Improving environmental exposure analysis using cumulative distribution functions and individual geocoding

Environmental justice and power plant emissions in the Regional Greenhouse Gas Initiative states

Policies to reduce greenhouse gases associated with electricity generation have been a major focus of public policy in the United States, but their implications for achieving environmental justice among historically overburdened communities inappropriately remains a marginal issue. In this study we address research gaps in historical and current ambient air emissions burdens in environmental justice communities from power plants participating in the Regional Greenhouse Gases Initiative (RGGI), the country's first market-based power sector emissions reduction program. We find that in RGGI states the percentage of people of color that live within 0-6.2 miles from power plants is up to 23.5 percent higher than the percent of the white population that lives within those same distance bands, and the percentage of people living in poverty that live within 0-5 miles from power plants is up to 15.3 percent higher than the percent of the population not living in poverty within those same distance bands. More importantly, the transition from coal to natural gas underway before RGGI formally started resulted in large increases in both the number of electric-generating units burning natural gas and total net generation from natural gas in environmental justice communities hosting electric-generating units, compared to other communities. Our findings indicate that power sector carbon mitigation policies' focusing on aggregate emissions reductions have largely benefitted non-environmental justice communities and have not redressed the fundamental problem of disparities in pollutant burdens between EJ and non-EJ communities. These must be directly addressed in climate change and carbon emissions mitigation policy.

The Number of Confirmed Cases of Covid-19 by using Machine Learning: Methods and Challenges

Covid-19 is one of the biggest health challenges that the world has ever faced. Public health policy makers need the reliable prediction of the confirmed cases in future to plan medical facilities. Machine learning methods learn from the historical data and make predictions about the events. Machine learning methods have been used to predict the number of confirmed cases of Covid-19. In this paper, we present a detailed review of these research papers. We present a taxonomy that groups them in four categories. We further present the challenges in this field. We provide suggestions to the machine learning practitioners to improve the performance of machine learning methods for the prediction of confirmed cases of Covid-19.

Can we predict the occurrence of COVID-19 cases? Considerations using a simple model of growth

This study aimed to present a simple model to follow the evolution of the COVID-19 (CV-19) pandemic in different countries. The cumulative distribution function (CDF) and its first derivative were employed for this task. The simulations showed that it is almost impossible to predict based on the initial CV-19 cases (1st 2nd or 3rd weeks) how the pandemic will evolve. However, the results presented here revealed that this approach can be used as an alternative for the exponential growth model, traditionally employed as a prediction model, and serve as a valuable tool for investigating how protective measures are changing the evolution of the pandemic.

Assessing the Influence of Socioeconomic Status and Air Pollution Levels on the Public Perception of Local Air Quality in a Mexico-US Border City

Air pollution in developing countries is a growing concern. It is associated with urbanization and social and economic structures. The understanding of how social factors can influence the perception and the potential impact of air pollution have not been addressed sufficiently. This paper addresses the social vulnerability and exposure to PM₁₀ association and its influence on the air quality perception of residents in Mexicali, a Mexico–US border city. This study used individual variables and population census data, as well as statistical and spatial analyses. A cluster of socially vulnerable populations with high exposure to coarse particulate matter (PM₁₀) was found in the city’s peripheral areas. The spatial distribution of the local perception of air quality varied by the exposure zones of the estimated PM₁₀ concentrations. Respondents living in very high exposure areas perceive air quality as “poor,” contrarily to a worse perception in areas of intermediate and lower exposure to PM₁₀. Proximity to stationary sources of pollution was associated with a poor perception of air quality. Results also indicate that low household income and poor air quality perceived at the place of residence negatively influences the perceived changes in the air quality over time. The knowledge of chronic health effects related to air pollution was scarce in the sampled population, especially in the areas with very high exposure and high social vulnerability. These findings can serve as a support in local air quality management.

Bivariate Spatial Pattern between Smoking Prevalence and Lung Cancer Screening in US Counties

Objectives: Lung cancer screening (LCS) with low-dose computed tomography (LDCT) has been a reimbursable preventive service covered by Medicare since 2015. Geographic disparities in the access to LDCT providers may contribute to the low uptake of LCS. We evaluated LDCT service availability for older adults in the United States (US) based on Medicare claims data and explored its ecological correlation with smoking prevalence. Materials and Methods: We identified providers who provided at least 11 LDCT services in 2016 using the Medicare Provider Utilization and Payment Data: Physician and Other Supplier Public Use File. We constructed a 30-mile Euclidian distance buffer around each provider’s location to estimate individual LDCT coverage areas. We then mapped the county-level density of LDCT providers and the county-level prevalence of current daily cigarette smoking in a bivariate choropleth map. Results: Approximately 1/5 of census tracts had no LDCT providers within 30 miles and 46% of counties had no LDCT services. At the county level, the median LDCT density was 0.5 (interquartile range (IQR): 0–5.3) providers per 1000 Medicare fee-for-service beneficiaries, and cigarette smoking prevalence was 17.5% (IQR: 15.2–19.8%). High LDCT service availability was most concentrated in the northeast US, revealing a misalignment with areas of high current smoking prevalence, which tended to be in the central and southern US. Conclusions: Our maps highlight areas in need for enhanced workforce and capacity building aimed at reducing disparities in the access and utilization of LDCT services among older adults in the US.

Influences of environmental exposures on individuals living with cystic fibrosis

INTRODUCTION

Natural, social, and constructed environments play a critical role in the development and exacerbation of respiratory diseases. However, less is known regarding the influence of these environmental/community risk factors on the health of individuals living with cystic fibrosis (CF), compared to other pulmonary disorders.

AREAS COVERED

Here, we review current knowledge of environmental exposures related to CF, which suggests that environmental/community risk factors do interact with the respiratory tract to affect outcomes. Studies discussed in this review were identified in PubMed between March 2019 and March 2020. Although the limited data available do not suggest that avoiding potentially detrimental exposures other than secondhand smoke could improve outcomes, additional research incorporating novel markers of environmental exposures and community characteristics obtained at localized levels is needed.

EXPERT OPINION

As we outline, some environmental exposures and community characteristics are modifiable; if not by the individual, then by policy. We recommend a variety of strategies to advance understanding of environmental influences on CF disease progression.

GIScience and cancer: State of the art and trends for cancer surveillance and epidemiology

Maps are well recognized as an effective means of presenting and communicating health data, such as cancer incidence and mortality rates. These data can be linked to geographic features like counties or census tracts and their associated attributes for mapping and analysis. Such visualization and analysis provide insights regarding the geographic distribution of cancer and can be important for advancing effective cancer prevention and control programs. Applying a spatial approach allows users to identify location-based patterns and trends related to risk factors, health outcomes, and population health. Geographic information science (GIScience) is the discipline that applies Geographic Information Systems (GIS) and other spatial concepts and methods in research. This review explores the current state and evolution of GIScience in cancer research by addressing fundamental topics and issues regarding spatial data and analysis that need to be considered. GIScience, along with its health-specific application in the spatial epidemiology of cancer, incorporates multiple geographic perspectives pertaining to the individual, the health care infrastructure, and the environment. Challenges addressing these perspectives and the synergies among them can be explored through GIScience methods and associated technologies as integral parts of epidemiologic research, analysis efforts, and solutions. The authors suggest GIScience is a powerful tool for cancer research, bringing additional context to cancer data analysis and potentially informing decision-making and policy, ultimately aimed at reducing the burden of cancer.

Human biomonitoring of polycyclic aromatic hydrocarbonsand metals in the general population residing near the municipal solid waste incinerator of Modena, Italy

BACKGROUND AND OBJECTIVES

A cross-sectional biomonitoring study was carried out to investigate exposure to incinerator emission in relation to the body burden of selected biomarkers in the population living around the plant.

METHODS

Approximately 500 people, aged 18-69 yrs, living within 4 km from the incinerator were randomly selected form the population register. Exposure was measured through fall-out maps of particulate matter (PM), used as tracer for incinerator emissions. Ten metabolized polycyclic aromatic hydrocarbons (PAHs), from naphthalene to chrysene, 1-hydroxypyrene and twelve metals (Cd, Cr, Cu, Hg, Ni, Pb, Ni, Zn, V, Tl, As, Sn) were measured in spot urine samples. Confounders, such as diet, smoking, traffic, occupation and personal characteristics were assessed by questionnaires and objective measurements, and included into multivariate linear regression models.

RESULTS

Metal concentrations in urine were in line with or higher than Italian reference limits, besides Cr and V with more than twofold concentrations. Metal levels did not show clear association to exposure categories. Most abundant PAHs were naphthalene (median 26.2 ng/L) and phenanthrene (7.4 ng/L). All PAHs, but benz[a]anthracene and 1-hydroxypyrene, were found in more than 52% of samples, and included in regression models. Significant associations between urinary PAHs and exposure were found, strong for fluorene, and weaker for naphthalene, fluoranthene and pyrene. Results were confirmed by sensitivity analyses. Correlation with variables reported in literature were observed.

CONCLUSIONS

The study indicates that the emissions were very low and highlights that specific urinary PAHs provided useful information about the internal dose arising from incinerator emission.

Survival functions for defining a clinical management Lost To Follow-Up (LTFU) cut-off in Antiretroviral Therapy (ART) program in Zomba, Malawi

BACKGROUND

While, lost to follow-up (LTFU) from antiretroviral therapy (ART) can be considered a catch-all category for patients who miss scheduled visits or medication pick-ups, operational definitions and methods for defining LTFU vary making comparisons across programs challenging. Using weekly cut-offs, we sought to determine the probability that an individual would return to clinic given that they had not yet returned in order to identify the LTFU cut-off that could be used to inform clinical management and tracing procedures.

METHODS

Individuals who initiated ART with Dignitas International supported sites (n = 22) in Zomba, Malawi between January 1 2007-June 30 2010 and were ≥ 1 week late for a follow-up visit were included. Lateness was categorized using weekly cut-offs from ≥1 to ≥26 weeks late. At each weekly cut-off, the proportion of patients who returned for a subsequent follow-up visit were identified. Cumulative Distribution Functions (CDFs) were plotted to determine the probability of returning as a function of lateness. Hazard functions were plotted to demonstrate the proportion of patients who returned each weekly interval relative to those who had yet to return.

RESULTS

In total, n = 4484 patients with n = 7316 follow-up visits were included. The number of included follow-up visits per patient ranged from 1-10 (median: 1). Both the CDF and hazard function demonstrated that after being ≥9 weeks late, the proportion of new patients who returned relative to those who had yet to return decreased substantially.

CONCLUSIONS

We identified a LTFU definition useful for clinical management. The simple functions plotted here did not require advanced statistical expertise and were created using Microsoft Excel, making it a particularly practical method for HIV programs in resource-constrained settings.

Spatial variation in mortality risk for hematological malignancies near a petrochemical refinery: A population-based case-control study

INTRODUCTION

The study investigated the geographic variation of mortality risk for hematological malignancies (HMs) in order to identify potential high-risk areas near an Italian petrochemical refinery.

MATERIAL AND METHODS

A population-based case-control study was conducted and residential histories for 171 cases and 338 sex- and age-matched controls were collected. Confounding factors were obtained from interviews with consenting relatives for 109 HM deaths and 267 controls. To produce risk mortality maps, two different approaches were applied and compared. We mapped (1) adaptive kernel density relative risk estimation for case-control studies which estimates a spatial relative risk function using the ratio between cases and controls' densities, and (2) estimated odds ratios for case-control study data using Generalized Additive Models (GAMs) to smooth the effect of location, a proxy for exposure, while adjusting for confounding variables.

RESULTS

No high-risk areas for HM mortality were identified among all subjects (men and women combined), by applying both approaches. Using the adaptive KDE approach, we found a significant increase in death risk only among women in a large area 2-6 km southeast of the refinery and the application of GAMs also identified a similarly-located significant high-risk area among women only (global p-value<0.025). Potential confounding risk factors we considered in the GAM did not alter the results.

CONCLUSION

Both approaches identified a high-risk area close to the refinery among women only. Those spatial methods are useful tools for public policy management to determine priority areas for intervention. Our findings suggest several directions for further research in order to identify other potential environmental exposures that may be assessed in forthcoming studies based on detailed exposure modeling.

Tools for address georeferencing - limitations and opportunities every public health professional should be aware of

Various address georeferencing (AG) tools are currently available. But little is known about the quality of each tool. Using data from the EPIPorto cohort we compared the most commonly used AG tools in terms of positional error (PE) and subjects' misclassification according to census tract socioeconomic status (SES), a widely used variable in epidemiologic studies. Participants of the EPIPorto cohort (n = 2427) were georeferenced using Geographical Information Systems (GIS) and Google Earth (GE). One hundred were randomly selected and georeferenced using three additional tools: 1) cadastral maps (gold-standard); 2) Global Positioning Systems (GPS) and 3) Google Earth, single and in a batch. Mean PE and the proportion of misclassified individuals were compared. Google Earth showed lower PE than GIS, but 10% of the addresses were imprecisely positioned. Thirty-eight, 27, 16 and 14% of the participants were located in the wrong census tract by GIS, GPS, GE (batch) and GE (single), respectively (p<0.001). Misclassification according to SES was less frequent but still non-negligible −14.4, 8.1, 4.2 and 2% (p<0.001). The quality of georeferencing differed substantially between AG tools. GE seems to be the best tool, but only if prudently used. Epidemiologic studies using spatial data should start including information on the quality and accuracy of their georeferencing tools and spatial datasets.

Geocoding large population-level administrative datasets at highly resolved spatial scales

Using geographic information systems to link administrative databases with demographic, social, and environmental data allows researchers to use spatial approaches to explore relationships between exposures and health. Traditionally, spatial analysis in public health has focused on the county, zip code, or tract level because of limitations to geocoding at highly resolved scales. Using 2005 birth and death data from North Carolina, we examine our ability to geocode population-level datasets at three spatial resolutions - zip code, street, and parcel. We achieve high geocoding rates at all three resolutions, with statewide street geocoding rates of 88.0% for births and 93.2% for deaths. We observe differences in geocoding rates across demographics and health outcomes, with lower geocoding rates in disadvantaged populations and the most dramatic differences occurring across the urban-rural spectrum. Our results suggest highly resolved spatial data architectures for population-level datasets are viable through geocoding individual street addresses. We recommend routinely geocoding administrative datasets to the highest spatial resolution feasible, allowing public health researchers to choose the spatial resolution used in analysis based on an understanding of the spatial dimensions of the health outcomes and exposures being investigated. Such research, however, must acknowledge how disparate geocoding success across subpopulations may affect findings.

Ensuring Confidentiality of Geocoded Health Data: Assessing Geographic Masking Strategies for Individual-Level Data

Public health datasets increasingly use geographic identifiers such as an individual's address. Geocoding these addresses often provides new insights since it becomes possible to examine spatial patterns and associations. Address information is typically considered confidential and is therefore not released or shared with others. Publishing maps with the locations of individuals, however, may also breach confidentiality since addresses and associated identities can be discovered through reverse geocoding. One commonly used technique to protect confidentiality when releasing individual-level geocoded data is geographic masking. This typically consists of applying a certain amount of random perturbation in a systematic manner to reduce the risk of reidentification. A number of geographic masking techniques have been developed as well as methods to quantity the risk of reidentification associated with a particular masking method. This paper presents a review of the current state-of-the-art in geographic masking, summarizing the various methods and their strengths and weaknesses. Despite recent progress, no universally accepted or endorsed geographic masking technique has emerged. Researchers on the other hand are publishing maps using geographic masking of confidential locations. Any researcher publishing such maps is advised to become familiar with the different masking techniques available and their associated reidentification risks.

Neighborhood walkability: field validation of geographic information system measures

BACKGROUND

Given the health benefits of walking, there is interest in understanding how physical environments favor walking. Although GIS-derived measures of land-use mix, street connectivity, and residential density are commonly combined into indices to assess how conducive neighborhoods are to walking, field validation of these measures is limited.

PURPOSE

To assess the relationship between audit- and GIS-derived measures of overall neighborhood walkability and between objective (audit- and GIS-derived) and participant-reported measures of walkability.

METHODS

Walkability assessments were conducted in 2009. Street-level audits were conducted using a modified version of the Pedestrian Environmental Data Scan. GIS analyses were used to derive land-use mix, street connectivity, and residential density. Participant perceptions were assessed using a self-administered questionnaire. Audit, GIS, and participant-reported indices of walkability were calculated. Spearman correlation coefficients were used to assess the relationships between measures. All analyses were conducted in 2012.

RESULTS

The correlation between audit- and GIS-derived measures of overall walkability was high (R=0.7 [95% CI=0.6, 0.8]); the correlations between objective (audit and GIS-derived) and participant-reported measures were low (R=0.2 [95% CI=0.06, 0.3]; R=0.2 [95% CI=0.04, 0.3], respectively). For comparable audit and participant-reported items, correlations were higher for items that appeared more objective (e.g., sidewalk presence, R=0.4 [95% CI=0.3, 0.5], versus safety, R=0.1 [95% CI=0.003, 0.3]).

CONCLUSIONS

The GIS-derived measure of walkability correlated well with the in-field audit, suggesting that it is reasonable to use GIS-derived measures in place of more labor-intensive audits. Interestingly, neither audit- nor GIS-derived measures correlated well with participants' perceptions of walkability.

Statistical air quality predictions for public health surveillance: evaluation and generation of county level metrics of PM2.5 for the environmental public health tracking network

Background

The Centers for Disease Control and Prevention (CDC) developed county level metrics for the Environmental Public Health Tracking Network (Tracking Network) to characterize potential population exposure to airborne particles with an aerodynamic diameter of 2.5 μm or less (PM_2.5). These metrics are based on Federal Reference Method (FRM) air monitor data in the Environmental Protection Agency (EPA) Air Quality System (AQS); however, monitor data are limited in space and time. In order to understand air quality in all areas and on days without monitor data, the CDC collaborated with the EPA in the development of hierarchical Bayesian (HB) based predictions of PM_2.5 concentrations. This paper describes the generation and evaluation of HB-based county level estimates of PM_2.5.

Methods

We used three geo-imputation approaches to convert grid-level predictions to county level estimates. We used Pearson (r) and Kendall Tau-B (τ) correlation coefficients to assess the consistency of the relationship, and examined the direct differences (by county) between HB-based estimates and AQS-based concentrations at the daily level. We further compared the annual averages using Tukey mean-difference plots.

Results

During the year 2005, fewer than 20% of the counties in the conterminous United States (U.S.) had PM_2.5 monitoring and 32% of the conterminous U.S. population resided in counties with no AQS monitors. County level estimates resulting from population-weighted centroid containment approach were correlated more strongly with monitor-based concentrations (r = 0.9; τ = 0.8) than were estimates from other geo-imputation approaches. The median daily difference was −0.2 μg/m³ with an interquartile range (IQR) of 1.9 μg/m³ and the median relative daily difference was −2.2% with an IQR of 17.2%. Under-prediction was more prevalent at higher concentrations and for counties in the western U.S.

Conclusions

While the relationship between county level HB-based estimates and AQS-based concentrations is generally good, there are clear variations in the strength of this relationship for different regions of the U.S. and at various concentrations of PM_2.5. This evaluation suggests that population-weighted county centroid containment method is an appropriate geo-imputation approach, and using the HB-based PM_2.5 estimates to augment gaps in AQS data provides a more spatially and temporally consistent basis for calculating the metrics deployed on the Tracking Network.

Disproportionate proximity to environmental health hazards: methods, models, and measurement

We sought to provide a historical overview of methods, models, and data used in the environmental justice (EJ) research literature to measure proximity to environmental hazards and potential exposure to their adverse health effects. We explored how the assessment of disproportionate proximity and exposure has evolved from comparing the prevalence of minority or low-income residents in geographic entities hosting pollution sources and discrete buffer zones to more refined techniques that use continuous distances, pollutant fate-and-transport models, and estimates of health risk from toxic exposure. We also reviewed analytical techniques used to determine the characteristics of people residing in areas potentially exposed to environmental hazards and emerging geostatistical techniques that are more appropriate for EJ analysis than conventional statistical methods. We concluded by providing several recommendations regarding future research and data needs for EJ assessment that would lead to more reliable results and policy solutions.

Children at risk: measuring racial/ethnic disparities in potential exposure to air pollution at school and home

STUDY OBJECTIVE

This paper addresses the environmental justice implications of children's health by exploring racial/ethnic disparities in potential exposure to air pollution, based on both school and home locations of children and three different types of pollution sources, in Orange County, Florida, USA.

METHODS

Using geocoded school and residence locations of 151 709 children enrolled in the public school system, distribution functions of proximity to the nearest source are generated for each type of air pollution source in order to compare the exposure potential of white, Hispanic, and black children. Discrete buffer distances are utilised to provide quantitative comparisons for statistical testing.

MAIN RESULTS

At any given distance from each type of pollution source, the cumulative proportion of Hispanic or black children significantly exceeds the corresponding proportion of white children, for both school and home locations. Regardless of race, however, a larger proportion of children are potentially exposed to air pollution at home than at school.

CONCLUSIONS

This study addresses the growing need to consider both daytime and nighttime activity patterns in the assessment of children's exposure to environmental hazards and related health risks. The results indicate a consistent pattern of racial inequity in the spatial distribution of all types of air pollution sources examined, with black children facing the highest relative levels of potential exposure at both school and home locations.

Error and bias in determining exposure potential of children at school locations using proximity-based GIS techniques

BACKGROUND

The widespread availability of powerful tools in commercial geographic information system (GIS) software has made address geocoding a widely employed technique in spatial epidemiologic studies.

OBJECTIVE

The objective of this study was to determine the effect of the positional error in geocoding on the analysis of exposure to traffic-related air pollution of children at school locations.

METHODS

For a case study of Orange County, Florida, we determined the positional error of geocoding of school locations through comparisons with a parcel database and digital orthophotography. We used four different geocoding techniques for comparison to establish the repeatability of geocoding, and an analysis of proximity to major roads to determine bias and error in environmental exposure assessment.

RESULTS

RESULTS INDICATE THAT THE POSITIONAL ERROR IN GEOCODING OF SCHOOLS IS VERY SUBSTANTIAL: We found that the 95% root mean square error was 196 m using street centerlines, 306 m using TIGER roads, and 210 and 235 m for two commercial geocoding firms. We found bias and error in proximity analysis to major roads to be unacceptably large at distances of < 500 m. Bias and error are introduced by lack of positional accuracy and lack of repeatability of geocoding of school locations.

CONCLUSIONS

These results suggest that typical geocoding is insufficient for fine-scale analysis of school locations and more accurate alternatives need to be considered.

Influence of geocoding quality on environmental exposure assessment of children living near high traffic roads

Background

The widespread availability of powerful geocoding tools in commercial GIS software and the interest in spatial analysis at the individual level have made address geocoding a widely employed technique in epidemiological studies. This study determined the effect of the positional error in street geocoding on the analysis of traffic-related air pollution on children.

Methods

For a case-study of a large sample of school children in Orange County, Florida (n = 104,865) the positional error of street geocoding was determined through comparison with a parcel database. The effect of this error was evaluated by analyzing the proximity of street and parcel geocoded locations to road segments with high traffic volume and determining the accuracy of the classification using the results of street geocoding. Of the original sample of 163,886 addresses 36% were not used in the final analysis because they could not be reliably geocoded using either street or parcel geocoding. The estimates of positional error can therefore be considered conservative underestimates.

Results

Street geocoding was found to have a median error of 41 meters, a 90^th percentile of 100 meters, a 95^th percentile of 137 meters and a 99^th percentile of 273 meters. These positional errors were found to be non-random in nature and introduced substantial bias and error in the estimates of potential exposure to traffic-related air pollution. Street geocoding was found to consistently over-estimate the number of potentially exposed children at small distances up to 250 meters. False positives and negatives were also found to be very common at these small distances.

Conclusion

Results of the case-study presented here strongly suggest that typical street geocoding is insufficient for fine-scale analysis and more accurate alternatives need to be considered.

A comparison of methods for calculating population exposure estimates of daily weather for health research

BACKGROUND

To explain the possible effects of exposure to weather conditions on population health outcomes, weather data need to be calculated at a level in space and time that is appropriate for the health data. There are various ways of estimating exposure values from raw data collected at weather stations but the rationale for using one technique rather than another; the significance of the difference in the values obtained; and the effect these have on a research question are factors often not explicitly considered. In this study we compare different techniques for allocating weather data observations to small geographical areas and different options for weighting averages of these observations when calculating estimates of daily precipitation and temperature for Australian Postal Areas. Options that weight observations based on distance from population centroids and population size are more computationally intensive but give estimates that conceptually are more closely related to the experience of the population.

RESULTS

Options based on values derived from sites internal to postal areas, or from nearest neighbour sites--that is, using proximity polygons around weather stations intersected with postal areas--tended to include fewer stations' observations in their estimates, and missing values were common. Options based on observations from stations within 50 kilometres radius of centroids and weighting of data by distance from centroids gave more complete estimates. Using the geographic centroid of the postal area gave estimates that differed slightly from the population weighted centroids and the population weighted average of sub-unit estimates.

CONCLUSION

To calculate daily weather exposure values for analysis of health outcome data for small areas, the use of data from weather stations internal to the area only, or from neighbouring weather stations (allocated by the use of proximity polygons), is too limited. The most appropriate method conceptually is the use of weather data from sites within 50 kilometres radius of the area weighted to population centres, but a simpler acceptable option is to weight to the geographic centroid.