Abstract:

Exposure to ambient air pollution increases mortality and morbidity, leading disabilities, and premature deaths. Air pollution has been identified as a leading cause of global disease burden, especially in low- and middle-income countries in 2015 (Global Burden of Diseases, Injuries and Risk Factors Study, 2015). This study explores the relation between mortality rates and particulate matter (PM) concentrations in the 50 Spanish regions for the period 2002-2017. Moreover, we estimated the premature deaths due to PM in Spain according to welfare and production losses in 2017. Random-effects models were developed to evaluate the relation between mortality rates and PM concentrations. The economic cost of premature deaths was assessed using the Willingness to Pay approach to quantify welfare losses and the Human Capital method to estimate production losses. _PM10 concentrations are positively related to mortality due to respiratory diseases and stroke. Based on 10,342 premature deaths in 2017, losses in welfare amount to EUR 36,227 million (3.1% of Spanish GDP). The economic value of current and future production losses reached EUR 229 million (0.02% of GDP). From a social perspective, air pollution is a public health concern that greatly impacts health and quality of life. Results highlight the need to implement or strengthen regulatory, fiscal, and health public policies to substantially benefit the population’s health by reducing their exposure to air pollution.

Keywords:
Particulate Matter; Premature Mortality; Health Risk; Economic Burden

Resumen:

La exposición a la contaminación atmosférica aumenta la mortalidad y la morbilidad, lo que conduce a la discapacidad y a la muerte prematura. La contaminación del aire se identificó como una de las principales causas de la carga mundial de enfermedades, sobre todo en países de ingresos bajos y medianos en el 2015 (Global Burden of Diseases, Injuries and Risk Factors Study, 2015). Este artículo explora la relación entre las tasas de mortalidad y la concentración de material particulado (PM) en las 50 regiones españolas desde el 2002 hasta el 2017. Además, se realizó una estimación de las muertes prematuras provocadas por PM en España en términos de bienestar y pérdidas de producción en el 2017. Se desarrollaron modelos de efectos aleatorios para estudiar la relación entre las tasas de mortalidad y las concentraciones de PM. El costo económico de las muertes prematuras se evaluó usando el enfoque “disposición a pagar” para monetizar las pérdidas de bienestar y el método del capital humano para estimar las pérdidas de producción. Las concentraciones de PM₁₀ están positivamente asociadas con la mortalidad por enfermedades respiratorias y accidente cerebrovascular. Con base en 10.342 muertes prematuras en el 2017, las pérdidas en el bienestar social ascendieron a EUR 36.227 millones (3,1% del PIB español). El valor económico de las pérdidas de producción presentes y futuras llegó a EUR 229 millones (0,02% del PIB). Desde un punto de vista social, la contaminación del aire es un problema de salud pública que tiene un gran impacto en la salud y en la calidad de vida. Los resultados ponen de manifiesto la necesidad de implementar o de fortalecer políticas públicas regulatorias, fiscales y de salud para obtener beneficios sustanciales para la salud con la reducción de la exposición.

Palabras-clave:
Material Particulado; Mortalidad Prematura; Riesgo a la Salud; Dificultades Económicas

Resumo:

A exposição à poluição do ar ambiente aumenta a mortalidade e a morbidade, levando a incapacidades e mortes prematuras. A poluição do ar foi identificada como uma das principais causas da carga global de doenças, principalmente em países de baixa e média renda em 2015 (Global Burden of Diseases, Injuries and Risk Factors Study, 2015). Este artigo explora a relação entre as taxas de mortalidade e a concentração de material particulado (PM) nas 50 regiões espanholas de 2002 a 2017. Além disso, foi realizada uma estimativa das mortes prematuras causadas por PM na Espanha em termos de bem-estar e perdas de produção em 2017. Modelos de efeitos aleatórios foram desenvolvidos para estudar a relação entre as taxas de mortalidade e as concentrações de PMP. O custo econômico das mortes prematuras foi avaliado usando a abordagem “disposição a pagar” para monetizar as perdas de bem-estar e o método do capital humano para estimar as perdas de produção. As concentrações de PM₁₀ estão positivamente associadas à mortalidade por doenças respiratórias e acidente vascular cerebral. Com base em 10.342 mortes prematuras em 2017, as perdas no bem-estar social subiram para EUR 36,227 bilhões (3,1% do PIB espanhol). O valor econômico das perdas de produção presentes e futuras atingiu os EUR 229 milhões (0,02% do PIB). Do ponto de vista social, a poluição do ar é um problema de saúde pública que tem grande impacto na saúde e na qualidade de vida. Os resultados evidenciam a necessidade de implementar ou fortalecer políticas públicas regulatórias, fiscais e de saúde para obter benefícios substanciais à saúde com a redução da exposição.

Palavras-chave:
Material Particulado; Mortalidade Prematura; Risco à Saúde Humana; Dificuldade Econômica

Introduction

Air pollution was identified as a leading cause of global disease burden, especially in low- and middle-income countries in 2015 ¹1. Institute for Health Metrics and Evaluation. Global Burden of Disease Collaborative Network (GRD 2015) risk factor results 1990-2015. Seattle: Institute for health Metrics and Evaluation; 2016.. The World Health Organization (WHO) estimates that around 4.2 million deaths were attributable to outdoor air pollution in 2016 ²2. World Health Organization. Burden of disease from ambient air pollution for 2016. https://www.who.int/airpollution/data/AAP_BoD_results_May2018_final.pdf (accessed on 11/Feb/2020).
https://www.who.int/airpollution/data/AA... . In Europe, estimates suggest that air pollution constitutes both a major cause of premature death and disease and the most important environmental and modifiable health risk ³3. Prüss-Üstün A, Wolf J, Corvalán C, Bos R, Neira M. Preventing disease through healthy environments: a global assessment of the burden of disease from environmental risks. Geneva: World Health Organization; 2016.^,44. GBD 2016 Risk Factors Collaborators. Global, regional, and national comparative risk assessment of 84 behavioural, environmental and occupational, and metabolic risks or clusters of risks, 1990-2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet 2017; 390:1345-422.. A vast body of literature confirms that air pollution significantly impacts the population’s health, with considerable economic effects due to increased premature mortality and morbidity ⁵5. European Environment Agency. Air quality in Europe - 2018 report. https://www.eea.europa.eu/publications/air-quality-in-europe-2018 (accessed on 04/Feb/2020).
https://www.eea.europa.eu/publications/a... ^,66. Newby DE, Mannucci PM, Tell GS, Baccarelli AA, Brook RD, Donaldson K, et al. Expert position paper on air pollution and cardiovascular disease. Eur Heart J 2015; 36:83-93b..

One of the main objectives of this study is to offer evidence of the adverse effects of air pollution on the population’s health. Among the different types of pollutants, this study focuses on the impacts of particulate matter (PM) exposure. The main reasons for choosing this pollutant are the following: firstly, in comparison with gaseous co-pollutants, clinical studies have shown that PM presents a higher impact on health ⁷7. Hamanaka RB, Mutlu GM. Particulate matter air pollution: effects on the cardiovascular system. Front Endocrinol (Lausanne) 2018; 9:680.. Secondly, regardless of the PM concentration, the literature suggests no safe level of exposure that would avoid affecting people’s health ⁸8. Manisalidis I, Stavropoulou D, Stavropoulos A, Bezirtzogtou E. Environmental and health impacts of air pollution: a review. Front Public Health 2020; 8:14.^,99. Anderson JO, Thundiyil JG, Stolbach A. Clearing the air: a review of the effects of particulate matter air pollution on human health. J Med Toxicol 2012; 8:166-75.^,1010. Ostro B. Outdoor air pollution: assessing the environmental burden of disease at national and local levels. http://www.who.int/quantifying_ehimpacts/publications/ebd5/en/ (accessed on 06/Feb/2020).
http://www.who.int/quantifying_ehimpacts... ^,1111. Colette A, Schucht S, Ciarelli G, Létinois L, Meleux F. Long-term air quality trends in Europe. Fine particulate matter (PM2.5) health impacts. Bilthoven: European Topic Centre on Air Pollution and Climate Change Mitigation; 2018. (ETC/ACM Technical Paper, 2017/4)..

A wide range of epidemiological studies support the assumption that PM exposure negatively affects human health. Mortality rates are often used to study the health impacts on populations subject to PM exposure ¹²12. Dockery DW, Pope 3rd CA, Xu X, Spengler JD, Ware JH, Fay ME, et al. An association between air pollution and mortality in six U.S. cities. N Engl J Med 1993; 329:1753-9.^,1313. Pope 3rd CA, Burnett RT, Thurston GD, Thun MJ, Calle EE, Krewski D, et al. Cardiovascular mortality and long-term exposure to particulate air pollution: epidemiological evidence of general pathophysiological pathways of disease. Circulation 2004; 109:71-7.^,1414. Jerrett M, Burnett RT, Ma R, Pope 3rd CA, Krewski D, Newbold KB. et al. Spatial analysis of air pollution and mortality in Los Angeles. Epidemiology 2005; 16:727-36.^,1515. Lepeule J, Laden F, Dockery D, Schwartz J. Chronic exposure to fine particles and mortality: an extended follow-up of the Harvard Six Cities study from 1974 to 2009. Environ Health Perspect 2012; 120:965-70.^,1616. Beelen R, Raaschou-Nielsen O, Stafoggia M, Andersen ZJ, Weinmayr G, Hoffmann B, et al. Effects of long-term exposure to air pollution on natural-cause mortality: an analysis of 22 European cohorts within the multicentre ESCAPE project. Lancet 2014; 383:785-95.. The effects of PM exposure are well documented for cardiovascular and respiratory system ¹³13. Pope 3rd CA, Burnett RT, Thurston GD, Thun MJ, Calle EE, Krewski D, et al. Cardiovascular mortality and long-term exposure to particulate air pollution: epidemiological evidence of general pathophysiological pathways of disease. Circulation 2004; 109:71-7.^,1717. Hoek G, Krishnan RM, Beelen R, Peters A, Ostro B, Brunekreef B, et al. Long-term air pollution exposure and cardio-respiratory mortality: a review. Environ Health 2013; 12:43.^,1818. Cesaroni G, Forastiere F, Stafoggia M, Andersen ZJ, Badaloni C, Beelen R, et al. Long term exposure to ambient air pollution and incidence of acute coronary events: prospective cohort study and meta-analysis in 11 European cohorts from the ESCAPE Project. BMJ 2014; 348:f7412.^,1919. Puett RC, Hart JE, Yanosky JD, Paciorek C, Schwartz J, Suh H, et al. Chronic fine and coarse particulate exposure, mortality, and coronary heart disease in the Nurses' Health Study. Environ Health Perspect 2009; 117:1697-701.^,2020. Nawrot TS, Perez L, Künzli N, Munters E, Nemery B. Public health importance of triggers of myocardial infarction: a comparative risk assessment. Lancet 2011; 377:732-40.^,2121. Shah ASV, Langrish JP, Nair H, McAllister DA, Hunter AL, Donaldson K. Global association of air pollution and heart failure: a systematic review and meta-analysis. Lancet 2013; 382:1039-48.^,2222. Atkinson RW, Carey IM, Kent AJ, van Staa TP, Anderson HR, Cook DG. Long-term exposure to outdoor air pollution and incidence of cardiovascular diseases. Epidemiology 2013; 24:44-53.^,2323. Combes A, Franchineau G. Fine particle environmental pollution and cardiovascular diseases. Metabolism 2019; 100 Suppl:153944.^,2424. Miller M, Newby D. Air pollution and cardiovascular disease: car sick. Cardiovasc Res 2020; 116:279-94.. New evidence also suggests a link between PM exposure, the cerebrovascular system, neurodevelopment, cognitive function, and metabolic diseases such as obesity and diabetes mellitus, which themselves configure risk factors for cardiovascular diseases ⁶6. Newby DE, Mannucci PM, Tell GS, Baccarelli AA, Brook RD, Donaldson K, et al. Expert position paper on air pollution and cardiovascular disease. Eur Heart J 2015; 36:83-93b.^,2525. World Health Organization. Burden of disease from ambient air pollution for 2012 - summary of results. http://www.who.int/phe/health_topics/outdoorair/databases/AAP_BoD_results_March2014.pdf (accessed on 05/Feb/2020).
http://www.who.int/phe/health_topics/out... .

In comparison to cardiovascular and respiratory diseases, the evidence linking stroke and PM exposure is less robust, as is the knowledge about the mechanisms underlying this relation. Nevertheless, an increasing number of epidemiological studies have found evidence in favor of the contribution of air pollution to stroke mortality. Maheswaran & Elliott ²⁶26. Maheswaran R, Elliott P. Stroke mortality associated with living near main roads in England and Wales: a geographical study. Stroke 2003; 34:2776-80. found that living near main roads is associated with a significant but small excessive risk of mortality from stroke. Stafoggia et al. ²⁷27. Stafoggia M, Cesaroni G, Peters A, Andersen ZJ, Badaloni C, Beelen R, et al. Long-term exposure to ambient air pollution and incidence of cerebrovascular events: results from 11 European cohorts within the ESCAPE project. Environ Health Perspect 2014; 122:919-25. conducted a multi-country study and reported that overall stroke incidence increased by 19% per 5μg/m³ increase in PM_2.5. Zhang et al. ²⁸28. Zhang R, Liu G, Jiang Y, Li G, Pan Y, Wang Y, et al. Acute effects of particular air pollution on ischemic stroke and hemorrhagic stroke mortality. Front Neurol 2018; 9:827. found a similar result for PM₁₀: stroke mortality increased by 49% per 10μg/m³ increases in PM₁₀ concentrations. Nevertheless, other studies have found no association between PM exposure and stroke ²⁹29. Villeneuve PJ, Chen L, Stieb D, Rowe BH. Associations between outdoor air pollution and emergency department visits for stroke in Edmonton, Canada. Eur J Epidemiol 2006; 21:689-700.^,3030. Henrotin JB, Besancenot JP, Bejot Y, Giroud M. Short-term effects of ozone air pollution on ischaemic stroke occurrence: a case-crossover analysis from a 10-year population-based study in Dijon, France. Occup Environ Med 2007; 64:439-45..

Estimates for 2010 showed that chronic obstructive pulmonary disease, acute lower respiratory illness, and lung cancer related to PM_2.5 concentrations, causing around 765,000 premature deaths worldwide ³¹31. Lelieveld J, Evans JS, Fnais M, Giannadaki D, Pozzer A. The contribution of outdoor air pollution sources to premature mortality on a global scale. Nature 2015; 525:367-71.. Medical studies have shown that PM exposure results in pulmonary oxidative stress and inflammation, which is associated with the development of asthma and chronic obstructive pulmonary disease ⁷7. Hamanaka RB, Mutlu GM. Particulate matter air pollution: effects on the cardiovascular system. Front Endocrinol (Lausanne) 2018; 9:680.^,3232. Zanobetti A, Schwartz J. The effect of fine and coarse particulate air pollution on mortality: a national analysis. Environ Health Perspect 2009; 117:898-903.^,3333. Ostro B, Broadwin R, Green S, Geng W-Y, Lipsett M. Fine particulate air pollution and mortality in nine California counties: results from CALFINE. Environ Health Perspect 2006; 114:29-33.^,3434. Hales S, Blakely T, Woodward A. Air pollution and mortality in New Zealand: cohort study. J Epidemiol Community Health 2012; 66:468-73..

Among the epidemiological studies on the association between PM exposure and mortality in Spain, Cárdaba et al. ³⁵35. Cárdaba M, Muñoz MF, Medina AA, Capitán MA, Vaquer FC, Gómez AA. Health impact assessment of air pollution in Valladolid, Spain. BMJ Open 2014; 4:e005999. estimated the burden of mortality from exposure to PM₁₀ and PM_2.5 in the municipality of Valladolid (Spain), pointing out a detrimental effect on mortality from exposure to these pollutants. Alonso et al. ³⁶36. Alonso Fuscel E, Martínez Rueda T, Cambra Contín K, Lopez Carrasco L, Boldo Pascua E, Zorrilla Torras B, et al. Evaluación en cinco ciudades españolas del impacto en salud de la contaminación atmosférica por partículas: proyecto europeo APHEIS. Rev Esp Salud Pública 2005; 79:297-308. studied the relation between PM₁₀ exposure and mortality (respiratory and cardiovascular diseases and all causes of death) in five Spanish municipalities from 2000 to 2003. In the long term, the number of total attributable deaths per year related to exposure over the 20µg/m³ limit amounted to 68 per 100,000 inhabitants.

The European Union (EU) and Spanish public authorities have recently applied regulatory policies that positively affected the task of improving air quality. This is the case of the Clean Air Package, proposed by the EU and adopted in 2013. Statistics have shown a 12% reduction in premature deaths from environmental PM from 2005 to 2010 but these reductions were relatively modest, and the authors believe that the EU should undertake more efforts in this field, including, e.g., reducing limit values for PM to the recommendations of the WHO guidelines ³⁷37. Miller TR. Variations between countries in values of statistical life. Journal of Transport Economics and Policy 2000; 34:169-88..

The Spanish government has also applied sectoral measures in line with energy and climate change national policies (e.g., reducing emissions associated with coal-fired power generation, encouraging energy efficiency and renewable energy innovations, transport mobility measures, among others). The National Atmospheric Pollution Control Programme 2019-2020 is the latest initiative in this field. Madrid, the capital of Spain, implemented the Air Quality and Climate Change Strategy (approved in 2017). Recent studies show the positive health impacts of reducing PM_2.5 and NO₂ regarding deaths from all foreseeable causes due to long-term exposure, with important health benefits related to that regional strategy ³⁸38. Izquierdo R, García Dos Santos S, Borge R, Paz D, Sarigiannis D, Gotti A, et al. Health impact assessment by the implementation of Madrid City air-quality plan in 2020. Environ Res 2020; 183:109021.. Fiscal instruments have also shown the efficacy of reducing the quantity of pollutants or improving the incorporation of clean technologies, for example, in industries (the same takes place for subsidies to reduce emissions). Nevertheless, Spain has remained relatively distant from the use of these instruments ³⁹39. Gago A, Labandeira X, Picos F, Rodríguez M. Environmental taxes in Spain: a missed opportunity. Atlanta: Andrew Young School of Policy Studies, Georgia State University; 2006. (International Studies Program, Working Paper 06-06).^,4040. Labandeira X, Labeaga JM, López-Otero X. New green tax reforms: ex-ante assessments for Spain. Sustainability 2019; 11:5640..

This study aims to analyze the relation between PM exposure and mortality rates due to chronic respiratory diseases; tracheal, bronchus, and lung cancer; stroke; and diabetes mellitus from 2001 to 2016 in Spain and to estimate the economic impact of premature death due to PM in Spain according to welfare and production losses in 2017, the most recent year for which data are available.

Data and methods

Health data

The burden attributable to PM was estimated using the criteria specified by the Global Burden of Disease (GBD) risk factors for the following diseases: chronic respiratory diseases; tracheal, bronchus, and lung cancer; stroke; and diabetes mellitus ¹. The GBD study estimated the burden of diseases attributable to 79 risk factors (including environmental air pollution) in 195 countries and territories from 1990 to 2018. Data is divided by age and sex.

Since a higher risk among exposed populations translates into a higher proportion of deaths attributed to air pollution, the three values of relative risk (upper, base, and lower) were applied to the number of deaths in Spain over the considered period. The data used to estimate premature deaths were obtained from the microdata in the death statistics according to cause of death provided by the Spanish National Statistics Institute ⁴¹41. Instituto Nacional de Estadística. Death statistics according to cause of death. https://www.ine.es/dyngs/INEbase/en/operacion.htm?c=Estadistica_C&cid=1254736176780&menu=ultiDatos&idp=1254735573175 (accessed on 18/Mar/2020).
https://www.ine.es/dyngs/INEbase/en/oper... . Deaths are registered following the 10th revision of the International Classification of Diseases (ICD-10). Data were divided by the following codes: J00-J99 (chronic respiratory diseases); C33-C34 (tracheal, bronchus, and lung cancer); I60-I69 (stroke); and E10-E14 (diabetes mellitus). Data provide annual information about the underlying cause of death and victims’ sex, age, and place of residence.

Figures 1 and 2 show the death rates by respiratory diseases and stroke in the Spanish regions in 2016, respectively. As can be observed, the eastern-Mediterranean regions show the lowest incidence of deaths from respiratory diseases. This is probably attributable to their drier and warmer Mediterranean climate. As for stroke deaths, some north-south regional patterns are also observed, with a higher incidence in the northern regions of Spain.

Pollutant data

According to the Spanish Ministry of Ecological Transition and Demographic Challenge, concentrations of PM are based on measurements performed by monitoring stations at fixed sampling points that are classified following two criteria: (a) type of area: urban, suburban, and rural stations; and (b) source of pollutants: traffic-oriented, industrial, and background stations ⁴²42. Ministerio para la Transición Ecológica y el Reto Demográfico. Statistic of air quality. https://www.miteco.gob.es/es/calidad-y-evaluacion-ambiental/temas/atmosfera-y-calidad-del-aire/calidad-del-aire/evaluacion-datos/datos/Datos_2001_2017.aspx (accessed on 14/Feb/2020).
https://www.miteco.gob.es/es/calidad-y-e... .

Covering all the Spanish territory, 600 stations evaluate its air quality. Although the number of monitoring stations and the places in which monitoring stations are located vary among Spanish regions, air quality is measured equally by each station. By region, the mean concentrations of PM_2.5 and PM₁₀ pollutants were obtained using the values reported by all monitoring stations every hour for the 365 days of the year. The information collected covers the period 2001-2016 according to data availability.

Focusing on the pollutants with the strongest evidence of effects on human health (PM_2.5 and PM₁₀), the Spanish territory is divided into several geographical zones. The division into zones considers upper and lower threshold values. This method ensures equivalence in air quality evaluations independently of the considered territorial scope, which could be considered a strength of the data used in this analysis.

The evolution of the percentage of zones in which PM_2.5 and PM₁₀ are equal or below the limits set by the EU Ambient Air Quality Directive is shown in Figure 3. This graph also plots the evolution of the mean concentration of both pollutants over the same years. Regarding PM_2.5 concentrations, established data are available only from 2009 onward. From 2010 onward, most Spanish regions reported concentrations below the annual limit values of PM_2.5 and PM₁₀. We found a 4.4μg/m³ decrease in annual mean concentrations of PM_2.5 between 2009 and 2016. Spain had always shown high levels of PM₁₀ and a significant part of it stems from natural sources, especially African air masses. The evolution in the annual mean concentration for PM₁₀ shows a similar decreasing trend, from 30.2μg/m³ in 2006 to 18.9μg/m³ in 2016. From 2014 to 2015, we found a slight increase in the mean concentration of this pollutant, especially derived from urban zones.

Regional particle concentration in 2016 in Spain is shown in Figure 4. Geographically, a higher concentration of polluting particles can be observed in Madrid and Barcelona, corresponding to the cities with greater development (higher GDP - gross domestic product), population, and other factors (such as the number of registered vehicles). The lowest rates of environmental pollution were found in the western regions bordering Portugal.

Control variables

Since our geographical delimitation consists of 50 Spanish regions, different control variables that can influence our outcome of interest were considered. These variables refer to regional development (proxied by GDP), ageing of the local population (proxied by the average age of their inhabitants), and available health resources (proxied by the number of doctors per inhabitant). Competencies in health in Spain are regionally devolved, producing differences in the provision and use of resources between regions. These variables were therefore included in our explanatory model. Data were obtained from the Spanish National Statistics Institute.

Econometric approach

To evaluate the relation between mortality rates and PM concentrations, pooled regressions and random-effects models were employed. Pooled regression consists of a standard ordinary least squares (OLS) regression without any cross-sectional or time effects. This kind of estimation is used to derive unbiased and consistent estimates of parameters even in the presence of time-constant attributes. This analysis combines cross-sectional data on the 50 Spanish regions over 16 years to determine its final model. Moreover, the Breusch-Pagan Lagrange multiplier test (LM) for random-effects was applied rejecting OLS regression. The general model has the following basic functional form:

in which i and t represent region and years, respectively; y _it , the mortality rate for a specific cause of death; x’ _it , a vector encompassing the explanatory variables; α, individual specific random effects; u _i , a group-specific random element; and e _it , a residual error. PM_2.5 and PM₁₀ concentrations (measured in μg/m³) are the variables of interest. A time trend is also included as a dummy variable to control for changes that affect our sample over time (for example, climate effects). Regression models with unbalanced two-way error component disturbances were used to avoid the missing data on pollutants affecting the robustness of our estimations.

Health data times series were complete for all periods but some missing values were missing for pollution data. Nevertheless, missing data failed to affect the robustness of our estimates since regression models with unbalanced two-way error component disturbances were used. Thus, our matrix gives variable structures for the incomplete data model with random effects.

Cost estimation methods

Considering the health impacts of air pollution exposure described in the introduction, the economic impacts of premature deaths were assessed by two different perspectives: the Human Capital (HC) approach to assess losses in labor productivity and the Willingness to Pay Approach (WTP) to evaluate the economic costs of premature mortality according to wellbeing ⁴³43. Jones-Lee MW. The value of life: an economic analysis. Chicago: University of Chicago; 1976.^,4444. Banzhaf HS. The value of statistical life: a meta-analysis of meta-analyses. Journal of Benefit-Cost Analysis 2022; 13:182-97.^,4545. Kouakou CRC, Poder TG. Willingness to pay for a quality-adjusted life year: a systematic review with meta-regression. Eur J Health Econ 2022; 23:277-99..

The WTP approach has become the standard method in high-income countries for evaluating mortality risks associated with air pollution ⁴⁶46. Organisation for Economic Co-Operation and Development. Valuing mortality risk reductions in regulatory analysis of environmental, health and transport policies: policy implications. Paris: Organisation for Economic Co-Operation and Development; 2011.^,4747. The World Bank; The Institute for Health Metrics and Evaluation. The cost of air pollution: strengthening the economic case for action. http://documents.worldbank.org/curated/en/781521473177013155/pdf/108141-REVISED-Cost-of-PollutionWebCORRECTEDfile.pdf (accessed on 20/Jan/2020).
http://documents.worldbank.org/curated/e... ^,4848. Roy R, Braathen N. The rising cost of ambient air pollution thus far in the 21st Century: results from the BRIICS and the OECD Countries. Paris: Organisation for Economic Co-operation and Development; 2017. (OECD Environment Working Papers, 124).. The approach is based on estimating mortality risks attributable to pollutant exposure, estimated based on the relative risks of death from such exposures and the prevalence of exposure in the populations under study ⁴⁶46. Organisation for Economic Co-Operation and Development. Valuing mortality risk reductions in regulatory analysis of environmental, health and transport policies: policy implications. Paris: Organisation for Economic Co-Operation and Development; 2011.. The Value of a Statistical Life (VSL) includes the full economic costs of premature mortality, consisting of two parts: the first of which (and the most important) refers to intangible losses and the other, to the consumption lost to premature deaths (material losses) ⁴³43. Jones-Lee MW. The value of life: an economic analysis. Chicago: University of Chicago; 1976..

The VSL concept is equivalent to the rate at which individuals are willing to exchange income and the risk of death:

in which MRSI indicates the marginal rate of substitution between income (reduced by the amount ΔWTP) and the risk of death (reduced by the amount ΔP) of individual i. The average of individual MRS provides the VSL:

The specific VSL used in this study was estimated by a benefit-transfer approach that assumed a VSL base value of USD 3.615 million (2005 prices) ¹¹11. Colette A, Schucht S, Ciarelli G, Létinois L, Meleux F. Long-term air quality trends in Europe. Fine particulate matter (PM2.5) health impacts. Bilthoven: European Topic Centre on Air Pollution and Climate Change Mitigation; 2018. (ETC/ACM Technical Paper, 2017/4). for the EU27. To estimate the VSL for Spain, the following expression was used:

in which VSL _Spain-2017 refers to the VSL for Spain in 2017; VSL _EU27 , the average base VSL estimate in 2005 for the EU27 countries; Y _Spain , the Spanish real GDP per capita according to purchasing power parity (PPP) in 2005; Y _EU27 , the average real GDP per capita at PPP in 2005 of the EU27 countries; PPP ₂₀₀₅ , the purchasing power parity-adjusted exchange rate in 2005 (EUR/USD); (1 + %ΔP _2005-2017 ), the inflation adjustment with the consumer price index of Spain between 2005 and 2017; and (1 + %ΔY _2005-2017 ), the income adjustment with growth in real GDP per capita in Spain from 2005 to 2017. According to an Organisation for Economic Co-operation and Development (OECD) study ⁴⁶46. Organisation for Economic Co-Operation and Development. Valuing mortality risk reductions in regulatory analysis of environmental, health and transport policies: policy implications. Paris: Organisation for Economic Co-Operation and Development; 2011., the income elasticity of VSL was assumed to equal 0.8 (central value). For estimates, data on prices, income, and PPP rates were obtained from the World Bank database ⁴⁹49. The World Bank. DataBank. https://databank.worldbank.org/home.aspx (accessed on 31/Mar/2020).
https://databank.worldbank.org/home.aspx... .

Using this formula, the VSL for Spain in 2017 was estimated at EUR 3.724 million (2017 prices). A sensitive analysis was then performed using the upper and lower values of relative risk. A reduction in life expectancy results in current and future production losses, which must be entirely incorporated into the year of death (2017 in our case). Different approaches can evaluate losses of production deriving from a premature dead or morbidity. The Theory of Human Capital is the most widely used approach placing a value on productivity costs in literature ⁵⁰50. Becker GS. Human capital. New York: Columbia University Press; 1964.^,5151. Grossman M. The demand for health: a theoretical and empirical investigation. New York: Columbia University Press; 1972.^,5252. Grossman M. The human capital model of the demand for health. In: Culyer AJ, Newhouse JP, editors. Handbook of health economics. Amsterdam: Elsevier; 2000. p. 348-408..

To simulate current and future flows for lost work-related income due to premature deaths caused by PM pollution, the following steps were followed:

1) Estimate the expected number of fatalities caused by PM pollution for each cause of death. Applying risk mortality rates, the number of deaths by gender, age, and region of residence were estimated ⁵³53. Instituto Nacional de Estadística. Economically active population survey. https://www.ine.es/dyngs/inebase/en/operacion.htm?c=Estadistica_C&cid=1254736176918&menu=ultiDatos&idp=1254735976595 (accessed on 18/Mar/2020).
https://www.ine.es/dyngs/inebase/en/oper... .

2) Quantify loss of production:

(a) A transition matrix for those who die by age intervals and gender were obtained over our time horizon until retirement age (if individuals survived).

(b) The estimated number of individuals in each gender and age interval is multiplied by average earnings, employment, and survival rates in Spanish statistics. The employment rate is obtained by culling data from the Economically Active Population Survey⁵³53. Instituto Nacional de Estadística. Economically active population survey. https://www.ine.es/dyngs/inebase/en/operacion.htm?c=Estadistica_C&cid=1254736176918&menu=ultiDatos&idp=1254735976595 (accessed on 18/Mar/2020).
https://www.ine.es/dyngs/inebase/en/oper... ^,5454. Instituto Nacional de Estadística. Annual wage structure survey. https://www.ine.es/dyngs/INEbase/es/operacion.htm?c=Estadistica_C&cid=1254736177025&menu=ultiDatos&idp=1254735976596 (accessed on 18/Mar/2020).
https://www.ine.es/dyngs/INEbase/es/oper... ^,5555. Instituto Nacional de Estadística. Life tables. https://www.ine.es/dyngs/INEbase/en/operacion.htm?c=Estadistica_C&cid=1254736177004&menu=ultiDatos&idp=1254735573002 (accessed on 18/Mar/2020).
https://www.ine.es/dyngs/INEbase/en/oper... .

(c) Productivity and discount rates are applied to the resulting matrix. We assumed a 1% increase in productivity - annual average growth in labor productivity from 2000 to 2017 in Spain (OECD, 2018 ⁵⁶56. Organisation for Economic Co-Operation and Development. Growth in GDP per capita, productivity and ULC. https://stats.oecd.org/Index.aspx?DataSetCode=PDB_GR# (accessed on 20/Mar/2020).
https://stats.oecd.org/Index.aspx?DataSe... ) - and a discount rate of 3% (Pinto & Sánchez ⁵⁷57. Pinto JL, Sánchez FI. Métodos para la evaluación económica de nuevas prestaciones. Madrid: Ministerio de Sanidad y Consumo; 2003.) to obtain the current value of future incomes.

(d) The current value of all cumulative gains obtained over our time horizon for each age interval is then aggregated.

Results

Regression results

Table 1 provides our descriptive statistics for the set of variables in our regression analysis. Of the chosen causes of death, respiratory diseases (diabetes) show the highest (lowest) average mortality rates over our set of regions and years, followed by stroke. Respecting average PM concentrations, no pollutant exceeded the recommended EU concentration levels.

Table 2 shows the results for our regressions on regional mortality rates due to the chosen diseases. As regressions show, mortality rates positively relate only to PM₁₀ concentrations, which are statistically significant for respiratory diseases and stroke (random-effects models). However, we found no significance for deaths due to stroke in our random effects estimates, which prohibits us to establish the consistency of the relation between variables. No studied pollutants show statistically significant relations with mortality due to diabetes and tracheal, bronchus, or lung cancer. Pir interpretation of their coefficients shows a small effect. Thus, if PM₁₀ concentrations increase by 10%, mortality rates due to respiratory diseases and stroke increase by 0.004 and 0.01 (per 100,000 inhabitants), respectively. R² was highly relevant in the six goodness-of-fit models we developed, showing lower values in diabetes models.

These results agree with those found in Alonso et al. ³⁶36. Alonso Fuscel E, Martínez Rueda T, Cambra Contín K, Lopez Carrasco L, Boldo Pascua E, Zorrilla Torras B, et al. Evaluación en cinco ciudades españolas del impacto en salud de la contaminación atmosférica por partículas: proyecto europeo APHEIS. Rev Esp Salud Pública 2005; 79:297-308., who analyzed the health impact of particulate air pollution in five Spanish cities. Specifically, they found an impact due to exposure to PM₁₀ on mortality from respiratory diseases and cardiac causes (0.7/100,000 and 0.4/100,000 people), based on the fraction of mortality attributable to pollution.

Economic impact

Table 3 shows the estimated cost of premature deaths due to PM in Spain during 2017, the latest year for which data is available (10,342 fatalities in the base scenario). For this year, the base case shows an estimated cost of EUR 36,226 million (3.1% of Spanish GDP) - an average annual cost of EUR 779 per inhabitant. Mortality due to respiratory diseases accounts for 53.4% of all welfare losses, followed by stroke (16.6%). We established a lower limit at EUR 11,515 million (1% of GDP) and the upper one at EUR 63,057 million (5.4% of GDP).

As expected, annual labor income losses from premature mortality are lower than total welfare losses. Table 4 shows the results obtained for labor productivity losses due to premature mortality. The base case (1% annual growth rate of productivity and a 3% annual discount rate) shows an estimated cost of EUR 229 million, equivalent to nearly 0.02% of the Spanish GDP. Production losses related to mortality due to tracheal, bronchus, and lung cancer represent more than 45% of forgone income.

We established a lower limit at EUR 194 million (0.017% GDP), estimating it by considering a 1% increment in productivity and adopting a 5% discount rate. We set an upper limit at 309.8 million euros, following a 0% discount rate (0.027% of GDP).

Discussion

Relevant epidemiological studies have shown strong evidence confirming the association between PM exposure and the risk of suffering respiratory and cardiovascular diseases, diabetes, lung cancer, among others. The WHO stated that air pollution configures a major risk for several diseases, leading to disabilities and premature deaths, including heart diseases and stroke as the most common reasons for premature death attributable to air pollution ²⁵25. World Health Organization. Burden of disease from ambient air pollution for 2012 - summary of results. http://www.who.int/phe/health_topics/outdoorair/databases/AAP_BoD_results_March2014.pdf (accessed on 05/Feb/2020).
http://www.who.int/phe/health_topics/out... . Evidence also suggests that the road transport sector now constitutes the leading cause of air pollution-related deaths in Europe ⁴⁸48. Roy R, Braathen N. The rising cost of ambient air pollution thus far in the 21st Century: results from the BRIICS and the OECD Countries. Paris: Organisation for Economic Co-operation and Development; 2017. (OECD Environment Working Papers, 124)..

This study shows evidence on the health impacts of exposure to PM and estimated the cost of these impacts according to premature deaths. Using mortality data due to those diseases with a high relative risk from air pollution, this study first examined the relation between mortality rates and PM concentrations in 50 Spanish regions. Our models also incorporate other variables the literature recognizes as determinants of health, such as income, population aging, and a proxy of healthcare resource availability. Analyses shows results that agree with previous studies. Mortality due to respiratory diseases and stroke positively relate only to PM₁₀ concentrations. Our results show no significant relation between PM and deaths due to diabetes or tracheal, bronchus, and lung cancer.

We assessed the economic cost of premature deaths using two approaches. Based on the WTP approach, our economic estimation was based on 10,342 premature deaths in Spain during 2017, resulting in a cost of around EUR 36,227 million (base case), a figure representing 3.1% of the Spanish GDP. This result hints at the dimensions of its impact if we compare it to labor productivity losses. Using the human capital method, the economic value of current and future production losses by premature deaths reaches EUR 229 million (0.02% of Spanish GDP).

Other studies found similar results. The WHO and OECD estimated the cost to society of PM for the countries in the WHO European Region in 2010 using the WTP methodology ⁵⁸58. World Health Organization; Organisation for Economic Co-Operation and Development. Economic cost of the health impact of air pollution in Europe: clean air, health and wealth. Copenhagen: WHO Regional Office for Europe; 2015.^,5959. World Health Organization. Ambient air pollution: a global assessment of exposure and burden of disease. https://apps.who.int/iris/bitstream/handle/10665/250141/9789241511353-eng.pdf?sequence=1 (accessed on 12/Dec/2019).
https://apps.who.int/iris/bitstream/hand... . Focusing on Spain, this study estimated 14,042 premature deaths due to ambient air pollution. Estimates have suggested that the economic cost of premature deaths for Spain in 2010 total 42,951 million dollars, about 2.8% of its GDP (3.1% in our case). Roy & Braathen ⁴⁸48. Roy R, Braathen N. The rising cost of ambient air pollution thus far in the 21st Century: results from the BRIICS and the OECD Countries. Paris: Organisation for Economic Co-operation and Development; 2017. (OECD Environment Working Papers, 124)., based on the epidemiological data in the Global Burden of Disease Study, have estimated the incidence of premature deaths due to environmental air pollution and the economic costs of these mortalities.

The World Bank and Institute for Health Metrics and Evaluation (IHME, United States) ⁴⁷47. The World Bank; The Institute for Health Metrics and Evaluation. The cost of air pollution: strengthening the economic case for action. http://documents.worldbank.org/curated/en/781521473177013155/pdf/108141-REVISED-Cost-of-PollutionWebCORRECTEDfile.pdf (accessed on 20/Jan/2020).
http://documents.worldbank.org/curated/e... estimate the economic costs of fatal health risks using the same evaluating method (WTP). Their estimates for Spain show that, in 2013, air pollution caused more than USD 49,331 million (2011 prices) in welfare losses, about 3.4% of Spanish GDP (3.1% in our case). Below total welfare losses, annual production losses from premature mortality are estimated to total USD 1,051 million (2011 prices), about 0.01% of its GDP (0.02% in our case).

To our knowledge, this is the first study to combine time with a regional cross-sectional dimension to evaluate the relation between mortality rates and PM concentrations, estimating the economic impact of premature deaths due to PM exposure.

Result extrapolation must consider a series of limitations. Regarding accuracy and availability, no PM concentration data is available before 2009 for the PM_2.5 pollutant. Moreover, variation in daily air pollutant levels often relates to weather conditions affecting pollutant dispersion and models fail to control these effects. Finally, the World Bank and IHME argue that ground monitoring is insufficient to provide a global coverage to estimate PM exposure that and satellite-based measurements may complement well areas without ground-level monitoring ⁴⁷47. The World Bank; The Institute for Health Metrics and Evaluation. The cost of air pollution: strengthening the economic case for action. http://documents.worldbank.org/curated/en/781521473177013155/pdf/108141-REVISED-Cost-of-PollutionWebCORRECTEDfile.pdf (accessed on 20/Jan/2020).
http://documents.worldbank.org/curated/e... . The use of indicators from monitoring stations offers an indirect way of measuring the population’s exposure to pollution. This limitation is common to this type of study, which aggregately analyzes the effects of risk factors from pollution impacts on health status.

We would have liked to evaluate individual-level data to adjust our analyses by other factors (modifiers) that could affect individual mortality. However, this information is unavailable in the microdata from the death statistics according to cause of death (which would offer us homogeneous regional information for the whole country).

Other limitations relate to the methods we used to quantify premature deaths. We would have preferred to use a specific VSL for air pollution in Spain ⁵⁵55. Instituto Nacional de Estadística. Life tables. https://www.ine.es/dyngs/INEbase/en/operacion.htm?c=Estadistica_C&cid=1254736177004&menu=ultiDatos&idp=1254735573002 (accessed on 18/Mar/2020).
https://www.ine.es/dyngs/INEbase/en/oper... . Secondly, the main critiques to the Human Capital approach refer to its possible overestimation of production losses. Total production loss will depend on the time companies spend re-establishing initial production level (friction period). Nevertheless, the Friction Cost method is under a great theoretical and empirical controversy because it contradicts some of the axioms in economic theory and the complexity of calculations. Despite these weaknesses, human capital remains one of the most commonly used methods to estimate premature death costs.

From a social perspective, air pollution is a public health concern that greatly impacts health and quality of life. Thus, exposure to air pollution should be a key public health priority for governments. Our results highlight the need to implement or strengthen different types of public policies to reduce air pollution and their effects on individuals’ physical health and wellbeing ⁶⁰60. Sofia D, Gioliella F, Lotrecchiano N, Giuliano A. Mitigation strategies for reducing air pollution. Environ Sci Pollut Res Int 2020; 27:19226-35.^,6161. Schneider R, Masselot P, Vicedo-Cabrera AM, Sera F, Blangiardo, Forlani C, et al. Differntial impact of government lockdown policies on reducing air pollution levels and related mortality in Europe. Sci Rep 2022; 12:726.. As an initial consideration and independently of the type of policies public authorities adopt, further quantitative information on the benefits of reducing exposure to these pollutants is needed.

Acknowledgments

The authors would like to thank the reviewers for their comments. They find these recommendations extremely useful, having undoubtedly helped to significantly improve this study.

References

Publication Dates

History