registries in New South Wales from 1990 to 2004. We compared four spatial models across seven birth defects: spina bifida, ventricular septal defect, OS atrial septal defect, patent ductus arteriosus, cleft lip and or palate, trisomy 21 and hypospadias. For three of the birth defects, the shared component model with a zero-inflated Poisson (ZIP) extension performed better than other simpler models, having a lower deviance information criteria (DIC). With spina bifida, the ratio of relative risk associated with the shared component was 2.82 (95% CI: 1.46-5.67). We found that shared component models are potentially beneficial, but only if there is a reasonably strong spatial correlation in effect for the study and referent outcomes.

Influences on the degree of preterm birth in New South Wales

Objective: To identify risk factors for preterm birth and determine if these vary by degree of prematurity.

THE EFFECTS OF LOW LEVEL AIR POLLUTION ON DAILY MORTALITY AND HOSPITAL ADMISSIONS IN SYDNEY, AUSTRALIA, 1994 TO 2000: ISEE-571

Introduction: PM levels in Sydney are low compared with most cities where time series studies of the acute effects of air pollution have been conducted. The 50th and 90th percentiles of daily average PM10 (particulate matter [pounds] 10mm) in Sydney are 16 and 26 mg/m3 respectively, and 8 and 14 mg/m3 for PM2.5 (particulate matter [pounds] 2.5mm). Time series studies in Sydney in the early 1990s demonstrated associations between particles (measured by light scatter using integrating nephelometers), nitrogen dioxide and ozone with daily mortality and hospital admissions. Light scatter (BSP) is considered to be a proxy measure for PM2.5. From 1994 air pollution data in Sydney is available from a larger number of monitoring sites and includes daily data on PM10 (measured by TEOM). Daily PM2.5 data (measured by TEOM) is available from 1997. Methods: We investigate associations between the three ambient particulate measures (BSP, PM2.5 and PM10) and the gaseous pollutants (nitrogen dioxide and ozone) with daily mortality and hospital admissions for all ages and the elderly (65+years) in the Sydney metropolitan area from 1994 to 2000 using time series analysis controlling for a range of confounders. We conducted sensitivity analyses to investigate the effect of different modeling approaches including the use of various methods for smoothing long term and seasonal trends, and weather parameters (penalized splines compared with loess and natural splines). Results: All three particulate measures (PM2.5, PM10 and BSP) were associated with all cause mortality, all cardiovascular mortality and all respiratory mortality. The magnitude of the PM2.5 effects were generally larger than those of BSP and PM10. Nitrogen dioxide was also associated with all cause mortality, all cardiovascular mortality and all respiratory mortality. Ozone was weakly associated with respiratory mortality. All three particulate measures were associated with all cardiac, IHD (ischaemic heart disease) and respiratory admissions, and weakly associated with COPD (chronic obstructive pulmonary disease). The magnitude of the effect for PM2.5 and BSP were generally larger than those for PM10. Nitrogen dioxide was also associated with all cardiac, IHD, all respiratory and COPD hospital admissions. Conclusions: The relatively low levels of particulate air pollution in Sydney were consistently associated with both daily mortality and hospital admissions. These particulate associations were generally strongest for fine particles (ie: PM2.5 and BSP) compared with PM10, and persist even at the relatively low particulate levels seen in Sydney, indicating no threshold concentrations are present. These results are consistent with the international literature. Nitrogen dioxide is also consistently associated with both daily mortality and hospital admissions. Further investigation is required to determine the effects of these specific pollutants, compared to the air pollution mix in Sydney.

TIME SCALE EFFECTS OF PARTICULATE AIR POLLUTION AND MORTALITY IN SYDNEY, AUSTRALIA, 1994 TO 2000.: ISEE-569

IntroductionThere is uncertainty about the extent of life-shortening due to the effects of air pollution on mortality demonstrated in time series studies. We implemented a methodology recently developed by Dominic F et al 2003 (Am. J. Epidemiol, in press) to investigate the extent of life shortening