Next Article in Journal
Uninterrupted Classroom Sitting is Associated with Increased Discomfort and Sleepiness Among College Students
Previous Article in Journal
Effects of FDI on the Efficiency of Government Expenditure on Environmental Protection Under Fiscal Decentralization: A Spatial Econometric Analysis for China
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
IJERPH
Volume 16
Issue 14
10.3390/ijerph16142497
ijerph-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

The Modifying Role of Socioeconomic Position and Greenness on the Short-Term Effect of Heat and Air Pollution on Preterm Births in Rome, 2001–2013

by
Federica Asta
*,
Paola Michelozzi
,
Giulia Cesaroni
,
Manuela De Sario
,
Chiara Badaloni
,
Marina Davoli
and
Patrizia Schifano
Department of Epidemiology, Lazio Regional Health Service, ASL Roma 1, Via Cristoforo Colombo, 112-00147 Rome, Italy
*
Author to whom correspondence should be addressed.
Int. J. Environ. Res. Public Health 2019, 16(14), 2497; https://doi.org/10.3390/ijerph16142497
Submission received: 27 May 2019 / Revised: 11 July 2019 / Accepted: 12 July 2019 / Published: 12 July 2019
Browse Figures
Review Reports Versions Notes

Abstract

:
Urban green spaces have been associated with health benefits, but few studies have evaluated the role of greenness on pregnancy outcomes. We examined how the association between short-term exposure to heat and air pollution on the probability of preterm delivery is affected by the spatial variation of socioeconomic position (SEP) and greenness. We analyzed a cohort of newborns in Rome, from April to October of 2001–2013, defining preterm as births between the 22nd and the 36th week of gestation. We used a time series approach, with maximum apparent temperature (MAT), PM10, NO2, and O3 as exposure variables. As greenness indicators, we considered maternal residential proximity to green spaces and the Normalized Difference Vegetation Index (NDVI) within a 100 m buffer around each woman’s residential address. We enrolled 56,576 newborns (5.1% preterm). The effect of a 1 °C increase in temperature on the daily number of preterm births was higher in women with low SEP (+2.49% (90% CI: 1.29–3.71)) and among those living within 100 m from green spaces (+3.33% (90% CI: 1.82–4.87)). No effect modification was observed for NDVI or PM10. SEP was an important effect modifier of the heat-preterm birth relationship. The role of greenness in modifying this association between heat and preterm delivery should be further investigated.

1. Introduction

Preterm birth children, representing the 5–9% of the total births in Europe, are more likely to experience adverse health outcomes throughout childhood and into adulthood. The rates of preterm birth are increasing in many countries, reaching 11% [1,2,3,4]. The etiology of preterm birth is still not well known, but it is thought that it could be due to multi-factorial characteristics such as age, maternal conditions like preeclampsia or chronic infections, genetic predisposition, and socio-economic and lifestyle factors [5,6,7].
Climate change has been widely regarded as the greatest global threat for human health in the 21st century. As extreme meteorological events (e.g., heat waves, floods, droughts) are predicted to increase in frequency and magnitude in the future, it is important to identify the potential health effects of these meteorological conditions among vulnerable subpopulations such as pregnant women [8,9,10]. In recent years, the effect of high or low temperatures on pregnancy outcomes have represented an important research issue, and studies in different countries have shown a positive association between ambient temperature and several pregnancy outcomes, in particular between heat and preterm births [11,12]. Furthermore, evidence in the literature suggests that preterm births are associated with other urban environmental exposures such as air pollution [8,13,14,15,16,17]. Although biological plausibility is recognized, mechanisms linking air pollution to preterm births are still being investigated [14,18,19] and probably differ for each pollutant [20]. Several studies have reported different effects; the most consistent evidence on short-term air pollution effects is related to particulate matter with an aerodynamic diameter of 10μm or less (PM10), with effect estimates comparable to those of heat but more persistent and delayed in time [21,22].
In a previous study, we analyzed the effect of short-term exposure to temperature and air pollutants during warm and cold seasons on the probability of preterm births [22]. We found a significant association with heat and PM10 during the warm season. Furthermore, specific subgroups of women, with low education level and of young age, were more susceptible. However, other important risk factors were not analyzed. When evaluating the relationship between air pollution, ambient temperature, and preterm births, the potential role of different measures of socioeconomic position (SEP) should also be considered. It is well recognized that socioeconomic inequality plays a key role in pregnancy outcomes and that low socioeconomic groups can also be more exposed to environmental stressors and be more vulnerable [23,24,25,26]. Individual and area-based SEP could have independent associations with the susceptibility of individuals or could modify individual susceptibility to other risk factors [26,27]. In this context, SEP could also have a potential modifying effect on the association between ambient temperature, air pollution, and preterm births.
Green spaces can mitigate the urban heat island, absorbing solar radiation and cooling through evapotranspiration, and can create differential heat patterns within a city, especially at night [28]. However, the literature on this mitigation effect on pregnancy outcomes is still limited. It is reasonable to suppose that the distribution of green spaces within a city is differential by area-based SEP, hence it is important to study these two variables together. Considering the biological plausibility and the lack of epidemiological evidence on this topic [4], this study aimed to assess the association of maternal short-term exposure to heat and air pollution with the probability of preterm birth by analyzing whether spatial variation of area-based SEP and greenness could modify these relationships in the city of Rome.

2. Materials and Methods

2.1. Study Population

Rome covers an area of about 1290 km2, and most of the inhabitants are located within the Main Ring Road (Figure 1A).
We previously collected data on all singleton live births in the city of Rome from 2001 and 2010, identified through the Certificate of Delivery Care Registry, which includes a wide range of information on maternal and fetal characteristics and clinical data on pregnancy and delivery. We followed the same inclusion criteria of our previous study [22], selecting pregnant women with singleton pregnancy that had a natural delivery or caesarean section with a spontaneous onset of labor that had reached at least 22 weeks of gestation.
We extended the study period up to 2013 by selecting only births occurred in the warm season (1 April to 31 October). Then, we geocoded each woman’s residence at the time of delivery (using ArcGis 9.1 (Esri, Redlands, CA, USA)) focusing only on residents within the Main Ring Road.

2.2. Outcome Definition

We defined preterm birth as a live birth between the 22nd and the 36th week of gestation. We computed gestational age as the difference between the date of delivery and the first day of the last menstrual period.

2.3. Meteorological Variables and Air Pollution

We used daily maximum apparent temperature (MAT), which is an index of discomfort that includes both air and dew-point temperatures [29,30]. Apparent temperature was calculated for all three-hourly data measurements, and then the maximum daily value was calculated. We used temperatures measured at the airport station (Rome Ciampino) closest to the city center, and for every day, we assigned the same value of MAT to all mothers.
We analyzed particulate matter with an aerodynamic diameter of 10 um (PM10), nitrogen dioxide (NO2), and ozone (O3) measured by the Lazio Environmental Protection Agency (http://www.arpalazio.gov.it/) at three monitoring sites (a background and two urban stations). We assigned daily values of PM10 (24-h mean), NO2 (24-h mean), and O3 (maximum eight-hour running mean) to all mothers.

2.4. Greenness Indicators

We used two measures of green space in order to address different aspects, considering as buffer sizes those ones reported in literature on the beneficial effects of greenness [26,31,32]. We used the distance from a major green area [33], considering residential proximity as a surrogate for the access to green spaces. We computed the minimum straight line distance between home addresses and the boundary of the nearest major green space, identified from a shapefile (provided to us courtesy of Municipality of Rome) including all green urban areas, sports and leisure facilities and parks within the Main Ring Road (Figure 1B). Then we classified women as living within 100 m, between 100 m and 500 m, and beyond 500 m from green spaces [34].
We used the Normalized Difference Vegetation Index (NDVI) as a measure of surrounding greenness and a surrogate for general outdoor greenness [35]. NDVI ranges from −1 to 1 (with higher numbers indicating more greenness) is based on land surface reflectance of visible (red) and near infrared radiation [33,36,37,38,39]. We downloaded a Landsat 8-OLI/TIRS remote sensing image of NDVI from the USGS database (www.glovis.usgs.gov), with a spatial resolution of 30 m × 30 m (Figure 1C), for the 17/07/2015, assuming this image was representative of the overall follow-up period exposure [40]. We were confident that this assumption could not bias our greenness indicator since Orioli et al. demonstrated that the correlations between NDVI summer values across the period from 2001 to 2013 plus 2015 were always very high (about 0.80), denoting no substantially changes in the amount of greenness in the city of Rome [40]. We selected a summer day, chosen as it had the least amount of cloud cover at the time of year when leaf development would be at a maximum. Finally, we calculated the average of NDVI in each 100 m buffer around of the mothers’ residential addresses and classified each woman using NDVI’s tertiles [34].

2.5. Socioeconomic Position

We considered a small area-level indicator of socioeconomic position (SEP): the index is based on data from the 2001 census in Rome, derived from a factor analysis that included education, occupation, home ownership, family composition, crowding, and immigration at census block level (500 inhabitants per block, on average) [41]. We categorized the original quintiles of this index in three levels: low (fourth and fifth quintiles), medium (third quintile) and high (first and second quintiles) (Figure 1D).

2.6. Statistical Analysis

We used Poisson generalized additive models to check the linearity of the relationships and to estimate the associations between environmental exposures and the daily proportion of preterm births. We included the daily number of pregnancies at risk as offset in the models. We adjusted the models for long-term trend and seasonality (considering natural splines with degrees of freedom equal to number of years minus one for the long-term and with degrees of freedom equal to number of months minus one for seasonality), and holidays (a categorical variable which assumes value “1” in national or local holiday days and “0” otherwise).
We first studied MAT and each pollutant separately, and we considered the adjusted models only when collinearity did not arise. We checked for collinearity among the exposures using the variance inflation factor (VIF) with values of VIF > 10 indicating presence of collinearity.
In our previous study, we studied the lag structure of the relationships with a distributed lag model (DLM), allowing a lag structure of up to 30 days [22]. In accordance to the previous study, we considered a lag of 0–2 days for MAT, NO2, and O3, while a lag of 12–22 days before delivery was considered for PM10. Each exposure was included in the model as a linear term at the chosen lag.
To identify whether greenness and small-area SEP modified the effect of MAT and air pollutants on the probability of preterm birth, we included an interaction term in the model between each potential effect modifier and the environmental indicators, considering a p-value less than 0.10 as threshold. We also stratified by each potential effect modifier considered in this study (green indicators and SEP) and maternal age and education.
We used SAS 9.2 (SAS Institute Inc., Cary, NC, USA) and ArcGis 9.1 (ESRI, Redlands, California, USA) to create datasets while we carried out the analyses with R software (R Foundation for Statistical Computing, Vienna, Austria) (www.r-project.org).

3. Results

We considered 56,576 singleton live births during the study period, from mothers delivering in Rome and residing within the Main Ring Road; 2910 (5.1%) were preterm. The maternal residential address was available for all women included in the study. In our cohort, 28,946 (51%) mothers lived within 100 to 500 m from a green space with 5.2% preterm births, while only 16% (9013) lived within 100 m of a green area and had a 5.4% of preterm births. No relevant differences in preterm birth distributions were found by NDVI (Table 1).
As shown in Table 1, while the distribution of maternal age and education level was similar across the different green space categories, a high SEP was more frequent among women living within 100 m and within 100–500 m from green spaces (43.5% and 46.2%, respectively) compared to those living beyond 500 m (34.8%). We found a similar trend when considering the tertiles of NDVI: a high SEP was more frequent among mothers living in areas with NDVI > 0.36 (48.0%), compared to the two lowest categories (38.4% and 40.0%, respectively).
The percentile distribution of daily MAT and air pollutants between 1st April and 31st October 2001–2013 is reported in Table 2. The MAT interquartile range (IQR) was 10.3 °C and the O3 showed the highest variability among the pollutants (Table 2). We observed moderate correlations between MAT and O3 (ρ = 0.6) and between PM10 and NO2 (ρ = 0.5) (Table S1).
The overall model showed that MAT was associated with an increased probability of preterm birth; as shown in Table 3, we observed a percent change of 2.00 (90% CI: 0.98–3.03) in the proportion of preterm births per 1 °C increase in MAT with a lag of 0–2 days. Regarding air pollutants, only PM10 was associated with an increased probability of preterm birth; for an IQR increase (11 μg/m3) in PM10 with a lag of 12–22 days before the delivery, the percent change in the proportion of preterm births was equal to 8.16 (90% CI: 2.38–14.26). For NO2 and O3, no significant effect on preterm births was observed (Table 3). No collinearity was observed among exposures.
Socioeconomic position and residential proximity to green spaces modified the effect of temperature in our analysis. We found a percent change of 3.33 (90% CI: 1.82–4.87) in the proportion of preterm births per 1 °C increase in MAT in women living within 100 m from green areas, while those living further away from green spaces (>500 m) showed a significantly lower effect of temperature on preterm births of 1.03% (90% CI: −0.19–2.28) compared to women living within 100 m (reference group) (Figure 2).
For SEP, women with low SEP were more susceptible to the effect of temperature on the probability of preterm birth (2.49% increase per 1 °C increase in MAT (90% CI: 1.29–3.71) than women with high SEP (1.01% increase per 1 °C increase in MAT (90% CI: −0.15–2.18)) (Figure 2). No effect modification by residential surrounding greenness was found.
Furthermore, none of the considered factors modified the effect of PM10 on the preterm birth probability (Figure 3). However, results suggested an increasing trend in PM10 effects with an increasing distance from green areas.
When we included a double interaction between residential proximity to green spaces and SEP in the model, the highest effect of temperature on the preterm birth probability was among women with low SEP and living within 100 m from a major green space. In this group, we detected a percent change of 5.28 (90% CI: 2.97–7.65; p-value of interaction term = 0.048) per 1 °C increase in MAT compared to women with high SEP living within 100 m (+2.2%; 90% CI: 0.2–4.2) (Table S2).
Considering the double interaction between residential proximity to green spaces and maternal age, we found a 4.4% increase (90% CI: 2.47–6.42) in the proportion of preterm births among women aged 30–36 years and living within 100 m from green areas for a 1 °C increase in MAT vs. a 1.3% increase (90% CI: −1.23–3.90) in women aged 37 and over living within 100 m (p-value of interaction term = 0.082; Table S3). No effect modification was found when we studied the double interaction between residential proximity to green spaces and education level.

4. Discussion

This study is one of the first to evaluate the association of maternal exposure to high ambient temperature and urban air pollutants with the probability of preterm birth by analyzing the spatial variation in SEP and green spaces as effect modifier. Although we considered a longer time series adding three years to previous analysis [22], our results confirmed that, in Rome maternal exposure to heat and PM10 was associated with a short-term increase in the probability of preterm birth during the warm season. A 1 °C increase in MAT was associated with a percent change of two in the daily proportion of preterm birth, consistent with our and other previous studies [10,11,22]. As for air pollutants, we found a significant effect of exposure to PM10 in the second and third week before delivery with an 8% increase in the daily proportion of preterm birth for an IQR increase of PM10 (11 μg/m3).
The effects of heat and air pollution on preterm births have been discussed in several studies and reviews [4,8,11,16,17,20,22]. In particular, the short-term effect of high temperature on pregnant women could be explained by several biological mechanisms: the secretion of oxytocin and prostaglandin (PGF2α) has been shown to increase under heat stress conditions and to induce labor [42,43]. In addition, the onset of labor could be induced by heat-related dehydration, reducing maternal fluid level, and consequently the fetus blood volume [44]. The underling biological mechanisms of the short-term effect of air pollution on preterm births were less clear [14], and seemed to involve both endothelial dysfunction and oxidative stress leading to placental inflammation [19,45].
Less is known on the individual characteristics that may worsen or mitigate these effects. Even if our study adopted a time series design, we were able to analyze socioeconomic position and closeness to major green areas at residential address of each mother as potential effect modifiers, considering residency from population registry data. We found an interaction between residential proximity to a major green space and temperature: women living within 100 m from a green area had a greater probability of preterm birth for every 1 °C increase in temperature compared to women living beyond 500 m. No effect modification was found when considering NDVI within 100 m from each residential address. Although there is a growing literature on the beneficial health effects of urban green spaces, existing evidence on pregnancy outcomes and surrounding greenness is still inconsistent [34,35,46,47]. In Rome, contrary to expectations, closeness to green spaces was associated with a higher probability of preterm birth. Our study did not seem to support the hypothesis of potential cooling effect due to greenness, but this could be due to the use of the same temperature level for all women. Furthermore, it should be taken into account that mitigation of temperatures and of the urban heat island occurs mostly at night, when women are indoor, with closed windows and air conditioning. All of these aspects could have impacted our results. The lack of literature on this topic limits comparison with other studies considering both meteorological parameters and the distance at which the mitigation is still effective [48,49]. A recent study focusing on heat-related mortality, conducted in Vietnam, highlighted the difficulty to choose the appropriate meteorological parameters when dealing with the impact of greenness on human health, given the complex interaction between vegetation and air temperature at urban levels [48]. Another aspect to consider when interpreting our findings could be the potential heterogeneous utilization patterns of the nearby green space between women living very close to a green area (less than 100 m) and those living farther. It could be that the former group, given the closeness to green spaces, might spend more time outdoors, thus increasing the level of heat exposure, compared to the latter. It is also possible that closeness to green spaces might be a proxy of other life styles variables we were not able to consider.
Our study suggested a potential mitigating effect of residential proximity to green spaces on PM10-preterm birth association. This is coherent with literature showing that vegetation could absorb particles decreasing the levels of air pollution, and with studies showing a lower traffic density nearby green areas [28]. We could not compare this finding with other studies because, to our knowledge, this was the first study to evaluate the role of greenness as modifier of temperature/pollution and preterm births relationship.
In our study the association between MAT and preterm birth was also modified by SEP with a higher effect in women with low SEP, equal to about 3% in the daily proportion of preterm birth for every 1 °C increase in MAT. A similar increased probability was found in women with primary school vs degree in our previous study in which we considered maternal education [22]. Although the results were quite similar, we believed that SEP as a composite index with its multidimensionality, could give us more comprehensive view on this aspect, compared to education level that represents only one dimension [26,27]. One explanation could be due to the potential worse health status of low SEP people due to limited access to health care and living in areas with poor housing and worse environmental conditions [26,50]. In particular, we identified a subgroup of women highly susceptible to the heat effects: those with low SEP and living very close to green areas (within 100 m) showed an increase of about 5% in the daily proportion of preterm births for every 1 °C increase in MAT. The literature suggests that subjects with low SEP spent more time in their local area, hence the time to local exposure was greater and they might use local green spaces close to their residence more [51]. Many other risk factors, not directly measured in this study, might be reasonably associated with a low SEP, such as lack of air conditioning, poor diet, and smoking during pregnancy, that should be considered [52]. Since our study was based on administrative databases, which represent a point of strength in terms of sample size, cost, and feasibility, the opportunity to consider other potential individual risk factors was limited.
Some limitations of this study have to be discussed. Since there was no universally accepted definition of urban green space, we decided to use two indicators of greenness—one derived from satellite data (NDVI) and one based on the distance between residential address and the nearest major green area provided by Municipality of Rome. For NDVI, we considered a single satellite image taken at a single day in July. Although we did not account for the potential temporal variation in greenness, a previous study conducted in Rome showed negligible changes in the quantity of greenness during the study period [40]. Neither of the chosen indicators allowed us to distinguish between different types of vegetation, vegetation volumes, and types of green space (street trees or gardens). The choice of buffer’s width around woman address of residence was another critical point: since we were interested to catch the cooling effect of greenness, we analyzed both green indicators within a small buffer size as suggested by literature [31,32,49]. Furthermore, we assumed that all people have equal access to the green space closer to the ward of residence, but this might not be true [53]. All of these features might play an important role in temperature mitigation, and they need to be accounted for in this kind of analysis [28,37].
Also, we used an indirect assessment of exposure: meteorological data were retrieved from a fixed monitoring weather station located in Rome (Ciampino Airport), and pollutants were derived from three fixed monitoring stations within the city. Therefore, an average measure of outdoor temperature and of air pollutants for all women residing in the municipality of Rome, within the Main Ring Road, was used. This could have induced non-differential misclassification of exposure because it was unrelated to the outcome, thus biasing the estimated effects toward the null value. Furthermore, it is important to take into account that personal exposure may be modified by the amount of time spent indoor, by the use of air conditioning, and by residential mobility, which we were, at this stage, unable to consider. Moreover, both effect modifiers (SEP and greenness) might have been affected by a potential misclassification due to the implicit assumption of residential stability during pregnancy. Our research relied on maternal residential addresses at the time of delivery, and we were not able to reconstruct residential mobility during pregnancy.
Finally, the different spatial resolutions between the main exposures (measured at the city scale) and the effect modifiers (measured at the individual scale) could have probably inflated the bias concerning the studied interactions. A major improvement in the analysis could be the use of the individual exposure levels through spatio-temporal models for both temperature and air pollutants to catch the true role of green areas and SEP as effect modifier.

5. Conclusions

The study confirms that maternal exposure to heat and PM10 is associated with a short-term increase in the preterm-birth probability. This association appears to be modified by socioeconomic position and closeness to green spaces. Contrary to expectations, living close to green areas seems to increase the preterm birth probability when exposed to heat, especially in women of low socioeconomic position.
It is interesting to note that, while the existing evidence confirms a general salutary impact of greenness on health, much remains to be clarified on the particular pathways involved and, above all how these may change by social context, subgroups, and pregnancy outcomes. For all these reasons, the role of greenness as potential modifier of hazardous exposures in urban areas should be further investigated.

Supplementary Materials

The following are available online at https://www.mdpi.com/1660-4601/16/14/2497/s1, Table S1: Pearson’s Correlation coefficients among the exposures at study. Table S2: Poisson regression estimated percent change in daily number of preterm births by 1 °C increase of MAT (Lag 0–2) for residential proximity to green spaces and SEP. Rome within the Main Ring Road, 2001–2013. Table S3: Poisson regression estimated percent change in daily number of preterm births by 1 °C increase of MAT (Lag 0–2) for residential proximity to green spaces and maternal age. Rome within the Main Ring Road, 2001–2013.

Author Contributions

Conceptualization, P.M. and P.S.; Data curation, C.B.; Formal analysis, F.A.; Methodology, F.A.; Project administration, P.S.; Supervision, P.M. and M.D.; Writing—original draft, F.A. and P.S.; Writing—review & editing, F.A., P.M., G.C., M.D.S., C.B., M.D. and P.S.

Funding

This research received no external funding.

Acknowledgments

We wish to thank Annibale Biggeri and Franco Merletti of Epidemiology at University of Turin for their contribute in the discussion of the methodology and results of this study and their critical revision the manuscript.

Conflicts of Interest

Authors have no conflicts of interest to disclose.

List of Abbreviations

MAT, maximum apparent temperature; PM10, particulate matter with an aerodynamic diameter of 10 µm or less; NO2, nitrogen dioxide; O3, ozone; NDVI, Normalized Difference Vegetation Index; DLM, distributed lag model; VIF, variance inflation factor; SEP, socioeconomic position; CI, confidence interval; IQR, interquartile range; pctl, percentile.

References

  1. Mathews, T.J.; MacDorman, M.F. Infant mortality statistics from the 2008 period linked birth/infant death data set. Natl. Vital Stat. Rep. 2012, 60, 1–27. [Google Scholar] [PubMed]
  2. Escobar, G.J.; McCormick, M.C.; Zupancic, J.A.F.; Coleman-Phox, K.; Armstrong, M.A.; Greene, J.D.; Eichenwald, E.C.; Richardson, D.K. Unstudied infants: Outcomes of moderately premature infants in the neonatal intensive care unit. Arch. Dis. Child. 2006, 91, F238–F244. [Google Scholar] [CrossRef] [PubMed]
  3. Basu, R.; Malig, B.; Ostro, B. High ambient temperature and the risk of preterm delivery. Am. J. Epidemiol. 2010, 172, 1108–1117. [Google Scholar] [CrossRef] [PubMed]
  4. Kloog, I. Air pollution, ambient temperature, green space and preterm birth. Curr. Opin. Pediatr. 2019, 31, 237–243. [Google Scholar] [CrossRef] [PubMed]
  5. Strauss, J.F.; Romero, R.; Gomez-Lopez, N.; Haymond-Thornburg, H.; Modi, B.P.; Teves, M.E.; Pearson, L.N.; York, T.P.; Schenkein, H.A. Spontaneous Preterm Birth: Advances Toward the Discovery of Genetic Predisposition; Elsevier Inc.: Amsterdam, The Netherlands, 2018; Volume 218, ISBN 2752013000. [Google Scholar]
  6. Goldenberg, R.L.; Culhane, J.F.; Iams, J.D.; Romero, R. Epidemiology and causes of preterm birth. Lancet 2008, 371, 75–84. [Google Scholar] [CrossRef]
  7. Lawn, J.E.; Gravett, M.G.; Nunes, T.M.; Rubens, C.E.; Stanton, C. Global report on preterm birth and stillbirth (1 of 7): Definitions, description of the burden and opportunities to improve data. BMC Pregnancy Childbirth 2010, 10. [Google Scholar] [CrossRef]
  8. Beltran, A.J.; Wu, J.; Laurent, O. Associations of meteorology with adverse pregnancy outcomes: A systematic review of preeclampsia, preterm birth and birth weight. Int. J. Environ. Res. Public Health 2013, 11, 91–172. [Google Scholar] [CrossRef]
  9. Vicedo-Cabrera, A.M.; Iñíguez, C.; Barona, C.; Ballester, F. Exposure to elevated temperatures and risk of preterm birth in Valencia, Spain. Environ. Res. 2014, 134, 210–217. [Google Scholar] [CrossRef]
  10. Vicedo-Cabrera, A.M.; Olsson, D.; Forsberg, B. Exposure to seasonal temperatures during the last month of gestation and the risk of preterm birth in stockholm. Int. J. Environ. Res. Public Health 2015, 12, 3962–3978. [Google Scholar] [CrossRef]
  11. Zhang, Y.; Yu, C.; Wang, L. Temperature exposure during pregnancy and birth outcomes: An updated systematic review of epidemiological evidence. Environ. Pollut. 2017, 225, 700–712. [Google Scholar] [CrossRef]
  12. Carolan-Olah, M.; Frankowska, D. High environmental temperature and preterm birth: A review of the evidence. Midwifery 2014, 30, 50–59. [Google Scholar] [CrossRef] [PubMed]
  13. Brauer, M.; Lencar, C.; Tamburic, L.; Koehoorn, M.; Demers, P.; Karr, C. A cohort study of traffic-related air pollution impacts on birth outcomes. Environ. Health Perspect. 2008. [Google Scholar] [CrossRef] [PubMed]
  14. Backes, C.H.; Nelin, T.; Gorr, M.W.; Wold, L.E. Early life exposure to air pollution: How bad is it? Toxicol. Lett. 2013, 216, 47–53. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  15. Strand, L.B.; Barnett, A.G.; Tong, S. The influence of season and ambient temperature on birth outcomes: A review of the epidemiological literature. Environ. Res. 2011, 111, 451–462. [Google Scholar] [CrossRef] [PubMed]
  16. Sapkota, A.; Chelikowsky, A.P.; Nachman, K.E.; Cohen, A.J.; Ritz, B. Exposure to particulate matter and adverse birth outcomes: A comprehensive review and meta-analysis. Air Qual. Atmos. Heal. 2012, 5, 369–381. [Google Scholar] [CrossRef]
  17. Stieb, D.M.; Chen, L.; Eshoul, M.; Judek, S. Ambient air pollution, birth weight and preterm birth: A systematic review and meta-analysis. Environ. Res. 2012, 117, 100–111. [Google Scholar] [CrossRef]
  18. O’Neill, M.S.; Osornio-Vargas, A.; Buxton, M.A.; Sánchez, B.N.; Rojas-Bracho, L.; Castillo-Castrejon, M.; Mordhukovich, I.B.; Brown, D.G.; Vadillo-Ortega, F. Air pollution, inflammation and preterm birth in Mexico City: Study design and methods. Sci. Total Environ. 2013, 448, 79–83. [Google Scholar] [CrossRef]
  19. Vadillo-Ortega, F.; Osornio-Vargas, A.; Buxton, M.A.; Sánchez, B.N.; Rojas-Bracho, L.; Viveros-Alcaráz, M.; Castillo-Castrejón, M.; Beltrán-Montoya, J.; Brown, D.G.; O’Neill, M.S. Air pollution, inflammation and preterm birth: A potential mechanistic link. Med. Hypotheses 2014, 82, 219–224. [Google Scholar] [CrossRef]
  20. Shah, P.S.; Balkhair, T. Air pollution and birth outcomes: A systematic review and on behalf of Knowledge Synthesis Group on Determinants of Preterm/LBW births. Environ. Int. 2011, 37, 498–516. [Google Scholar] [CrossRef]
  21. Lee, S.J.; Hajat, S.; Steer, P.J.; Filippi, V. A time-series analysis of any short-term effects of meteorological and air pollution factors on preterm births in London, UK. Environ. Res. 2008, 106, 185–194. [Google Scholar] [CrossRef]
  22. Schifano, P.; Lallo, A.; Asta, F.; De Sario, M.; Davoli, M.; Michelozzi, P. Effect of ambient temperature and air pollutants on the risk of preterm birth, Rome 2001–2010. Environ. Int. 2013, 61, 77–87. [Google Scholar] [CrossRef] [PubMed]
  23. Pickett, K.E.; Pearl, M. Multilevel analyses of neighbourhood socioeconomic context and health outcomes: A critical review. J. Epidemiol. Commun. Health 2001, 55. [Google Scholar] [CrossRef] [PubMed]
  24. Samet, J.M.; White, R.H. Urban air pollution, health, and equity. J. Epidemiol. Commun. Health 2004, 58, 3–5. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  25. Laurent, O.; Bard, D.; Filleul, L.; Segala, C. Effect of socioeconomic status on the relationship between atmospheric pollution and mortality. J. Epidemiol. Commun. Health 2007, 61. [Google Scholar] [CrossRef] [PubMed]
  26. Dadvand, P.; Wright, J.; Martinez, D.; Basagaña, X.; McEachan, R.R.C.; Cirach, M.; Gidlow, C.J.; de Hoogh, K.; Gražulevičiene, R.; Nieuwenhuijsen, M.J. Inequality, green spaces, and pregnant women: Roles of ethnicity and individual and neighbourhood socioeconomic status. Environ. Int. 2014, 71, 101–108. [Google Scholar] [CrossRef] [PubMed]
  27. Ross, C.E.; Mirowsky, J. Neighborhood Socioeconomic Status and Health: Context or Composition? City Commun. 2008, 7, 163–179. [Google Scholar] [CrossRef]
  28. Markevych, I.; Schoierer, J.; Hartig, T.; Chudnovsky, A.; Hystad, P.; Dzhambov, A.M.; de Vries, S.; Triguero-Mas, M.; Brauer, M.; Nieuwenhuijsen, M.J.; et al. Exploring pathways linking greenspace to health: Theoretical and methodological guidance. Environ. Res. 2017, 158, 301–317. [Google Scholar] [CrossRef]
  29. Kalkstein, L.S.; Valimont, K.M. An Evaluation of Summer Discomfort in the United State Using a Relative Climatological Index. Bull. Am. Meteorol. Soc. 1986. [Google Scholar] [CrossRef]
  30. O’Neill, M.S.; Zanobetti, A.; Schwartz, J. Modifiers of the temperature and mortality association in seven US cities. Am. J. Epidemiol. 2003, 157, 1074–1082. [Google Scholar] [CrossRef]
  31. Doick, K.J.; Peace, A.; Hutchings, T.R. The role of one large greenspace in mitigating London’s nocturnal urban heat island. Sci. Total Environ. 2014, 493, 662–671. [Google Scholar] [CrossRef]
  32. Dzhambov, A.M.; Dimitrova, D.D.; Dimitrakova, E.D. Association between residential greenness and birth weight: Systematic review and meta-analysis. Urban For. Urban Green. 2014, 13, 621–629. [Google Scholar] [CrossRef]
  33. McEachan, R.R.C.; Prady, S.L.; Smith, G.; Fairley, L.; Cabieses, B.; Gidlow, C.; Wright, J.; Dadvand, P.; van Gent, D.; Nieuwenhuijsen, M.J. The association between green space and depressive symptoms in pregnant women: Moderating roles of socioeconomic status and physical activity. J. Epidemiol. Commun. Health 2016, 70, 253–259. [Google Scholar] [CrossRef] [PubMed]
  34. Abelt, K.; McLafferty, S. Green streets: Urban green and birth outcomes. Int. J. Environ. Res. Public Health 2017, 14, 771. [Google Scholar] [CrossRef] [PubMed]
  35. WHO. Urban Green Spaces and Health; WHO: Copenhagen, Denmark, 2016; Volume 80. [Google Scholar]
  36. Weier, J.; Herring, D. Measuring Vegetation (NDVI & EVI): Feature Articles Measuring Vegetation (NDVI & EVI): Feature Articles; NASA Earth Observatory: Washington, DC, USA, 2011.
  37. Payam, D.; Jordi, S.; Xavier, B.; Ferran, B.; Aitana, L.; Ana, F.-S.; Marisa, E.; Raquel, G.-E.; Michelle A., M.; Mark, J. Nieuwenhuijsen surrounding greenness and pregnancy outcomes in four spanish birth cohorts. Environ. Health Perspect. 2012, 120, 1481–1487. [Google Scholar]
  38. Dadvand, P.; Villanueva, C.M.; Font-Ribera, L.; Martinez, D.; Basagaña, X.; Belmonte, J.; Vrijheid, M.; Gražulevičienė, R.; Kogevinas, M.; Nieuwenhuijsen, M.J. Risks and benefits of green spaces for children: A cross-sectional study of associations with sedentary behavior, obesity, asthma, and allergy. Environ. Health Perspect. 2014. [Google Scholar] [CrossRef]
  39. Hystad, P.; Davies, H.W.; Frank, L.; Van Loon, J.; Gehring, U.; Tamburic, L.; Brauer, M. Residential greenness and birth outcomes: Evaluating the influence of spatially correlated built-environment factors. Environ. Health Perspect. 2014. [Google Scholar] [CrossRef] [PubMed]
  40. Orioli, R.; Antonucci, C.; Scortichini, M.; Cerza, F.; Marando, F.; Ancona, C.; Manes, F.; Davoli, M.; Michelozzi, P.; Forastiere, F.; et al. Exposure to residential greenness as a predictor of cause-specific mortality and stroke incidence in the Rome Longitudinal Study. Environ. Health Perspect. 2019, 127. [Google Scholar] [CrossRef]
  41. Cesaroni, G.; Badaloni, C.; Romano, V.; Donato, E.; Perucci, C.A.; Forastiere, F. Socioeconomic position and health status of people who live near busy roads: The Rome Longitudinal Study (RoLS). Environ. Health 2010, 9, 1–12. [Google Scholar] [CrossRef]
  42. Kelly, A.J.; Malik, S.; Smith, L.; Kavanagh, J.; Thomas, J. Vaginal prostaglandin (PGE2 and PGF2a) for induction of labour at term. Cochrane Database Syst. Rev. 2009. [Google Scholar] [CrossRef]
  43. Lee, S.E.; Park, I.S.; Romero, R.; Yoon, B.H. Amniotic fluid prostaglandin F2 increases even in sterile amniotic fluid and is an independent predictor of impending delivery in preterm premature rupture of membranes. J. Matern. Neonatal Med. 2009. [Google Scholar] [CrossRef]
  44. Stan, C.M.; Boulvain, M.; Pfister, R.; Hirsbrunner-Almagbaly, P. Hydration for treatment of preterm labour. Cochrane Database Syst. Rev. 2013. [Google Scholar] [CrossRef] [PubMed]
  45. Veras, M.M.; Damaceno-Rodrigues, N.R.; Caldini, E.G.; Ribeiro, A.A.C.M.; Mayhew, T.M.; Saldiva, P.H.N.; Dolhnikoff, M. Particulate urban air pollution affects the functional morphology of mouse placenta1. Biol. Reprod. 2008, 79, 578–584. [Google Scholar] [CrossRef] [PubMed]
  46. Kabisch, N.; van den Bosch, M.; Lafortezza, R. The health benefits of nature-based solutions to urbanization challenges for children and the elderly—A systematic review. Environ. Res. 2017, 159, 362–373. [Google Scholar] [CrossRef] [PubMed]
  47. Glazer, K.B.; Eliot, M.N.; Danilack, V.A.; Carlson, L.; Phipps, M.G.; Dadvand, P.; Savitz, D.A.; Wellenius, G.A. Residential green space and birth outcomes in a coastal setting. Environ. Res. 2018, 163, 97–107. [Google Scholar] [CrossRef] [PubMed]
  48. Hondula, D.M.; Davis, R.E.; Georgescu, M. Clarifying the connections between green space, urban climate, and heat-related mortality. Am. J. Public Health 2018, 108, S62–S63. [Google Scholar] [CrossRef] [PubMed]
  49. Takebayashi, H. Influence of urban green area on air temperature of surrounding built-up area. Climate 2017, 5, 60. [Google Scholar] [CrossRef]
  50. Zanobetti, A.; O’Neill, M.S.; Gronlund, C.J.; Schwartz, J.D. Susceptibility to mortality in weather extremes: Effect modification by personal and small-area characteristics. Epidemiology 2013, 24, 809–819. [Google Scholar] [CrossRef] [PubMed]
  51. Dadvand, P.; Basagaña, X.; Sartini, C.; Figueras, F.; Vrijheid, M.; de Nazelle, A.; Sunyer, J.; Nieuwenhuijsen, M.J. Climate extremes and the length of gestation. Environ. Health Perspect. 2011, 119, 1449–1453. [Google Scholar] [CrossRef]
  52. Lee, A.C.K.; Maheswaran, R. The health benefits of urban green spaces: A review of the evidence. J. Public Health 2011, 33, 212–222. [Google Scholar] [CrossRef]
  53. Mitchell, R.; Astell-Burt, T.; Richardson, E.A. A comparison of green space indicators for epidemiological research. J. Epidemiol. Commun. Health 2011, 65, 853–858. [Google Scholar] [CrossRef] [Green Version]
Ijerph 16 02497 g001 550
Figure 1. Distribution of greenness and socioeconomic position within the Main Ring Road in Rome. (A) Map of Rome with high traffic roads; (B) Major green space within the Main Ring Road; (C) Spatial distribution of the Normalized Difference Vegetation Index (NDVI) within the Main Ring Road according to tertiles; (D) Map of Rome within the Main Ring Road by socioeconomic position (SEP).
Figure 1. Distribution of greenness and socioeconomic position within the Main Ring Road in Rome. (A) Map of Rome with high traffic roads; (B) Major green space within the Main Ring Road; (C) Spatial distribution of the Normalized Difference Vegetation Index (NDVI) within the Main Ring Road according to tertiles; (D) Map of Rome within the Main Ring Road by socioeconomic position (SEP).
Ijerph 16 02497 g001
Ijerph 16 02497 g002 550
Figure 2. Percent change * in daily preterm births by 1 °C increase of MAT (Lag 0–2). * Effect estimates derived by independent Poisson regression models including one covariate per time. Ref.: reference category. p: p-Value of interaction terms of each modality compared with the reference category.
Figure 2. Percent change * in daily preterm births by 1 °C increase of MAT (Lag 0–2). * Effect estimates derived by independent Poisson regression models including one covariate per time. Ref.: reference category. p: p-Value of interaction terms of each modality compared with the reference category.
Ijerph 16 02497 g002
Ijerph 16 02497 g003 550
Figure 3. Percent change * in daily preterm births by 1 μg/m3 increase of PM10 (Lag 12–22). * Effect estimates derived by independent Poisson regression models including one covariate per time. Ref.: reference category. P: p-Value of interaction terms of each modality compared with the reference category.
Figure 3. Percent change * in daily preterm births by 1 μg/m3 increase of PM10 (Lag 12–22). * Effect estimates derived by independent Poisson regression models including one covariate per time. Ref.: reference category. P: p-Value of interaction terms of each modality compared with the reference category.
Ijerph 16 02497 g003
Table
Table 1. Characteristics of the birth cohort in the metropolitan area within the Main Ring Road (Rome).
Table 1. Characteristics of the birth cohort in the metropolitan area within the Main Ring Road (Rome).
Variable Entire Cohort
n (%)		Residential Proximity to Green SpacesResidential Surrounding Greenness
≤100 m
n (%)		100–500 m
n (%)		>500 m
n (%)		NDVI T1
(≤0.28)
n (%)		NDVI T2
(0.29–0.36)
n (%)		NDVI T3
(>0.36)
n (%)
Study population 56,5769013 (15.9)28,946 (51.2)18,617 (32.9)19,115 (33.8)19,239 (34.0)18,222 (32.2)
Preterm birth 2910 (5.1)487 (5.4)1491 (5.2)932 (5.0)989 (5.2)944 (4.9)977 (5.1)
Maternal Age (years)
 <3013,644 (24.1)2156 (23.9)6850 (23.7)4638 (24.9)4590 (24.0)4608 (24.0)4446 (24.4)
 30–3630,485 (53.9)4846 (53.8)15,707 (54.3)9932 (53.4)10,315 (54.0)10,443 (54.3)9727 (53.4)
 ≥3712,447 (22.0)2011 (22.3)6389 (22.1)4047 (21.7)4210 (22.0)4188 (21.8)4049 (22.2)
Education Level
 Primary school 18,660 (33.0)2899 (32.2)9426 (32.6)6335 (34.0)6429 (33.6)6343 (33.0)5888 (32.3)
 High school24,268 (42.9)3936 (43.7)12,329 (42.6)8003 (43.0)8022 (42.0)8415 (43.7)7831 (43.0)
 Degree13,648 (24.1)2178 (24.2)7191 (24.8)4279 (23.0)4664 (24.4)4481 (23.3)4503 (24.7)
Socioeconomic position
 High23,795 (42.1)3924 (43.5)13,384 (46.2)6484 (34.8)7347 (38.4)7704 (40.0)8744 (48.0)
 Medium12,349 (21.8)1985 (22.0)5972 (20.6)4392 (23.6)4862 (25.4)3804 (19.8)3683 (20.2)
 Low20,432 (36.1)3104 (34.4)9590 (33.1)7738 (41.6)6906 (36.1)7731 (40.2)5795 (31.8)
Table
Table 2. Description of environmental exposures during the period April–October in Rome, 2001–2013.
Table 2. Description of environmental exposures during the period April–October in Rome, 2001–2013.
ExposureMin25° pctl50° pctl75° pctlMax
Warm Season (April–October)
Meteorological
 daily maximum apparent temperature (°C) 4.120.625.830.939.7
Air pollution
 PM10 (μg/m3) 7.024.030.338.0181.7
 Ozone (μg/m3)7.778.694.8111.4199.2
 NO2 (μg/m3)11.942.153.063.5110.1
pctl = percentile.
Table
Table 3. Percent change in daily number of preterm by temperature and air pollutants. Rome, 2001–2013.
Table 3. Percent change in daily number of preterm by temperature and air pollutants. Rome, 2001–2013.
Environmental Exposures a% Change per Unit Increase (90% CI)% Change per IQR (90% CI)IQR
Warm Season (April–October)
MAT (Daily maximum apparent temperature) (Lag 0–2) b2.00 (0.98, 3.03)22.74 (10.59, 36.21)10.3
Air pollutants c
 PM10 (μg/m3) (Lag 12–22)0.73 (0.22, 1.25)8.16 (2.38, 14.26)10.7
 Ozone (μg/m3) (Lag 0–2)0.03 (−0.19, 0.25)0.53 (−3.40, 4.61)30.4
 NO2 (μg/m3) (Lag 0–2)0.20 (−0.14, 0.55)6.39 (−4.27, 18.22)18.2
a Each model was adjusted for long term trend, seasonality and holidays. b Percent change from model adjusted for PM10 (lag 12–22). c Percent changes derived from independent models including one pollutant per time, adjusted for MAT (lag 0–2). IQR = interquartile range.

Share and Cite

MDPI and ACS Style

Asta, F.; Michelozzi, P.; Cesaroni, G.; De Sario, M.; Badaloni, C.; Davoli, M.; Schifano, P. The Modifying Role of Socioeconomic Position and Greenness on the Short-Term Effect of Heat and Air Pollution on Preterm Births in Rome, 2001–2013. Int. J. Environ. Res. Public Health 2019, 16, 2497. https://doi.org/10.3390/ijerph16142497

AMA Style

Asta F, Michelozzi P, Cesaroni G, De Sario M, Badaloni C, Davoli M, Schifano P. The Modifying Role of Socioeconomic Position and Greenness on the Short-Term Effect of Heat and Air Pollution on Preterm Births in Rome, 2001–2013. International Journal of Environmental Research and Public Health. 2019; 16(14):2497. https://doi.org/10.3390/ijerph16142497

Chicago/Turabian Style

Asta, Federica, Paola Michelozzi, Giulia Cesaroni, Manuela De Sario, Chiara Badaloni, Marina Davoli, and Patrizia Schifano. 2019. "The Modifying Role of Socioeconomic Position and Greenness on the Short-Term Effect of Heat and Air Pollution on Preterm Births in Rome, 2001–2013" International Journal of Environmental Research and Public Health 16, no. 14: 2497. https://doi.org/10.3390/ijerph16142497

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop