Household Air Pollution and Health: A Multi-country LPG Intervention Trial (HAPIN)

National Institutes of Health (NIH), National Heart, Lung, and Blood Institute (NHLBI), Bill and Melinda Gates Foundation, Berkeley Air Monitoring Group, Colorado State University, Global LPG Partnership, Harvard University, Johns Hopkins University, London School of Hygiene and Tropical Medicine, Asociacion Benefica Prisma, Sri Ramachandra University, Universidad del Valle, Guatemala, Universidad Peruana Cayetano Heredia, University of Georgia, Washington University School of Medicine, University of Rwanda, Eagle Research Center, University of Oxford

This study is a randomized controlled trial of liquefied petroleum gas (LPG) stove and fuel distribution in 3,200 households in four countries (India, Guatemala, Peru, and Rwanda). Following a common protocol, each intervention site will recruit 800 pregnant women (aged 18-34 years, 9 - <20 weeks gestation), and will randomly assign half their households to receive LPG stoves and an 18-month supply of LPG. Control households are anticipated to continue to cook primarily with solid biomass fuels, and will receive compensation based on a uniform set of trial-wide principles, customized to each site based on formative research. The mother will be followed along with her child until the child is 1 year old. The researchers estimate that 15% of households will have a second, non-pregnant older adult woman (aged 40 to <80 years) who will also be enrolled at baseline and followed during the 18-month follow-up period. To optimize intervention use, the researchers will implement behavior change strategies informed by previous experiences and formative research in Year 1. This study will assess cookstove use, conduct repeated personal exposure assessments of household air pollution, and collect dried blood spots and urinary samples for biomarker analysis and biospecimen storage. The primary outcomes are low birth weight, severe pneumonia incidence, and stunting of the child, and blood pressure in the older adult woman. Secondary outcomes include preterm birth and development in the child, maternal blood pressure during pregnancy, and endothelial function, respiratory impairment, atherosclerosis, carcinogenic metabolites, and quality of life in the older adult woman. Participants in India, Guatemala and Rwanda will be followed until the child is 5 years old to assess the longer-term effects of the intervention.

Globally, nearly 3 billion people rely on solid fuels for cooking and heating, the vast majority in low- and middle-income countries (LMICs). The resulting household air pollution (HAP) is the third leading risk factor in the 2010 global burden of disease, accounting for an estimated 4.3 million deaths annually, largely among women and young children. Previous interventions have provided cleaner biomass-based cookstoves, but have failed to reduce exposure to levels that produce meaningful health improvements. There have been no large-scale field trials with liquefied petroleum gas (LPG) cookstoves, likely the cleanest scalable intervention. The aim of this study is to conduct a randomized controlled trial of LPG stove and fuel distribution in 3,200 households in four LMICs (India, Guatemala, Peru, and Rwanda) to deliver rigorous evidence regarding potential health benefits across the lifespan. Each intervention site will recruit 800 pregnant women (aged 18-34 years, 9 - <20 weeks gestation), and will randomly assign half their households to receive LPG stoves and an 18-month supply of LPG. Control households are anticipated to continue to cook primarily with solid biomass fuels, and will receive compensation based on a uniform set of trial-wide principles, customized to each site based on formative research. The mother will be followed along with her child until the child is 1 year old. In households with a second, non-pregnant older adult woman (aged 40 to <80 years) the researchers will also enroll and follow her during the 18-month follow-up period in order to assess cardiopulmonary, metabolic, and cancer outcomes. To optimize intervention use, the researchers will implement behavior change strategies. This study will assess cookstove use, conduct repeated personal exposure assessments to HAP (PM2.5, black carbon, carbon monoxide), and collect dried blood spots and urinary samples for biomarker analysis and biospecimen storage on all participants at multiple time points. The primary outcomes are low birth weight, severe pneumonia incidence, and stunting of the child, and blood pressure in the older adult woman. Secondary outcomes include preterm birth and development in the child, maternal blood pressure during pregnancy, and endothelial function, respiratory impairment, atherosclerosis, carcinogenic metabolites, and quality of life in the older adult woman. This study will address the following specific aims: (1) using an intent-to-treat analysis, determine the effect of a randomized LPG stove and fuel intervention on health in four diverse LMIC populations using a common protocol; (2) determine the exposure-response relationships for HAP and health outcomes; and (3) determine relationships between LPG intervention and both targeted and exploratory biomarkers of exposure/health effects. This study will provide evidence, including costs and implementation strategies, to inform national and global policies on scaling up LPG stoves among vulnerable populations. Ultimately, this will facilitate deeper policy-level discussions as well as identify requirements for initiating and sustaining HAP interventions globally. The intervention delivery occurs when the child is one year of age. The researchers will continue to follow participants in India, Guatemala and Rwanda until the child is 5 years old to assess the longer-term effects of the intervention. Previous evidence suggests that the benefits of reduced exposure during the first, critical year of development will continue even if the intervention ends. The researchers will continue using methods employed during the HAPIN trial period. The HAPIN trial provides a unique context in which to address these questions, particularly given the successful intervention and exposure reduction. Participants are well-characterized and health and exposures to air pollution are being documented. Critically, because of its experimental design of the trial, continued follow-up of the cohort will provide rigorous causal inferences about the effects of this 500-day intervention over the most important period of early childhood development.

Obstetrics, Pregnancy Outcomes, Pediatrics, Atherosclerosis, Cancer, Cardiovascular Disease, Public Health, Environmental Air Pollutants

Participants randomized to the experimental arm will receive a liquefied petroleum gas (LPG) cookstove and 18-month supply of LPG.

Participants in the control group will not receive a liquefied petroleum gas (LPG) stove and will continue using traditional cooking methods (open fire or traditional stoves), or the cooking method of their choice. Control households will receive compensation based on a uniform set of trial-wide principles, customized to each site based on formative research.

The intervention consists of a high-quality locally-available liquefied petroleum gas (LPG) stove having at least two burners, a continuous supply of LPG fuel for 18 months, and the promotion of stove use on an exclusive basis for cooking. The intervention will be provided free of charge to all intervention households upon enrollment. On a weekly basis, study staff will examine stove condition, perform any repairs necessary, and measure and record weight of LPG tanks in order to anticipate need for refills.

Birth weight will be assessed by a trained nurse or health worker within 24 hours of birth. Infants will be weighed naked or in a pre-weighed blanket. Weight will be measured to the nearest 10 g using a digital electronic scale, if performed by the study field staff; otherwise, hospital medical records will be used.

The number of times a child has severe pneumonia over their period of follow-up during the first year of life will be assessed. HAPIN pneumonia criteria are adapted from the WHO classification of childhood pneumonia (2014) and there are 3 algorithms for HAPIN case criteria: 1) the presence of cough and/or difficult breathing and at least 1 general danger sign plus evidence of pneumonia on lung imaging (i.e., lung ultrasound or chest x-ray), or 2) the presence of cough and/or difficult breathing and hypoxemia (measured either via pulse oximetry (SpO2), or observing a child requiring advanced respiratory support (i.e., intubation and mechanical ventilation, non-invasive ventilation with continuous or bi-level positive airway pressure support, or high-flow nasal cannula oxygen), or 3) children who die prior to evaluation but their death is attributed to pneumonia by verbal autopsy. Cases of pneumonia are recorded children present to HAPIN health facilities with respiratory symptoms.

The primary outcome measured is stunting at one year of age, defined as a length-for-age z-score (LAZ) that is 2 standard deviations below the median of the growth standard. Infant length will be assessed at birth and quarterly thereafter, until the child is 12 months old. Z-scores will be calculated using the 2006 World Health Organization (WHO) Multi-Growth Reference Standard (MGRS).

Systolic blood pressure will be assessed in the older adult women in the intervention and control arms using automatic sphygmomanometers (Omron HEM-907XL; Osaka, Japan). The study team will use the procedures adapted from previously validated methods and cardiovascular outcome studies, following recommendations for the American Heart Association and the European Society of Hypertension.

Linear growth of children will be assessed in centimeters of height from the time of birth until 60 months of age.

Child development will be assessed with the Caregiver Reported Early Childhood Development Instrument (CREDI). The CREDI is a population-level measure of early childhood development (ECD) for children from 0-2 years of age. The CREDI assesses 5 domains of child development: 1) motor development (fine and gross motor), 2) language development (expressive and receptive language), 3) cognitive development (executive function, problem solving and reasoning, and pre-academic knowledge), 4) socio-emotional development (emotional and behavioral self-regulation, emotional knowledge, and social competence), and 5) mental health (internalizing and externalizing behaviors). The CREDI long form has 117 items and the number of questions answered depends on the age of the child. Responses of "yes" are coded as 1 and "no" is coded as 0; certain items are reverse coded. Total raw scores increase by age (with developmental progression), and higher scores indicate increased development.

The MDAT measures gross motor (39 items), fine motor (42 items), language/cognition (40 items) and social skills (36 items). Originally developed and validated in rural Malawi, it has now been used in over 25 countries with more than 8,000 children as both a clinical and research tool. The MDAT is a continuous test with start and stop rules. Most items are administered directly to the child and items that are not easily observed (e.g., child speaks in full sentences; child understands sharing with others; child can dress self) are administered by parent report. Children receive either a pass or fail for each item, and summed pass scores can produce a composite score as well as domain-specific scores. Total scores range from 0 to 157 where higher scores indicate greater neurodevelopment.

Blood pressure will be assessed in the pregnant women in the intervention and control arms using automatic sphygmomanometers (OMRON HEM-907XL; Osaka, Japan). After delivery, blood pressure will be measured in the new mothers when the child is 24, 36, 48 and 60 months old. The study team will use the procedures adapted from previously validated methods and cardiovascular outcome studies, following recommendations for the American Heart Association and the European Society of Hypertension.

Diastolic blood pressure will be assessed in the older adult women and new mothers in the intervention and control arms using automatic sphygmomanometers (Omron HEM-907XL; Osaka, Japan). The study team will use the procedures adapted from previously validated methods and cardiovascular outcome studies, following recommendations for the American Heart Association and the European Society of Hypertension.

Mean arterial pressure will be assessed in the older adult women and new mothers in the intervention and control arms using automatic sphygmomanometers (Omron HEM-907XL; Osaka, Japan). Mean arterial pressure is calculated as DBP+(SBP-DBP)/3, where SBP=systolic blood pressure and DBP=diastolic blood pressure.

Pulse pressure will be assessed in the older adult women and new mothers in the intervention and control arms using automatic sphygmomanometers (Omron HEM-907XL; Osaka, Japan). pressure. Pulse pressure is the difference between systolic blood pressure and diastolic blood pressure.

Pregnant women will have ultrasounds at Baseline and during gestation weeks 24-28 and gestation weeks 32-36 to measure fetal growth outcomes. Specifically, we will evaluate head circumference (HC), abdominal circumference (AC), femur length (FL) and estimated fetal weight (EFW) during gestation. We will compare (i) z-scores of individual fetal growth measurements (HC, AC, FL, EFW) at the 2 growth ultrasound visits between intervention and control participants (separately at 24-28 wks gestation and 32-36 wks gestation); (ii) differences in proportions of the 2.5th percentiles of each of these measurements evaluated separately at 24-28 and 32-36 weeks gestation; (iii) Z-score trajectories of HC, AC, FL and EFW as a function of gestational age and intervention; and (iv) prevalence of small for gestational age (SGA) during the fetal period through birth as measured by WHO INTERGROWTH 21st standards.

Cumulative incidence of WHO non-severe pneumonia (2014 definition and 2013 definition) during the first year of life. Cases of pneumonia are recorded whenever children present to HAPIN health facilities with respiratory symptoms.

Cumulative incidence of WHO non-severe pneumonia (2014 definition and 2013 definition) during the first year of life. Cases of pneumonia are recorded whenever children present to HAPIN health facilities with respiratory symptoms.

Cumulative incidence of WHO non-severe pneumonia during the first year of life, as defined in the second edition of the "Pocket book of hospital care for children" (2013). Cases of pneumonia are recorded whenever children present to HAPIN health facilities with respiratory symptoms.

Cumulative incidence of WHO severe pneumonia during the first year of life, as defined in the second edition of the "Pocket book of hospital care for children" (2013). Cases of pneumonia are recorded whenever children present to HAPIN health facilities with respiratory symptoms.

Cumulative incidence of hypoxemic pneumonia during the first year of life. Cases of pneumonia are recorded whenever children present to HAPIN health facilities with respiratory symptoms.

Cumulative incidence of lung ultrasound or chest radiograph pneumonia during the first year of life. Cases of pneumonia are recorded whenever children present to HAPIN health facilities with respiratory symptoms.

Brachial artery reactivity testing (BART) measures endothelial function via flow-mediated dilatation to reactive hyperemia following the release of arm blood-flow occlusion. In this test, baseline artery diameter is measured, then a blood pressure cuff is inflated to induce distal arm ischemia for 5 minutes and after releasing the pressure, the post-occlusion brachial artery diameter is measured. The ratio of post- to pre-occlusion artery diameter represents endothelial function where lower values indicate worse endothelial function. (Peru only)

The carotid intima-media thickness test (CIMT) is used to determine the extent of carotid atherosclerotic vascular disease. The test measures the thickness of the inner two layers of the carotid artery and can detect plaque build up prior to physical symptoms being experienced. The carotid ultrasound will be performed with a portable ultrasound by trained sonographers.

Adult respiratory health and well-being will be assessed with the St. George Respiratory Questionnaire (SGRQ). The SGRQ measures impaired health and perceived well-being among individuals with chronic airway disease. The SGRQ has sections assessing symptoms, activities that cause breathlessness or are limited because of breathlessness, and the impacts of respiratory problems on employment, sense of control of health, panic, stigmatization, medication use, side effects of therapies, expectations for health and disturbances of daily life. The questionnaire includes multiple choice, true/false and open-ended questions.

The Short Form 36 survey (SF-36) is a standardized, preference-based 36 item questionnaire evaluating quality of life. The survey has 8 sections (vitality, physical functioning, bodily pain, general health perceptions, physical role functioning, emotional role functioning, social role functioning, and mental health). Possible scores range from 0 (lowest quality of life) to 100 (highest quality of life).

Weight will be measured in the pregnant women/new mothers, the older adult women, and the children. Weight is measured in kilograms (kg). Weight in pregnant women will be measured at baseline, 24-28 weeks gestation, and 32-36 weeks gestation, and in new mothers when the child is 24- and 36-months old. In older adult women, it will be measured at baseline, 3, 6, 9, 12 and 18 months post-randomization, and when the child is 24-months old. Weight in children will be measured at birth, and at 3, 6, 9, 12, 24, 36, 48 and 60 months of age.

BMI will be calculated for the pregnant women, the older adult women, and the children. BMI is calculated as weight in kilograms divided by height in meters (m) squared (kg/m²). BMI in pregnant women will be calculated at baseline, 24-28 weeks gestation, and 32-36 weeks gestation, and in new mothers when the child is 24- and 36-months old. In older adult women, it will be calculated at baseline, 3, 6, 9, 12 and 18 months post-randomization, and when the child is 24-months old. Weight in children will be calculated at birth, and at 3, 6, 9, 12, 24, 36, 48 and 60 months of age.

Height in women will be measured in centimeters. Height in pregnant women will be measured at baseline, 24-28 weeks gestation, and 32-36 weeks gestation, and in new mothers when the child is 24- and 36-months old. In older adult women, it will be measured at baseline, 3, 6, 9, 12 and 18 months post-randomization, and when the child is 24-months old. This measurement will be used to compute the body mass index.

Multiple exposure biomarkers will be measured: 3-OH Cotinine, 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol (NNAL), levoglucosan, 8OH-deoxyguanosine (8OHdG), and Volatile Organic Chemicals (VOC) metabolites. Exposure biomarkers (especially for children whose urine may be limited) will be prioritized as follows: polycyclic aromatic hydrocarbon (PAH) biomarkers, levoglucosan, volatile organic chemicals (VOC) biomarkers, heavy metals, and tobacco-related biomarkers. Urinary biomarkers will be measured in pregnant women at baseline, 24-28 weeks gestation, and 32-36 weeks gestation, and in new mothers when the child is 24-months old. Biomarkers will be measured in older adult women at baseline, 3, 6, 9, 12 and 18 months post-randomization. Biomarkers will be measured in children at 3, 6, 12 and 24 months of age.

The main biomarkers to be measured from the dried blood spots is: inflammation markers, endothelial markers of cardiovascular disease, oxidative stress markers, Hb, HbA1C, tumor-associated antigen antibodies, cytochrome P450, p53 TAA, lipids, metabolomics, MiRNA, heavy metals. DBS biomarkers will be measured in pregnant women at baseline, 24-28 weeks gestation, and 32-36 weeks gestation, and in new mothers when the child is 24-months old. DBS biomarkers will be measured in older adult women at baseline, 3, 6, 9, 12 and 18 months post-randomization. DBS biomarkers will be measured in children at 3, 6, 12 and 24 months of age.

Lung function measurements will be made at 3 years of age using the forced oscillation technique (FOT) with the Tremoflo C-100 device with disposable mouthpieces. FOT is a technique that can identify early changes in the airways. The FOT device measures the relationship between externally applied pressure waves and the resulting air flow to measure respiratory impedance. Values produced at high frequencies correspond to the proximal and large airways, and values produced at low frequencies correspond to distal and small airways.

Personal monitoring equipment will be used to assess exposure to fine particulate matter (PM2.5) over a 24-hour period in intervention and control participants (pregnant women, older adult women, and children). Exposure for pregnant women will be measured at baseline, 24-28 weeks gestation, 32-36 weeks gestation, and 24 months after delivery. Exposure in the child will be measured at 3, 6, 12, 24, 36, 48 and 60 months of age. Exposure in the older adult women will be measured at baseline, 3, 6, 12 and 18 months post-randomization, and at 24 months after delivery. Additionally, PM2.5 in the home will be measured 6 months, 12 months and 24 months after delivery of the child.

Personal monitoring equipment will be used to assess exposure to carbon monoxide (CO) over a 24-hour period in intervention and control participants (pregnant women, older adult women, and children). Exposure for pregnant women will be measured at baseline, 24-28 weeks gestation, 32-36 weeks gestation, and 24 months after delivery. Children will be measured up to 12 months after delivery, and at 24, 36, 48 and 60 months of age. Older adult women will be measured up to 12 months after delivery, and at 24 months after delivery. Additionally, CO in the home will be measured 6 months, 12 months and 24 months after delivery of the child.

Personal monitoring equipment will be used to assess exposure to black carbon (BC) over a 24-hour period in intervention and control participants (pregnant women, older adult women, and children). Exposure for pregnant women will be measured at baseline, 24-28 weeks gestation, 32-36 weeks gestation, and 24 months after delivery. Children will be measured up to 12 months after delivery, and at 24, 36, 48 and 60 months of age. Older adult women will be measured up to 12 months after delivery, and at 24 months after delivery. Additionally, BC in the home will be measured 6 months, 12 months and 24 months after delivery of the child.

The 10-item Cohen's Perceived Stress Scale (PPS) assesses the way an individual appraises their life events as stressful (e.g., "In the last month, how often have you felt difficulties were piling up so high that you could not overcome them?") (Cohen, 1983). Likert-level responses ranged from 0 (never) to 4 (very often), meaning a high PPS score would result in a high level of perceived stress. A Spanish language PPS created and tested and found to be valid and reliable (Vallijo et al., 2018).

Death of a woman while pregnant or within 42 days of termination of pregnancy irrespective of the duration and site of the pregnancy.

Any fetal deaths occurring at or after 20 weeks' gestation OR indicated on Serious Adverse Event form OR on Pregnant Woman Medical History form OR on verbal autopsy form.

Blood pressure will be assessed in the children using automatic sphygmomanometers (OMRON HEM-907XL; Osaka, Japan). The study team will use the procedures adapted from previously validated methods and cardiovascular outcome studies, following recommendations for the American Heart Association and the European Society of Hypertension.

Two buccal cell scrapes will be collected by gently scraping the buccal mucosa on both sides of the mouth with a small plastic collection spoon. Nasal turbinate brush samples can be collected using a soft cytobrush on each turbinate. Collection is gentle and causes no discomfort to study participants. Both samples will be processed in the laboratory according to procedures detailed in the protocol. This will occur in the older adult women in an NCI substudy.

For the oral rinse, participants will vigorously rinse their mouth and the rinsates are collected in a centrifuge tube. The tube is centrifuged and the pellet and supernatant are removed to separate cryovials, labeled and frozen. This will occur in the older adult women in an NCI substudy.

Two buccal cell scrapes will be collected by gently scraping the buccal mucosa on both sides of the mouth with a small plastic collection spoon. A 5-mL venous blood sample will be collected in an ethylenediaminetetraacetic acid (EDTA) vacutainer tube by standard clinical venipuncture of a cubital vein. Both samples will be transported and processed in the laboratory according to procedures detailed in the protocol. This will occur in the older adult women in an NCI substudy.

A 5-mL venous blood sample will be collected in an EDTA vacutainer tube by standard clinical venipuncture of a cubital vein. The sample will be transported and processed in the laboratory according to procedures detailed in the protocol. This will occur in the older adult women in an NCI substudy.

Inclusion Criteria for Pregnant Women: Confirmed pregnancy (hCG positive blood or urine test) Aged 18 to <35 years (via self-report) Uses biomass stove predominantly Lives in study area 9 - <20 weeks gestation confirmed by ultrasound Singleton pregnancy (one fetus) Viable fetus with normal fetal heart rate (120-180 beats per minute) at time of ultrasound Continued pregnancy at the time of randomization confirmed by self-report Agrees to participate with informed consent Exclusion Criteria for Pregnant Women: Currently smokes cigarettes or other tobacco products Plans to move permanently outside study area in the next 12 months Uses LPG stove predominantly, or is likely to use LPG predominantly in the near future Inclusion Criteria for Older Adult Woman in the Same Household: Aged 40 to <80 years (via self-report) Exclusion Criteria for Older Adult Woman in the Same Household: Currently smokes cigarettes or other tobacco products Pregnant (by self-report) Plans to move out of her current household in the next 12 months

Clasen TF, Chang HH, Thompson LM, Kirby MA, Balakrishnan K, Diaz-Artiga A, McCracken JP, Rosa G, Steenland K, Younger A, Aravindalochanan V, Barr DB, Castanaza A, Chen Y, Chiang M, Clark ML, Garg S, Hartinger S, Jabbarzadeh S, Johnson MA, Kim DY, Lovvorn AE, McCollum ED, Monroy L, Moulton LH, Mukeshimana A, Mukhopadhyay K, Naeher LP, Ndagijimana F, Papageorghiou A, Piedrahita R, Pillarisetti A, Puttaswamy N, Quinn A, Ramakrishnan U, Sambandam S, Sinharoy SS, Thangavel G, Underhill LJ, Waller LA, Wang J, Williams KN, Rosenthal JP, Checkley W, Peel JL; HAPIN Investigators. Liquefied Petroleum Gas or Biomass for Cooking and Effects on Birth Weight. N Engl J Med. 2022 Nov 10;387(19):1735-1746. doi: 10.1056/NEJMoa2206734. Epub 2022 Oct 10.

Johnson M, Pillarisetti A, Piedrahita R, Balakrishnan K, Peel JL, Steenland K, Underhill LJ, Rosa G, Kirby MA, Diaz-Artiga A, McCracken J, Clark ML, Waller L, Chang HH, Wang J, Dusabimana E, Ndagijimana F, Sambandam S, Mukhopadhyay K, Kearns KA, Campbell D, Kremer J, Rosenthal JP, Checkley W, Clasen T, Naeher L; the Household Air Pollution Intervention Network (HAPIN) Trial Investigators. Exposure Contrasts of Pregnant Women during the Household Air Pollution Intervention Network Randomized Controlled Trial. Environ Health Perspect. 2022 Sep;130(9):97005. doi: 10.1289/EHP10295. Epub 2022 Sep 16.

Simkovich SM, Thompson LM, Clark ML, Balakrishnan K, Bussalleu A, Checkley W, Clasen T, Davila-Roman VG, Diaz-Artiga A, Dusabimana E, Fuentes LL, Harvey S, Kirby MA, Lovvorn A, McCollum ED, Mollinedo EE, Peel JL, Quinn A, Rosa G, Underhill LJ, Williams KN, Young BN, Rosenthal J; HAPIN Investigators. A risk assessment tool for resumption of research activities during the COVID-19 pandemic for field trials in low resource settings. BMC Med Res Methodol. 2021 Apr 12;21(1):68. doi: 10.1186/s12874-021-01232-x.

Hengstermann M, Diaz-Artiga A, Otzoy-Sucuc R, Laura Maria Ruiz-Aguilar A, Thompson LM; HAPIN Investigators. Developing Visual Messages to Support Liquefied Petroleum Gas Use in Intervention Homes in the Household Air Pollution Intervention Network (HAPIN) Trial in Rural Guatemala. Health Educ Behav. 2021 Oct;48(5):651-669. doi: 10.1177/1090198121996280. Epub 2021 Mar 18.

Sambandam S, Mukhopadhyay K, Sendhil S, Ye W, Pillarisetti A, Thangavel G, Natesan D, Ramasamy R, Natarajan A, Aravindalochanan V, Vinayagamoorthi A, Sivavadivel S, Uma Maheswari R, Balakrishnan L, Gayatri S, Nargunanathan S, Madhavan S, Puttaswamy N, Garg SS, Quinn A, Rosenthal J, Johnson M, Liao J, Steenland K, Piedhrahita R, Peel J, Checkley W, Clasen T, Balakrishnan K. Exposure contrasts associated with a liquefied petroleum gas (LPG) intervention at potential field sites for the multi-country household air pollution intervention network (HAPIN) trial in India: results from pilot phase activities in rural Tamil Nadu. BMC Public Health. 2020 Nov 26;20(1):1799. doi: 10.1186/s12889-020-09865-1.

Simkovich SM, Thompson LM, Clark M, Balakrishnan K, Bussalleu A, Checkley W, Clasen T, Davila-Roman V, Diaz-Artiga A, de Las Fuentes L, Harvey S, Kirby M, Lovvorn A, McCollum E, Peel J, Quinn A, Rosa G, Underhill L, Williams K, Young B, Rosenthal J. A Risk Assessment Tool for Resumption of Research Activities During the COVID-19 Pandemic. Res Sq. 2020 Nov 12:rs.3.rs-103997. doi: 10.21203/rs.3.rs-103997/v1. Preprint.

Williams KN, Thompson LM, Sakas Z, Hengstermann M, Quinn A, Diaz-Artiga A, Thangavel G, Puzzolo E, Rosa G, Balakrishnan K, Peel J, Checkley W, Clasen TF, Miranda JJ, Rosenthal JP, Harvey SA; Household Air Pollution Intervention Network (HAPIN) trial Investigators; HAPIN Investigators. Designing a comprehensive behaviour change intervention to promote and monitor exclusive use of liquefied petroleum gas stoves for the Household Air Pollution Intervention Network (HAPIN) trial. BMJ Open. 2020 Sep 29;10(9):e037761. doi: 10.1136/bmjopen-2020-037761.

Clasen T, Checkley W, Peel JL, Balakrishnan K, McCracken JP, Rosa G, Thompson LM, Barr DB, Clark ML, Johnson MA, Waller LA, Jaacks LM, Steenland K, Miranda JJ, Chang HH, Kim DY, McCollum ED, Davila-Roman VG, Papageorghiou A, Rosenthal JP; HAPIN Investigators. Design and Rationale of the HAPIN Study: A Multicountry Randomized Controlled Trial to Assess the Effect of Liquefied Petroleum Gas Stove and Continuous Fuel Distribution. Environ Health Perspect. 2020 Apr;128(4):47008. doi: 10.1289/EHP6407. Epub 2020 Apr 29.

Steenland K, Pillarisetti A, Kirby M, Peel J, Clark M, Checkley W, Chang HH, Clasen T. Modeling the potential health benefits of lower household air pollution after a hypothetical liquified petroleum gas (LPG) cookstove intervention. Environ Int. 2018 Feb;111:71-79. doi: 10.1016/j.envint.2017.11.018. Epub 2017 Nov 26.