Available via license: CC BY 4.0
Content may be subject to copyright.
Meisner C, Gjorgjev D, Tozija F. Estimating health impacts and economic costs of air pollution in the Republic
of Macedonia (Original research). SEEJPH 2015, posted: 07 April 2015. DOI 10.12908/SEEJPH-2014-45
1
ORIGINAL RESEARCH
Estimating health impacts and economic costs of air pollution in the
Republic of Macedonia
Craig Meisner1, Dragan Gjorgjev2,3, Fimka Tozija2,3
1 The World Bank, Washington, DC, USA;
2 Institute of Public Health, Skopje, Republic of Macedonia
3 Medical Faculty, Skopje, Republic of Macedonia
Corresponding author: Craig Meisner, Senior Environmental Economist, The World Bank,
MSN MC7-720;
Address: 1818 H Street, NW, Washington, DC 20433, USA;
Telephone: 202-473-6852; E-mail: cmeisner@worldbank.org
Meisner C, Gjorgjev D, Tozija F. Estimating health impacts and economic costs of air pollution in the Republic
of Macedonia (Original research). SEEJPH 2015, posted: 07 April 2015. DOI 10.12908/SEEJPH-2014-45
2
Abstract
Aim: This paper assesses the magnitude of health impacts and economic costs of fine
particulate matter (PM) air pollution in the Republic of Macedonia.
Methods: Ambient PM10 and PM2.5 monitoring data were combined with population
characteristics and exposure-response functions to calculate the incidence of several health
end-points known to be highly influenced by air pollution. Health impacts were converted to
Disability-Adjusted Life Years (DALYs) and then translated into economic terms using three
valuation approaches to form lower and higher bounds: the (adjusted) Human Capital
Approach (HCA), Value of a Statistical Life (VSL) and the COI (cost of illness).
Results: Fine particulate matter frequently exceeds daily and annual limit values and
influences a person‟s day-to-day health and their ability to work. Converting lost years of life
and disabilities into DALYs - these health effects represent an annual economic cost of
approximately €253 million or 3.2% of GDP (midpoint estimate). Premature death accounts
for over 90% of the total health burden since this represents a loss of total life-long income. A
reduction of even 1μg/m3 in ambient PM10 or PM2.5 would imply 195 fewer deaths and
represent an economic savings of €34 million per year in reduced health costs.
Conclusion: Interventions that reduce ambient PM10 or PM2.5 have significant economic
savings in both the short and long run. Currently, these benefits (costs) are „hidden‟ due to
the lack of information linking air quality and health outcomes and translating this into
economic terms. Policymakers seeking ways to improve the public‟s health and lessen the
burden on the health system could focus on a narrow set of air pollution sources to achieve
these goals.
Keywords: air pollution, health and economic costs, particulate matter.
Conflicts of interest: None.
Acknowledgements: The authors would like to first acknowledge the financial support of the
Green Growth and Climate Change Analytic and Advisory Support Program launched in
2011, with funding support from the World Bank and the Governments of Norway and
Sweden. We would also like to thank our local Macedonian counterparts at the Institute of
Public Health and the Ministry of Environment and Physical Planning for their willingness to
collect and share data. We would also like to thank the Finnish Meteorological Institute
(FMI) for their guidance and suggestions on earlier drafts of this work. FMI is currently
working with the MoEPP in strengthening their air quality monitoring network through an
EU-sponsored Twinning Project.
Meisner C, Gjorgjev D, Tozija F. Estimating health impacts and economic costs of air pollution in the Republic
of Macedonia (Original research). SEEJPH 2015, posted: 07 April 2015. DOI 10.12908/SEEJPH-2014-45
3
Introduction
According to the Global Burden of Disease (2010) estimates (1), the crude mortality rate
from ambient particulate matter (PM) pollution in Macedonia was 80.6 deaths per 100,000 in
2010. In comparable neighboring states such as Serbia, it was 71.8 deaths per 100,000; in
Croatia it was 69.4 per 100,000; in Hungary 92.0 per 100,000; and 70 per 100,000 in
Slovakia. The total Disability-Adjusted Life Years (DALYs) attributable to PM were about
1,480 per 100,000 in Macedonia (but, up to 1,600 in Hungary) (1).
The main sources of this ambient condition were the use of solid fuel for heating households
in the winter, as well as the impact of industry and traffic. Uncontrolled urbanization is also a
significant source of particulate matter. In 2009, an average annual concentration of 90µg/m3
was registered in Skopje. Compounding the situation, poor air circulation is another reason
why the capital city of Skopje has one of the worse air conditions in winter.
Air pollution is also significant throughout the European region, with only nine of the 34
Member States reporting PM10 levels below the annual WHO air quality guideline (AQG) of
20μg/m3. Almost 83% of the population in these cities is exposed to PM10 levels exceeding
the AQG levels (2).
Results from a recent project Improving Knowledge and Communication for Decision-making
on Air Pollution and Health in Europe (Aphekom), which uses a traditional health impact
assessment method, indicated that average life expectancy in the most polluted cities could be
increased by approximately 20 months if long-term PM2.5 concentrations were reduced to
WHO guidelines (3). One recent study in Macedonia found that an increase of PM10 by
10μg/m3 above the daily maximum permitted level (50μg/m3) was associated with a 12%
increase in cardiovascular disease (2).
Methods
To estimate the health impacts and economic costs of air pollution, the approach required
overlaying data from multiple sources. The method used ambient air quality data
[information received from the Ministry of Environment and Physical Planning (MoEPP)]
for PM10 and PM2.5, health statistics – annual deaths by disease type; frequency of chronic
bronchitis, asthma, infant mortality; and health cost data (information received from the
Institute of Public Health and Health Insurance Fund), exposure-response functions from
health studies (information from international and local literature) and population
characteristics – age groups, gender, urban/rural population (information from the state
Statistics Bureau). These data were combined for a municipal (city) - level analysis.
The approach to estimating health impacts and economic costs encompassed five steps:
Collection of monitored, ambient concentration data on PM10 and PM2.5
Calculation of exposed population
Exposure-response functions
Calculation of physical health impacts (mortality, morbidity, DALYs)
Monetizing health impacts
Collection of monitored data on fine particulate matter
Currently, the Ministry of Environment and Physical Planning (MoEPP) has a network of 19
Automatic Monitoring Stations: seven in Skopje, two in Bitola, two in Veles and one in
Kicevo, Kumanovo, Kocani, Tetovo, Kavadarci, village Lazaropole, and two near the OKTA
oil refinery (near the villages of Miladinovci and Mrsevci). Stations measure SO2, NO2, CO,
PM10, PM2.5, ozone, benzene, toluene, ethyl benzene and BTX – although some stations do
not measure all pollutants [monitored PM2.5 measurements began in November, 2011 in
Meisner C, Gjorgjev D, Tozija F. Estimating health impacts and economic costs of air pollution in the Republic
of Macedonia (Original research). SEEJPH 2015, posted: 07 April 2015. DOI 10.12908/SEEJPH-2014-45
4
Karpos and Centar. In cases where PM2.5 is not actually monitored, observed PM10 is
adjusted by the ratio PM2.5/PM10. The ratio, based on recent observations, is estimated at
0.71 in the case of Macedonia; and is within ranges found in other international studies. See
Ostro (4) for a discussion]. This information is available electronically through their air
quality portal (available at: http://airquality.moepp.gov.mk/?lang=en).
Calculation of exposed population
Population information for 2010 was used focusing on the working population as well as
vulnerable segments of society (for example, those under the age of five or older than 65 are
considered more vulnerable to the effects of air pollution – that is more prone to develop
acute or chronic respiratory ailments).
Exposure-response functions
The selection of exposure-response functions was based on epidemiological research between
PM10 and PM2.5 and mortality and morbidity. For mortality, exposure-response functions for
long-term exposure to PM2.5 were (4):
Relative risks (RR) were calculated as:
Cardiopulmonary (CP) mortality: RR =[(X+1)/(X0 +1)]0.15515
Lung cancer (LC) mortality: RR = exp[0.23218 (X-X0)]
ALRI mortality in under-five children: RR = exp[0.00166 (X-X0)]
with: X = current annual average PM2.5 concentration for CP and LC among adults, and PM10
concentrations for ALRI among children and X0 = target or baseline PM2.5 concentration.
Information on the crude death rate (CDR), CP, LC and ALRI data were used to set the
mortality baseline. For morbidity, exposure-response coefficients (annual cases per 100,000
population) for PM10 from Ostro (4,5) and Abbey et al. (6) were applied. Ostro (4) reflects a
review of worldwide studies, and Abbey et al., (6) provides estimates of chronic bronchitis
associated with particulates (PM10).
A baseline for PM concentrations
A baseline level (natural background concentration) for PM2.5 = 7.5 µg/m3, as suggested by
Ostro (4), was used (some argue that the baseline should be set at zero since the literature
does not support the existence of a concentration level of which there are no observable
effects. However a baseline of zero is not realistic since natural background concentrations
hover between 10-15 μg/m3 in Macedonia – and one would only look at investments which
could reduce ambient concentrations to this level (i.e. at least from a benefit-cost standpoint
of weighing alternative investments).
Given a PM2.5/PM10 ratio of 0.71 using observations in Macedonia, the baseline level for
PM10 is 10.6 µg/m3. These baseline concentrations were applied to both large and
medium/small urban areas.
Calculation of physical health impacts (mortality, morbidity, DALYs)
Using the population information and the exposure-response functions, mortality and
morbidity impacts were calculated through the conversion of impacts to DALYs (DALYs =
sum of years of potential life lost due to premature mortality and the years of productive life
lost due to disability). The DALY method weights illnesses by severity: a mild illness or
disability (e.g. morbidity effects) represents a small fraction of a DALY and a severe illness
represents a larger fraction (e.g. mortality = 1 DALY). Weights used in this context were
adapted from Larsen (7) and are presented in Table 1.
Meisner C, Gjorgjev D, Tozija F. Estimating health impacts and economic costs of air pollution in the Republic
of Macedonia (Original research). SEEJPH 2015, posted: 07 April 2015. DOI 10.12908/SEEJPH-2014-45
5
Table 1. Estimated health impacts of air pollution, urban and rural, 2010
(Source: World Bank, 2012)
Health impacts
DALYs /10,000 cases
CP mortality (PM2.5)
80,000
LC mortality (PM2.5)
80,000
ALRI mortality (PM10)
340,000
Chronic bronchitis (PM10)
22,000
Hospital admissions (PM10)
160
Emergency room visits (PM10)
45
Restricted activity days (PM10)
3
Lower respiratory illness in children (PM10)
65
Respiratory symptoms (PM10)
0.75
Total
Monetizing health impacts
To create a set of bounds three alternative valuation approaches were used: the (adjusted)
Human Capital Approach (HCA) [the adjusted version avoids the issue of assigning a value
of zero to the lives of the retired and the disabled since the traditional approach is based on
foregone earnings. It avoids this issue by assigning the same value – per capita GDP – to a
year of life lost by all persons, regardless of age], Value of a Statistical Life (VSL) and the
COI (cost of illness). The HCA estimates the indirect cost of productivity loss through the
value of an individual‟s future earnings. Thus, one DALY corresponds to one person‟s
contribution to production, or GDP per capita. This method provides a realistic lower bound
for the loss of one DALY. The VSL measures the willingness-to-pay (WTP) to avoid death –
using actual behavior on the tradeoffs between risks and money. The VSL is calculated by
dividing the marginal WTP to reduce the risk of death by the size of the risk reduction.
Measured this way, the value of one DALY corresponds to the VSL divided by the number of
discounted years lost because of death. The VSL typically forms an upper bound measure of
health damages. The COI approach estimates the direct treatment costs associated to different
health end-points (e.g. hospitalization, restricted activity days, and doctor visits). Mortality
was valued using HCA as a lower bound and the VSL as an upper bound. For morbidity
effects the COI was estimated as a lower bound and willingness-to-pay to avoid a case of
illness was applied as a higher bound of cost (WTP was assumed to be two times the COI).
Results
Air quality data support the finding that particulate matter is one of the most serious concerns
in the country. Ambient PM10 concentrations frequently exceeded the EU standard of
40μg/m3 over the years (Figure 1).
Using information on ambient PM10 and PM2.5 in conjunction with the methods outlined
above, it is estimated that in Macedonia 1,350 deaths occur annually from cardiopulmonary
disease and lung cancer (Table 2). These deaths are considered „premature‟ in the sense that
air pollution contributed to their early demise – since many factors actually influence a
persons‟ lifespan (e.g. smoking, exposure to the outdoors, job, etc.). Particulate matter can
also influence a person‟s day-to-day health and their ability to work. In 2011, levels of PM10
and PM2.5 were primarily responsible for 485 new cases of chronic bronchitis, 770 hospital
admissions, and 15,200 emergency visits.
Meisner C, Gjorgjev D, Tozija F. Estimating health impacts and economic costs of air pollution in the Republic
of Macedonia (Original research). SEEJPH 2015, posted: 07 April 2015. DOI 10.12908/SEEJPH-2014-45
6
Figure 1. Annual average PM10 concentration at each automatic monitoring station in
μg/m3 (Source: Ministry of Environment and Physical Planning, 2012)
What do these translate to in terms of a total cost to society? Converting lost years of life and
disabilities to DALYs (or disability-adjusted life years) - these health effects represent an
annual economic cost of €253 million or 3.2% of GDP (Table 2). Note that premature death
accounts for over 90% of the total health cost since the loss of life is a loss of total (future)
income. People also suffer from the day-to-day consequences of respiratory diseases. It is
estimated that several thousand work-years are lost annually from chronic bronchitis,
asthma, hospital admissions and days of restricted activity.
These estimates are consistent with other recent studies – such as Kosovo where annual
deaths were estimated to be in the range of 805-861 from cardiovascular disease and lung
cancer (8). It should be noted that our estimates are mid-points (middle) with lower and
higher ranges reflecting different assumptions made on the PM2.5/PM10 ratio and the
population‟s exposure to air pollution.
What are the potential benefits of reducing particulate matter? If Macedonia were to lower
PM10 and PM2.5 to EU limit values this would avoid over 800 deaths and thousands of days in
lost productivity – representing a health cost savings of €151 million per year (Table 3). A
reduction of even 1μg/m3 in ambient PM10 and PM2.5 would result in 195 fewer deaths (1,648
fewer DALYs) and imply an economic savings of €34 million per year in reduced health
costs.
PM10 concentration (ug/m3)
Skopje
Bitola
Veles
Tetovo
Kumanovo
Kavadarci
Kocani
Kicevo
Rural
EU std
Meisner C, Gjorgjev D, Tozija F. Estimating health impacts and economic costs of air pollution in the Republic
of Macedonia (Original research). SEEJPH 2015, posted: 07 April 2015. DOI 10.12908/SEEJPH-2014-45
7
Table 2. Number of annual cases, DALYs per year and economic cost in million Euros,
2011 (Source: authors’ calculations)
Health impact
Annual
cases*
Total DALYs
per year
Annual economic
cost (€ million)
Cardiopulmonary & lung cancer mortality (PM2.5)
1,351
10,809
232.0
ALRI† mortality (PM10)
1
17
0.1
Chronic bronchitis (PM10)
485
1,066
3.0
Hospital admissions (PM10)
770
12
0.4
Emergency room visits (PM10)
15,200
68
0.9
Restricted activity days (PM10)
3,213,000
964
8.6
Lower respiratory illness in children (PM10)
22,400
146
1.5
Respiratory symptoms (PM10)
10,197,000
765
6.8
Total
13,847
253.3
* Mid-point estimates using a baseline for PM10 = 15 µg/m3 and PM2.5 = 7.5 µg/m3
† ALRI: Acute Lower Respiratory Infections.
Table 3. The potential health ‘savings’ associated with reductions in PM10 and PM2.5
(€ million) [Source: authors’ calculations]
Level of reduction in ambient
PM10 and PM2.5 (μg/m3)*
Reduced DALYs
Annual health savings
(€ million)
0
0
0.0
1
1,648
34.1
5
4,894
98.9
10
6,636
133.6
15
8,059
161.5
20
9,275
184.9
EU standards met†
7,840
151.5
* Example reductions were equally applied to both PM10 and PM2.5 at the same time.
† PM10 = 40 µg/m3 and PM2.5 = 20 µg/m3.
Discussion
There is significant evidence of the effects of short-term exposure to PM10 on respiratory
health, but for mortality, and especially as a consequence of long-term exposure, PM2.5 is a
more robust risk factor than the coarse part of PM10 (particles in the 2.5–10 μm range). All-
cause daily mortality is estimated to increase by 0.2 - 0.6% per 10 μg/m3 of PM10 (9).
Furthermore, it has been estimated that exposure to PM2.5 reduces life expectancy by about
8.6 months on average in the European Region. Results from the study “Improving
Knowledge and Communication for Decision-making on Air Pollution and Health in Europe”
(Aphekom), which uses traditional health impact assessment methods, indicates that average
life expectancy in the most polluted cities could increase by approximately 20 months if long-
term PM2.5 concentrations were reduced to WHO annual guidelines (10).
Monitored PM10 and PM2.5 concentrations have repeatedly exceeded EU standards in
Republic of Macedonia and have contributed to short-term and chronic respiratory disease.
This study estimated an annual (mid-point) loss of approximately 1,350 lives with thousands
of lost-productive days, indirectly costing the economy up wards of €253 million or
3.2% of GDP in 2011. The specific exposure-response functions used in this study were
Meisner C, Gjorgjev D, Tozija F. Estimating health impacts and economic costs of air pollution in the Republic
of Macedonia (Original research). SEEJPH 2015, posted: 07 April 2015. DOI 10.12908/SEEJPH-2014-45
8
borrowed from the international literature – however the orders of magnitude have been
shown to be robust in many developing country applications after adjusting for local
conditions (4,5,7,8).
From a policy standpoint, it is important to note that these estimated costs are generally
“hidden” since they are not normally quantified, and benchmarked to the value of economic
activity that generated the pollution (i.e. GDP). Likewise the distribution of this burden is
shared between the general public and the health care system – so total costs are not
transparent. The results should motivate policy makers to be more focused on preventative
measures, among them, local green options to reduce particulate matter including energy
efficiency, fuel switching and the adoption of cleaner technologies. The benefits from such
actions should find their way into the benefit-cost analysis of associated investments since the
health “savings” could offset the investment costs of greening interventions.
References
1. Institute for Health Metrics and Evaluation. Global Burden of Disease, 2010.
http://www.healthdata.org/search-gbd-data?s (accessed: February 2, 2015).
2. Kochubovski M, Kendrovski V. Monitoring of the ambient air quality (PM10) in
Skopje and evaluation of the health effects in 2010. JEPE 2012;13:789-96.
3. World Health Organization (WHO). WHO Air quality guidelines, particulate matter,
ozone, nitrogen dioxide and sulphur oxide; Geneva, Switzerland; 2006.
4. Ostro B. Outdoor Air Pollution - Assessing the environmental burden of disease at
national and local levels. Environmental Burden of Disease, Series No. 5, Geneva:
WHO; 2004 (62p).
5. Ostro B. Estimating the Health Effects of Air Pollution: A Method with an application
to Jakarta. Policy Research Working Paper No. 1301, Washington, D.C.: The World
Bank; 1994.
6. Abbey DE, Lebowitz MD, Mills PK, Petersen FF, Beeson WL, Burchette RJ. Long-
term ambient concentrations of particulates and oxidants and development of chronic
disease in a cohort of nonsmoking California residents. Inhal Toxicol 1995;7:19-34.
7. Larsen B. Colombia. Cost of Environmental Damage: A Socio-Economic and
Environmental Health Risk Assessment. Final Report prepared for the Ministry of
Environment, Housing and Land Development of Republic of Colombia; 2004.
8. World Bank. Kosovo Country Environmental Analysis: Cost Assessment of
Environmental Degradation, Institutional Review, and Public Environmental
Expenditure Review, Washington, DC. The World Bank; 2012.
http://documents.worldbank.org/curated/en/2013/01/17485553/kosovo-country-
environmental-analysis-kosovo-country-environmental-analysis-cea (accessed:
February 2, 2015).
9. Samoli E, Peng R, Ramsay T, Pipikou M, Touloumi G, Dominici F, et al. Acute
effects of ambient particulate matter on mortality in Europe and North America:
Results from the APHENA Study. Environ Health Perspect 2008;116:1480-6.
10. World Health Organization (WHO) – Regional office for Europe. Health effects of
particulate matter: Policy implications for countries in Eastern Europe, Caucasus and
Central Asia. Copenhagen, Denmark; 2013.
___________________________________________________________
© 2015 Meisner et al; This is an Open Access article distributed under the terms of the Creative Commons
Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution,
and reproduction in any medium, provided the original work is properly cited.