Health impact assessment of decreases in PM₁₀ and ozone concentrations in the Mexico City Metropolitan Area. A basis for a new air quality management program

Evaluación de impacto en salud ante reducciones de PM₁₀ y ozono en la Zona Metropolitana del Valle de México. Base para un nuevo programa de calidad del aire

Horacio Riojas-Rodríguez, MSc, PhD,⁽¹⁾ Urinda Álamo-Hernández, MSc,⁽¹⁾José Luis Texcalac-Sangrador, MSc,⁽¹⁾ Isabelle Romieu, MD, MPH, ScD.⁽¹⁾

(1) Instituto Nacional de Salud Pública. Cuernavaca, Morelos, México.

Corresponding author

Abstract

Objective. To conduct a health impact assessment (HIA) to quantify health benefits for several PM and O₃ air pollution reduction scenarios in the Mexico City Metropolitan Area (MCMA). Results from this HIA will contribute to the scientific support of the MCMA air quality management plan (PROAIRE) for the period 2011-2020.

Materials and methods. The HIA methodology consisted of four steps: 1) selection of the air pollution reduction scenarios, 2) identification of the at-risk population and health outcomes for the 2005 baseline scenario, 3) selection of concentration-response functions and 4) estimation of health impacts.

Results. Reductions of PM₁₀ levels to 20 μg/m³ and O₃ levels to 0.050ppm (98 µg/m³) would prevent 2300 and 400 annual deaths respectively. The greatest health impact was seen in the over-65 age group and in mortality due to cardiopulmonary and cardiovascular disease.

Conclusion. Improved air quality in the MCMA could provide significant health benefits through focusing interventions by exposure zones.

Key words: air pollution; health impact assessment; Mexico City.

Resumen

Material y métodos. La metodología de EIS comprende cuatro pasos: 1) selección de los escenarios de reducción, 2) identificación de la población en riesgo y de los eventos en salud para el año basal 2005, 3) selección de las funciones de concentración-respuesta y 4) estimación del impacto en la salud.

Resultados. Reducciones de PM₁₀ a 20μg/m³ y de O₃ a 0.050ppm (98 µg/m³) evitarían, respectivamente, cerca de 2 300 y 400 muertes por año. El mayor impacto se observa en el grupo de más de 65 años y en la mortalidad por causas cardiopulmonares y cardiovasculares.

Conclusiones. Mejorar la calidad del aire en la ZMVM podría reflejar importantes beneficios para la salud focalizados por zonas o áreas de exposición.

Palabras clave: contaminación del aire; evaluación del impacto en la salud; Ciudad de México.

A large amount of scientific literature documents the different health effects due to exposure to atmospheric pollutants.^1-3 The results of research conducted worldwide serves to support local governments and international organizations in implementing public policies protective of human health.⁴

Ozone (O₃) and particulate matter (PM₁₀ and PM_2.5) have received special attention because of the health effects they pose to exposed populations^5,6 and because they are the two criteria pollutants whose concentrations have remained elevated over time in the Mexico City Metropolitan Area (MCMA). Acute and chronic exposures to these pollutants are associated with increased mortality caused by cardiovascular and respiratory disease and increased morbidity from cardiovascular and respiratory disease,^1,6-8 certain types of cancer,^9,10 and reproductive, developmental and neurological effects.^2,11 Exposure to air pollutants during pregnancy and early stages of life is associated with premature birth, delayed intrauterine growth, low birth weight, early death syndrome and infant mortality.^12,13 The groups most susceptible to the health effects of air pollution are children, adults over 65 years of age, persons with chronic diseases such as asthma, chronic obstructive pulmonary disease, and cardiac ischemia, and pregnant women because of prenatal exposure of the fetus.^3,14

A variety of studies have documented the relationship between decreased exposure to air pollution and decreases in population mortality and morbidity as well as increases in life expectancy.^15,16 Findings from epidemiological studies have supported the implementation of health protective policies among populations exposed to air pollution.^4,17

Over the past 20 years, the health impact assessment (HIA) has served as a methodological tool used by decision-makers in diverse countries, to quantify the impact of interventions on air pollution and human health.^18-20 It has been used in several studies throughout the world, particularly in Europe and the United States,^19-23 and its use has enabled the estimation of economic and health impacts of diverse air quality management measures.^8,24-27

This work documents the results of MCMA of HIA from three different reduction scenarios for PM₁₀ and O₃ concentrations; and was conducted according to the research methods recommended by the World Health Organization (WHO) and the United States Environmental Protection Agency (US-EPA), using the best calculations available to date, according to the literature reviewed.

We selected this methodology because it has been developed and validated in similar scenarios. This methodology is the standard approach for linking epidemiolgical evidence with new air quality standards in Europe and the United States.⁴⁹

Materials and methods

The four steps of an HIA include: 1) selection of the air pollution reduction scenarios, 2) identification of the at-risk population and health outcomes for the baseline scenario, 3) selection of concentration–response functions (CRF) and 4) estimation of health impacts.^18,19,21

Study area

The study area was defined by crossing two spatial layers of information (figure 1) using a Geographic Information System (GIS). The first layer represented PM₁₀ and O₃ monitoring stations administered by the Atmospheric Monitoring System (SIMAT, by its acronym in Spanish) of the Mexico City Ministry of Environment (SMA-GDF, by its acronym in Spanish). The second layer was the boundary of Mexico City. A grid of 500m² resolution was overlaid on the two layers in GIS. As some monitoring stations were located beyond the boundaries of Mexico City and in the neighboring State of Mexico, the final study area included 33 municipalities in the State of Mexico in their entirety.

Population and cartographic information was obtained from the National Statistics and Geography Institute (INEGI, by its acronym in Spanish) using data from the 2005 population census. The spatial information was processed and analyzed using ArcGIS Desktop software, release 9.3 and its Spatial Analyst module (Environmental Systems Research Institute, 2009).

Air pollution scenarios

Health effects were evaluated for three hypothetical PM₁₀ and O₃ pollution reduction scenarios. For annual average PM₁₀, the following concentrations were used: 1) the WHO recommendation of 20µg/m³, 2) the European Union recommendation of 40µg/m³, and 3) the Mexican Federal standard of 50µg/m³. For annual average O₃, the following concentrations were used: 1) 0.05 ppm, 2) 0.06 ppm and 3) a 5% reduction from the 2005 annual average.

Estimating population exposure to pollutants in the study area

PM₁₀ and O₃ dispersion was calculated based on the grid delimiting the study area and using inverse distance weighting (IDW) as a geostatistical calculation method.³⁵ Concentration gradients in the study area were classified using the ArcGIS manual classification method. The study area was classified into 5 zones for PM₁₀ and 4 zones for O₃ (figure 1).

Because the spatial distribution of air pollution does not align with the spatial distribution of the population, we used a population-weighted annual average concentration to estimate population exposure. This weighting approach aims to quantify population exposure by linking the concentration of a pollutant in a particular geographic area with the population residing in that same area.^36,37 To estimate the population-weighted annual average concentration, the GIS layer with concentration zones was overlaid with a layer containing municipal-level population data. The population-weighted annual average concentration for the study area was computed with the following formula:

where:

C_p= Population-weighted concentration for the study area

C_i= Concentration in municipality i

Ni= Number of persons in municipality i

The value obtained for the indicator represents an estimate of the average exposure of the population in each municipality.

At-risk population and health outcomes for the baseline scenario

This HIA uses health outcomes which have been consistently reported as having an association with PM₁₀ and O₃ exposure and for which the Mexican Health System reports statistics. Table I presents the population groups included in the study, the health outcomes selected for HIA estimations, and the corresponding International Classification of Diseases (ICD) 9 and 10 codes.

The year 2005 was chosen for the baseline scenario since it was the year for which the most recent, highest quality population and health data were available. Data for population, total mortality and cause-specific mortality were provided by Inegi. Data for hospital admissions were obtained from the National Health Information System (SINAIS, by its Spanish acronym).

Selection of concentration-response functions (CRF)

The CRFs used in this study correspond to the relative risks (RR), or a pooled RR, found in epidemiological studies that associate air pollution with health effects. We used CRFs from studies conducted in the MCMA, including the ESCALA project, and recent international epidemiological studies and meta-analyses, which provide reliable estimates (table I).^{31,33,34,38-47}

Estimation of health impacts

The benefits of the air pollution reduction scenarios are expressed in attributable, or preventable, cases per year for each health outcome evaluated (total mortality, mortality from respiratory disease, etc.). These are estimated using the attributable population fraction, which is a function of the CRFs and the fraction of exposed persons in the study population.²¹ In this context, attributable population fraction is defined as the reduction in incidence that would be observed if the population exposure were decreased to the levels proposed in the reduction scenarios. This approach assumes that air pollution is only a part of the causal model for each health outcome.

The formula for the attributable population fraction (AFpop) of the total population is:

where p_p represents the fraction of the population exposed to the environmental factor under consideration and RR is the concentration-response function for the change in exposure being evaluated. When the entire population has been exposed, as in the majority of air pollution studies, the p_p is equal to 1 and the above formula is simplified to the attributable fraction of those who have been exposed (AF_exp):

Multiplying the AF_exp by the total number of cases observed in the baseline scenario and using the different concentration scenarios, we can estimate the number of preventable cases as follows:

where:

• ∆conc is the change in PM₁₀ or O₃ concentration in each scenario,

• CRFcorrected is the CRF expressed as relative risk, per unit change in pollutant concentration,

• Number of cases in MCMA is the number of cases observed in 2005 for the outcome evaluated.

All of the stages described above present a series of assumptions and uncertainties associated with the following aspects: a) selection and frequency of health effects, b) the CRFs, c) the selection of air pollution indicators and d) the exposure distribution in the population. Therefore, the results obtained are only an approximation of what could be expected if air quality was improved. The discussion below describes the expected influence of each uncertainty factor. In addition, a sensitivity analysis was performed to evaluate uncertainty with regard to the CRFs.

Generally, a sensitivity analysis estimates results using different thresholds for the estimated point, or suggests different assumptions and examines the variation in the results. For this study, a sensitivity analysis was performed according to the method proposed by Kunzli and Perez,²¹ in which the effect of variability in CRFs (upper and lower 95% confidence interval limits) was evaluated. At least two CRF sources were used for the estimates whenever possible.

The project was evaluated and approved by research and ethics committees at the National Institute of Public Health (INSP, by its Spanish initials).

Results

The study area obtained by crossing information in the GIS included Mexico City and 33 municipalities representing 68.9% of the population of the State of Mexico. The total population for the study area in the year 2005 was 18 419 138 inhabitants.

The highest concentrations of PM₁₀ were found in the northern part of the study area, while the highest O₃ concentrations were found in the southeast. Figure 1 summarizes the annual average concentration for the study area. The municipalities of Jaltenco and Tlahuac had the highest annual average concentrations of PM₁₀ and O₃, 70.1 µg/m³ and 0.069 ppm, respectively. Based on these results, the following estimates were performed.

Evaluation of the impact on total and cause-specific mortality (short-term)

The estimates for preventable cases, or deaths, vary with the CRF used. Table II presents total preventable deaths per study zone for each pollution reduction scenario and CRF used. The results indicate that, using the CRF from the ESCALA study and a PM₁₀ concentration of 20 µg/m³, it would be possible to prevent approximately 2300 deaths per year in the entire study area; that is, 13 deaths per 100 000 inhabitants. If a reduction to 40µg/m³ was achieved roughly 1 040 deaths would be prevented per year, or 6 deaths per 100 000 inhabitants. If the Mexican standard of 50 µg/m³ was met, approximately 400 deaths would be prevented per year, or 2 per 100 000 inhabitants. Disaggregating by zone the most number of deaths would be prevented by reducing PM₁₀ concentrations in the center (Z2PM₁₀) and northwest zones (Z3PM₁₀).

The results indicate that the greatest benefit in mortality would be the group over 65 years of age. Using the baseline population of 982 876 for this age group and CRFs from the ESCALA project, approximately 1400 (95%CI:1 029-2 184) and 300 (95%CI:195-393) deaths would be avoided annually for the most extreme PM₁₀ and O₃ reduction scenarios. Using the same CRFs, achieving the Mexican standard for PM₁₀ and decreasing O₃ concentrations by 5% would achieve a reduction of 212 (95%CI 158-340) and 68 (95%CI:45- 091) annual deaths respectively.

Using the baseline population of 288 694 for children under 1 year of age in the study area, 265 (95%CI:55-468) deaths would be prevented by reducing PM₁₀ levels to 20 µg/m3 and 47 (95%CI:10-84) deaths would be avoided by complying with the Mexican PM₁₀ standard of 50 µg/m3.

Table III presents the estimates for mortality due to specific causes in the entire study area using the best available CRFs. The greatest benefit is seen for cardio-pulmonary and cardiovascular mortality.

Long-term estimates indicate that achieving the 20, 40 and 50 µg/m³ PM₁₀ scenarios would prevent roughly 10 500 (95%CI:6563-14282), 4,800 (95%CI:2984-6704) and 1 800 (95%CI:1836-1125) deaths, respectively, over a period of 15 years in the population over 30 years of age, using a baseline population of 8 105 538 persons of that age in the study area.

Evaluation of the impact on hospitalizations due to specific causes

In the Mexico City population, approximately 1 650 (95%CI:1415-1902), 660 (95%CI:563-760) and 150 (95%CI:129-175) hospitalizations due to respiratory disease could be prevented per year, for the PM₁₀ reduction scenarios of 20 µg/m3, 40 µg/m³, 50 µg/m³ respectively. For the same scenarios, 800 (95%CI:570-1 063), 320 (95%CI:225-421) and 70 (95%CI:51-97) hospitalizations due to cardiovascular disease could be prevented. The greatest decrease in hospitalizations would occur in the group over 65 years of age. In this age group, under the most extreme reduction scenario for PM₁₀, 360 (95%CI:292-427) hospitalizations due to respiratory disease and 700 (95%CI:548-864) hospitalizations due to cerebrovascular disease would be prevented.

Meanwhile, the results indicate that reducing the annual maximum moving average for O₃ to 0.05 ppm would prevent approximately 2000 annual (95%CI:249-3608) hospitalizations due to respiratory disease and 430 (95%CI:185-612) due to cardiovascular disease. Reducing the annual average O₃ concentration by 5% below baseline level would prevent approximately 440 (95%CI:54-809) annual hospitalizations due to respiratory disease and 90 (95%CI:40-133) hospitalizations due to cardiovascular disease.

Sensitivity analysis

The sensitivity analysis evaluates uncertainty in CRFs, specifically the effect of CRF variability on calculations and the effect of using CRFs other than those from the ESCALA project, which are considered to be more reliable for the study area. Figure 2 presents the percentage change in the central estimates for the different effects evaluated. The uncertainty margin associated with the 95% CI is between +/- 14% and +/- 95%, with an average of 51%. The greatest uncertainty corresponds to the estimate of preventable deaths from respiratory disease by reducing O₃ concentrations, while for total preventable deaths and deaths among persons over 65 years of age the uncertainty margin is less than 45%.

Discussion

According to the estimates based on CRFs from the ESCALA project, we can see that if PM₁₀ levels were to comply with the current limits established by Mexican standards, total mortality in the study area could be reduced by 0.5% (400 deaths, 2 per 100 000). With the strictest measures for reducing PM₁₀ concentrations to comply with current California EPA standards, greater benefits could be obtained and total annual mortality in the study area could be reduced by 3% (2 300 deaths, 13 per 100 000). Based on this scenario, the estimates for impact on long-term mortality (15 years) indicate that 10500 deaths could be avoided among persons over 30 years of age.

Although the benefit is less when reducing O₃ levels, estimates indicate that it would be possible to prevent approximately 400 deaths if the annual 8-hour moving average for O₃ were reduced to 0.05 ppm.

A comparative analysis of the results in each of the concentration zones indicates that the impact of reducing particulates is greater in center (Z2PM₁₀) and northwest (Z3PM₁₀) zones and the impact of reducing O₃ occurs primarily in center (Z2O₃) and north (Z1O₃) zones. It is important to remember that the specific calculations for the zones take into account the mortality rate, the exposed population and concentrations of PM₁₀ and O₃ weighted for the population. Therefore, a greater number of cases are found in zones with a larger exposed population, higher mortality rate and higher PM₁₀ or O₃ concentrations.

With regard to this last point, the annual average exposure to particulate matter varies among the 5 zones, with a difference of 20 micrograms per cubic meter between the zones with the greatest (southwest) and the lowest (northeast) concentrations. These differences occur primarily because a large amount of emissions are generated by mobile sources in the central MCMA region and by activity on unpaved roads in the periphery, while industrial emissions are generated in the northern Mexico City municipalities.⁴⁸ The construction material and iron industries generate roughly 400 tons of PM₁₀ per year in the municipality of Tlalnepantla alone. In Ecatepec, the most significant emissions are generated by metal smelting and forging, and in Acolman by electricity generation. All of these municipalities are located in the northern part of the MCMA. In addition, in the central and northern districts of Mexico City, PM₁₀ emissions can be as much as 80 tons per year, primarily from heavy vehicular activity, along with industrial activity in some cases. O₃, on the other hand, tends to be concentrated in the southern and southeastern zones. This is mainly because the dominant air stream transports ozone precursors and ozone, concentrating them in these zones.⁴⁸

The results show that improving air quality in the MCMA area would provide considerable health benefits. Nevertheless, interpretation of these results should take into account the assumptions and uncertainties associated with the data and methods used. The validity of the results depends mainly on the quality of population and health data, as well as PM₁₀ and O₃ concentrations and the CRFs used. In our case, with the exception of morbidity data, which may be underestimated, the information comes from reliable information and solid data sources.

Most of the assumptions and uncertainties may have resulted in an underestimation of benefits. This underestimation could be due restricting the analysis to O₃ and PM₁₀ without considering effects of and interactions with other pollutants, such as PM_2.5, NO₂, SO₂, COVs, polycyclic aromatic compounds and CO. In Mexico, the standard for O₃ establishes an annual limit of 0.08 ppm, which should not be exceeded more than 5 times per year as an 8-hour moving average. Nevertheless, using only an annual average for ozone could have underestimated the impacts on health since the average does not reflect that the limit of 0.08 ppm was exceeded on more than 59% of the days of the year in 2005.⁴⁸ The annual average for ozone was chosen for this study, rather than the established standard, so as to not overestimate health impact, especially for zones that are further away from air monitoring stations.

It is also likely that benefits were underestimated by not evaluating effects such as: changes in pulmonary function; lung cancer; school absenteeism; days with restricted activities; visits to emergency rooms; effects on morbidity from certain illnesses such as chronic bronchitis, asthma or lung cancer; and effects on the reproductive system such as complications in pregnancy, low birth weight, premature birth and delayed cognitive development.

The CRF is one of the most important components for calculating health impacts. The sensitivity analysis indicates that, depending on the health outcome under evaluation, a 95% CI for CRFs results in a margin of error of 14 to 95%. Another uncertainty associated with the CRFs is their applicability to different populations and locations. Therefore, for most of the calculations, preference was given to those that included CRFs generated in the MCMA, and especially those estimated in the ESCALA project, since it included data from 1997 to 2005. Further, the ESCALA project incorporated the most recently developed statistical methods which yield the most accurate and reliable calculations for time-series analyses. Using these CRFs reduces the uncertainty since epidemiological and statistical methods have improved in recent years, allowing for the extrapolation of results and the use of indicators in other populations, as is the case of Mexico City. The sensitivity analysis also evaluated the use of alternate CRFs, finding that in most scenarios the central estimates were underestimated. This finding indicates that the benefit would be expected to be greater when other CRFs are used for the estimate. More complex sensitivity analyses using probability models to include other uncertainties were not developed since they require additional data that is not available for the study area.

Finally, an economic evaluation of the health effects presented may be complementary to this study. For mortality, we have avoided assigning costs to expected mortality because we consider determining the rate and number of events to be sufficient in terms of benefits. For morbidity, the study could be complemented by assigning a value to each hospitalization event in order to obtain and estimate of the preventable costs in the health sector.

In conclusion, air pollution continues to have an impact on public health in the MCMA and the results suggest that reducing the current levels of pollutants such as ozone and PM₁₀ would prevent hundreds of deaths and hospitalizations. The spatial distribution of concentrations in the metropolitan area enables identifying and implementing focused interventions in the geographic area, specifically by concentration zones. Even with the limitations and uncertainties in the calculations performed, the HIA is a useful tool for the design of public policies. It is recommended that existing standards be reviewed and adjusted in the context of current scientific evidence and international guidelines. It is also recommended that HIAs be performed immediately for other pollutants such as PM_2.5 and for other cities in the country which may have air monitoring networks.

Acknowledgements

This study had the financial support of the Mexico City Government (Instituto de Ciencia y Tecnología del Distrito Federal). The authors thank Irma Soyachi, Alan Castillo Ferraez and Ana Elizabeth Sandoval Loredo for their assistance in the preparation of this document.

References

1. Samet J, Krewski D. Health effects associated with exposure to ambient air pollution. J Toxicol Environ Health Part A 2007;70:227-242.         [ Links ]

2. Curtis L, Rea W, Smith-Willis P, Fenyves E, Pan Y. Adverse health effects of outdoor air pollutants. Environ Int 2006;32:815-830.         [ Links ]

3. Brunekreef B, Holgate ST. Air pollution and health. Lancet 2002;360:1233-1242.         [ Links ]

4. WHO. Air quality guidelines global update 2005. Germany: World Health Organization, 2006.         [ Links ]

5. 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;8:166-175.         [ Links ]

6. Jerrett M, Burnett RT, Pope CA 3rd, Ito K, Thurston G, Krewski D, et al. Long-term ozone exposure and mortality. N Engl J Med 2009;360:1085-1095.         [ Links ]

7. Katsouyanni K, Samet JM, Anderson HR, Atkinson R, Le Tertre A, Medina S, et al. Air pollution and health: a European and North American approach (APHENA). Res Rep Health Eff Inst 2009;5-90.         [ Links ]

8. Analitis A, Katsouyanni K, Dimakopoulou K, Samoli E, Nikoloulopoulos AK, Petasakis Y, et al. Short-term effects of ambient particles on cardiovascular and respiratory mortality. Epidemiol 2006;17:230-233.         [ Links ]

9. Pope CA 3rd, Burnett RT, Thun MJ, Calle EE, Krewski D, Ito K, et al. Lung cancer, cardiopulmonary mortality, and long-term exposure to fine particulate air pollution. JAMA 2002;287:1132-1141.         [ Links ]

10. Krewski D, Burnett R, Jerrett M, Pope CA, Rainham D, Calle E, et al. Mortality and long-term exposure to ambient air pollution: ongoing analyses based on the American Cancer Society cohort. J Toxicol Environ Health Part A 2005;68:1093-1109.         [ Links ]

11. Gallus S, Negri E, Boffetta P, McLaughlin JK, Bosetti C, La Vecchia C. European studies on long-term exposure to ambient particulate matter and lung cancer. Eur J Cancer Prev 2008;17:191-194.         [ Links ]

12. Lacasaña M, Esplugues A, Ballester F. Exposure to ambient air pollution and prenatal and early childhood health effects. Eur J Epidemiol 2005;20:183-199.         [ Links ]

13. Wigle DT, Arbuckle TE, Walker M, Wade MG, Liu S, Krewski D. Environmental hazards: evidence for effects on child health. J Toxicol Environ Health Part B 2007;10:3-39.         [ Links ]

14. Kampa M, Castanas E. Human health effects of air pollution. Environ Pollut 2008;151:362-367.         [ Links ]

15. Leksell I, Rabl A. Air pollution and mortality: quantification and valuation of years of life lost. Risk Anal 2001;21:843-857.         [ Links ]

16. Pope CA 3rd, Ezzati M, Dockery DW. Fine-particulate air pollution and life expectancy in the United States. N Engl J Med 2009;360:376-386.         [ Links ]

17. APHEIS. Health impact assessment of air pollution and communication strategy: third-year report. Saint-Maurice Cedex: Institut de Veille Sanitaire, 2005.         [ Links ]

18. Harris-Roxas B, Harris E. Differing forms, differing purposes: a typology of health impact assessment. Environmental Impact Assessment Review 2011;31:396-403.         [ Links ]

19. Boldo E, Linares C, Lumbreras J, Borge R, Narros A, García-Pérez J, et al. Health impact assessment of a reduction in ambient PM(2.5) levels in Spain. Environ Int 2011;37:342–8.         [ Links ]

20. Baccini M, Biggeri A, Grillo P, Consonni D, Bertazzi PA. Health impact assessment of fine particle pollution at the regional level. Am J Epidemiol 2011; 174:1396-1405.         [ Links ]

21. Pérez L, Sunyer J, Künzli N. Estimating the health and economic benefits associated with reducing air pollution in the Barcelona metropolitan area (Spain). Gac Sanit 2009;23:287–94.         [ Links ]

22. Remy S, Nawrot T, Fierens F, Petit P, Vanderstraeten P, Nemery B, et al. Health impact of urban air pollution in Belgium. Air Qual Atmos Health 2011;4:243–6.         [ Links ]

23. Boldo E, Medina S, Le Tertre A, Hurley F, Mücke H-G, Ballester F, et al. Apheis: Health impact assessment of long-term exposure to PM(2.5) in 23 European cities. Eur J Epidemiol 2006;21:449-458.         [ Links ]

24. Jusot JF, Lefranc A, Cassadou S, D'Helf-Blanchard M, Eilstein D, Chardon B, et al. Estimating mortality attribuable to PM₁₀ particles in 9 French cities participating in the European programme Apheis. Sante Publique 2006;18:71-84.         [ Links ]

25. Yorifuji T, Yamamoto E, Tsuda T, Kawakami N. Health impact assessment of particulate matter in Tokyo, Japan. Arch Environ Occup Health 2005;60:179-185.         [ Links ]

26. Forsberg B, Hansson H-C, Johansson C, Areskoug H, Persson K, Järvholm B. Comparative health impact assessment of local and regional particulate air pollutants in Scandinavia. Ambio 2005;34:11-19.         [ Links ]

27. Corbett S. The art of the possible: experience and practice in health impact assessment in New South Wales. N S W Public Health Bull 2005;16:116-118.         [ Links ]

28. Mckinley G, Zuk M, Höjer M, Avalos M, González I, Iniestra R, et al. Quantification of local and global benefits from air pollution control in Mexico City. Environ Sci Technol 2005;39:1954-1961.         [ Links ]

29. Bell ML, Davis DL, Gouveia N, Borja-Aburto VH, Cifuentes LA. The avoidable health effects of air pollution in three Latin American cities: Santiago, São Paulo, and Mexico City. Environ Res 2006;100:431-440.         [ Links ]

30. Molina L. Air quality in the Mexico Megacity: an integrated assessment. United States of America: Kluwer Academic Publishers, 2002;vol.2:384.         [ Links ]

31. Herman C, Borja-Aburto VH, Kees D, Muñoz-Cruz R, Brander L, Cropper M, et al. Improving air quality in metropolitan Mexico City: an economic valuation. México: The World Bank, Comision Ambiental Metropolitana, 2002;62.         [ Links ]

32. Borja-Aburto VH, Rosales-Castillo JA, Torres-Meza VM, Corey G, Olaiz-Fernandez G. Evaluation of health effects of pollution. Ancillary benefits and costs of greenhouse gas mitigation. France: OECD, 2000: 592.         [ Links ]

33. Carbajal-Arroyo L, Miranda-Soberanis V, Medina-Ramón M, Rojas-Bracho L, Tzintzun G, Solís-Gutiérrez P, et al. Effect of PM(10) and O(3) on infant mortality among residents in the Mexico City Metropolitan Area: a case-crossover analysis, 1997-2005. J Epidemiol Commu Health 2011;65:715-721.         [ Links ]

34. Romieu I, Gouveia N, Cifuentes LA, de León AP, Junger W, Vera J, et al. Multicity study of air pollution and mortality in Latin America (the ESCALA study). Res Rep Health Eff Inst, 2012;171:5-86.         [ Links ]

35. Lloyd CD. Local models for spatial analysis. United States of America: CRC Press, 2006;vol.1:244.         [ Links ]

36. Ivy D, Mulholland JA, Russell AG. Development of ambient air quality population-weighted metrics for use in time-series health studies. J Air Waste Manag Assoc 2008;58:711-720.         [ Links ]

37. Zuk M, Tzintzun-Cervantes G, Rojas-Bracho L. Tercer almanaque de datos y tendencias de la calidad del aire en nueve ciudades mexicanas. México: INE-SEMARNAT, 2007; 116.         [ Links ]

38. Anderson HR, Atkinson RW, Peacock JL, Marston L, Konstantinou K. Meta-analysis of time-series studies and panel studies of particulate matter (PM) and ozone (O3). Copenhagen: World Health Organization Regional Office for Europe, 2004.         [ Links ]

39. O'Neill MS, Bell ML, Ranjit N, Cifuentes LA, Loomis D, Gouveia N, et al. Air pollution and mortality in Latin America: the role of education. Epidemiology 2008;19:810-819.         [ Links ]

40. Künzli N, Kaiser R, Medina S, Studnicka M, Chanel O, Filliger P, et al. Public-health impact of outdoor and traffic-related air pollution: a European assessment. Lancet 2000;356:795-801.         [ Links ]

41. Rosales-Castillo JA, Torres-Meza VM, Olaiz-Fernández G, Borja-Aburto VH. Acute effects of air pollution on health: evidence from epidemiological studies. Salud Publica Mex 2001;43:544-555.         [ Links ]

42. McKinley G, Zuk M, Hojer M, Avalos M, González I, Hernández M, et al. The local benefits of global air pollution control in Mexico City. Final Report of the Second Phase of the Integrated Environmental Strategies Program in Mexico. México DF: Instituto Nacional de Ecología, Instituto Nacional de Salud Pública México, 2003.         [ Links ]

43. Abbey DE, Petersen F, Mills PK, Beeson WL. Long-term ambient concentrations of total suspended particulates, ozone, and sulfur dioxide and respiratory symptoms in a nonsmoking population. Arch Environ Health 1993;48:33-46.         [ Links ]

44. Ito K, De Leon SF, Lippmann M. Associations between ozone and daily mortality: analysis and meta-analysis. Epidemiol 2005;16:446-457.         [ Links ]

45. Levy JI, Chemerynski SM, Sarnat JA. Ozone exposure and mortality: an empiric bayes metaregression analysis. Epidemiol 2005;16:458-468.         [ Links ]

46. O'Neill MS, Loomis D, Borja-Aburto VH. Ozone, area social conditions, and mortality in Mexico City. Environ Res 2004;94:234-242.         [ Links ]

47. Bell ML, Dominici F, Samet JM. A meta-analysis of time-series studies of ozone and mortality with comparison to the national morbidity, mortality, and air pollution study. Epidemiol 2005;16:436-445.         [ Links ]

48. SMA-GDF. Inventario de emisiones de contaminantes criterio de la Zona Metropolitana del Valle de México 2006. Ciudad de México: SMA-GDF, 2008.         [ Links ]

49. Künzli N, Kaiser R, Medina S, Studnicka M, Chanel O, Filliger P, et al. Public-health impact of outdoor and traffic-related air pollution: a European assessment. Lancet 2000;356(9232):795-801.         [ Links ]

Received on: March 19, 2014
Accepted on: September 1, 2014

Corresponding author:
Dr. Isabelle Romieu.
Instituto Nacional de Salud Pública. Av. Universidad 655,
col. Santa María Ahuacatitlán. 62100 Cuernavaca, Morelos, México.
E-mail: iromieu@correo.insp.mx

Declaration of conflict of interests. The authors declare that they have no conflict of interests.