Spatial identification of environmental health hazards potentially associated with adverse birth outcomes

Abstract

Reduced birth weight (RBW) and reduced head circumference (RHC) are adverse birth outcomes (ABOs), often linked to environmental exposures. However, spatial identification of specific health hazards, associated with these ABOs, is not always straightforward due to presence of multiple health hazards and sources of air pollution in urban areas. In this study, we test a novel empirical approach to the spatial identification of environmental health hazards potentially associated with the observed RHC and RBW patterns. The proposed approach is implemented as a systematic search, according to which alternative candidate locations are ranked based on the strength of association with the observed birth outcome patterns. For empirical validation, we apply this approach to the Haifa Bay Area (HBA) in Israel, which is characterized by multiple health hazards and numerous sources of air pollution. We identified a spot in the local industrial zone as the main risk source associated with the observed RHC and RBW patterns. Multivariate regressions, controlling for personal, neighborhood, and geographic factors, revealed that the relative risks of RHC and RBW tend to decline, other things being equal, as a function of distance from the identified industrial spot. We recommend the proposed identification approach as a preliminary risk assessment tool for environmental health studies, in which detailed information on specific sources of air pollution and air pollution dispersion patterns is unavailable due to limited reporting or insufficient monitoring.

Appendices

Appendix 1

Table 2 Descriptive statistics of the research variables

Appendix 2

Wind adjustment

Several empirical studies implemented adjustments for wind frequency and directions, considering that air pollution may spread over larger areas by wind (see inter alia, Dore et al. 2006; Xu and Akhtar 2009; Zoë et al. 2012; Chen and Ye 2015). There are several approaches used for wind adjustments, some of which are based on a seasonal analysis of the data (Dore et al. 2006) while others are based on the analysis of the amount of precipitation, solar radiation, maximum and minimum temperatures (Chen and Ye 2015), and average wind speed (Zoe et al. 2012).

For wind adjustment, we used the wind frequency rose of the study area, plotted at a one-degree angular resolution and featuring the average annual distribution of wind frequencies for each one-degree angle. Wind frequency data for the study were obtained from two AQMSs, “Igud Arim,” located in the central part of industrial area, and “Kiryat Tiv’on,” located in the southeastern part of the study area and used for a sensitivity test (see Fig. 2 and text for explanations).

Since the probability of wind from point j to point i (w_ji) was assumed to be random, we used the probability density function (PDF) instead of a fixed distribution function. PDF describes the relative likelihood for the random variable, e.g., wind frequency to take on a given value (w_ji):

$$ {PDF}_w\left({w}_{ji},{\lambda}_j\;\right)={\lambda}_j\bullet {e}^{-{\lambda}_{\mathrm{j}}\bullet {W}_{ji}},\forall {W}_{ji}\in \left(0,1\right),{\lambda}_{\mathrm{j}}=\frac{1}{\overline{w_{ji}}},\mathrm{for}\ \overline{w_j}=\frac{1}{n}{\sum}_{i=1}^n{w}_{ji}. $$

(A2.1)

where w_ji is the annual wind frequency from point j to point i, $\overline{w_j}$ is the average annual wind frequency² from point j, n is a number of i points, and λ_j > 0 is the parameter of the distribution, also known as the rate parameter.

Next, we adjusted distances between j and i (dist_ji) for wind frequency and direction as follows:

$$ \overset{\sim }{dist_{ji}}=T\left({dist}_{ji}\left|{W}_{ji}\right.\right)={dist}_{ji}\bullet {PDF}_w\left({W}_{ji},{\uplambda}_{\mathrm{j}}\;\right). $$

(A2.2)

where $\overset{\sim }{dist_{ij}}$ = distance between i and j adjusted by wind frequency (W_ji) between the points (measured as e.g., annual or seasonal averages of directional wind frequencies), and T(dist_ji| W_ji) is a distance transformation function (exponential transformation was used). According to this transformation, distances between points j and i with frequent winds are reduced, while distances between points with infrequent winds remain unchanged.

Appendix 3

Table 3 Data on the emission inventory of main industrial enterprises located in the study area (source, IMEP 2017)