collaboration.worldbank.org · Web viewPrevalence rates of solid fuel use for cooking in Guatemala,...

Benefits and Costs of Cookstove Options for Household Air Pollution Control

Guatemala, Honduras and Nicaragua

Prepared for the World Bank

Bjorn Larsen1

Consultant

November 2016

1 Email: [email protected] Web: http://bjorn-larsen.com

http://bjorn-larsen.com/

mailto:[email protected]

Table of Content

Acronyms

Executive Summary

1. Household use of solid biomass fuels for cooking

2. Household exposure to PM2.5

3. Health effects and cost of household PM2.5

4. Pollution control interventions

5. Post-intervention PM2.5 exposures

6. Health benefits of interventions

7. Non-health benefits of interventions7.1. Fuel savings7.2. Cooking time savings

8. Cost of interventions8.1. Cost of stoves8.2. Cost of LPG fuel8.3. Cost of stove maintenance and stove promotion programs8.4. Annualized cost shares

9. Benefit-cost ratios9.1. Valuation of health benefits9.2. Benefits and costs9.3. Sensitivity analysis

10. Summary and conclusions

References

Annex 1. Health effects of particulate matter pollution

Annex 2. Valuation of health benefits

Acronyms

AF Attributable fractionALRI Acute lower respiratory infectionAQG Air quality guidelineCB Chronic bronchitisCBV Cerebrovascular diseaseCI Confidence IntervalCOPD Chronic obstructive pulmonary disease CP Cardiopulmonary diseaseDHS Demographic and Health SurveyGBD Global burden of diseaseGDP Gross domestic productHAP Household air pollutionICS Improved cookstoveIHD Ischemic heart diseaseLAC Latin America and the Caribbean LC Lung cancerLPG Liquefied petroleum gasµg/m3 microgram per cubic meterOR Odds ratioPM Particulate matterPPP Purchasing power parityRR Relative riskSD Standard deviationVSL Value of statistical lifeYLD Year lost to diseaseWHO World Health Organization

Executive SummaryThe use of solid fuels for cooking causes serious household air pollution (HAP). 2.9 million people died globally in 2015 from harmful exposure to PM2.5 emissions from cooking with solid fuels according to the Global Burden of Disease Project (Forouzanfar et al, 2016). This makes HAP one of the leading health risk factors in developing countries. 41% of households globally relied on solid fuels for cooking in 2010.

Over half of the population in Guatemala, Honduras and Nicaragua cook with solid fuels. Major health effects of HAP from theses fuels are estimated in this report for the three countries, and benefits and costs of two widely promoted cookstove interventions for control of HAP are assessed.

Solid fuel use

Prevalence rates of solid fuel use for cooking in Guatemala, Honduras and Nicaragua stood at 51-63% in 2012, only behind Haiti in Latin America and the Caribbean (LAC) region. The use of solid fuels declined at an annual rate of 1 percentage point in Nicaragua and 0.7 percentage points in Honduras from 1990 to 2012. In Guatemala, the proportion of the population using solid fuels has remained nearly the same (figure E1).

Figure E1. Use solid fuels as primary cooking fuels (% of population), 1990-2012

Nicaragua Honduras Guatemala0%

10%20%30%40%50%60%70%80%90% 76%

66% 62%54% 51%

63%

19902012

Source: Produced from data in World Bank (2016a).

In Honduras - the country for which most detailed data are available - the vast majority of the population cooks in the house, particularly among households cooking with electricity, LPG, and kerosene. Cooking in a separate building or outdoors is, however, quite prevalent among the population cooking with wood.

Less than 4% of the population in Honduras using solid fuels cooks with a closed stove with chimney. This is generally the most effective method of removing solid fuel use related smoke from the household environment. As many as 40% of the population using solid fuels cooks with an open stove with no chimney and 57% cook with an open stove with chimney (table E1). A study from Peru found that concentrations of PM2.5 were as high in households with a

chimney over an open fire as in households cooking on open fire without a chimney (Pollard et al, 2014).

Table E1. Type of cookstove among population cooking with solid fuels in Honduras (% of population), 2011-12

In houseIn separate

building Outdoors TotalOpen stove with no chimney 18.1% 9.8% 11.6% 39.6%Closed stove with chimney 2.8% 0.5% 0.2% 3.4%Closed stove with no chimney 0.1% 0.0% 0.0% 0.1%Open stove with chimney 42.4% 9.6% 4.8% 56.8%Total 63.4% 19.9% 16.7% 100.0%

Source: Produced from the Honduras DHS 2011-12 (SdS et al, 2013).

Household exposure to PM2.5

Measurement studies in five countries in LAC – including Guatemala, Honduras and Nicaragua - found average PM2.5 concentrations of 130-1020 µg/m3 in kitchens of households that used wood for cooking over open fire or unimproved stoves. The same studies found that average concentrations declined to 50-340 µg/m3 after installation of improved cookstoves, often with an attached chimney.

Several studies in LAC have also measured personal PM2.5 exposure of the person cooking - averaging 116-260 µg/m3 before and 58-100 µg/m3 after the installation of improved cookstoves. This is in stark contrast to WHO’s PM2.5 outdoor annual air quality guideline (AQG) of 10 µg/m3 and evidence of serious health effects at PM2.5 concentrations even below the AQG (Forouzanfar et al, 2016).

Cooking with solid fuels also affects the community. PM2.5 concentrations of about 80 µg/m3

were found at the outdoor patio of the dwellings, and about 60 µg/m3 at the community plaza in rural Mexico (Zuk et al, 2007). In Honduras, 8-hour daytime average PM2.5 concentrations of 215 µg/m3 were found outside dwellings using improved chimney stoves and 358 µg/m3 outside dwellings cooking on open fire or traditional stoves (Clark et al, 2010).

Other studies point to the role of the quality and maintenance of cookstoves and chimneys for effective reduction of PM2.5 in the household environment.

Health effects and costs of HAP

Over 11,000 deaths and 32 million days of disease are caused by HAP in Guatemala, Honduras and Nicaragua each year according to this report (table E2). This is estimated based on current prevalence rates of solid fuel use for cooking, and health risks of PM2.5 exposure as evidenced by the Global Burden of Disease Project (2015) (Forouzanfar et al, 2016).

Table E2. Estimated annual cases of morality and days of disease attributable to HAP, 2015Deaths Days lived with disease (million)

Guatemala 6,369 17.2Honduras 3,098 8.0Nicaragua 1,843 7.3

Total 11,310 32.5Source: Estimates by the author.

These health effects are monetized as a cost to society by using economic valuation methods. The cost of mortality is estimated by applying a “value of statistical life” or VSL to the number of deaths, using the methodology in World Bank (2016b). The cost of morbidity is estimated as 50% of average wage rates for each day of disease. The estimated cost of health effects totaled nearly US$ 2.9 billion in 2015, and is equivalent to 2.3 - 3.1% of GDP in the three countries (table E3).

Table E3. Annual cost of health effects of HAP, 2015Guatemala Honduras Nicaragua

GTQ billion

USD million

HNL billion

USD million

NIO billion

USD million

Cost of health effects 15.2 1,988 12.7 576 8.0 295

% of GDP in 2015 3.12% 3.12% 2.86% 2.86% 2.33% 2.33%Source: Estimates by the author.

Pollution control interventions

Two cookstove interventions for HAP control are assessed in this report: (i) improved biomass cookstoves (ICS), and (ii) LPG stoves. Several major health and non-health benefits, and private and public (promotion programs) costs are estimated (table E4).

Table E4. Interventions and associated benefits and costsInterventions Benefits CostsImproved biomass cookstoves (ICS)LPG stoves

Health improvementsBiomass fuel savingsCooking time savings

StoveStove maintenanceLPG fuelStove promotion programs

Expected reduction in household member exposure to PM2.5 is 30-40% for ICS and 50-75% from LPG, depending on cooking location. If the entire community adopts the interventions, expected reduction in exposure is increased to 40-50% for ICS and 65-88% for LPG.

Benefits and costs of interventions

Benefits. Health benefits account for 40-71% of total estimated benefits of interventions, depending on household cooking location. The remaining benefits are evenly split between fuelwood savings and cooking time savings. Health benefits as a share of total benefits are highest for households cooking in the house and lowest for households cooking outdoors.

Stove cost. Cost of improved biomass cookstoves commonly promoted in Guatemala, Honduras and Nicaragua cost in the range of US$ 115-190 including cost of local material, labor and installation. An average cost of ICS of US$ 150 is applied in this report. The same cost is applied for a modern LPG stove, including fuel tank and regulator hose. Useful life of 4 years is applied for ICS and 10 years for LPG stoves. The cost is for ICS and LPG stoves that have multiple burners so that cooking with the traditional, unimproved stove or open fire can be avoided. Cost of interventions also includes stove maintenance and repairs at an annual rate of 5% of initial cost of improved cookstoves and LPG stoves.

LPG fuel cost. LPG consumption is assumed to be 30 kg per person per year for households that switch to exclusive use LPG for cooking. The local price of LPG was in the range of US$ 0.85-1.25 per kg in 2015 in the three countries.

Program cost. Achieving adoption of LPG and improved stoves for cooking requires promotion, community participation, and behavioral change programs. Such programs cost money and is part of the cost of achieving targets. Programs are expected to cost US$ 10 per household in the first year (promotion and monitoring), and US$ 2 per household per year in subsequent years (monitoring).

Cost shares. For the ICS intervention, the cost of stove accounts for 78% of annualized cost, maintenance cost for 13%, and cost of promotion programs for 9%. For the LPG intervention, LPG fuel cost accounts for 79-84% of total annualized cost, while cost of stove accounts for 11-15%, maintenance cost for 3-5%, and promotion program cost for about 2%. Costs are annualized using a discount rate of 10%.

Benefit-cost ratios of interventions

ICS provide benefits that are 2.0 to 5.7 times their cost, depending on the country and household cooking location (in house, separate building, or outdoors) (table E5). Benefit-cost ratios are highest in Guatemala, mainly because of higher valuation of health benefits associated with higher GDP per capita than in the other two countries.

Benefit-cost ratios for cooking with LPG are lower than for ICS and in the range of 1.4 to 3.2, reflecting the cost of LPG fuel (table E6). But health benefits of cooking with LPG are more than twice as large as cooking with an ICS.

Clean energies, such as LPG, are therefore the only option for effectively combatting health effects of solid fuels. In other words, ICS may be the most efficient solution as reflected by the high benefit-cost ratios, but not the most effective solution.

Table E5. Benefit-cost ratios of Improved cookstove (ICS) intervention, 2015Guatemala Honduras Nicaragua

In house 5.7 3.6 2.5Separate building 5.0 3.2 2.2Outdoors 4.3 2.8 2.0

Source: Estimates by the author.

Table E6. Benefit-cost ratios of LPG interventions, 2015Guatemala Honduras Nicaragua

In house 3.2 2.1 1.9Separate building 2.7 1.9 1.6Outdoors 2.2 1.6 1.4


Electricity is an alternative to LPG, widely used for cooking in urban areas of Honduras. Cooking with electricity provides the same health benefits of PM2.5 reductions as cooking with LPG. The cost of cooking with electricity would be the same as cost of cooking with LPG if electricity prices were US$ 0.08-0.115 per kWh in the three countries. Benefit-cost ratios of cooking with electricity would therefore be similar to those of LPG at these electricity prices.

The use of solid cooking fuels by one household affects surrounding households. Smoke is vented out of one household for so to enter the dwellings of others and also pollutes the ambient outdoor air. There are therefore benefits from stove promotion programs being community focused with the aim of achieving “unimproved stove free” and eventually “solid fuel free” communities along the lines of community lead sanitation programs and open defecation free communities. Benefit-cost ratios estimated in this paper with full community adoption of ICS or LPG are 15-20% higher than partial adoption of interventions, reflecting reduced community pollution and 30-33% larger health benefits of full adoption.

Promotion programs need also demonstrate and encourage proper use and good maintenance and repairs of stoves.

Promotion of large-scale adoption of cleaner cookstoves has had limited success in Guatemala, Honduras and Nicaragua so far. Some of the factors influencing adoption rates are:

i) High initial cost of ICS and LPG stoves, and high cost of LPG fuel; ii) Need tailoring to consumers’ preferences for stove characteristics; iii) Need installment financing; iv) Need well-targeted information campaigns.

1. Household use of solid biomass fuels for cooking

As many as 41% of households globally relied on solid fuels for cooking in 2010 (Bonjour et al, 2013). The prevalence of solid fuel use is still high in several countries in Latin America and Caribbean (LAC), including Guatemala, Honduras and Nicaragua (figure 1.1).

The use of solid fuels for cooking causes serious household air pollution (HAP), both indoors and outdoors. An estimated 2.9 million people died in 2015 from harmful exposure to PM2.5 emissions from household use of solid fuels according to the Global Burden of Disease (GBD) Project (Forouzanfar et al, 2016). This makes HAP one of the leading health risk factors in developing countries.

Figure 1.1 Prevalence of solid fuel use as primary cooking fuels, 2010

Source: Presented in Smith et al (2014).

The use of solid fuels as primary cooking fuels in Latin America and the Caribbean (LAC) varies from less than one percent of the population in a handful of countries to over 90% in Haiti. Prevalence rates of 51-63% in Guatemala, Honduras and Nicaragua were among the highest in the region in 2012 (figure 1.2).

Figure 1.2. Use solid fuels as primary cooking fuels in LAC (% of population), 2012

Argentina

Uruguay

St. Lu

cia

Ecuad

orBraz

il

Costa Rica

Dominica

Venezu

ela

Guyana

CubaChile

Dominican Rep

Surin

ame

Jamaic

aBeli

ze

Colombia

Panam

a

Mexico

El Salv

adorBolivi

aPeru

Paragu

ay

Honduras

Nicarag

ua

Guatemala Haiti

0%20%40%60%80%

100%

Note: Countries in LAC in 2012 with GDP per capita < US$ 16,000 or solid fuel use > 1%. Source: World Bank (2016a).

The use of modern fuels (e.g., LPG, natural gas, electricity) instead of solid fuels as primary cooking fuels generally increases with GDP per capita as evident in figure 1.3 for countries in the LAC region.2 Countries above the solid line have a higher prevalence of modern fuel use than predicted by their GDP per capita, and countries below the line have a lower prevalence than predicted by their GDP per capita.

The prevalence of modern fuel use in Honduras and Nicaragua is very close to the rate predicted by their GDP per capita. In Guatemala, however, the prevalence of modern fuel use is 29 percentage points lower than predicted by GDP per capita (37% vs. 66%). Paraguay and Peru also have low prevalence rates of modern fuel use relative to their GDP per capita, while rates in Bolivia and Guyana are much higher than predicted.

Figure 1.3. Use of modern fuels as primary cooking fuels in relation to income in LAC (% of population), 2012

0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,0000%

20%

40%

60%

80%

100%f(x) = 0.369503481124336 ln(x) − 2.33119375187975R² = 0.737488329332021

GDP per capita (US$ in 2012)


The use of solid fuels as primary cooking fuels has declined at an annual rate of 1 percentage point in Nicaragua and 0.7 percentage points in Honduras from 1990 to 2012. In Guatemala, however, data indicates that the proportion of the population using solid fuels has remained nearly the same.

The declines in solid fuel use prevalence in Nicaragua and Honduras are the 10 th and the 19th

largest among 33 countries in LAC from 1990 to 2012. All the countries in LAC experienced larger declines than in Guatemala over this time period, according to data in World Bank (2016a).

Figure 1.4. Use solid fuels as primary cooking fuels (% of population), 1990-2012

2 The strongest relationship in LAC between prevalence rates of modern fuel use for cooking and GDP per capita is for countries with GDP per capita < US$ 8,000, with R2=0.74 (19 countries). The relationship is also strong at higher levels of income, with R2=0.68 for countries with GDP per capita < 16,000 (31 countries). The relationship between Purchasing Power Parity (PPP) adjusted GDP per capita and modern fuel use is similar to that of GDP per capita at market prices with R2=0.70.

Nicaragua Honduras Guatemala0%

10%20%30%40%50%60%70%80%90%

76%66% 62%

54% 51%

63%

19902012


The use of solid fuels for cooking varies markedly between urban and rural areas in Honduras, Guatemala and Nicaragua, as in many other countries (figure 1.4). Prevalence was 22-37% of the urban population and as high as 83-91% of the rural population in 2012. Consequently 71-77% of the population using solid fuels as primary cooking fuels resides in rural areas.

Figure 1.4. Urban and rural use solid fuels as primary cooking fuels (% of population), 2012

Honduras Guatemala Nicaragua0%

10%20%30%40%50%60%70%80%90%

100%

51%63%

54%

22%

37%27%

83%90% 91%

TotalUrbanRural


The Demographic and Health Survey (DHS) administered periodically in a majority of low- and middle-income countries in LAC often contains additional information on household cooking fuels, cooking location and cookstove. The most recent DHS survey was in 2011-12 in Honduras, 2014-15 in Guatemala, and 2011-12 in Nicaragua. The final survey report and data for Guatemala have not been released as of this time, and the previous DHS survey was administered as far back as 1998-99. The Nicaragua DHS 2011-12 contains limited information on household cooking. But the Honduras DHS 2011-12 contains valuable additional information (SdS et al, 2013).

Over 56% of the population used solid fuels (wood, charcoal) as their primary cooking fuel in Honduras in 2011-12 according to the Honduras DHS (table 1.1).3 The prevalence of solid fuel use in rural areas was over three times higher than in urban areas. The main fuels in urban areas were LPG and electricity.

Table 1.1. Primary cooking fuel in Honduras (% of population), 2011-12

Urban Rural TotalElectricity 29.8% 5.1% 17.1%LPG 38.3% 8.9% 23.3%Kerosene 4.9% 0.3% 2.5%Charcoal 0.1% 0.2% 0.1%Wood 25.8% 85.0% 56.1%No cooking 1.1% 0.6% 0.9%Total 100.0% 100.0% 100.0%


The vast majority of the population cooked in the house. Cooking in a separate building or outdoors is, however, quite prevalent in rural areas and largely among the population cooking with wood. However, nearly 2/3rd of the population cooking with wood does so in the house (tables 1.2-3).

Table 1.2. Cooking location in urban and rural areas of Honduras (% of population), 2011-12

Urban Rural TotalIn house 83.9% 72.3% 78.0%In separate building 7.9% 16.5% 12.4%Outdoors 8.1% 11.2% 9.7%Total 100.0% 100.0% 100.0%


Table 1.3. Cooking location by cooking fuel in Honduras (% of population), 2011-12

In houseIn separate

building Outdoors TotalElectricity 16.9% 0.3% 0.1% 17.3%LPG/gas/biogas 22.7% 0.6% 0.1% 23.5%Kerosene 2.4% 0.1% 0.1% 2.5%Charcoal 0.1% 0.0% 0.0% 0.1%Wood 35.9% 11.3% 9.4% 56.6%Total 78.0% 12.4% 9.7% 100.0%

Note: The percentages in the column “Total” differ somewhat from the “Total” in previous table because households reporting “No cooking” are excluded. Source: Produced from the Honduras DHS 2011-12 (SdS et al, 2013).

3 The following tables on Honduras are produced from analysis of the household member data of the Honduras DHS 2011-12. These data are available at http://dhsprogram.com.

The Honduras DHS 2011-12 also reports on characteristics of the cookstove in households cooking with solid fuels, i.e., if the stove is open or closed and if the stove has a chimney. Nearly 40% of the population cooks with an open stove with no chimney and nearly 57% cook with an open stove with chimney (table 1.4). No information is reported on the effectiveness of a chimney over an open stove in terms of venting the smoke out of the building or cooking area, but a study from Peru found that concentrations of PM2.5 were as high in households with a chimney over an open fire as in households cooking on open fire without a chimney (Pollard et al, 2014). Only 3.5% of the population using solid fuels cooks with a closed stove.

About 3/4th of the population with an open stove with chimney cooks in the house, while “only” half of the population with no chimney does so.

Table 1.4. Type of cookstove among population cooking with solid fuels in Honduras (% of population), 2011-12

In houseIn separate

building Outdoors TotalOpen stove with no chimney 18.1% 9.8% 11.6% 39.6%Closed stove with chimney 2.8% 0.5% 0.2% 3.4%Closed stove with no chimney 0.1% 0.0% 0.0% 0.1%Open stove with chimney 42.4% 9.6% 4.8% 56.8%Total 63.4% 19.9% 16.7% 100.0%


2. Household exposure to PM2.5

Air concentrations of PM2.5 from the use of solid biomass cooking fuels often reach several hundred micrograms per cubic meter (µg/m3) in the kitchen, and well over one hundred micrograms in the living and sleeping environments. These are findings from measurement studies around the world including in several Latin American countries (WHO, 2014).

Measurement studies in five countries in Latin America found 24-48 hours average PM2.5 kitchen concentrations of 130-1020 µg/m3 among households that used wood for cooking over open fire or unimproved stoves. The same studies found that average concentrations declined to 50-340 µg/m3 in the same kitchens after installation of improved cookstoves, often with an attached chimney (table 2.1).

Several studies also measured 24-48 hours personal PM2.5 exposure of the person cooking. Personal exposures averaged 116-260 µg/m3 before and 58-100 µg/m3 after the installation of improved cookstoves (table 2.2).

Some studies point to the community effects of cooking with solid fuels. A study in rural Mexico measured outdoor PM2.5 concentrations. Median 48-hours concentrations of about 80 µg/m3 were found at the outdoor patio of the dwellings, and about 60 µg/m3 at the community

plaza with very little difference in concentration before and after installation of improved cookstoves with chimney (Zuk et al, 2007). A study in a semi-urban and a rural community in Honduras using solid fuels for cooking found 8-hour daytime average PM2.5 concentrations of 215 µg/m3 outside dwellings using improved chimney stoves and 358 µg/m3 outside dwellings cooking on open fire or traditional stoves (Clark et al, 2010).

Other studies point to the role of technologies and maintenance of the improved stoves and chimneys. Over 20% of the households in a study in Peru had a chimney above the fire. PM2.5 concentrations were, however, not statistically lower in these households than in households without chimney, suggesting that most of the smoke escaped into the kitchen from the open fire before reaching the chimney and/or that the chimneys were ineffective in venting the smoke (Pollard et al, 2014). Hartinger et al (2013), also in Peru, found that proper operation, maintenance and repair of stoves after installation was essential for reducing indoor PM2.5 concentrations, even in the first year. In Honduras, PM2.5 personal exposure levels and indoor concentrations decreased distinctly in relation to improvements in the quality of cookstoves (Clark et al, 2010).

Table 2.1. Average PM2.5 kitchen concentrations in households using solid fuels for cooking in Latin America (µg/m3)

Country Unimproved stoves Improved stoves SourceGuatemala Open fire 900 340 Improved chimney stove Northcross et al (2010)

HondurasOpen fire/UCS 1002 266 Justa chimney stove Clark et al (2009)*

Honduras Open fire 310 60 ECO-Stove Lam et al (2012)Mexico Open fire 257 101 Patsari chimney stove Cynthia et al (2010)

Mexico Open fire 1020 350 Patsari chimney stoveCynthia et al (2008); Masera et al (2007)

Mexico Open fire 658 255 Patsari chimney stove Zuk et al (2007)Nicaragua Open fire 1801 416 Eco-Stove with chimney Clark et al (2013)Nicaragua Open fire 514 53 Closed Eco-Stove Terrado and Eitel (2005)Nicaragua Open fire 639 121 Semi-open Eco Stove Terrado and Eitel (2005)Peru Open fire 130 Pollard et al (2014)Peru Various 173-207 50-84 Improved stove Fitzgerald et al (2012)Peru Various 189 136 OPTIMA Hartinger et al (2013)

Peru Open fire 680 200Inkawasina chimney stove Winrock (2008)

Peru Open fire 380 130 HNP 3-pot metal stove Li et al (2011)Peru Open fire 320 110 BGC 3-pot metal stove Li et al (2011)

Notes: Measurements are 24- to 48-hour averages. UCS=unimproved cookstove. * 8-hour daytime average.

Table 2.2. Average PM2.5 personal exposure in households using solid fuels for cooking in Latin America (µg/m3)

Country Unimproved stoves Improved stoves SourceGuatemala Open fire 264 102 Plancha chimney stove McCracken et al (2007)*Guatemala Open fire 273 174 Plancha chimney stove McCracken et al (2007)**Guatemala Open fire 200 70 Plancha chimney stove McCracken et al (2013)*

Honduras Open fire/UCS 198 74 Justa chimney stove Clark et al (2009)Mexico Open fire 156 78 Patsari chimney stove Cynthia et al (2010)Mexico Open fire 240 160 Patsari chimney stove Cynthia et al (2008)Nicaragua Open fire 374 49 Closed Eco-Stove Terrado and Eitel (2005)Nicaragua Open fire 355 96 Semi-open Eco Stove Terrado and Eitel (2005)Peru Various 116-126 58-68 Improved stove Fitzgerald et al (2012)Peru Open fire 190 80 HNP 3-pot metal stove Li et al (2011)Peru Open fire 150 70 BGC 3-pot metal stove Li et al (2011)

Notes: Measurements are 24- to 48-hour averages, except for in Honduras (8-hour daytime average). UCS=unimproved cookstove. * Control group (open fire) versus intervention group (chimney stove). ** Before (open fire) versus after (chimney stove) intervention. Also see McCracken et al (2011).

The studies of PM2.5 kitchen concentrations and personal exposure of the person cooking are summarized in table 2.3. Improved cookstoves reduced kitchen concentrations by over 65% and personal exposure by well over 55%.

Personal exposure is the indicator of importance in terms of health effects of household PM2.5. In household cooking over an open fire, the median4 personal exposure of the person cooking was 200 µg/m3, or 20 times WHO’s annual air quality guideline for outdoor PM2.5. The median personal exposure after installation of an improved cookstove was nearly 80 µg/m3 or still 8 times WHO’s annual guideline.

Table 2.3. PM2.5 kitchen concentrations and personal exposure in households using solid fuels for cooking in Latin America (µg/m3)

StatNo of

studiesOpen fire (µg/m3)

Improved stove (µg/m3)

Reductions from Improved stove

Kitchen concentrations

Mean 14 633 186 68%Median 14 577 133 66%

Personal exposure

Mean 11 229 92 57%Median 11 200 78 58%

Source: Calculated from studies presented above.

PM2.5 concentrations vary substantially in relation to type of solid fuel, cooking location, type of stove and ventilation practices, cooking duration, and structure of dwelling. And household members’ personal exposure to PM from combustion of solid fuels depends additionally on their activity patterns inside and outside the household environment.

In estimating the health effects of household air pollution, and the benefits and costs of cookstove options in the sections that follow in this report, exposure of adult women is used as a reference point for personal exposure. This is because the person cooking in the household is most often a woman, and the exposure measurement studies discussed above are in reference to the person cooking. Exposures of adult men and young children are set at 60-85% of adult women’s exposure, as adult men and young children generally spend less time in the household

4 The median of the mean exposures in the studies referenced in this section.

environment and the kitchen than adult women (Smith et al, 2014). In terms of cooking location, cooking in the house is used as reference location as the personal exposure studies presented above reflect this location. Personal exposures from cooking outdoors or in a separate building are set at 60-80% of exposure from cooking in the house (table 2.4). The exposure levels reflect that a portion of biomass smoke from outdoor cooking or cooking in a separate building enters the indoor living and sleeping areas.

Table 2.4. Relative exposure levels by household member and cooking locationHousehold member (H) Location (L)

1 Adult women 100% In house 100%2 Adult men 60% Separate building 80%3 Children < 5 years 85% Outdoors 60%


An average exposure level of 200 µg/m3 is applied to adult women cooking in the house with solid fuels over open fire or traditional cookstove. This is the median exposure level of the cook in table 2.3. Average exposure levels of adult men and children under five years of age, and in various cooking locations are calculated in relation to the exposure level of adult women cooking in the house by applying the relative exposure factors in table 2.4. So for instance, the exposure level of adult men in a household cooking outdoors is 200 µg/m3 * G2 * L3 = 200 µg/m3 * 60% * 60% = 72 µg/m3 (table 2.5; figure 2.1).

Table 2.5 Long term personal PM2.5 exposure by cooking location in households using traditional cookstove with solid fuels (µg/m3)

Adult women Adult men Children < 5

In house 200 120 170

Separate building 160 96 136

Outdoors 120 72 102Source: Estimates by the author.

Figure 2.1. Long term personal PM2.5 exposure in households using solid fuels for cooking (µg/m3)

Adult women Adult men Children < 50

20406080

100120140160180200

In houseSeparate buildingOutdoors


3. Health effects and cost of household PM2.5

Health effects of long term exposure to PM2.5 in the household environment from the burning of solid fuels include: (i) ischemic heart disease (IHD), (ii) cerebrovascular disease (stroke), (iii) lung cancer (LC), and (iv) chronic obstructive pulmonary disease (COPD) among adult women and men, and (v) acute lower respiratory infections (ALRI) among children under five years of age and adult women and men. These are all major health effects evidenced by the Global Burden of Disease (GBD) Project (Forouzanfar et al, 2016), and figure 3.1 shows how the risk of these five health effects in terms of mortality increases with increasing levels of PM2.5 exposure.

Figure 3.1 Relative risk of mortality from long term PM2.5 exposure

5 20 35 50 65 80 95 110

125

140

155

170

185

200

215

230

1.001.101.201.301.401.501.601.701.801.902.002.102.20

IHDStrokeCOPDLCALRI

PM2.5 (µg/m3)

Note: Age-weighted relative risks. Source: Produced from Forouzanfar et al (2016).

Estimated prevalence rates of solid fuel use for cooking range from 49% in Honduras to 63% in Guatemala in 2015 (table 3.1). These estimates are based on trends from 1990 to 2012 presented in section 1. Solid fuel use by cooking location in Honduras is in proportion to the prevalence rates in the Honduras DHS 2011-12 household survey (SdS et al, 2013). Rates by cooking location in Guatemala and Nicaragua are proportion to rates in Honduras, as no recent data have been identified from the two countries.

Table 3.1. Estimated use of solid fuels for cooking by cooking location in 2015 (% of population)

In houseIn separate

building Outdoors TotalHonduras 31% 10% 8% 49%Nicaragua 33% 10% 9% 52%Guatemala 40% 13% 10% 63%


The solid fuel use prevalence rates in table 3.1, PM2.5 exposure levels presented in the previous section, and the relative risks of health effects in figure 3.1 are combined to estimate the health effects of household PM2.5 air pollution from the use of solid fuels. The results

show that 13-19% of all IHD and stroke, and 25-33% of all COPD, lung cancer, and ALRI in the countries are from household PM2.5 air pollution (table 3.2).5 The percentage of total disease specific mortality attributable to household air pollution is highest in Guatemala, as this is the country with the highest prevalence rate of solid fuel use. The attributable percentages translate to annual deaths of 1,843 in Nicaragua; 3,098 in Honduras; and 6,369 in Guatemala in 2015 (table 3.3).

Table 3.2. Estimated morality attributable to PM2.5 household air pollution (% of total mortality), 2015Guatemala Honduras Nicaragua

Ischemic heart disease (IHD) 19% 15% 16%

Cerebrovascular disease (stroke) 16% 13% 14%

COPD 33% 29% 30%

Lung cancer 30% 25% 26%

ALRI 36% 32% 33%Source: Estimates by the author.

Table 3.3. Estimated annual cases of morality attributable to PM2.5 household air pollution, 2015Guatemala Honduras Nicaragua

Ischemic heart disease (IHD) 1530 1437 736

Cerebrovascular disease (stroke) 652 576 317

COPD 446 533 307

Lung cancer 191 101 78

ALRI 3551 450 405

Total 6369 3098 1843Source: Estimates by the author.

The health effects also include days of illness or disease associated with the five major health outcomes. Days lived with disease range from 7 million in Nicaragua to 17 million in Guatemala in 2015 (table 3.4).6

Table 3.4. Estimated days lived with disease attributable to PM2.5 household air pollution, 2015Guatemala Honduras Nicaragua

Years lost to disease 6,297 2,914 2,656

Days lived with disease 17,238,083 7,978,198 7,269,452Source: Estimates by the author.

These health effects can be monetized as a cost to society by using economic valuation methods. The cost of mortality is estimated by applying a “value of statistical life” or VSL to the number of deaths. The VSL is estimated using the methodology in World Bank (2016b), and ranges from 70 times GDP per capita in Honduras and Nicaragua to 76 times GDP per capita in

5 See annex 1 for methodological details.6 Days lived with disease (DD) is estimated as follows: DD= yld * M * 365 / W where “yld” is “years lost to disease per death” for each country from the Global Burden of Disease (GBD) Project, M is number of annual deaths from household air pollution, and W is “disability weight” of disease (here on average 0.15).

Guatemala (see annex 2). The cost of morbidity is estimated as 50% of average wage rates for each day of illness. The estimated cost of health effects is equivalent to 2.3% of GDP in Nicaragua, 2.9% of GDP in Honduras, and 3.1% of GDP in Guatemala in 2015 (table 3.5).7

Table 3.5. Annual cost of health effects of household PM2.5 air pollution, 2015Guatemala Honduras Nicaragua

GTQ billion

USD million

HNL billion

USD million

NIO billion

USD million

Mortality 14.4 1,880 11.9 538 7.3 269

Morbidity 0.8 107 0.8 37 0.7 26

Total 15.2 1,988 12.7 576 8.0 295

% of GDP in 2015 3.12% 3.12% 2.86% 2.86% 2.33% 2.33%Source: Estimates by the author.

4. Pollution control interventions

Benefits and costs of two common household cookstove interventions for household air pollution (HAP) control among households cooking with biomass fuels over open fire or traditional, unimproved cookstove are assessed in this paper:

(1) Improved biomass cookstove (ICS); and(2) LPG stove (LPG).

Stoves that are assessed have a minimum of two burners so that households are less likely to continue to use their traditional stove for their cooking needs. Electric stove is compared to LPG, as electricity is widely used in urban areas in Honduras albeit rarely used in Guatemala and Nicaragua (see section 1).

Examples of improved cookstoves distributed in the countries are reported in table 4.1.

Table 4.1. Examples of improved cookstoves distributed in Guatemala, Honduras and NicaraguaStove Description

Guatemala Noya stove Portable metal stove with a rocket elbow in the combustion chamber and chimney. Saves 60% fuelwood. Costs US$ 160.

Onil stove Semi-portable cement stove with an iron plancha, chimney, and a rocket elbow in the combustion chamber. Saves 67% fuelwood. Costs US$ 125.

Honduras JUSTA stove Several stove variations, made of bricks or cement. Stove has rocket elbow in the combustion chamber and chimney, and a continuous plancha with pots not exposed to direct fire. Stove is large enough for multiple pots and area for making tortillas. Saves about 50% fuelwood. Costs US$120-165.

Environfit Metal stove with chimney. Cost US$190.Nicaragua Ecofogon Stove is made from brick, with and without chimney. Also a metal version. The

stoves save 50% fuelwood. Costs US$115.Mifogon Metal stove with one open pot for cooking with open fire and a plancha

7 The lower cost as a percent of GDP in Nicaragua is mainly due to lower baseline death rates than in Guatemala and Honduras.

Source: Based on information in PAHO Household Air Pollution Country Profiles.

Assessment of interventions is undertaken with respect to:

(1) Health benefits of reduced PM2.5 exposure; (2) Non-health benefits (i.e., fuel savings and cooking time savings); (3) Stove and fuel costs of interventions; (4) Stove promotion programs and stove maintenance; and(5) Comparison of benefits and costs of interventions (i.e., benefit-cost ratios).

Each of the cookstove interventions are assessed in three cooking locations:

(1) Cooking in the house;(2) Cooking in a separate building; and(3) Cooking outdoors.

The interventions are assessed for two targeting scenarios in terms of community adoption rates of the cookstove interventions:

(1) Partial adoption with continued high community pollution from cooking; and (2) Full adoption with substantially lower community pollution.

Household use of solid fuels has community effects. Smoke from fuel burning enters dwellings of other households as well as contributes to outdoor ambient air pollution. An improved stove with chimney, or simply venting of smoke through a hood from any stove or open fire, may be effective for the household installing these devices, but contributes to increased outdoor ambient pollution and indoor pollution in nearby dwellings. Only “smokeless” fuels and technologies prevent this problem of externalities.

To achieve the maximum benefits per unit of expenditure on household energy and stove interventions, all households would need to participate, and thus achieve a “solid fuel free” community or, alternatively, an “unimproved stove free” community. This concept may be applicable to rural areas where communities consist of a cluster of households and each community is spatially separated from one another, and is similar to an “open defecation free” community in the sanitation sector, often promoted and achieved through community-lead or total sanitation programs.

5. Post-intervention PM2.5 exposures

The use of improved cookstoves (ICS) or LPG (or electricity) for cooking is expected to substantially reduce household members’ exposure to PM2.5. The review of the personal

exposure studies in Latin America in section 2 before and after installation of an ICS indicated a median reduction in exposure of over 55% from about 200 µg/m3 to 80 µg/m3.

The exposure reductions were measured within relatively short time after the installation of the ICS. Exposure reductions over the life of the ICS is likely to be somewhat less as the quality of the ICS deteriorates over time.

A 40% exposure reduction from an ICS over its lifetime is therefore likely to be more realistic and is applied here to households cooking in the house.8 Exposure reductions from an ICS for households cooking in a separate building or outdoors may be less than for households cooking in the house. This is because the relative contribution to exposure from pollution originating from other households cooking with solid fuels in the community is larger for households cooking in a separate building or outdoors than for households cooking in the house. Thus exposure reductions of 35% and 30% are applied to households cooking in a separate building and outdoors, respectively.9

The exposure reductions from an ICS also depend on the rate of adoption of ICS in the community. An addition 15% reduction in exposure is applied in the scenario of full adoption, i.e., all households adopt an ICS.

Table 5.1 summarizes the exposure reductions from ICS. These reductions are relative to the exposure levels using traditional cookstoves (TCS) presented in table 2.5, and are applied to adult women, men and children.10

Table 5.1. Household member exposure reduction from the use of ICSPartial Community Adoption Full community adoption

In house 40% 50%Separate building 35% 45%Outdoors 30% 40%

Source: The author.

As to LPG, its combustion results in very little PM emissions and is therefore considered a relatively clean cooking fuel. Studies have however found that household PM2.5 concentrations often remain as high as 40-60 µg/m3, presumably mainly due to the community pollution from neighboring households using solid fuels. It is therefore stipulated here that exposure levels associated with cooking with LPG stoves (or electricity) are on average 50 µg/m3. This exposure level is applied to adult women and children, and is independent of

8 A 40% reduction over the life of the ICS reflects a linear deterioration in exposure reduction from 55% in the first year to 25% in the fourth year, after which time the stove is either replaced or receives a major overhauled.9 These exposure reductions in relation to cooking location give in fact a very similar percentage reduction in exposure from own pollution across cooking locations, after subtracting exposure resulting from community pollution. 10 Example for cooking in separate building: Reduction with full community adoption is 1-(1-35%)*(1-15%)=45%.

cooking location. A somewhat lower exposure level of 35 µg/m3 is applied to adult men, as this household member group often spends considerable time away from the immediate community, and presumably in locations with less pollution.

In the case of full community adoption of LPG stoves (or electricity), a personal exposure level of 25 µg/m3 is applied to all household members. This exposure originates from other sources of PM2.5 in the household environment and non-solid fuel related outdoor ambient PM2.5.

Personal exposure levels in households using LPG stoves may decline to levels below 25 µg/m3. Joon et al (2011) found a 24-hour average PM2.5 exposure for the cook of 25 µg/m3 among rural households using LPG in Haryana, India. Titcombe and Simcik (2011) measured an average PM2.5 personal exposure of 14 µg/m3 in households in the southern highlands of Tanzania cooking indoors with LPG.

Pre- and post-intervention levels of personal exposure to PM2.5 are presented in table 5.2 and figure 5.1 and reflect the exposure reductions from ICS and levels associated with LPG stoves (or electricity) discussed above. The exposure levels are broad averages and will vary substantially across individual households.

Table 5.2. Household member air pollution exposure by intervention and cooking location (µg/m3)

Pre-Intervention Post-Intervention

TCS ICS

- Partial Adoption

ICS- Full

Adoption

LPG- Partial

Adoption

LPG- Full

AdoptionAdult female

Outdoors 120 84 71 50 25

Separate building 160 104 88 50 25

In house 200 120 102 50 25

Adult maleOutdoors 72 50 43 35 25


In house 120 72 61 35 25

ChildrenOutdoors 102 71 61 50 25


In house 170 102 87 50 25Note: TCS = Traditional cookstove (biomass over open fire or unimproved stove); ICS = Improved Cook Stove; LPG = Liquefied Petroleum Gas. Source: The author.

Figure 5.1. Average household exposure levels (µg/m3)*

In house Separate building Outdoors0

20

40

60

80

100

120

140

160

180

TCS ICS ICS - FullLPG - PartialLPG - Full

* Average of women, men and children. Note: TCS = Traditional cookstove (biomass over open fire or unimproved stove); ICS = Improved Cook Stove; LPG = Liquefied Petroleum Gas. Source: The author.

6. Health benefits of interventions

Health benefits of moving from pre-intervention to post-intervention exposure levels are estimated by using the integrated PM2.5 exposure-health response methodology from the GBD 2015 Project presented in annex 1 and health risks presented in figure 3.1.

Estimated percentage reductions in health effects of interventions are presented in table 6.1. The reductions – or health benefits - are larger among households cooking in the house than among households cooking in separate building or outdoors, due to larger exposure reductions in the former group of households. Percentage reductions in the case of full community adoption of interventions are also substantially larger than in the case of partial adoption due to reduced community pollution from full adoption. Moreover, LPG stoves (or electricity) provide more than twice as high health benefits as ICS.

In total, over 3,000 deaths can be avoided annually by full community adoption of ICS and over 6,800 by full adoption of LPG in Guatemala, Honduras and Nicaragua.

It should also be noted that the percentage reduction in health effects of ICS is substantially smaller than the percentage reduction in PM2.5 exposure, as seen by comparing tables 5.1 and 6.1. This is because of the non-linear relationship between exposure level and health risks, as seen in figure 3.1.

Table 6.1 Reduction in health effects from cookstove interventions Guatemala Honduras Nicaragua

ICS LPG ICS LPG ICS LPG

In house 24% 55% 20% 46% 21% 48%

In separate building 20% 44% 16% 37% 17% 38%

Outdoors 15% 31% 12% 26% 13% 27%

Partial community adoption* 22% 49% 18% 41% 19% 43%

Full community adoption* 29% 65% 24% 54% 25% 56%Annual deaths avoided (full community adoption of intervention) 1835 4134 733 1672 463 1038

* Weighted average by cooking location. Source: Estimates by the author.

7. Non-health benefits of interventions

Switching to an improved cookstove (ICS) or to LPG (or electricity) also has non-health benefits. Main benefits are reduced fuelwood consumption, whether self-collected or purchased, and reduced cooking time. The magnitude of these benefits will depend on current cooking arrangements, type of improved stove, household cooking patterns, and household member valuation of time savings.

7.1 Fuel savings

Common energy conversion efficiencies of unimproved stoves or cooking over open fire are in the range of 13-18% for wood and 9-12% for agricultural resides and dung. Reported efficiencies of improved biomass cookstoves are 23-40% for wood and 15-19% for agricultural residues (Malla and Timilsina, 2014). This means that efficiency gains from using an improved stove instead of an unimproved stove or open fire generally exceed 25% and can be more than 100% depending on type of stoves, cooking practices and type of food cooked. Consequently, biomass fuel savings therefore generally exceed 20% and can be nearly 70% using wood.

It is here assumed that average fuelwood savings are 50% from the use of an ICS instead of an unimproved stove or open fire, on the low end of the savings reported for stoves commonly promoted in Guatemala, Honduras and Nicaragua (table 4.1). Exclusive use of LPG (or electricity) results in 100% savings of fuelwood.

Many urban households purchase some or all of the fuelwood they use for cooking while the majority of rural households collects these fuels themselves. It is important to impute a value of these self-collected fuels. A common approach is to impute a value based on the amount of time households spend on biomass fuel collection.

A collection time of 30 minutes per household per day among households using unimproved biomass stoves or open fire, and a value of time equal to 50% of average female wage rate is applied for valuation of household fuel collection. A female wage rate is applied as most fuel

collection is carried out by women (or children).11 The estimated value of biomass fuel savings are presented in table 7.1.12 The difference in the value of fuel savings per household across the three countries is due to the difference in GDP per capita and labor income share of GDP.

Table 7.1 Estimated value of household fuel savings in local currencies, 2015Guatemala Honduras Nicaragua

Biomass fuel collection time (minutes/household/day) 30 30 30

Female rural wage rate (per hour)* 10 23 20

Value of time (% of wage rate) 50% 50% 50%Value of biomass fuel collection for households using traditional stove or open fire (per household per year) 871 2,130 1,807Value of biomass fuel savings (per household per year)

Improved cookstove (50% savings)LPG stove (100% savings)

435871

1,0652,130

9031,807

Value of biomass fuel savings (per household per year)Improved cookstove (50% savings)LPG stove (100% savings)

US$ 57US$ 114

US$ 48US$ 96

US$ 33US$ 66

* Weighted by the share of rural and urban households using solid fuels for cooking. Source: Estimates by the author.

7.2 Cooking time savings

Households in developing countries typically spend 3-5 hours per day on cooking. Hutton et al (2006) report that it takes 11-14% less time to boil water with a Rocket stove (improved cookstove) or LPG stove than over open fire. Habermehl (2007) reports that monitoring studies have found that cooking time declined by 1.8 hours per day with the use of a Rocket Lorena stove. One-quarter of this time, or 27 minutes, is considered time savings by Habermehl, as the person cooking often engages in multiple household activities simultaneously. Siddiqui et al (2009) report that daily fuel burning time for cooking in a semi-rural community outside Karachi was 30 minutes less in households using natural gas than in households using wood, and that time spent in the kitchen was 40 minutes less. Jeuland and Pattanayak (2012) assumes that an improved wood stove saves around 10 minutes per day and that LPG saves one hour per day in cooking time.

Garcia-Frapolli et al (2010) report that cooking time from using the improved Patsari chimney stove in Mexico declined by about 1 hour per household per day. Effectively 15-30 minutes of this time is saved as the person cooking often engages in multiple household activities simultaneously.

11 Rural wage rate is assumed a half of urban wage rates. Wage rates are estimated based on GDP per capita and labor force participation from World Bank (2016a) and labor income share of GDP from PENN World Tables 80.12 This is likely a somewhat underestimate of the value of biomass fuels used by households as households also purchase some of their fuels at a higher price than reflected in the valuation of collection time.

It is here applied a cooking time saving of 15 minutes per day from the use of an improved cookstove and 30 minutes from the use of LPG stove (or electricity) compared to an unimproved cookstove or open fire. As for fuel collection time savings, a value of time equal to 50% of female wage rates are applied to estimate the value of cooking time savings. Annual values of time savings per household are presented in table 7.2.

Table 7.2. Estimated value of cooking time savings in local currencies per household per year, 2015*Guatemala Honduras Nicaragua

Improved biomass cookstove 435 (57) 1,065 (48) 903 (33)LPG stove 871 (114) 2,130 (96) 1,807 (66)

* US$ in parentheses. Source: Author’s estimates.

8. Costs of interventions

8.1 Cost of stoves

Cost of improved biomass cookstoves varies depending on fuel and emission efficiency, durability, materials, and technology. The stoves commonly promoted in Guatemala, Honduras and Nicaragua cost in the range of US$ 115-190 including cost of local material, labor and installation (table 4.1).

A cost of ICS of US$ 150 is applied here, reflecting a midpoint of promoted stoves. The same cost is applied for a modern LPG stove with a fuel tank and regulator hose. Useful life of 4 years is applied for ICS and 10 years for LPG stoves (table 8.1). The cost is for ICS and LPG stoves that have multiple burners so that cooking with the traditional, unimproved stove or open fire can be avoided. Cost of stoves is annualized using a discount rate of 10%.

Table 8.1. Cost of stoves in local currencies, 2015Improved Biomass

CookstoveLPG Stove

Guatemala 1,150 1,150Honduras 3,300 3,300Nicaragua 4,100 4,100Useful life of stove (years) 4 10

Source: The author.

8.2 Cost of LPG fuel

It is here assumed that LPG consumption is 30 kg per person per year for households that exclusively use LPG for cooking. This is in line with estimates for several countries in Latin America, Asia and Africa (Kojima et al, 2011). The local price of LPG was in the range of US$ 0.85-1.25 per kg in 2015 in Nicaragua, Honduras and Guatemala according to PAHO Household Air Pollution Country Profiles.13 Annual cost of LPG fuel per household for exclusive cooking 13 http://www.paho.org/hq/index.php?option=com_content&view=article&id=11772&Itemid=41769&lang=en

with LPG is presented in table 8.2.

Table 8.2. Estimates of LPG fuel cost in local currencies, 2015Guatemala Honduras Nicaragua

LPG fuel (kg/person/year) 30 30 30LPG fuel (kg/household/year) 150 150 150LPG cost per kg 9.5 (US$1.25) 25.3 (US$1.14) 23.3 (US$0.85)LPG fuel cost per household per year 1,430 3,800 3,500

Source: The author.

8.3 Cost of stove maintenance and stove promotion programs

Cost of interventions also includes stove maintenance and repairs of improved cookstoves and LPG stoves. Annual cost of maintenance and repair is assumed to be 5% of initial stove cost.

Achieving adoption of modern energy and improved stoves for cooking requires promotion, community participation, and behavioral change programs. Such programs cost money and is part of the cost of achieving targets. Program cost increases on the margin as increased intensity and scale of programs are needed to achieve an increasing share of the population switching to modern energy or improved stoves.

Programs are assumed to cost US$ 10 per household in the first year (promotion and monitoring), and US$ 2 per household per year in subsequent years (monitoring).14 Program costs are annualized using a discount rate of 10%.

8.4 Annualized cost shares

For the ICS intervention, the cost of stove accounts for 78% of annualized cost, maintenance cost for 13%, and cost of promotion programs for 9%. For the LPG intervention, LPG fuel cost accounts for 79-84% of total annualized cost, while cost of stove accounts for 11-15%, maintenance cost for 3-5%, and promotion program cost for about 2%.

9. Benefit-cost ratios

9.1 Valuation of health benefits

Household air pollution control is not likely to instantaneously provide full benefits for health outcomes that develop over long periods of PM2.5 exposure, i.e., for heart disease, stroke, chronic obstructive pulmonary disease (COPD) and lung cancer. It is therefore assumed that reduced incidence of and deaths from these diseases are gradually realized over ten years. For acute lower respiratory infections (ALRI), however, full health benefits are realized in the same

14 Garcia-Frapolli et al (2010) apply a similar cost for maintenance and repair of a Patsari stove in the Purepecha region of Mexico, and a total program cost of US$ 25 per stove.

year as PM2.5 exposure reduction. This means that over a time horizon of 20 years the present value of health benefits, as a percentage of full health benefits presented in section 6, are 70% in Honduras, 74% in Nicaragua, and 87% in Guatemala at a discount rate of 10%. 15 The higher present value in Guatemala is due to the higher share of ALRI mortality than in Honduras and Nicaragua (see table 3.3). Health benefits of cleaner cooking in terms of avoided deaths and associated illness are monetized by using the same valuation methods as in section 3 and annex 2.

Health benefits account for 40-71% of total estimated benefits. The remaining benefits are evenly split between fuelwood savings and cooking time savings. Health benefits as a share of total benefits are highest for households cooking in the house and lowest for households cooking outdoors. The health benefit share is substantially higher in Guatemala than in Honduras and Nicaragua. This is mainly due to the differences in GDP per capita and therefore the “value of statistical life” that is applied to calculate the value of health benefits in terms of avoided deaths (see annex 2).

Table 9.1 Health benefits of interventions (% of total benefits)Guatemala Honduras Nicaragua

Cooking location/cookstove ICS LPG ICS LPG ICS LPGIn house 65% 68% 52% 55% 53% 56%Separate building 60% 63% 46% 50% 48% 51%Outdoors 54% 54% 40% 41% 41% 42%Community adoption ratePartial adoption 62% 65% 49% 52% 50% 53%Full adoption 68% 71% 56% 59% 57% 60%


9.2 Benefits and costs

Benefits and costs of interventions are compared by using their ratio. A benefit-cost ratio (BCR) greater than one indicates that benefits exceed the costs of interventions. The ratio can be calculated as the present value of benefits over the present value of costs, or as annualized benefits over annualized costs. A discount rate of 10% is used in the calculations.

BCRs of improved biomass cookstoves (ICS) and LPG stoves are presented in tables 9.2-3 for cooking in three locations and for partial and full community adoption of interventions. The BCRs by location reflect partial community adoption of interventions.

BCRs are highest in Guatemala and lowest in Nicaragua. This is mainly due to the differences in GDP per capita and therefore the “value of statistical life” that is applied to calculate the value of health benefits in terms of avoided deaths (see annex 2).

15 The present value of health benefits are 78%, 81%, and 91% respectively at a discount rate is 3%.

BCRs are largest for households cooking in the house and smallest for households cooking outdoors. However, even for households cooking outdoors the BCRs of interventions are all greater than one, and ranges from 2.0 to 4.3 for ICS and 1.4 to 2.2 for LPG. BCRs with full community adoption of ICS or LPG are 15-20% higher than partial adoption of interventions, reflecting reduced community pollution and 30-33% larger health benefits of full adoption.

BCRs of ICS are substantially larger than for LPG. This is mainly because of the cost of LPG fuel. However, the benefits of cooking with LPG are substantially larger than the costs (BCRs of 1.4-3.2), with health benefits of LPG more than twice as large as for ICS. Clean energies, such as LPG, are therefore the only option for effectively combatting health effects of solid fuels, especially when achieved community-wide. In other words, ICS may be the most efficient solution but not the most effective solution.

An alternative to LPG is electricity, widely used for cooking in urban areas of Honduras. Cooking with electricity provides the same health benefits of PM2.5 reductions as cooking with LPG. At LPG prices in the three countries in 2015 as reported in table 8.2, the cost of cooking with electricity would be the same as the cost of cooking with LPG if electricity prices were US$ 0.08 per kWh in Nicaragua, US$ 0.10 per kWh in Honduras, and US$0.115 in Guatemala. 16 BCRs of cooking with electricity would therefore be the same as for LPG at these electricity prices.17

Table 9.2 Benefit-cost ratios of Improved cookstove (ICS) intervention, 2015Guatemala Honduras Nicaragua

Cooking locationIn house 5.7 3.6 2.5Separate building 5.0 3.2 2.2Outdoors 4.3 2.8 2.0Community adoption ratePartial adoption 5.3 3.4 2.4Full adoption 6.4 3.9 2.7


Table 9.3 Benefit-cost ratios of LPG interventions, 2015Guatemala Honduras Nicaragua

Cooking locationIn house 3.2 2.1 1.9Separate building 2.7 1.9 1.6Outdoors 2.2 1.6 1.4Community adoption ratePartial adoption 2.9 2.0 1.7Full adoption 3.5 2.3 2.0


16 Reflecting stove energy efficiencies of 55% for LPG stoves and 65% for electric stoves.17 This assumes that other costs (stove cost, stove maintenance, promotion programs) and non-health benefits (fuelwood savings, cooking time savings) are similar.

9.3 Sensitivity analysis

The predominant cost component of the ICS intervention is the cost of stove (78% of total cost), while LPG fuel is the predominant cost component of the LPG intervention (79-84% of total cost) (see section 8.4). A doubling of the cost of an ICS or LPG fuel will therefore reduce the BCRs by nearly 50%. Similarly, halving the useful life of the ICS from 4 to 2 years will reduce the BRC by 40%, while increasing the life to 6 years will increase the BCR by 30%. A doubling of the cost of stove maintenance or promotion program cost for ICS and LPG, or a doubling of the cost of the LPG stove, will have only a small effect on BCRs.

In terms of benefits, health benefits account for 40-71% depending on the country. Thus both health and non-health benefits have a substantial impact on BCRs. Halving the health benefits reduce the BCRs by 25-35%. A doubling of the fuelwood collection time from 0.5 to 1 hour per household per day increases the BCR by 15-25%. A halving of the fuelwood collection time to 0.25 hours per day reduces the BCR by about 7-13%. A doubling or halving of the cooking time savings has a similar effect on the BCRs.

The BCRs are also influenced by pre- and post-intervention household air pollution exposure level. A 20% reduction in the pre-intervention PM2.5 exposure level (and the same percentage PM2.5 reduction effectiveness of interventions) reduces the BCRs for ICS by 1-3% and the BCRs for LPG by 9-12%. A 10 percentage point lower PM2.5 reduction effectiveness of ICS reduces the BCR by 12-15%.

The BCRs are relatively insensitive to discount rate. Applying a discount rate of 3% instead of 10% increases the BCRs only marginally by 13-17% for ICS and by 6-9% for LPG.

10. Summary and conclusions

Prevalence rates of solid fuel use for cooking in Guatemala, Honduras and Nicaragua stood at 51-63% in 2012, only behind Haiti in Latin America and the Caribbean (LAC) region.

Measurement studies in LAC have found 24-48 hours average personal PM2.5 exposure of the person cooking with solid fuels in the range of 116-260 µg/m3 before and 58-100 µg/m3 after the installation of improved biomass cookstoves (ICS). This is many times higher than WHO’s annual air quality guideline (AQG) of 10 µg/m3 for outdoor PM2.5.

13-19% of all ischemic heart disease (IHD) and stroke, and 25-33% of all chronic obstructive pulmonary disease (COPD), lung cancer and acute lower respiratory infections (ALRI) in Guatemala, Honduras and Nicaragua are from HAP according to estimates in this paper. This translates to annual deaths from HAP of 1,843 in Nicaragua; 3,098 in Honduras; and 6,369 in

Guatemala in 2015. The percentage of total mortality attributable to HAP is highest in Guatemala, as this is the country with the highest prevalence rate of solid fuel use.

An assessment of benefits and costs is undertaken for two interventions: i) improved biomass cookstoves (ICS); and ii) LPG stoves. Benefits assessed are health improvements, biomass fuel savings, and cooking time savings. Costs assessed are stove purchase, stove maintenance, LPG fuel purchase, and costs of stove promotion programs.

In total, over 3,000 deaths can be avoided annually by full community adoption of ICS and over 6,800 by full adoption of LPG.

ICS provide benefits that are 2.0 to 6.4 times their cost, depending on the country, household cooking location (in house, separate building, or outdoors), and community adoption rate of ICS. Benefit-cost ratios for cooking with LPG are lower than for ICS and in the range of 1.4 to 3.5, reflecting the cost of LPG fuel. But health benefits of cooking with LPG are more than twice as large as cooking with an ICS. Clean energies, such as LPG, are therefore the only option for effectively combatting health effects of solid fuels. In other words, ICS may be the most efficient solution as reflected by the high benefit-cost ratios, but not the most effective solution.

An alternative to LPG is electricity, widely used for cooking in urban areas of Honduras. Cooking with electricity provides the same health benefits of PM2.5 reductions as cooking with LPG. The cost of cooking with electricity would be the same as cost of cooking with LPG if electricity prices were US$ 0.08-0.115 per kWh in the three countries. Benefit-cost ratios of cooking with electricity would therefore be similar to those of LPG at these electricity prices.

The use of solid cooking fuels by one household affects surrounding households. Smoke is vented out of one household for so to enter the dwellings of others and also pollutes the ambient outdoor air. There are therefore benefits from stove promotion programs being community focused with the aim of achieving “unimproved stove free” and eventually “solid fuel free” communities along the lines of community lead sanitation programs and open defecation free communities. Benefit-cost ratios estimated in this paper with full community adoption of ICS or LPG are 15-20% higher than partial adoption of interventions, reflecting reduced community pollution and 30-33% larger health benefits of full adoption.

Benefit-cost ratios of interventions depend on pre-intervention PM2.5 personal exposure levels, and the magnitude of PM2.5 reductions achieved by the interventions. This is influenced by multiple factors, such as characteristics of dwellings, cooking location, cooking practices, and activity patterns of household members. These factors can be positively modified by stove promotion programs to enhance the benefits of cleaner cookstoves.

Post-intervention PM2.5 exposure levels are also influenced by the condition of ICS. Promotion programs need therefore demonstrate and encourage proper use and good maintenance and repairs of stoves.

Promotion of large-scale adoption of cleaner cookstoves has had limited success in Guatemala, Honduras and Nicaragua so far. Some of the factors influencing adoption rates are:

i) High initial cost of ICS and LPG stoves, and high cost of LPG fuel; iii) Need tailoring to consumers’ preferences for stove characteristics; iv) Need installment financing; andv) Need well-targeted information campaigns.

ReferencesBaumgartner, J., Schauer, J., Ezzati, M., et al. 2011. Indoor air pollution and blood pressure in adult women living in rural China. Environmental Health Perspectives, 119(10): 1390-95.

Berrueta, V., Edward, R., and Masera, O. 2008. Energy performance of wood-burning cookstoves in Michoacan, Mexico. Renewable Energy, 33(5), 859–870.

Bonjour, S., Adair-Rohani, H., Wolf, J., Bruce, N.G., Mehta, S., et al. 2013. Solid fuel use for household cooking: Country and regional estimates for 1980-2010. Environmental Health Perspectives, 121(7): 784-90.

Burnett, RT., Pope, CA III., Ezzati, M., Olives, C., Lim, SS., et al. 2014. An integrated risk function for estimating the global burden of disease attributable to ambient fine particulate matter exposure. Environmental Health Perspectives, 122: 397-403.

Clark, M., Bachand, M.,, Heiderscheidt, J., Yoder, S., Luna, B., et al. 2013. Impact of a cleaner-burning cookstove intervention on blood pressure in Nicaraguan women. Indoor Air, 23(2): 105-14.

Clark, M., Reynolds, S., Burch, J., et al. 2010. Indoor air pollution, cookstove quality, and housing characteristics in two Honduran communities. Environmental Research, 10: 12-18.

Clark, M., Peel, J., Burch, J., Nelson, T., Robinson, M., Conway, S., Bachand, A., and Reynolds, S. 2009. Impact of improved cookstoves on indoor air pollution and adverse health effects among Honduran women. International Journal of Environmental Health Research, 19(5): 357-368.

Cynthia, AA., Edwards, RD., Johnson, Rosas, IA., et al. 2010. Indoor particle size distributions in homes with open fires and improved Patsari cook stoves. Atmospheric Environment, 44: 2881-86.

Cynthia, AA., Edwards, RD., Johnson, M., Zuk, M., et al. 2008. Reduction in personal exposures to particulate matter and carbon monoxide as a result of the installation of a Patsari improved cook stove in Michoacan Mexico. Indoor Air, 18(2): 93-105.

Desai, MA., Mehta, S., and Smith, K., 2004: Indoor Smoke from Solid Fuels: Assessing the Environmental Burden of Disease at National and Local Levels. Environmental Burden of Disease Series, No. 4. World Health Organization.

Dherani, M., Pope, D., Mascarenhas, M., Smith, K., Weber, M., and Bruce, N. 2008. Indoor air pollution from unprocessed solid fuel use and pneumonia risk in children aged under five years: a systematic review and meta-analysis. Bulletin of the World Health Organization, 86:390-98.

Fitzgerald, C., Aguilar-Villalobos, M., Eppler,. AR., et al. 2012. Testing the effectiveness of two improved cookstove interventions in the Santiago de Chuco Province of Peru. Sci Total Environ, 420: 54-64.

Forouzanfar, M., Afshin, A., Alexander, L., Anderson, HR., et al. 2016. Global, regional, and national comparative risk assessment of 79 behavioural, environmental and occupational, and metabolic risks or clusters of risks, 1990–2015: a systematic analysis for the Global Burden of Disease Study 2015. Lancet, 388: 1659-724.

Forouzanfar, MH., Alexander, L., Anderson, HR., et al. 2015. Global, regional, and national comparative risk assessment of 79 behavioural, environmental and occupational, and metabolic risks or clusters of risks in 188 countries, 1990–2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet, 386: 2287–323.

GACC. 2014. Guatemala country action plan for clean cookstoves and fuels – annexes. Global Alliance for Clean Cookstoves. www.cleancookstoves.org

Garcia-Frapolli, E., Schilmann, A., Berrueta, V., et al. 2010. Beyond fuelwood savings: Valuing the economic benefits of introducing improved biomass cookstoves in the Purépecha region of Mexico. Ecological Economics, 69(12): 2598–2605.

Habermehl, H. 2007. Economic evaluation of the improved household cooking stove dissemination programme in Uganda. German GTZ.

Hartinger, SM., Commodore, AA., Hattendorf, J., Lanata, CF., et al. 2013. Chimney stoves modestly improved indoor air quality measurements compared with traditional open fire stoves: results from a small-scale intervention study in rural Peru. Indoor Air, 23(4): 342-52.

Helen, St G., Aguilar-Villalobos, M., Adetona, O., et al. 2015. Exposure of pregnant women to cookstove-related household air pollution in urban and periurban Trujillo, Peru. Arch Environ Occup Health, 70(1):10-8. doi: 10.1080/19338244.2013.807761.

Hutton, G., Rehfuess, E., Tedioso, F., and Weiss, S. 2006. Evaluation of the costs and benefits of household energy and health interventions at global and regional levels. World Health Organization.

Jeuland, M. and Pattanayak, S. 2012. Benefits and costs of improved cookstoves: Assessing the implications of variability in health, forest and climate impacts. PLoS ONE 7(2): e30338. doi:10.1371/journal.pone.0030338.

Joon, V., Kumari, H., Chandra, A., and Bhattacharya, M. 2011. Predicting exposure levels of respirable particulate matter (PM2.5) and carbon monoxide for the cook from combustion of cooking fuels. International Conference on Chemistry and Chemical Process 2011. IPCBBE vol. 10: 229-32. Singapore.

Kojima, M., Bacon, R., and Zhou, X. 2011. Who uses bottled Gas? Evidence from households in developing countries. Policy Research Working Paper 5731. World Bank.

Kurmi, OP., Semple, S., Simkhada, P., Smith, WC., and Ayres, JG. 2010. COPD and chronic bronchitis risk of indoor air pollution from solid fuel: a systematic review and meta-analysis. Thorax, 65: 221-228.

Lam N, Africano S, Humphrey S, Pennise D. Indoor Air Quality Impact of an Improved Chimney Wood Stove in Peri- Urban Honduras. 2012.

Li, Z., Sjodin, A., Romanoff, L., Horton, K., et al. 2011. Evaluation of exposure reduction to indoor air pollution in stove intervention projects in Peru by urinary biomonitoring of polycyclic aromatic hydrocarbon metabolites. Environment International, 37: 1157–1163.

Lim, S.S., Vos, T., Flaxman, A.D., Danaei, G., et al. 2012. A comparative risk assessment of burden of disease and injury attributable to 67 risk factors and risk factor clusters in 21 regions, 1990-2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet, 380: 2224-60.

Lindhjem, H., Narvud, S., Braathen, N.A., and Biausque, V. 2011. Valuing mortality risk reductions from environmental, transport, and health policies: A global meta-analysis of stated preference studies. Risk Analysis, 31(9): 1381-1407.

Malla, S. and Timilsina, G. 2014. Household cooking fuel choice and adoption of improved cookstoves in developing countries: A review. Policy Research Working Paper 6903. World Bank.

Masera, O., Edwards, R., Armendariz-Arnez, C., et al. 2007. Impact of Patsari improved cookstoves on indoor air quality in Michoacan, Mexico. Energy for Sustainable Development, 11(2): 45-56.

McCracken, JP., Schwartz, J., Diaz, A., Bruce, N., Smith, KR. 2013. Longitudinal relationship between personal CO and personal PM2.5 among women cooking with woodfired cookstoves in Guatemala. PLOS ONE, 8(2): e55670.

McCracken J, Smith KR, Stone P, Díaz A, Arana B,Schwartz J. 2011. Intervention to lower household wood smoke exposure in Guatemala reduces ST-Segment depression on electrocardiograms. Environ Health Perspect, 119: 1562-68.

McCracken, JP., Smith, KR., Díaz, A., Mittleman, MA., Schwartz, J. 2007. Chimney stove intervention to reduce long-term wood smoke exposure lowers blood pressure among Guatemalan women. Environmental Health Perspectives, 115(7): 996-1001.

Mehta, S., Shin, H., Burnett, R., North, T., and Cohen, A. 2013. Ambient particulate air pollution and acute lower respiratory infections: a systematic review and implications for estimating the global burden of disease. Air Qual Atmos Health, 6: 69-83.

Northcross, A., Chowdhury, Z., McCracken, J., Canuz, E., and Smith, K. 2010. Estimating personal PM2.5 exposures using CO measurements in Guatemalan households cooking with wood fuel. J Environ Monitoring, 12: 873-78.

OECD. 2012. Mortality Risk Valuation in Environment, Health, and Transport Policies. Organisation for Economic Co-operation and Development. Paris, France.

Po, JYT., FitzGerald, JM., and Carlsten, C. 2011. Respiratory disease associated with solid biomass fuel exposure in rural women and children: systematic review and meta-analysis. Thorax, 66: 232-239.

Pollard, S., Williams, D’A., Breysse, P., Baron, P., et al. 2014. A cross-sectional study of the determinants of indoor environmental exposures in households with and without chronic exposure to biomass fuel smoke. Environmental Health, 13: 21.

Pope CA III, Burnett RT, Turner, M, et al. 2011. Lung cancer and cardiovascular disease mortality associated with ambient air pollution and cigarette smoke: shape of the exposure-response relationships. Environmental Health Perspectives, 119(11): 1616-21.

Pope CA III, Burnett RT, Krewski D, et al. 2009. Cardiovascular mortality and exposure to airborne fine particulate matter and cigarette smoke: shape of the exposure-response relationship. Circulation, 120: 941-948.

Pope CA III, Burnett RT, Thun MJ, Calle, E., Krewski, D., Ito, K., and Thurston, G. 2002. Lung cancer, Cardiopulmonary mortality, and Long-term exposure to Fine particulate air pollution. Journal of the American Medical Association, 287: 1132-1141.

Secretaría de Salud [Honduras], Instituto Nacional de Estadística (INE) e ICF International. 2013.Encuesta Nacional de Salud y Demografía 2011-2012. Tegucigalpa, Honduras: SS, INE e ICF International.

Servals. 2012. User feed-back for Servals TLUD biomass gasifier stoves from Sundarbans (West Bengal, India). Servals Automation Pvt Ltd. Chennai, India.

Shin, H., Cohen, A., Pope III, C., Ezzati, M., et al. 2013. Critical issues in combining disparate sources of information to estimate the global burden of disease attributable to ambient fine particulate matter exposure. Working Paper prepared for Methods for Research Synthesis: A Cross-Disciplinary Workshop. October 3, 2013. Harvard Center for Risk Analysis.

Siddiqui, AR., Lee, K., Bennett, D., Yang, X., Brown, KH., Bhutta, ZA., and Gold, EB. 2009. Indoor carbon monoxide and PM2.5 concentrations by cooking fuels in Pakistan. Indoor Air, 19:75-82.

Smith KR, Bruce N, Balakrishnan K, Adair-Rohani H, Balmes J, Chafe Z, Dherani M, Hosgood DH, Mehta S, Pope D, Rehfuess E, et al. 2014. Millions Dead: How Do We Know and What Does It Mean? Methods Used in the Comparative Risk Assessment of Household Air Pollution. Annu. Rev. Public Health, 35: 185-206.

Smith, K., Mehta, S., and Feuz, M. 2004. Indoor air pollution from household use of solid fuels. In: Ezzati, M. et al . (eds): Comparative quantification of health risks: global and regional burden of disease attributable to selected major risk factors. World Health Organization.

Terrado, E. and Eitel, B. 2005. Pilot commercialization of improved cookstoves in Nicaragua. ESMAP Technical Paper 085. World Bank. Washington DC. USA.

Titcombe, M., and Simcik, M. 2011. Personal and indoor exposure to PM2.5 and polycyclic aromatic hydrocarbons in the southern highlands of Tanzania: a pilot-scale study. Environmental Monitoring and Assessment, 180: 461-476.

WHO. 2014. WHO guidelines for indoor air quality: household fuel combustion. World Health Organization. Geneva, Switzerland.

WHO. 2009. Estimated deaths and DALYs attributable to selected environmental risk factors, by WHO member states, 2004. World Health Organization. Downloadable at:http://www.who.int/quantifying_ehimpacts/national/countryprofile/intro/en/index.html

WHO. 2004. Comparative quantification of health risks: Global and regional burden of disease attributable to selected major risk factors. World Health Organization. Geneva, Switzerland.

Winrock. 2008. Peru healthy kitchen/healthy stove pilot project. Winrock International under cooperative agreement with USAID. December 2008.

World Bank and Institute for Health Metrics and Evaluation. 2016. The Cost of Air Pollution: Strengthening the Economic Case for Action. Washington, DC: World Bank.

World Bank. 2016a. World Development Indicators. World Bank. Washington DC. USA.

World Bank. 2016b. Methodology for valuing the health impacts of air pollution: Discussion of challenges and proposed solutions. Prepared by Urvashi, N. and Sall, C. World Bank. Washington DC. USA.

Zuk, M., Rojas, L., Blanco, S., Serrano, P., et al. 2007. The impact of improved wood-burning stoves on fine particulate matter concentrations in rural Mexican homes. J Exposure Sci and Environ Epidemiology, 17:224-32.

Annex 1. Health effects of particulate matter pollution

Health effects of PM exposure include both premature mortality and morbidity. The methodologies to estimate these health effects have evolved as the body of research evidence has increased.

1.1 Outdoor particulate matter air pollution

Over a decade ago, Pope et al (2002) found elevated risk of cardiopulmonary (CP) and lung cancer (LC) mortality from long term exposure to outdoor ambient PM2.5 in a study of a large population of adults 30 or more years of age in the United States. CP mortality includes mortality from respiratory infections, cardiovascular disease, and chronic respiratory disease. The World Health Organization used the study by Pope et al when estimating global mortality from outdoor ambient air pollution (WHO 2004; 2009). Since then, recent research suggests that the marginal increase in relative risk of mortality from PM2.5 declines with increasing concentrations of PM2.5 (Pope et al 2009; 2011). Pope et al (2009; 2011) derive a shape of the PM2.5 exposure-response curve based on studies of mortality from active cigarette smoking, second-hand cigarette smoking (SHS), and outdoor ambient PM2.5 air pollution.

1.2 Household particulate matter air pollution

Combustion of solid fuels for cooking (and in some regions, heating) is a major source of household air pollution (HAP) in developing countries. Concentrations of PM2.5 often reach several hundred micrograms per cubic meter (µg/m3) in the kitchen and living and sleeping environments. Combustion of these fuels is therefore associated with an increased risk of several health outcomes, such as acute lower respiratory infections (ALRI), chronic obstructive pulmonary disease (COPD) and chronic bronchitis (CB), and lung cancer (LC). The global evidence is summarized in meta-analyses by Desai et al (2004), Smith et al (2004), Dherani et al (2008), Po et al (2011), and Kurmi et al (2010). Risks of health outcomes reported in these meta-analyses are generally point estimates of relative risks of disease (with confidence intervals) from the use of fuel wood, coal and other biomass fuels18 relative to the risks from use of liquid fuels (e.g., LPG).

A randomized intervention trial in Guatemala found that cooking with wood using an improved chimney stove, which greatly reduced PM2.5 exposure, was associated with lower systolic blood pressure (SBP) among adult women compared to SBP among women cooking with wood on open fire (McCracken et al, 2007). Baumgartner et al (2011) found that an increase in PM2.5 personal exposure was associated with an increase in SBP among a group of women in rural households using biomass fuels in China. These studies provide some evidence that PM air 18 Other biomass fuels used for cooking is mostly straw/shrubs/grass, agricultural crop residues and animal dung.

pollution in the household environment from combustion of solid fuels contributes to cardiovascular disease.

1.3 An integrated exposure-response function

The Global Burden of Disease (GBD) Project takes Pope et al (2009; 2011) some steps further by deriving an integrated exposure-response (IER) relative risk function (RR) for disease outcome, k, in age-group, l, associated with exposure to fine particulate matter pollution (PM2.5) both in the outdoor and household environments:

RR(x)kl=1 for x < xcf (A1.1a)

RR(x)kl=1+α kl(1−e¿¿−βkl (x−xcf )ρkl)¿ for x ≥ xcf (A1.1b)

where x is the ambient concentration of PM2.5 in µg/m3 and xcf is a counterfactual concentration below which it is assumed that no association exists between PM2.5 exposure and assessed health outcomes (theoretical minimum risk exposure level). The function allows prediction of RR over a very large range of PM2.5 concentrations, with RR(xcf+1) ~ 1+αβ and RR(∞) = 1 + α being the maximum risk (Shin et al, 2013; Burnett et al, 2014).

The parameter values of the risk function are derived based on studies of health outcomes associated with long term exposure to ambient particulate matter pollution, second hand tobacco smoking, household solid cooking fuels, and active tobacco smoking (Burnett et al, 2014). This provides a risk function that can be applied to a wide range of ambient PM2.5

concentrations around the world as well as to high household air pollution levels of PM 2.5 from combustion of solid fuels.

The health outcomes assessed in the GBD Project are ischemic heart disease (IHD), cerebrovascular disease (stroke), lung cancer, chronic obstructive pulmonary disease (COPD), and acute lower respiratory infections (ALRI) (Lim et al, 2012; Mehta et al, 2013; Smith et al, 2014; Forouzanfar et al, 2015; Forouzanfar et al, 2016). The risk functions for IHD and cerebrovascular disease are age-specific with five-year age intervals from 25 years of age, while singular age-group risk functions are applied for lung cancer (≥ 25 years), COPD (≥ 25 years), and ALRI for children and adults in GBD 2013 and 2015.

An xcf between 2.4 and 5.9 µg/m3 is applied in the GBD 2015 Project (Forouzanfar et al, 2016).

The population attributable fraction of disease from PM2.5 exposure is then approximated by the following expression:

PAF=∑i=1

n

P i[RR( x i+ xi−1

2 )−1]/¿¿ (A1.2)

where Pi is the share of the population exposed to PM2.5 concentrations in the range xi-1 to xi.19 This attributable fraction is calculated for each disease outcome, k, and age group, l. The disease burden (D) in terms of annual cases of disease outcomes due to PM2.5 exposure is then estimated by:

D=∑k=1

t

∑l=1

s

mklPAF kl (A1.3)

where mkl is the total annual number of cases of disease, k, in age group, l, and PAFkl is the population attributable fraction of these cases of disease, k, in age group, l, due to PM2.5 exposure.

The potential impact fraction is applied to estimate the reduction in disease burden from a change in the population exposure distribution that can result from an intervention to control PM2.5 exposure levels among a sub-set of the population:

PIF=¿ (A1.4)

where P’i is the population exposure distribution after the intervention. The reduction in annual

cases of disease outcomes is then estimated by:

∆D=∑k=1

t

∑l=1

s

mkl PIFkl (A1.5)

This approach is applied in this paper to estimate the reduction in the disease burden from improved household cooking options.

Annex 2. Valuation of health benefits

2.1 Valuation of mortality

The predominant measure of the welfare cost of a premature death used by economists is the value of statistical life (VSL), estimated from individuals’ willingness-to-pay (WTP) for mortality risk reductions.

As high quality VSL studies only are available from a minority of countries globally, a commonly used approach to estimate VSL in a country that lacks such studies is to use a benefit transfer

19 With a non-linear RR function, the precision of the calculation of PAF increases as x i-xi-1 approaches zero, or “n” approaches infinity.

(BT) based on meta-analyses of VSL studies from other countries. The World Bank (2016b) presents such a benefit transfer methodology for valuing mortality from air pollution, drawing on the empirical literature of VSL, especially OECD (2012). The methodology is applied in the recent publication by the World Bank and IHME (2016) on the global cost of air pollution. The proposed benefit transfer function is:

VSLc ,n=VSLOECD∗(Y c ,nY OECD

)∈

(A2.1)

where VSLc,n is the estimated VSL for country c in year n, VSLOECD is the average base VSL in the sample of OECD countries with VSL studies (US$ 3.83 million), Yc,n is GDP per capita in country c in year n, and YOECD is the average GDP per capita for the sample of OECD countries (US$ 37,000), and ɛ an income elasticity of 1.2 for low- and middle-income countries and 0.8 for high income countries. All values are in purchasing power parity (PPP) prices. VSLc,n must therefore be converted to local currency using PPP exchange rates, available in the World Development Indicators by the World Bank.

Table A2.1. GDP per capita and estimated Value of Statistical Life (VSL) in 2015Unit Guatemala Honduras Nicaragua

GDP per capita US dollar 3,903 2,496 2,087US dollar (PPP) 7,707 5,084 5,190

VSL National currency GTQ 2,260,049 HNL 3,838,339 NIO 3,975,220US dollar (PPP) 582,910 353,837 362,686

US dollar 295,246 173,690 145,843Source: GDP per capita is from World Bank (2016a). VSL is estimate by the author.

2.2 The Value of Statistical Life

VSL is calculated based on individuals’ valuation of changes in mortality risk. Everyone in society is constantly facing a certain risk of dying. Examples of such risks are occupational fatality risk, risk of traffic accident fatality, and environmental mortality risks. It has been observed that individuals adjust their behavior and decisions in relation to such risks. For instance, individuals demand a higher wage (a wage premium) for a job that involves a higher occupational risk of fatal accident than in other jobs, individuals may purchase safety equipment to reduce the risk of death, and/or individuals and families may be willing to pay a premium or higher rent for properties (land and buildings) in a cleaner and less polluted neighborhood or city.

Through the observation of individuals’ choices and willingness to pay for reducing mortality risk (or minimum amounts that individuals require to accept a higher mortality risk), it is possible to estimate the value to society of reducing mortality risk, or, equivalently, measure the welfare cost of a particular mortality risk.

For instance, it may be observed that a certain health hazard has a mortality risk of 2.5/10,000. This means that 2.5 individuals die from this hazard for every 10,000 individuals exposed. If each individual on average is willing to pay US$ 40 for eliminating this mortality risk, then every 10,000 individuals are collectively willing to pay US$ 400,000. Dividing this amount by the risk gives the VSL of US$ 160,000. Mathematically this can be expressed as follows:

VSL = WTPAve * 1/ R (A2.2)

where WTPAve is the average willingness-to-pay per individual for a mortality risk reduction of magnitude R. In the illustration above, R=2.5/10 000 (or R=0.00025) and WTPAve= US$ 40. Thus, if 10 individuals die from the health risk illustrated above, the cost (C) to society is

C = 10* VSL = 10* US$ 0.16 million = US$ 1.6 million (3)

The main approaches to estimating VSL are through revealed preferences and stated preferences of people’s WTP for a reduction in mortality risk or their willingness-to-accept (WTA) an increase in mortality risk. Most of the studies of revealed preferences are hedonic wage studies, which estimate labor market wage differentials associated with differences in occupational mortality risk. Most of the stated preference studies rely on contingent valuation methods (CVM), which in various forms ask individuals about their WTP for mortality risk reduction.

Studies of VSL have been carried out in many countries and several meta-analyses of these studies have been conducted in the last 2-3 decades. Meta-analyses assess characteristics that determine VSL, such as household income, size of risk reduction, other individual and household characteristics, and often characteristics of the methodologies used in the original WTP studies.

Most of the meta-analyses of VSL are entirely or predominantly based on hedonic wage studies. A meta-analysis prepared for the OECD was, however, exclusively based on stated preference studies, arguably of greater relevance for valuation of mortality risk from environmental factors such as air pollution than hedonic wage studies. These stated preference studies are from a database of more than 1,000 VSL estimates from multiple studies in over 30 countries, including in developing countries (Lindhjem et al, 2011; OECD, 2012).

2.3 Valuation of morbidity

Estimating the cost of morbidity requires much less accessible baseline health data than estimating mortality. It also includes data on work absenteeism and frequency and cost of medical treatment. The willingness to pay to avoid pain and suffering can also be added to this cost, but estimates are generally not available for most countries.

One option is to use the estimates of years lost to disease (YLD) from the relevant illnesses associated with household air pollution (HAP) available from the GBD Project for each country. YLD can then be converted to days of illness by applying the disability weights in the GBD Project. And the cost of days of illness can then be estimated.

In this report, the cost of a day of illness is estimated as 50% of average labor income. This figure is also applied to individuals without income, because illness prevents most of these individuals from undertaking household work and other activities with a social value.

collaboration.worldbank.org · Web viewPrevalence rates of solid fuel use for cooking in Guatemala,...

Documents

Transcript of collaboration.worldbank.org · Web viewPrevalence rates of solid fuel use for cooking in Guatemala,...