MONITORING INDOOR AIR POLLUTION

Photo by Susmita Dasgupta

By

SUSMITA DASGUPTA MAINUL HUQ M. KHALIQUZZAMAN CRAIG MEISNER KIRAN PANDEY DAVID WHEELER OCTOBER 2004

 Introduction Indoor air pollution from burning wood, animal dung and other bio-fuels is a major cause of acute respiratory infections (ARI), one of the most important causes of childhood death in developing countries (Murray and Lopez, 1996). Acute lower respiratory infection (ALRI), the most serious type of ARI, accounts for 20% of the estimated 12 million annual childhood deaths under five, and about 10% of perinatal deaths (WHO, 2001; Bruce, 1999). Nearly all of these deaths occur in developing countries, with the heaviest losses in Asia (42% of total deaths) and Africa (28%) (Murray and Lopez, 1996). Through its effect on respiratory infection, indoor air pollution (IAP) is estimated to cause between 1.6 and 2 million deaths per year in developing countries (Smith, 2000). The majority of these deaths are in poor households, with approximately 1 million being children (Smith, 1993; Smith, et al., 1993; Smith and Mehta, 2000). Table 1 provides estimates of health damage from IAP by region. Table 1: Annual Disease Burden From Indoor Air Pollution (Early 1990’s) Region

Deaths (‘000)

China India Sub-Saharan Africa Other Asia & Pacific Islands Mid-East and North Africa Latin America & Caribbean Total

516.5 496.1 429.0 210.7 165.8 29.0 1,800.0

Illness Incidence (‘000,000) 209.7 448.4 350.7 306.4 64.2 58.2 1,400 .0

DALYs1 (‘000,000) 9.3 16.0 14.3 6.6 5.6 0.9 53.0

Source: World Bank (2002), drawing on Smith and Mehta (2000) and Von Schrinding, et al., (2001).

The magnitude of IAP’s estimated impact has prompted the World Bank (2001) and other international development institutions to identify the reduction of indoor air pollution as a critical objective for the coming decade. To promote a better understanding of IAP, the World Bank’s research department recently completed an IAP study, in Bangladesh, monitoring key determinants of IAP, and using the latest air monitoring technology and a national household survey. The purpose of this note is to summarize the study’s experience with monitoring IAP exposure and to act as a guide for World Bank task managers and policy makers in the identification and targeting of suitable interventions.

1

DALYs, or disability-adjusted life years, combine life-years lost from premature death and fractional years of healthy life lost due to illness and disability (Murray and Lopez 1996).

 Minute Particles, Major Problems: By definition, indoor air pollution is a complex mixture of small and large particles. Recent epidemiological studies have reported that exposure to particulates, particularly small particulates, is strongly associated with respiratory illness and death. Small particles are likely to be more dangerous, since they can be inhaled deeply into the lungs and settle in areas where natural clearance mechanisms, like coughing, cannot remove them. The current scientific consensus is that most respiratory health damage comes from the inhalation of respirable particles whose diameter is less than 10 microns (PM10), with recent attention focusing on even finer particles (PM2.5).  Monitoring IAP Exposure: Exposure to particulates depends on the level of emissions and the length of time that particulates remain in indoor air.  In particular, the extent and duration of smoke in the kitchen, and the amount of smoke leaking from the kitchen and infiltrating into other rooms (or other portions of the house or outdoors) depends highly on the type of fuel2, efficiency of the stove3, duration of cooking, location of the kitchen4, the extent of ventilation5, and materials used in construction6 (hence porosity) of walls and roofs of the kitchen.  In addition, the overall decay function of indoor pollution depends on ventilation characteristics of the house, such as the number of rooms, number, size and placement of doors and windows, and the materials used in the construction of walls and roofs, etc. Previous IAP studies identified several potential determinants of exposure to indoor air pollution, including fuel type, time spent cooking, structural characteristics of houses, and household ventilation practices (opening of windows and doors, etc.) (World Bank, 2002; Brauer and Saxena, 2002; Moschandreas et al., 2002; Freeman and Sanez de Tajeda, 2002). All of these factors may be important in any developing country context, where households exhibit a significant diversity in cooking fuels, stove types, cooking locations, and quality of ventilation.

2

Fuel types could include: gas, electricity, kerosene, fire wood, twigs, leaves ,dung, agricultural residues, sawdust, etc. 3 Efficiency of stove could include: traditional stove or improved. 4 Kitchen locations could include: no separate kitchen from the rest of the house, separate but attached kitchen, separate but detached kitchen, outside or open kitchen, etc. 5 Extent of ventilation could include: number/size/location of doors and windows and how long they are open during and after cooking. 6 Construction materials could include: brick, mud, tin, thatch roof and walls.

Roadmap to measuring Indoor Air Pollution: Selection of 1. Household Characteristics: (list of household characteristic by which IAP may vary) - fuel type - kitchen type - construction material 2. Monitored Households: (separate households into groups defined by these characteristics, and select samples independently from each group. )

IAP Monitoring Monitoring PM10 and PM2.5 over a period of 24-hours in at least two locations for each household (kitchen and living room)

IAP Determinants Survey Collect information on household structural characteristics, cooking and ventilation behavior and socio-demographics.

Time-Activity Pattern Survey Collect individual information on time spent at different locations within and outside the household.

Estimation of IAP concentration & exposure 1) Regression analysis of the determinants of PM10 and PM2.5 concentration. 2) Estimation of exposure using time-activity patterns and average concentration.

National Survey and Extrapolation 1) Collect information on key determinants of IAP and time-activity patterns of the nationally representative sample. 2) Use regression estimates to extrapolate IAP concentration and exposure to the national sample.

3a. Developing a Sampling Strategy: Monitoring indoor air quality is often time consuming and requires a sampling strategy which will be representative of the extent of the problem. One highly recommended method is to stratify the sample of households for indoor air monitoring. This method has the triple advantage of being cost-effective, diverse and representative.7 The concept of stratification is basically to select only those households in which key determinants vary; and to select a representative sample of households possessing those characteristics. A typical starting point for identifying stratifying variables (e.g such as fuel use pattern, kitchen type and location, construction materials) is from the Ministry of Energy and/or the Ministry of Housing census. In the absence of detailed secondary data, preparation of a complete list of combinations, even through on-the-ground reconnaissance, is one of the first steps. A typical stratification (or grouping) of households in the study area would be by cooking fuel, kitchen type and location and construction material of the house. Representative samples would then be selected and independently drawn from each group, with the number of selected households for each combination being large enough to avoid small sample-related problems. In each selected household, residential/ indoor air should be monitored continuously, over the course of a 24-hour period, in at-least two locations: the kitchen and living room. 3b. Monitoring IAP (Minute Particle) Concentrations The World Bank study in Bangladesh used two types of equipment: real-time monitors that recorded PM10 at 2-minute intervals, and air samplers that measured 24-hour average PM10 and PM2.5 concentrations. Each of the instruments are described below. 1. The real-time monitoring instrument used was the Thermo Electric Personal DataRAM (pDR-1000) (Thermo Electron, 2004). The pDR-1000, or PD-RAM, uses a light scattering photometer (nephelometer) to measure airborne particle concentrations.8 At each location, the instrument operated continuously, without intervention, for a 24hour period to record PM10 concentrations at 2-minute intervals. With data-logging enabled, the instrument automatically tags and time-stamps the data collected, and stores it for subsequent retrieval, printing, or graphing through a computer. Fixed Cost: $3,303 (machine) $675.00 (for 2 batteries) (Sept., 2003) Variable Cost: Recharging of batteries For more information: Thermo Electric Personal DataRAM (pDR-1000) 7

Due to clustering of similar houses, a random sample may not capture the diversity commonly observed in developing countries. 8 The operative principle is real-time measurement of light scattered by aerosols, integrated over as wide a range of angles as possible.

2. The second instrument used in the study was the Airmetrics MiniVol Portable Air Sampler (Airmetrics, 2004), a more conventional device that samples ambient air for 24 hours. While the MiniVol is not a reference method sampler, it gives results that closely approximate data from U.S. Federal Reference Method samplers. The MiniVols were programmed to draw air at 5 liters/minute through PM10 and PM2.5 particle size separators (impactors) and then through filters. The particles were caught on the filters, and weighed pre- and post exposure with a microbalance. The air samplers for PM10 and PM2.5 also operated for a 24-hour period. Fixed Cost: $1,890.00 MiniVol sampler (Aug., 2003) Variable Costs: (As of Aug., 2003) a) Impactor Grease: $44.00 The impactor is cleaned every seventh run (or more often if heavy soiling is observed); b) Pure Teflon Filters: $280.00/box (50 filters/box) Especially recommended if chemical analysis for non-carbon based compounds is to follow gravimetric analysis; c) Petrislides: $44.00 box (100/box) Plastic cassettes for shipping and storage of filters; d) Weighing of filters pre- and post- exposure. For more information: Airmetrics Minivol Both instruments are silent, unobtrusive and easy to operate. At each location, the instruments operate continuously without personal intervention for a 24-hour period, and the readings of PD-RAM and MiniVol air sampler provided a detailed record of IAP concentration in each household. Calibration and Quality Control: Before a monitoring program is undertaken, all sampling and analysis equipment must be properly calibrated. Performing periodic audits is also essential to ensure the integrity and assess the accuracy of sampling data. For important lessons learned from the field on the operation of the air monitors, see Appendix I. Laboratory Analysis: Pure Teflon filters used to collect particulates were weighed preand post-exposure with a microbalance accurate to one microgram by AirMetrics (QA as per US EPA Region 10 protocol).

3c. Monitoring Determinants of IAP Concentration (IAP Determinants Survey) In addition to the actual air monitoring, a determinants survey should also be conducted. The purpose of this survey is to get a reliable understanding of all major sources of minute particulates in residential indoor air (as well as the exposure setting (where exposure occurs) and duration). Structured questionnaires should collect information on the characteristics of the house, house ventilation related factors (e.g., number, size, location of doors and windows, length of time doors and windows are kept open during and after cooking), and other possible sources of smoke (cigarettes smoked, lanterns lit, mosquito coils, incense burnt, etc.). In addition, all relevant socio-economic information should be elicited from all household occupants. 3d. Monitoring IAP Exposure (Time-Activity Pattern Survey) In addition to the actual air monitoring and the household characteristics survey above, it is also very important for investigators to better understand the time-activity patterns of individuals within the house, one of the key elements of exposure. Individuals’ exposure may vary significantly because they spend very different amounts of time in cooking areas, living areas, and outside the house. For the purposes of analysis, exposure can be analyzed at two levels: differences within households attributable to family roles, and differences across households attributable to income and education. Ideally, detailed time-activity records would be required by all members of a family (infants, children, adolescents, adults, elderly- all by gender) during their daily rounds of along with socio-economic characteristics to monitor IAP exposure.

3e. Analysis of Determinants of IAP and Estimation of Exposure Conduct a regression analysis to explore the relationships between PM10, PM2.5 concentrations, fuel choices and a large set of variables that describe household cooking and ventilation practices, structure characteristics and building materials. Regression analysis would identify the key determinants of IAP. Combine monitored PM10 and PM2.5 concentrations at different locations with the timeactivity patterns (from the Time-Activity Survey) to estimate individuals’ exposure to IAP. For application of the methodology, see (Dasgupta et al, 2004a; Dasgupta et al, 2004b).

4. Regional or Nation-wide Extrapolation As a final step to analyzing the impact of IAP from a broader perspective (e.g. for a region or nationally), the estimates of IAP concentration and exposure from sample households can be extrapolated using household characteristics from a regional/ national

survey. This will measure the seriousness of IAP pollution and aid in the identification of “hotspots” in larger rural, peri-urban or urban areas. Policymakers can then cater suitable interventions in these areas. Selection of Nation-wide Representative Sample of Households For the selection of regional or nation-wide representative households, a random sampling technique across the total population would be recommended. Collection of Data Necessary for Extrapolation A survey needs to be administered to a sample of regional or nation-wide representative households to collect information on key IAP determinants identified during estimation (such as cooking and ventilation behavior, structural characteristics and building materials, socio-economic characteristics of the households). For application of the methodology, see (Dasgupta et al, 2004a; Dasgupta et al, 2004b). References Brauer, M. and S. Saxena, 2002, “Accessible Tools for Classification of Exposure to Particles,” Chemosphere, 49: 1151-1162. Bruce, N., 1999, Lowering Exposure of Children to Indoor Air Pollution to Prevent ARI: the Need for Information and Action, Capsule Report (3), Environmental Health Project, Arlington VA. Dasgupta, S., M. Huq., M. Khaliquzzaman, K. Pandey, and D.Wheeler, 2004, “Indoor Air Quality for Poor Families: New Evidence from Bangladesh,” World Bank, Development Research Group Working Paper No. 3393, August. Dasgupta, S., M. Huq., M. Khaliquzzaman, K. Pandey, and D.Wheeler, 2004, “Who Suffers from Indoor Air Pollution?: Evidence from Bangladesh,” World Bank, Development Research Group Working Paper , September. Freeman, N. C. G. and S. Saenz de Tejada, 2002, “Methods for Collecting Time/Activity Pattern Information Related to Exposure to Combustion Products,” Chemosphere, 49: 979-992. Moschandreas, D.J., J. Watson, P. D’Aberton, J. Scire, T. Zhu, W. Klein, and S. Saxena, 2002, “Methodology of Exposure Modeling,” Chemosphere, 49: 923-946. Murray, C. and A. Lopez (eds.), 1996, The Global Burden of Disease, Cambridge MA: Harvard School of Public Health, WHO, World Bank.

Smith, K., 2000, “National Burden of Disease in India From Indoor Air Pollution,” Proceedings of the National Academy of Science USA, (97) 13286-13293. Smith, K. and S. Metha, 2000, The Global Burden of Disease from Indoor Air Pollution in Developing Countries: Comparison of Estimates, Prepared for the WHO/USAID Global Technical Consultation on Health Impacts of Indoor Air Pollution in Developing Countries. Smith, K., 1993, “Fuel Combustion, Air Pollution Exposure, and Health: the Situation in Developing Countries," Annual Review of Environment and Energy, 18:526-566. WHO, 2001, Informal Consultation on Epidemiologic Estimates for Child Health 11-12 June 2001, Department of Child and Adolescent Health Development, WHO, Geneva (http://www.who.int/child-adolescent-health/New Publications/Overview/). World Bank, 2002, “India: Household Energy, Indoor Air Pollution, and Health,” ESMAP / South Asia Environment and Social Development Unit, November. World Bank, 2001, Making Sustainable Commitments: An Environment Strategy for the World Bank, July.

APPENDIX I: EXPERIENCE FROM THE FIELD CALIBRATION AND QUALITY CONTROL Technical Advisor: Dr. M. Khaliquzzaman, Group Leader of Bangladesh Atomic Energy Commission's Program on Air Pollution (Retired)

Calibration and quality control Technical Advisor: Dr. M. Khaliquzzaman, Group Leader of Bangladesh Atomic Energy Commission's Program on Air Pollution (Retired)

In the measurements for air quality parameters, a lot of uncertainties of different origins are involved and this has led to the emergence of Quality Control (QC) and Quality Assurance (QA) as major issues in such measurements. Worldwide experiences have shown that the validation of the methods used for any major work in this area must be established through well defined procedures. In such cases, just good knowledge of the fundamentals of the techniques are not necessarily sufficient to get good data. Meticulous attention to the details is needed on the part of the operators to ensure quality results. Quality of the data from measurements refers to the attributes such as usefulness for the intended purpose, absence of errors and worth the price of gathering the data. It is

essential that the measurements conform to defined accuracy and precision. Quality Control refers to a system of activities whose purpose is to control the quality of the data so that the data meets the need of the users. The aim is to provide quality that is adequate, satisfactory and economic. Another very important considerations in measurements is Quality Assurance which refers to a system of activities to provide the assurance that the data obtained from the measurement meets defined standards of quality with a stated level of confidence. Quality Assurance is achieved through Quality Assessment. The system of activities that fall under Quality Control are:  Routine operations based on SOP.  Establishment of calibration of equipment.  System maintenance and support.  Data review and management. In order to achieve proper routine operation, it is essential to train the operation staff sufficiently so that they are technically competent and written standard operations procedures (SOP) must be made available to them. The calibration of the equipment are needed to ensure precision and accuracy of the data obtained. It should be noted that precision refers to reproducibility in repetitive measurements and accuracy refers to nearness of the values obtained in measurements to actual values. Precision is usually determined through repetitive measurements. Determination of accuracy is more involved which may include interchange of operators, interchange of equipment, use of independent techniques, use of control charts and collaborative test among others.

PD-RAM Calibration In real-time measurements using PD-RAM, the purpose in most cases is to compare PM levels in two or more different locations. So, the concern here is precision rather than accuracy. The numbers obtained from the PD-RAMs depend on the calibration factor inserted during setup procedure. The units come with calibration settings from the factory. It is essential to perform co-located measurements using all the units to be used in measurements. These measurements should be done for a sufficiently long time so that statistical fluctuations are low. For example, a set of measurements over a period of 24 hours with 10 minutes’ logging interval may be appropriate. The scatter plot of data obtained from two units should show a slope value of unity (or very close such as 0.95 or 1.05 etc), low intercept (close to zero) and high R-squared value (>0.95). In case, a significantly different value from unity for the slope is obtained calibration of one of the unit has to be adjusted. The calibration of the unit which provide values closer to filter based monitors should be left unchanged and others should be adjusted to yield unit slope in the plots. In case the intercepts are significantly different from zero, the zeroing of the units should be done using the procedure in the instrument manual using the zero air

pouch supplied by the manufacturer. The calibration process should be repeated every two weeks or so. If the concentration measured is low, then the plots may show too much scatter in the plots. In such cases, a number of data points may be added to reduce statistical fluctuations and analysis may be done using the new data obtained by the combination process. If PD-RAM data are to be used for exposure determination, the PD-RAMs data must be calibrated by comparison with the MiniVols data in co-located statistically valid measurements.

MiniVol Sampler Calibration and Operation The MiniVol samplers are known to yield PM data which closely match USFRM samplers. It is essential to maintain the designed flow rate to maintain the cut off diameter of the particulate in the impactor at the inlet of the samplers. The flow rate of all the samplers should be checked using a NIST traceable flow calibration unit before the start of any sampling campaign, so as to ensure proper flow rate during data collection. Inspection and maintenance of the pump components must be done regularly to ensure proper pump operation at all times. Changes in the sound level is a good indicator for the malfunction of the pumps. One of the most common malfunction mode of the MiniVols is the partial failure of the pump diaphragms. This can be repaired easily by replacing the diaphragms. After any repair, the pump flow rate must be checked with a flow calibrator. Although details of operation of the samplers are given in the instruction manual supplied by the manufacturers, it is not convenient for the operators to follow this manual. A simplified SOP has to be provided for them which is more convenient. A sample SOP is given below.

SOP for PM Sampling with MiniVol Samplers At the Office Loading Filter in the Filter Assembly 1. Clean and Grease Impactor : Apply grease solution using dropper to the surface after it is cleaned with solvent. This should be done on first use and then every seventh day or when the impactor surface looks dirty. 2. Select a pre-weighted filter paper and remove cover from petri slide. 3. Using forceps, install the new filter in the filter holder. 4. Place the filter cassette in the filter holder. 5. Replace pre-separator adapter and screw it down. 6. Place an identifying tag on the filter holder with filter number. 7. Place the assembly in a clean plastic bag. Retrieval of Sample Filter from the Filter Assembly 1. Unscrew the filter holder and remove filter cassette

2. 3. 4. 5.

Locate the petri slide with filter number which matches the number on the holder. Use cassette separator (P/N 600-007) to remove top half of the filter cassette. Using forceps remove exposed filter and place it into the petri slide and store it. Remove ID tag from the filter holder assembly.

Battery Charging 1. 2. 3. 4.

Connect the charging plug of the charger to the battery pack. Plug the charger to an AC outlet. The green LED will light up indicating charging is complete. Disconnect battery when fully charged. It may take up to 18 hours for full charging. Measure the battery voltage with VOM to make sure the full charging.

Making Grease Solution and Application 1. Add one inch length of grease to 30 ml solvent in a glass bottle and shake vigorously. 2. Unmate impactor from pre-seperator. 3. Rinse impactor with solvent using from the squeeze bottle and wipe with a tissue. 4. Let the impactor dry. 5. Put 2/3 drops of grease solution using a dropper on the impactor surface and allow it to dry. 6. Place the impactor back in the assembly.

AT THE SAMPLING SITE

Starting Sampling 1. Place the sampler and freshly charged battery pack on a firm level surface. 2. Remove the clean Filter holder from the plastic transport bag. Attach the assembly at the sampler top quick connect. 3. Record the information on Field Data Sheet ( Site, date, number of filter, battery ID, Sampler ID, Temperature, Pressure etc). 4. Place the sampler on the charged battery pack matching the banana jacks and clamp the two side latches. 5. Lift the pump and timer assembly and rest it on the edge of the sampler casing. Hold the assembly by top cap and take care not to pull the wires. 6. Take the beginning flow rate. To do this press ON/AUTO/OFF button to start the pump. The LCD display, the horizontal bar should move to ON.

7. Check that flow meter reads 5 while it is vertical. If not adjust the control to make flow 5. Low flow cut off adjustment is to set at 90 degree. Record the information on Field Data Sheet. 8. Press ON/AUTO/OFF button twice to stop pump. 9. Program the timer to turn the sampler on and off. Enter these data on the field data sheet. 10. Press ON/AUTO/OFF button to set the timer to ‘Auto’. 11. Place the pump and timer assembly in the sampler body and replace the bale assembly bar. 12. Hoist the pump to sampling position. Retrieving Samples 1. Place the sampler assembly-Battery on a firm level surface on removal from the sampling position. 2. Unscrew the cap of the sampler and lift the pump and timer assembly and rest it on the edge of the sampler casing. Hold the assembly by top cap and take care not to pull the wires. 3. Check the sampler face plate for any error condition and recor it on the field data sheet. 4. Verify the correct time and day of the week on time LCD. 5. Record Elapsed Time on the Elapsed Time Totalizer. 6. Obtain end flow rate: Press ON/AUTO/OFF button twice to start the pump. With flowmeter in a vertical position, record flow rate to the nearest 10th of liter/min. Press ON/AUTO/OFF button twice to stop the pump. 7. Remove the filter holder assembly and put it in the plastic bag. 8. FOLLOW THE SAMPLING START PROCEDURE FOR NEW SAMPLING Error Codes ( To be noted when needed) B: Battery Failure D: Damaged filter F: Low Flow M: Missing Field Data S: Sampler Malfunction T: Timer Malfunction Supplies Required 1. Teflon tipped forceps 2. Screw driver set

3. 4. 5. 6. 7. 8. 9.

Solvent Measuring cylinder Squeeze bottle Grease solution Dropper VOM (Volt Ohm Meter) Electronic Barometer ( Pressure and Temperature).

LESSONS LEARNED Consultant in charge of Field IAP Concentration Monitoring: Mr. Mainul Huq, CEO, Development Policy Group, Dhaka, Bangladesh (email: [email protected])  MiniVol and PD-RAM are sophisticated pieces of equipment. The operator should have sound knowledge about the systems, otherwise, the person would neither be able to diagnose the problems nor fix them.  As air flow rate in the MiniVol air sampler is a function of temperature and barometric pressure, it is absolutely necessary to perform the flow rate calibration before each sampling project. If temperature and/or pressure changes considerably, the process of calibration has to be repeated.  In order to get the exact air flow rate from the MiniVol air sampler, operators should handle the adjustment with great precision. Sometimes, failing this, one needs to use a manometer.  High humidity level, fog, and heavy rainfall affect the air flow rate which may prematurely shut down the MiniVol sampler. In such an event, the experiment needs to be repeated.  When the filter is to be retrieved from the filter holder (MiniVol), a trained person is required so that no change in the amount of collected particles on the filter takes place.  With repeated use, pump valves and diaphragms become dirty and get worn out (MiniVol). They should be checked periodically and cleaned/replaced as necessary.  Protect the PD-RAM from rain. Although the manual says there should not be any problem, the reality is once they were soaked in rainy water, they are likely to provide wrong readings.  Do not install PD-RAM on a high place from where it can fall. The equipment is not as sturdy as it looks.

 When a MiniVol / PD-RAM is installed in a kitchen, it should be installed on the cook side of the stove.  The MiniVol / PD-RAM should not be placed near a door or window otherwise the result will be biased.  Try to avoid placing the PD-RAM right on the floor; if the floor gets washed for some reason, the equipment could be damaged.  When installed in a kitchen or other room, the relevant persons should understand not to touch or handle the equipment in any way. Careless or untrained handling might cause low flow or inflict damage to the equipment.  Security of the equipment from theft, tempering, and unauthorized handling is a necessary condition before installing any equipment.