Household Air Pollution: the Scope of the Problem. The Health Impacts of Humanity s Oldest Occupation

Transcription

1 Household Air Pollution: the Scope of the Problem The Health Impacts of Humanity s Oldest Occupation Kirk R. Smith Professor of Global Environmental Health University of California Berkeley

2 The Environmental Health Risk Transition Household, e.g. poor sanitation Severity Community, e.g. Urban Pollution Global, e.g. greenhousegas emissions Increasing Wealth Shifting Environmental Burdens Local Immediate Global Delayed Smith, 1990

3 Four points Why we do not call it indoor air pollution anymore Although poverty is the problem, poverty alleviation is not the answer Important role as a risk factor for NCDs in the poorest populations globally A fundamental shift in thinking about exposure-response for combustion mixtures

4 thousand years ago, hearths became a regular feature in human habitation On the earliest evidence for habitual use of fire Roebroeks and Villa, PNAS, 2011

5 Million Peope World Population Using Solid Fuels World Pop Solid Fuel Users

6 Biomass Cooking in History Today, ~40% use solid fuels, about 2.7 billion people Although the percentage is dropping, the absolute number is still rising. Indeed, there are more people using solid fuels today for cooking than the total world population in 1950 Or any year previously

7 A problem that has lasted one-third of a million years and is showing no sign of quickly going away by itself.

8 The three major solid fuels

9 Toxic Pollutants in Biomass Fuel Smoke from Simple (poor) Combustion Small particles, CO, NO 2 Hydrocarbons 25+ saturated hydrocarbons such as n-hexane 40+ unsaturated hydrocarbons such as 1,3 butadiene 28+ mono-aromatics such as benzene & styrene 20+ polycyclic aromatics such as benzo( )pyrene Oxygenated organics 20+ aldehydes including formaldehyde & acrolein 25+ alcohols and acids such as methanol 33+ phenols such as catechol & cresol Many quinones such as hydroquinone Source: Naeher et al, J Inhal Tox, 2007 Semi-quinone-type and other radicals Chlorinated organics such as methylene chloride and dioxin

10 Health-Damaging Air Pollutants From Typical Woodfired Cookstove in India. Typical Health-based Standards Wood: 1.0 kg Per Hour in 15 ACH 40 m3 kitchen Typical Indoor Concentrations Carbon Monoxide: 150 mg/m3 Particles 3.3 mg/m3 Benzene 0.8 mg/m3 1,3-Butadiene 0.15 mg/m3 Formaldehyde 0.7 mg/m3 10 mg/m3 0.1 mg/m mg/m mg/m3 0.1 mg/m3 Best single indicator IARC Group 1 Carcinogens

11 First person in human history to have her exposure measured doing the oldest task in human history How much exposure? Kheda District, Gujarat, 1981

12 ~70% use solid fuels Estimated PM2.5 Exposure for solid fuel using households USEPA Standard 15 ug/m3 WHO Guideline ug/m3 24 Hrs PM2.5 Concentration (mic.g/m3) ~285 ug/m3 average Unknown Preliminary result from CRA Balakrishnan et al.

13 ALRI/ Pneumonia Diseases for which we have epidemiological studies COPD Lung cancer (coal) These three diseases were included in the 2004 Comparative Risk Assessment Managed and published by WHO First ever comprehensive risk assessment with consistent rules of evidence and common databases

14 2000 ARI Perinatal HIV Depression Cancer 1/6 is lung Diarrhea Heart (Is chaem ic) Child Clus ter Malnutrition Stroke Road Traffic Malaria TB Maternal COPD World DALYs Congenital Percent of Total

15 Global Burden of Disease from T op 10 Risk Factors plus selected other risk factors Percent of All DALYs 0.0% 2.0% 4.0% 6.0% 8.0% 10.0% Underweight Unsafe sex Blood pressure Tobacco Alcohol Unsafe water/sanitation Child cluster vaccination* Cholesterol Lack of Malaria control* Indoor smoke from solid fuels 1.6 million premature deaths/year Overweight Occupational hazards (5 kinds) Road traffic accidents* Physical inactivity Lead (Pb) pollution Urban outdoor air pollution Climate change Chernobyl per month 0.8 million premature deaths/year 0.12 million premature deaths/year

16 Global Burden of Disease Database and Comparative Risk Assessment World Health Organization Being completely updated For 2011 release For household air pollution: New exposure assessment modeling New outcome estimates based on meta-analyses ALRI, COPD, Lung Cancer Low birth weight, cataracts, cardiovascular

17 ALRI/ Pneumonia Low birth weight Stillbirth Diseases for which we have epidemiological studies COPD Lung cancer (coal) Lung cancer (biomass) Blindness (cataracts, opacity) CV disease Blood pressure ST-segment These diseases were included in the 2004 Comparative Risk Assessment These additional diseases are included in the 2011 Comparative Risk Assessment

18 Risk of COPD: Vented vs. unvented coal stoves Xuan Wei County China, retrospective cohort, , 20,453 subjects 81% added chimneys Chapman et al. Br Med J 2005; 331: 1050.

19 Preliminary CRA Effect Estimates Health Outcome Sex Age Level of Outcome Risk Estimate ALRI M & F < 60 mo Ia 1.78 (1.45 to 2.18) ALRI: M&F < 60 mo Ib 2.3 (95% CI?) exposure/response COPD F >15 yr Ia 2.7 (1.95 to 3.75) COPD M >15 yo Ia 1.9 (1.15 to 3.13) Lung Cancer (coal) F > 15 yr Ia 1.98 (1.16 to 3.36) Lung Cancer (coal) M > 15 yr Ia* 1.38 Cataract F > 30 yr Ia 2.45 (1.61 to 3.73) Cataract M > 30 yr Ia? LBW (OR) M & F Perinatal Ia 1.52 (1.25 to 1.80) LBW (mean weight) M & F Perinatal Ia 93.1g (64.6, 121.6) Lung Cancer (biomass) F > 15 yr Ia 1.81 (1.07 to 3.06) Lung Cancer (biomass) M > 15 yr Ia 1.26 (1.04 to 1.52) CVD F > 30 yr Ib 1.3 to 1.4 (95% CI) CVD M > 30 yr Ib* 1.16

20 Story of Two Conferences Air pollution conference High exposures to large vulnerable population No more health effects work needed International health conference Still doubt about causality Need to know exact benefit to be expected Where are your randomized controlled trials?

21 History of an RCT ~1980: Case reports of health effects in South Asia 1981: First measurements of pollution levels in India 1984: International meeting to decide on needed research Chose randomized controlled trial (RCT) of ALRI : Unfunded proposals to do RCT in Nepal 1990: WHO establishes committee to find best sites : Criteria established and site visits made 1992: Highland Guatemala chosen : Pilot studies to establish data needed for proposal does stove work and do people use it? : Unfunded proposals 2001: NIEHS funding secured : Fieldwork completed 2011: Main results published 25+ years from deciding to conduct RCT to results!

22 RESPIRE Randomized trial (n=518) Impact on pneumonia up to 18 months of age Traditional open 3-stone fire: kitchen 48-hour PM 2.5 levels of μg/m 3 Chimney wood stove, locally made and popular with households

23 RESPIRE Results (Randomized Exposure Study of Pollution Indoors and Respiratory Effects) Intention-to-Treat analysis of the RCT forthcoming, but under embargo Will present preliminary results of the exposure-response analysis, which is most relevant to this audience

24 Tubito Tubito

25 Kitchen CO concentration (ppm) Guatemala RCT: Kitchen Concentrations (c) Control open fire Effect of Chimney Stove On Kitchen CO Levels Intervention Chimney Stove Factor of ~10 less h measurements Smith, et al, Months

26 Child CO exposure (ppm) Infant Exposures (a) Control Intervention h measurements Effect of Chimney Stove On Infant Exposures - 2x less Smith, et al, Time relative to intervention (months)

27 (b) ention group rol group Children with chimney stove ~ 50% reduction in exposure on on average Children with open fire Chimney stove did not protect all children (ppm) (ug/m3 PM 2.5 ) Child Average of Log of CO (ppm) -2-1 Predicted Long

28 Preliminary Adjustments for Exposure-Response Model Adjusted for child s age (quadratic), sex, birth interval less than 2 yr (yes/no), mother s age (quadratic), maternal education and paternal education (none/primary/ secondary), secondhand tobacco smoke exposure (yes/no), latrine (yes/no), piped water (yes/no), electricity (yes/no), kerosene lamp (yes/no), wood-fired sauna (yes/no), bedroom in kitchen (yes/no), roof type (metal sheet/tiles/straw), earth floor (yes/no), asset index (linear over range 0 to 6), animal ownership index (linear over range 0 to 4), crowding index (people per room), altitude (5 categories), occupation (farm other land/farm own land/other), and season (cold dry, warm wet, warm dry).

29 Preliminary MD-diagnosed (C) Pneumonia Approximate Mean PM2.5 exposure in 100s of ug/m3

30 MD-diagnosed (D) Severe Pneumonia Preliminary

31 X-ray-confirmed (E) Pneumonia Preliminary Approximate Mean PM2.5 exposure in 100s of ug/m3

32 Severe (F) X-ray-confirmed Pneumonia Preliminary Approximate Mean PM2.5 exposure in 100s of ug/m3

33 RESPIRE: Pneumonia Reductions with Exposure Reduction Preliminary Results RESPIRE - Guatemala

34 Kitchen concentrations down by 10x, but children exposure down by only 2x, because --Time-activity: the kids do not spend their entire day in the kitchen --Household (or neighborhood ) pollution: a chimney does not reduce smoke, but just shifts it outside into the household environment, where the difference between intervention and control households was less

35 20-month average ground-level PM2.5 from satellite data Large areas of rural India and China have high ambient air pollution much from household fuel

36 NASA INTEX_B Database Percent PM 2.5 emissions from households 0% 53% 30% Percent of PM2.5 emissions from households Chafe, 2010; data from 36 NASA INTEX_B 2006

37 Climate Warming in 2020 Under Present Trends Household Biomass Unger et al. PNAS, 2010

38 Heart Disease and Combustion Particle Doses Solid Fuel Zone From Mind the Gap, Smith/Peel, 2010 and Pope et al., 2009

39 Relative Risk Generalized Exposure-Response: Outdoor Air, SHS, and Smoking IHD risks from combustion particles Annual average PM2.5 in ug/m Solid Fuel Zone CRA, 2011

40 Annual Incidence Generalized Exposure-Response: Outdoor Air, SHS, and HAP Pneumonia from combustion particles Annual average PM2.5 in ug/m Solid Fuel Solid Fuel Zone Zone CRA, 2011

41 ALRI Rate (per 100 Child-Yr) MD-diagnosed Acute Lower Respiratory Infection (A) Open fire Where we Want to Be! Chimney stove Approximate Mean CO PM2.5 Exposure exposure (ppm) in 100s of ug/m3 RESPIRE- Guatemala

42 Combustion Particles and their accompanying toxic side-kicks Cause more health damage than any other environmental contaminant Worst thing to do is stick burning stuff in your mouth ~5 million deaths Not so great to have other people sticking in their mouths nearby ~ 500k deaths Bad even to have poorly burning stuff in your city ~ 1 million deaths The oldest burning practice, however --poorly combusted fuels in the home -- is still the cause of more ill-health than any other particle source except smoking ~ 1.6 million deaths

43 The Classic Epidemiological Transition Infectious Diseases Non- Communicable Diseases Time

44 Epidemiological Transition: All Ages both female male Poor Communicable Non-communicable Injuries Rich Purchasing Power Parity Ranking Data from WHO GBD database Garcia, 2011

45

46 What NCD risk factor is shared by all everyone in the bottom two billion? Diet (fat, etc.)? Physical inactivity (obesity, etc)? Smoking? Appropriate infectious agents? No, but there is one So ubiquitous, in fact, that it is one of the best quick indicators of poverty?

47 Households using biomass or coal to cook Comparative Risk Assessment (CRA) preliminary, Adair, et al.

48 Percent Solid Fuel Use Distribution of Household Cooking Fuel by Income in India million households Lowest Highest 5 Income Quintal Solid Fuel LPG + Kero NFHS, 2006

49 If it doesn t take fifty years, it isn t worth doing.* First Royal Commission on Air Pollution in London in 1315; recommendation (ban coal burning) taken up in John Snow in 1854; still one-third of world population without adequate sanitation/water Surgeon General s Report in 1964; Framework Convention on Tobacco Control came into force in 2005 (still not all countries signed up) *Attributed to Albert Einstein

50 The Environmental Health Risk Transition Household, e.g. poor sanitation Severity Community, e.g. Urban Pollution Global, e.g. greenhousegas emissions Increasing Wealth Shifting Environmental Burdens Local Immediate Global Delayed Smith, 1990

51 Many thanks Publications and presentations available at: Or just Google Kirk R. Smith

52 Age-Adjusted Death Rates Compared to USA India Rwanda World 10.0 Ratio to USA 1.0 All Causes CDs NCDs Cancer CVD Chronic Resp All Causes Selected NCD Groups Cervical cancer Stroke IHD COPD LBW-PTB Selected Causes 0.1 CDs include communicable, maternal, and perinatal causes

53 Relative Risk 10 SHS HAP ATS ALRI LBW CVD OAP Cataracts TB Dose PM2.5 - mg/day

54 Relative Risk 100 ATS ALRI COPD LBW 10 Lung cancer OAP SHS HAP CVD Cataracts TB Dose PM2.5 - mg/day

55 COPD Reduction with Household Air Pollution 10-year Intervention in India Wilkinson, Smith, et al Chronic Obstructive Pulmonary Disease

56 Health Benefits Upon Completion, 2020 Wilkinson, Smith, et al. 2009