An investigation of the environmental determinants of asthma hospitalizations: An applied spatial approach

doi:10.1016/j.apgeog.2013.11.011

Applied Geography

Volume 47, February 2014, Pages 10-19

https://doi.org/10.1016/j.apgeog.2013.11.011 Get rights and content

Highlights

•
We perform a nationwide spatial study to assess asthma–environment relationship.
•
High NO₂, low NDVI, and high temperatures increase asthma hospitalizations.
•
The urban coverage influences the relationship between asthma and environment.
•
Urbanization poses a higher risk of complications for asthmatic people.

Abstract

Several previous studies have connected asthma exacerbations with environmental factors such as pollutants. However, the majority do not analyze the information spatially. The objective of this study was to evaluate the relationship between asthma hospital admissions and several environmental variables in mainland Portugal using spatial data from remote sensing and spatial modeling. A set of five environmental variables were considered: near-surface air temperature (T_a) from the temperature profile of the Moderate Resolution Imaging Spectroradiometer (MODIS); relative humidity (RH) from meteorological station data interpolated by kriging; vegetation density from MODIS Normalized Difference Vegetation Index (NDVI); and space-time estimates of nitrogen dioxide (NO₂) and particulate matter less than 10 μm (PM₁₀), both from Land-Use Regression (LUR) models based on data from air quality stations. Districts were aggregated into three groups based on their percent urban cover, and the municipality was chosen as the sampling unit to assess the relationship between asthma hospital admission rates and environmental variables by season for the years 2003–2008. In the most urban group, T_a, NDVI, and NO₂ had consistent relationships with asthma in all seasons (Pearson correlation coefficients ranging from 0.351 to 0.600, −0.376 to −0.498, and 0.405 to 0.513, respectively). The associations in the other groups were very weak or non-existent. The percentage of urban cover influences the relationship between the environment and asthma. The results suggest that asthmatic people living in highly urbanized and sparsely vegetated areas are at a greater risk of suffering severe asthma attacks that lead to hospital admissions.

Introduction

Asthma is a chronic inflammatory disorder of the airways that affects people of all ages throughout the world. The chronic inflammation is associated with a hyper-responsiveness of the airways that leads to recurrent episodes of wheezing, breathlessness, chest tightness, and coughing, particularly at night or in the early morning. The disease is prevalent in all age groups (GINA Global Initiative for Asthma, 2011, Sa-Sousa et al., 2012) and affects approximately 300 million people worldwide, causing 250 thousand deaths per year (GINA, 2011). The disease can be exacerbated by environmental factors such as allergens, air pollution or weather changes (Portnov, Reiser, Karkabi, Cohen-Kastel, & Dubnov, 2012) and infectious factors such as viruses and bacteria (GINA, 2011). When uncontrolled, asthma can place severe limits on daily life and is sometimes fatal.

In recent years, several studies have analyzed how asthma is exacerbated by pollutants (Wilhelm et al., 2008) such as ozone (O₃), particulate matter (PM) with aerodynamic diameters less than 10 μm or 2.5 μm (PM₁₀ or PM_2.5, respectively), nitrogen dioxide (NO₂), carbon monoxide (CO), and sulfur dioxide (SO₂). However, the results of these studies do not agree (Akinbami et al., 2010, Delamater et al., 2012), and thus, the role of pollutants in asthma exacerbations remains controversial (GINA, 2011). The discrepancies among studies could be explained by variations in study design and the modeling methods adopted (Akinbami et al., 2010). In addition, the choice of outcome (e.g., prevalence, emergency department visits, hospitalizations, and mortality), spatio-temporal scale (e.g., data aggregation level, temporal resolution), and exposure modeling (e.g., ambient monitor data, personal exposure data) have the potential to influence the observed relationship (Delamater et al., 2012).

The majority of asthma studies are not spatially explicit. Non-spatial studies have relied on data gathered at a single monitoring station or on an average value from multiple monitoring stations, which may lead to misclassifications of exposure (Chen, Mengersen, & Tong, 2007). The exacerbation of a chronic disease is controlled by epidemiological factors that operate over a range of spatial and temporal scales to produce spatially and temporally complex patterns of disease incidence (Graham, Atkinson, & Danson, 2004). Therefore, it is important to incorporate spatial information in the study of environment-related diseases (Delamater et al., 2012). In addition to spatial prediction techniques (e.g., inverse distance weighting and kriging), remote sensing has been widely used in health sciences, particularly in the study of infectious diseases (Maxwell, Meliker, & Goovaerts, 2010). Remotely sensed data have the major advantage of providing synoptic and frequent overviews of the Earth's surface, whereas the distribution of ground-based measurements is usually sparse and uneven. Additionally, using these data avoids expensive and time-consuming monitoring campaigns.

The objective of this study was to examine the association between severe asthma exacerbations requiring hospital admission and weather conditions, vegetation density, and air pollution in mainland Portugal by using both remotely sensed and modeled spatial data. Specifically, the near-surface air temperature (T_a), relative humidity (RH), Normalized Difference Vegetation Index (NDVI), and the air pollutants NO₂ and PM₁₀ were analyzed from 2003 to 2008 in a retrospective ecological study.

Section snippets

Study area

Mainland Portugal (Fig. 1) has a total area of approximately 89,000 km² and is subdivided into three major administrative levels: 18 districts, 278 municipalities, and 4050 sub-municipalities (as of the study period). Between 2003 and 2008, the total population ranged from 9,991,654 to 10,135,309.

Data sources

The data sources used to obtain the variables T_a, RH, NDVI, NO₂, and PM₁₀ are briefly described below.

Results

The validation results for the monthly averages of the environmental variables determined in this study are presented in Table 4. T_a achieved an RMSE within the 1–2 °C range, which is generally accepted as accurate for remote sensing based T_a estimations (Benali, Carvalho, Nunes, Carvalhais, & Santos, 2012). RH reached an acceptable R² (cross-validation), and the pollutants achieved an R² within the range that is usually obtained (0.54–0.81) (Ryan & LeMasters, 2007). All of the four variables

Discussion

This study shows that in highly urbanized districts, hospital admissions due to asthma exacerbations are associated with high levels of NO₂, low NDVI values, and high temperatures throughout the year. It is important to note that there was no expectation of high correlations with the environmental variables because the exacerbation of asthma symptoms depends on several other external factors (e.g., indoor pollution, viral infections) and host factors (e.g., genetics).

There was a positive

Conclusion

To our knowledge, this is the first nationwide study that evaluates the association between asthma and environmental factors through the use of spatial information. The work presented here shows that the relationship between the environment and asthma is influenced by the percentage of urban cover. In highly urbanized districts, hospital admissions due to asthma exacerbation are associated with high levels of NO₂, low NDVI values, and high temperatures throughout all seasons. Therefore, it can

Acknowledgments

The authors wish to thank the Portuguese Ministry of Health's Authority for Health Services (Administração Central do Sistema de Saúde – ACSS) for providing access to national hospital admissions data. The authors would also like to express gratitude to the Geo-Space Sciences Research Center and to the Center for Research in Health Technologies and Information Systems for providing the conditions necessary to carry out this study.

References (41)

L.J. Akinbami et al.
The association between childhood asthma prevalence and monitored air pollutants in metropolitan areas, United States, 2001–2004
Environmental Research
(2010)
A. Benali et al.
Estimating air surface temperature in Portugal using MODIS LST data
Remote Sensing of Environment
(2012)
A.J. Chauhan et al.
Personal exposure to nitrogen dioxide (NO₂) and the severity of virus-induced asthma in children
The Lancet
(2003)
L. Chen et al.
Spatiotemporal relationship between particle air pollution and respiratory emergency hospital admissions in Brisbane, Australia
Science of the Total Environment
(2007)
P.L. Delamater et al.
An analysis of asthma hospitalizations, air pollution, and weather conditions in Los Angeles County, California
Science of the Total Environment
(2012)
A.J. Graham et al.
Spatial analysis for epidemiology
Acta Tropica
(2004)
S.E. Grineski et al.
Climate change and environmental injustice in a bi-national context
Applied Geography
(2012)
L. von Hertzen et al.
Disconnection of man and the soil: reason for the asthma and atopy epidemic?
Journal of Allergy and Clinical Immunology
(2006)
Y. Shimada et al.
Analysis of indoor PM_2.5 exposure in Asian countries using time use survey
Science of the Total Environment
(2011)
C. Vancutsem et al.
Evaluation of MODIS land surface temperature data to estimate air temperature in different ecosystems over Africa
Remote Sensing of Environment
(2010)

H. Zhang et al.

Analysis of land use/land cover change, population shift, and their effects on spatiotemporal patterns of urban heat islands in metropolitan Shanghai, China

Applied Geography

(2013)

R.L. Atmar et al.

Respiratory tract viral infections in inner-city asthmatic adults

Archives of Internal Medicine

(1998)

S.M. Babin et al.

Pediatric patient asthma-related emergency department visits and admissions in Washington, DC, from 2001–2004, and associations with air quality, socio-economic status and age group

Environmental Health

(2007)

J.D. Bury

Climate and chest disorders

British Medical Journal

(1972)

M. Caetano et al.

CORINE land cover 2006 for continental Portugal

(2009)

A.J. Chauhan et al.

Air pollution and infection in respiratory illness

British Medical Bulletin

(2003)

M.S. Davis et al.

Cold weather exercise and airway cytokine expression

Journal of Applied Physiology

(2005)

J.C. Emberlin

Aerobiology

B. Erbas et al.

Air pollution and childhood asthma emergency hospital admissions: estimating intra-city regional variations

International Journal of Environmental Health Research

(2005)

T.G. Farr et al.

The shuttle radar topography mission

Reviews of Geophysics

(2007)

Cited by (46)

Integration of Landsat 8 (OLI) and MODIS images to monitor suspended particles and evaluate the spatial pattern of air pollution
2024, Atmospheric Pollution Research
Air pollution has become one of the disturbances of daily life in urban areas due to its negative and repulsive effect in current conditions. On the other hand, it has a long-term negative impact on the behavior and quantity of climate change scenarios at the local, national and global levels. In this research, the integration of Landsat 8 (OLI) satellite images and Moderate Resolution Imaging Spectroradiometer (MODIS) sensor (to estimate the ozone permeability and the water vapor permeability coefficients) using the Second Simulation of a Satellite Signal in the Solar Spectrum (6S) radiation transfer model was used to determine Aerosol Optical Depth (AOD). For this purpose, instead of using the look-up table and approximating the AOD values, the components of the 6S radiative transfer model were estimated with a combined approach and the AOD values were retrieved without using the look-up table approach. One of the main problems of past research is to identify a suitable conversion model to connect the refractive index to the composition of aerosols. Therefore, the main innovation of this research is focused on a simplified, high resolution AOD retrieval method for Landsat 8 (OLI) images over mixed surfaces which does not require a LUT. Considering the relationship between AOD values and various indicators of air quality measurement, in this work, AOD estimation has been done with the aim of helping to predict these indicators. In this research, 24,569 square kilometers from a part of two states of Maryland and Virginia in the United States of America were investigated in order to estimate AOD. In order to evaluate the results of AOD retrieval, 5 Aerosol Robotic Network (AERONET) ground stations were used for comparison and their accuracy was evaluated. The results show that the proposed method has been able to extract the amount of aerosol with acceptable accuracy. According to the results obtained in this research, the root mean square errors (RMSE) is equal to 0.1036, the adjusted R² coefficient (AdjR²) is equal to 0.8032 and the coefficient of determination (R²) is equal to 0.8079. It can be concluded that the integration of Landsat 8 (OLI) images with MODIS images has been able to provide more suitable results.
Asthma hospitalizations: A call for a national strategy to fight health inequities
2023, Pulmonology
Issue 1 - “Update on adverse respiratory effects of outdoor air pollution” Part 2): Outdoor air pollution and respiratory diseases: Perspectives from Angola, Brazil, Canada, Iran, Mozambique and Portugal
2022, Pulmonology
Citation Excerpt :
Globally, results have shown that PM1091–97,99,100 and SO290–92,97,99,100 were most frequently positively associated with respiratory disease-related outcomes (however, PM10 levels were inversely correlated with hospital admissions for COPD, in one study99). With a slightly lower frequency, PM2.5,92,93,96,97,101, CO,90,92,95,97,98,100 and NO291,92,95,97,100,102 have also been shown to be positively correlated with respiratory disease outcomes. O3 was a special situation since it showed positive correlations with respiratory outcomes in four studies: with all respiratory causes, asthma, and COPD,91 acute upper respiratory infections (AURI) and chronic lower respiratory diseases (CLRD),95 and COPD and “allied conditions”,98 and respiratory mortality in those > 65 years-old, as well as ER visits in those < 14 years-old and those > 65 years-old100; however, in contrast, it showed negative correlations in two studies: with asthma97 and with all respiratory causes.101
To analyse the GARD perspective on the health effects of outdoor air pollution, and to synthesise the Portuguese epidemiological contribution to knowledge on its respiratory impact.
Ambient air pollution has deleterious respiratory effects which are more apparent in larger, densely populated and industrialised countries, such as Canada, Iran, Brazil and Portugal, but it also affects people living in low-level exposure areas. While low- and middle-income countries (LMICs), are particularly affected, evidence based on epidemiological studies from LMICs is both limited and heterogeneous. While nationally, Portugal has a relatively low level of air pollution, many major cities face with substantial air pollution problems. Time series and cross-sectional epidemiological studies have suggested increased respiratory hospital admissions, and increased risk of respiratory diseases in people who live in urban areas and are exposed to even a relatively low level of air pollution.
Adverse respiratory effects due to air pollution, even at low levels, have been confirmed by epidemiological studies. However, evidence from LMICs is heterogeneous and relatively limited. Furthermore, longitudinal cohort studies designed to study and quantify the link between exposure to air pollutants and respiratory diseases are needed. Worldwide, an integrated approach must involve multi-level stakeholders including governments (in Portugal, the Portuguese Ministry of Health, which hosts GARD-Portugal), academia, health professionals, scientific societies, patient associations and the community at large. Such an approach not only will garner a robust commitment, establish strong advocacy and clear objectives, and raise greater awareness, it will also support a strategy with adequate measures to be implemented to achieve better air quality and reduce the burden of chronic respiratory diseases (CRDs).
Exposure to urban greenspace and pathways to respiratory health: An exploratory systematic review
2022, Science of the Total Environment
Urban greenspace may have a beneficial or adverse effect on respiratory health. Our objective was to perform an exploratory systematic review to synthesise the evidence and identify the potential causal pathways relating urban greenspace and respiratory health.
We followed PRISMA guidelines on systematic reviews and searched five databases for eligible studies during 2000–2021. We incorporated a broad range of urban greenspace and respiratory health search terms, including both observational and experimental studies. Screening, data extraction, and risk of bias, assessed using the Navigation Guide criteria, were performed independently by two authors. We performed a narrative synthesis and discuss suggested pathways to respiratory health.
We identified 108 eligible papers (n = 104 observational, n = 4 experimental). The most common greenspace indicators were the overall greenery or vegetation (also known as greenness), green land use/land cover of physical area classes (e.g., parks, forests), and tree canopy cover. A wide range of respiratory health indicators were studied, with asthma prevalence being the most common. Two thirds (n = 195) of the associations in these studies were positive (i.e., beneficial) with health, with 31% (n = 91) statistically significant; only 9% (n = 25) of reported associations were negative (i.e., adverse) with health and statistically significant. The most consistent positive evidence was apparent for respiratory mortality. There were n = 35 (32%) ‘probably low’ and n = 73 (68%) ‘probably high’ overall ratings of bias. Hypothesised causal pathways for health benefits included lower air pollution, more physically active populations, and exposure to microbial diversity; suggested mechanisms with poorer health included exposure to pollen and other aeroallergens.
Many studies showed positive association between urban greenspace and respiratory health, especially lower respiratory mortality; this is suggestive, but not conclusive, of causal effects. Results underscore the importance of contextual factors, greenspace metric employed, and the potential bias of subtle selection factors, which should be explored further.
Modeling air pollution-related hospital admissions employing remote sensing and geographical information systems
2021, Atmospheric Environment
Citation Excerpt :
In fact, we found a negative association between maximum PBL height and hospital admissions in all the regression models, suggesting that as PBL height decreases the aerosols tend to concentrate in the lower layers of the atmosphere, which could increase hospital admissions. Regarding remote sensing parameters, several studies have found significant associations between vegetal covering, land temperature or elevation with air pollutants (Yang et al., 2017; Song et al., 2018; Wang et al., 2020) and air pollution-related hospital admissions (Ayres-Sampaio et al., 2014; Heo and Bell, 2019; Alvarez-Mendoza et al., 2020). Regarding the NDVI and the LST, this relationship can be easily explained since, at the study area latitude, both these parameters decrease during the cold season when respiratory diseases have the highest prevalence (Carreras et al., 2015).
Land-use regression models and remote sensing data have been widely employed to forecast atmospheric aerosol levels. Recently, these methodologies have been combined to predict the influence of this pollutant on human health. However, traditional land-use regression models do not often consider the complex interactions between predictors, and most of these do not include socioeconomic variables. Thus, in the present study, we aimed to estimate suspended particle-related hospital admissions by employing remote sensing, meteorological, environmental, and demographic parameters. In this cohort study, we analyzed 1,612,049 hospital admissions from Córdoba city, Argentina, from 2005 to 2011, and developed several regression and machine learning land-use models to compare their predictive powers. We found that childhood was the age group with the highest number of hospital admissions related with upper respiratory tract diseases. When predicting population-normalized hospital admissions, the machine learning models, in particular the generalized boosted machine, revealed a better performance than regression models, exhibiting the lowest root mean square error (0.4264) in the test data set. This model also achieved the best R²_adj (0.6088) when plotting predicted vs. reported normalized cases. The most important predictors were the meteorological variables, followed by the aerosol optical depth and the planet boundary layer height. Some other predictors, such as educational level, land value, and unsatisfied basic needs, showed less relevance but enhanced the model's prediction power. Furthermore, the predictive power increased after a 1-day lag in hospital admissions (RMSE = 0.4121), highlighting the importance of meteorological and environmental variables in the onset of respiratory diseases.
Spatial estimation of chronic respiratory diseases based on machine learning procedures—an approach using remote sensing data and environmental variables in quito, Ecuador
2020, Applied Geography
Citation Excerpt :
During the last few years, remote sensing data have increasing been used in monitoring, spatial predictive modeling, surveillance, and risk assessment with respect to human health (Viana et al., 2017). These human health studies have also been associated with air pollution spatial modeling, which is connected to some vector-borne (Estallo, Benitez, Lanfri, Scavuzzo, & Almirón, 2016) and respiratory diseases (Ayres-Sampaio et al., 2014). In this context, spatial models relying on remote sensing data have been developed to identify different air pollutants, the most common of which are particulate matter less than 2.5 and 10 μm (PM2.5 and PM10, respectively) (Alvarez-Mendoza, Teodoro, Torres, et al., 2019; Alvarez et al., 2016), nitrogen dioxide (NO2) (C. Zheng et al., 2018), tropospheric ozone (O3) (Alvarez-Mendoza, Teodoro, & Ramirez-Cando, 2019), sulfur dioxide (SO2) (Ibrahim Sameen et al., 2014), and carbon dioxide (CO2) (Chejarla, Maheshuni, & Mandla, 2016).
Over the last few years, the use of remote sensing data in different applications such as estimation of air pollution concentration and health applications has become very popular and new. Thus, some studies have established a possible relationship between environmental variables and respiratory health parameters. This study proposes to estimate the prevalence of Chronic Respiratory Diseases, where there is a relationship between remote sensing data (Landsat 8) and environmental variables (air pollution and meteorological data) to determine the number of hospital discharges of patients with chronic respiratory diseases in Quito, Ecuador, between 2013 and 2017. The main objective of this study is to establish and evaluate an alternative LUR model that is capable of estimate the prevalence of chronic respiratory diseases, in contrast with traditional LUR models, which typically assess air pollutants. Moreover, this study also evaluates different analytic techniques (multiple linear regression, multilayer perceptron, support vector regression, and random forest regression) that often form the basis of spatial models. The results show that machine learning techniques, such as support vector machine, are the most effective in computing such models, presenting the lowest root-mean-square error (RMSE). Additionally, in this study, we show that the most significant remote sensing predictors are the blue and infrared bands. Our proposed model is a spatial modeling approach that is capable of determining the prevalence of chronic respiratory diseases in the city of Quito, which can serve as a useful tool for health authorities in policy- and decision-making.

View all citing articles on Scopus

View full text

An investigation of the environmental determinants of asthma hospitalizations: An applied spatial approach

Highlights

Abstract

Introduction

Section snippets

Study area

Data sources

Results

Discussion

Conclusion

Acknowledgments

Environmental Research

Remote Sensing of Environment

The Lancet

Science of the Total Environment

Science of the Total Environment

Acta Tropica

Applied Geography

Journal of Allergy and Clinical Immunology

Science of the Total Environment

Remote Sensing of Environment

Applied Geography

Respiratory tract viral infections in inner-city asthmatic adults

Archives of Internal Medicine

Pediatric patient asthma-related emergency department visits and admissions in Washington, DC, from 2001–2004, and associations with air quality, socio-economic status and age group

Environmental Health

Climate and chest disorders

British Medical Journal

CORINE land cover 2006 for continental Portugal

Air pollution and infection in respiratory illness

British Medical Bulletin

Cold weather exercise and airway cytokine expression

Journal of Applied Physiology

Aerobiology

Air pollution and childhood asthma emergency hospital admissions: estimating intra-city regional variations

International Journal of Environmental Health Research

The shuttle radar topography mission

Reviews of Geophysics