Characteristics of included (quasi-)experimental studies
| First author (year) | Country/setting | Study design | Participants | Sample size | Study time frame | Type of intervention | Control situation | Outcomes | |||
| Level of outcome assessment | Eligible outcomes | Definition of outcome | Follow-up | ||||||||
| Aung et al 42 (2016) | India, rural | RCT | Households with one female cook >25 years who was neither pregnant at enrolment nor a current or previous smoker | N=187; C=91; I=96 | Sept 2011–Aug 2012 | CDM approved single-pot ‘rocket-style’ biomass cookstove | Three-stone stoves | Household | 24-hour PM2.5 exposure | 37 mm Teflon filters placed downstream of a cyclone with a 2.5 µm aerodynamic-diameter cut point connected to a battery-operated pump | 1 year |
| Bensch and Peters22 (2012) | Senegal, rural | C-RCT (household level) | Households with no previous access to improved cookstoves | N=253 households; C=155 households; I=98 households | Nov 2009– Nov 2010 | Portable clay/metal firewood stove (Jambar) | Three-stone stoves; traditional metal wood stoves | Household | Respiratory system disease (cough, asthma, difficulty breathing); eye problems | Self-reported symptoms | 1 year |
| Bensch and Peters23 (2015) | Senegal, rural | C-RCT (household level) | Households with no previous access to improved cookstoves | N=253 households; C=155 households; I=98 households | Nov 2009–Nov 2010 | Portable clay/metal firewood stove (Jambar) | Three-stone stoves and traditional metal wood stoves | Household; individual (women responsible for cooking) | Respiratory system disease (cough, asthma, difficulty breathing); eye problems | Self-reported symptoms | 1 year |
| Burwen and Levine24 (2012) | Ghana, rural | RCT | One woman per household (most frequently responsible for cooking) | N=768; C=366; I=402 | Feb– May 2009 | Improved cookstove (unspecified) | Traditional L-shaped or U-shaped three-stone fire | Individual | CO exposure during cooking; cough; difficulty breathing; chest pain; excessive mucus | CO: Gastec 1 DL CO Passive Diffusion tubes; self-reported symptoms | 8 months |
| Diaz et al 25 (2007) | Guatemala, rural | RCT | Pregnant women or families with child <4 months | Group A: N=300; C=147; I=153 Group B: N=204; C=98; I=106 | Oct 2002–Dec 2004 | Plancha stove | Traditional open fire | Individual | Sore eyes; exhaled CO | Sore eyes assessed with a health questionnaire, reassessed every 6 months; exhaled CO measured with a Micro Medical Micro CO monitor | 12 months (group A); 18 months (group B) |
| Hanna et al 26 (2012) | India, rural | C-RCT (household level) | Eligibility criteria not reported | N=2651 households; first 1/3 within each village received stoves at start of project, second 1/3 received stove 2 years after the first wave and remaining households at end of the study | Sep 2006–Mar 2007; May 2009–Apr 2010 | ICS (ARTI) | Traditional stove | Individual (primary cooks, women and children) | Exhaled CO; lung function; symptoms (cough, phlegm, wheezing, sore eyes, tightness in the chest); infant mortality; stillbirths and miscarriages | CO: Micro Medical CO monitor; health symptoms and outcomes were self-reported; lung function measured with spirometry | 4 years |
| Hartinger et al 40 (2016) | Peru, rural | C-RCT (community level) | Children aged 6–35 months from households using solid fuels | N=503; C=253; I=250 | Sept 2008–Jan 2010 | OPTIMA-improved ventilated solid-fuel stove | Unventilated traditional stove or open fire | Individual | ARI; ALRI; chronic cough in children <36 months | ARI: child w/cough and/or difficulty breathing. ALRI: child w/ cough and/or difficulty breathing and raised resp. rate | 12 months |
| Jamali et al 21 (2017) | Pakistan, rural | NRCT | Households and one woman (main cook) per household | Sindh: N=292; C=209; I=83 Punjab: N=313; C=179; I=134 | Mar– Sept 2014 | Chulhas | Traditional three-stone stove | Kitchen and individual (women responsible for cooking) | Respiratory symptoms (chest tightness, shortness of breath, phlegm, asthma, cough); lung function (PEF); eye symptoms; 24-hour PM2.5 and CO exposure | Symptoms: self--reported. Lung function: Philips Respironics peak flow meter; PM2.5: RTI MicroPEM; CO: QRAE II multigas monitor | 3 months |
| Jary et al 27 (2014) | Malawi, rural | RCT (feasibility study) | Non-smoking women in Ntcheu district who cooked on traditional open wood fires, but wished to purchase a chitetezo stove | N=51; C=26; I=25 | Nov–Dec 2011 | Chitetezo stove | Traditional open wood fire | Individual | Exhaled CO; cough, mucus, shortness of breath, wheezing; burning/watery eyes | PM2.5: Sidepak Monitor CO: personal CO monitors Symptoms were self-reported | 7 days |
| Ludwinski et al 43 (2011) | Guatemala, rural | Cluster NRCT (household level) | Households with open fires or stoves in poor condition | N=73 households (477 members); C=45 households (284 members); I=28 households (193 members) | Aug 2008–Aug 2009 | Onil stove | Traditional open fire | Individual | Health outcomes (number of days having cough and eye irritation) in women and children | Health outcomes assessed through an interview | 1 year |
| McCracken et al 30 (2007) | Guatemala, rural | RCT | Women ≥38 years living in households participating in RESPIRE | N=238; C=115; I=123 | Jul 2003–Dec 2005 | Plancha stove | Open fire | Individual | 24-hour average PM2.5 exposure | Measured by air sampler pump | 3 years and 2 months |
| McCracken et al 28 (2009) | Guatemala, rural | RCT | Main study: 509 children 0–18 months Validation study: 70 children randomly selected from main study population | Main study N=509; 1932 observations, validation study N=70; 270 observations | Jan 2003–May 2004 | Plancha stove | Open fire | Individual | 48-hour CO exposure | Measured by Gastec 1 DL passive diffusion tubes worn by the children during 48-hour periods | 1.5 years |
| McCracken et al 29 (2013) | Guatemala, rural | RCT | Women ≥38 years living in households participating in RESPIRE | N=238; C=115; I=123 | Jul 2003–Dec 2005 | Plancha stove | Open fire | Individual | 24-hour CO and PM2.5 exposure | PM2.5: gravimetric measure of 24- hour personal exposure using Teflon filter CO: span-gas calibrated passive electrochemical data logger | 1 year |
| Mortimer et al 41 (2017) | Malawi, rural | C-RCT (community level) | Households with at ≥1 child aged ≤4.5 years | N=10 750; C=5350; I=5400 | Dec 2013–Feb 2016 | Two cleaner burning biomass-fuelled cookstoves | Open fires | Individual | WHO IMCI- defined pneumonia in children <5 years; WHO IMCI-defined severe pneumonia; death due to pneumonia | Pneumonia: cough or difficulty breathing and fast breathing; severe pneumonia: pneumonia plus chest in-drawing, stridor or general danger sign (inability to drink or breast feed, vomiting, convulsions, lethargy or unconsciousness) | 2 years and 2 months |
| Piedrahita et al 31 (2017) | Ghana, rural | C-RCT (household level) | Households using biofuels as main cooking fuel sources; ≥1 child <5 years and one woman aged 18–55 years | N=200; C=50; I=150. 50 given two locally made Gyapa rocket stoves (Gyapa/Gyapa), 50 given two Philips HD4012 LS stoves (Philips/Philips), 50 given one of each (Gyapa/Philips) | Nov 2013– Jan 2016 | Gyapa rocket stoves, Philips HD4012 LS stoves | Three-stone stoves | Individual (women and children) and kitchen | 48-hour carbonaceous PM2.5 exposure | Elemental and organic carbon in μg/m3 measured by sampling packs (backpacks for children, waist packs for adults) using quartz filters | 2 years |
| Riojas-Rodriguez et al 32 (2011) | Mexico, rural | RCT | Subsample of 63 women from Romieu et al (2009)33 RCT | N=63 women, C=20, I=43 | Feb 2005–Jun 2006 | Patsari stove | Open wood fire | Individual | 8-hour CO exposure | Continuous data--logging electrochemical CO monitors | 10 months |
| Romieu et al 33 (2009) | Mexico, rural | RCT | Households/women using open wood fire and having a child <5 years | N=668 households, C=330, I=338 | Feb 2005–Jun 2006 | Patsari stove | Open wood fire | Individual | Resp. symptoms (phlegm, cough, wheezing, chest tightness); non-resp. symptoms (eye burning, watery eyes); lung function | Clinical symptoms were assessed by a team of local nurses; spirometry was performed to determine lung function | 10 months |
| Rosa et al 34 (2014) | Rwanda, rural | C-RCT (household level) | Head of household >18 years; no household members were community health workers | N=566; C=281; I=285 PM2.5 was measured in N=121; C=61; I=60 | 1 year and 7 months | EcoZoom Dura stove | Stone fires | Household | 24-hour average PM2.5 exposure | PM2.5: a semicontinuous, light scattering nephelometer (Berkeley particle and temperature sensor) | 5 months |
| Schilmann et al 35 (2015) | Mexico, rural | RCT | Fuelwood-using households/women with child <4 years in six rural communities in Purepecha region | N=668; C=330; I=338 | Feb 2005– Jun 2006 | Patsari stove | Open wood fire | Individual | Resp. infections in children <5 years | LRI: fast breathing and cough or difficult breathing in previous 15 days and/or observed by fieldworker URI: ≥2 of the following: cough, phlegm, nasal congestion or secretion and sore throat in previous 15 days and/or observed by fieldworker | 10 months |
| Smith et al 36 (2010) | Guatemala, rural | RCT | Children living in homes with open fires or chimney stoves in 23 villages in San Marcos highlands; children were in utero or ≤4 months | N=534; C=265; I=269 | Oct 2002– Dec 2004 | Plancha stove | Open fire or chimney stove | Individual and kitchen | 48-hour CO exposure | Personal CO: passive diffusion GASTEC tubes Kitchen CO: HOBO/onset CO monitor | Until child was 18 months |
| Smith et al 37 (2011) | Guatemala, rural | RCT | Households with a pregnant woman or a child <4 months | N=534; C=265; I=269 | Oct 2002– Dec 2004 | Plancha stove | Open fire | Individual | Clinical, radiological or fieldworker-assessed pneumonia | Fieldworker-assessed pneumonia: child with cough and/or difficulty breathing meeting criteria for referral to study physician as possible ALRI | Until child was 18 months |
| Smith-Sivertsen et al 38 (2009) | Guatemala, rural | RCT | Pregnant women or families with child <4 months | Group A: N=300; C=147; I=153 Group B: N=204; C=98; I=106 | Oct 2002– Dec 2004 | Plancha stove | Traditional open fire | Individual | 48-hour CO exposure; chronic lung symptoms (cough, phlegm, wheeze, tightness in the chest); lung function | CO: GASTEC CO passive diffusion tubes Chronic lung symptoms assessed via questionnaire Lung function: ATS criteria and spirometry | 12 months (group A); 18 months (group B) |
| Thompson et al 39 (2011) | Guatemala, rural | RCT | Households using open wood fires for cooking and having a pregnant woman or a child <4 months | N=266; C=120; I=134 | Oct 2002– Dec 2004 | Plancha stove | Open fire | Individual | 48-hour CO exposure during pregnancy; LBW (<2500 g); birth weight | CO: passive diffusion colorimetric CO tubes; birth weight: weighed by team staff within 48 hours after birth | Until child was born |
| Zhou et al 44 (2006) | China, rural | CBA | Households that: (i) lived in the area for ≥1 year; (ii) used coal and/or biomass as main energy source; (iii) had a female member >18 years and a child <14 years | Gansu: N=1009; C=509; I=500; Guizhou: N=1023; C=523; I=500; Shaanxi: N=1089; C=508; I=581 | March 2003– April 2005 | Gansu: unspecified stove; Guizhou: air circular stove; Shaanxi: unspecified stove | Gansu: brick/clay stove; Guizhou: coal stove; Shaanxi: brick/clay biomass/coal stove | Household | 24-hour respirable particles (RPM) 24-hour CO exposure | RPM: nylon cyclone equipped with a 37 mm diameter poly-vinyl-chloride filter CO: long-term diffusion tubes | 12 months |
Specifications of improved cookstoves are reported in web supplementary appendix pages 15–16.
ALRI, acute lower respiratory tract infection; ARI, acute respiratory tract infection; ARTI, Appropriate Rural Technology Institute; ATS, American Thoracic Society; CBA, controlled before–after study; CDM, clean development mechanism; CO, carbon monoxide; C-RCT, cluster randomised controlled trial; IMCI, Integrated Management of Childhood Illness; ICS, improved cook stoves; LBW, low birth weight; LRI, lower respiratory tract infection; NRCT, non-randomised controlled trial; PEF, peak expiratory flow; PM, particulate matter; RCT, randomised controlled trial; URI, upper respiratory tract infection