Effects of a liquefied petroleum gas stove intervention on pollutant exposure and adult cardiopulmonary outcomes (CHAP): study protocol for a randomized controlled trial
Fandiño-Del-Rio et al. Trials
Effects of a liquefied petroleum gas stove intervention on pollutant exposure and adult cardiopulmonary outcomes (CHAP): study protocol for a randomized controlled trial
Magdalena Fandiño-Del-Rio 1 2
Dina Goodman 1
Josiah L. Kephart 1 2
Catherine H. Miele 2
Kendra N. Williams 0 1
Mitra Moazzami 1
Elizabeth C. Fung 1
Kirsten Koehler 2
Victor G. Davila-Roman 5
Kathryn A. Lee 1
Saachi Nangia 1
Steven A. Harvey 0
Kyle Steenland 4
Gustavo F. Gonzales 3
William Checkley 0 1
0 Department of International Health, Bloomberg School of Public Health, Johns Hopkins University , Baltimore, MD , USA
1 Division of Pulmonary and Critical Care, School of Medicine, Johns Hopkins University , 1830 E. Monument St. Room 555, Baltimore, MD 21205 , USA
2 Department of Environmental Health and Engineering, Bloomberg School of Public Health, Johns Hopkins University , Baltimore, MD , USA
3 Department of Biological and Physiological Sciences, Faculty of Sciences and Philosophy, Universidad Peruana Cayetano Heredia , Lima , Peru
4 Department of Environmental Health, Rollins School of Public Health, Emory University , Atlanta, GA , USA
5 Cardiovascular Division, Department of Medicine, Washington University , St. Louis, MO , USA
Background: Biomass fuel smoke is a leading risk factor for the burden of disease worldwide. International campaigns are promoting the widespread adoption of liquefied petroleum gas (LPG) in resource-limited settings. However, it is unclear if the introduction and use of LPG stoves, in settings where biomass fuels are used daily, reduces pollution concentration exposure, improves health outcomes, or how cultural and social barriers influence the exclusive adoption of LPG stoves. Methods: We will conduct a randomized controlled, field intervention trial of LPG stoves and fuel distribution in rural Puno, Peru, in which we will enroll 180 female participants aged 25-64 years and follow them for 2 years. After enrollment, we will collect information on sociodemographic characteristics, household characteristics, and cooking practices. During the first year of the study, LPG stoves and fuel tanks will be delivered to the homes of 90 intervention participants. During the second year, participants in the intervention arm will keep their LPG stoves, but the gas supply will stop. Control participants will receive LPG stoves and vouchers to obtain free fuel from distributors at the beginning of the second year, but gas will not be delivered. Starting at baseline, we will collect longitudinal measurements of respiratory symptoms, pulmonary function, blood pressure, endothelial function, carotid artery intima-media thickness, 24-h dietary recalls, exhaled carbon monoxide, quality-of-life indicators, and stove-use behaviors. Environmental exposure assessments will occur six times over the 2-year follow-up period, consisting of 48-h personal exposure and kitchen concentration measurements of fine particulate matter and carbon monoxide, and 48-h kitchen concentrations of nitrogen dioxide for a subset of 100 participants. (Continued on next page)
(Continued from previous page)
Discussion: Findings from this study will allow us to better understand behavioral patterns, environmental exposures,
and cardiovascular and pulmonary outcomes resulting from the adoption of LPG stoves. If this trial indicates that LPG
stoves are a feasible and effective way to reduce household air pollution and improve health, it will provide important
information to support widespread adoption of LPG fuel as a strategy to reduce the global burden of disease.
Trial registration: ClinicalTrials.gov, ID: NCT02994680, Cardiopulmonary Outcomes and Household Air Pollution (CHAP)
Trial. Registered on 28 November 2016. Keywords: Cookstove, LPG, Indoor air pollution, Household air pollution, Personal exposure, Biomass fuel, Cardiopulmonary outcomes, Behavior change, Exclusive adoption
Household air pollution (HAP), caused by the combustion
of biomass fuels (typically wood, dung, and agricultural
crop waste), is a leading contributor to the global burden of
disease and the largest environmental risk factor for
preventable disease [
]. HAP from biomass fuel smoke was
estimated to be responsible for 4.3 million deaths in 2012
]. It has been associated with chronic bronchitis, chronic
obstructive pulmonary disease (COPD) [
], lung cancer
], childhood pneumonia , acute lower respiratory
], cardiovascular events [
], and low
]. Globally, nearly three billion
individuals, 40% of all households and 90% of rural households
use biomass fuels as their primary source of domestic
energy . Burning of biomass fuels for cooking is most
pervasive in low- and middle-income countries (LMIC).
Women and children have the highest risks of exposure to
biomass fuel smoke, accounting for 60% of premature
deaths from HAP in 2012 [
]. HAP may also have
important effects on both cardiovascular and pulmonary health.
Cardiovascular disease (CVD) is the leading cause of
morbidity and mortality worldwide [
]. More than 80%
of CVD deaths occur in LMICs, where age-adjusted
rates of cardiovascular-related deaths can be up to five
times higher than in high-income countries [
2030, CVD will be responsible for 75% of deaths
worldwide, more than 24 million deaths, and will account for
more deaths in LMICs than infectious diseases, maternal
and perinatal conditions, and nutritional disorders
]. Targeting preventable causes of CVD in
LMICs, such as HAP, is imperative. To date, no field
intervention trial has sufficiently demonstrated that
reducing HAP can decrease CVD morbidity, leaving
policy-makers reluctant to implement cleaner stove
programs as a strategy to address CVD risk.
Cardiovascular disease related to biomass fuel smoke exposure
There is a growing understanding that pulmonary
inflammation caused by HAP affects cardiovascular health [
However, less is known about the mechanism by which
HAP leads to CVD or the true burden of biomass fuel
smoke exposure on CVD risk [
]. Cross-sectional and
longitudinal studies have found strong associations between
ambient air pollution, measured by particulate matter (PM)
levels, and an increased risk of cardiovascular-related death
]. Chronic exposure to biomass fuels has been
associated with a thicker carotid artery intima-media complex,
higher prevalence of atherosclerotic plaques, and higher
blood pressure [
]. PM less than 2.5 μm in diameter
(PM2.5), including ultrafine particles, are small enough to
deposit in alveoli, initiate inflammatory cascades, and enter
the pulmonary circulation [
]. The Women’s Health
Initiative Observational Study followed more than 65,000
women from urban areas for 6 years and found that an
increase of 10 μg/m3 in ambient PM2.5 was associated with
a 24% increased risk of cardiovascular events and a 76%
increased risk of cardiovascular-related deaths [
little is known about the response curve of HAP on
cardiovascular outcomes [
]. Investigators in Guatemala found
that when biomass fuel smoke-related PM exposure was
reduced by 50% via an improved biomass-burning stove, a
reduction in blood pressure within a few months was found
in a sample of women [
]. PM is believed to affect the
circulatory system through increased blood pressure,
inflammation, propagation of coagulation, and increased
blood viscosity [
]. Particularly, endothelial
dysfunction and progression of atherosclerosis are believed
to be an integral link between PM exposure and worse
cardiovascular health . There is clear evidence that
ambient air pollution causes both acute [
] and chronic
endothelial dysfunction [
], but the relationship with
HAP has not been fully explored in intervention trials.
Respiratory disease related to biomass fuel smoke exposure
The adverse health effects of HAP on respiratory health,
and specifically its contribution to COPD, have been well
]. A previous study in rural Peru found
that 55% of COPD prevalence can be attributed to daily
exposure to biomass fuel smoke . Among women,
biomass fuel smoke exposure was associated with a doubling
in the odds of COPD [
] and chronic bronchitis [
similar to that described in other studies [
]. Interventions in
rural Mexico revealed a consistent relationship, with
improved wood stoves being associated with reductions in
respiratory symptoms [
]. However, there is still an
important gap in our understanding of how biomass fuel
smoke exposure affects lung function decline . Previous
findings suggest a log-linear relationship between HAP and
lower respiratory infections, i.e., a steep slope for PM2.5
levels below 100 μg/m3 and a less pronounced slope
]. This implies that very low levels of exposure
need to be achieved to observe health benefits.
Previous research on improved stoves
Improved cookstoves were introduced as a cost-effective
strategy to reduce HAP [
16, 44, 52–55
] through more
efficient combustion of biomass fuels and ventilation.
Most modifications include closed combustion chambers
and chimneys [
]. Although improved cookstoves
have achieved important reductions in HAP, both
personal exposure and kitchen concentrations of
environmental pollutants remain several-fold higher than World
Health Organization (WHO)-recommended levels [
and the effect of these achieved reductions on health
outcomes has not always been evident [
]. For example,
a randomized field trial of improved biomass fuel stoves
in Guatemala did not lower physician-diagnosed
pneumonia (primary outcome) in the intention-to-treat
analysis (RR = 0.84, 95% CI 0.63–1.13) between children in
the intervention and control arms [
]. Similarly, a
recent, large-scale cluster-randomized trial in Malawi did
not reduce childhood pneumonia with an improved
biomass-burning stove intervention [
]. The lack of
health effects for both of these trials is likely attributed
to the continued use of polluting stoves and an inability
of the intervention to achieve reductions in personal
posure below the WHO intermediate targets (35 μg/m )
of household air quality. Prior exposure-response
analyses have found that the greatest risk reductions occur
with much lower exposure levels that are unlikely to be
achieved with improved biomass-burning stoves [
]. For example, in RESPIRE, the exposure-response
analyses found that an expected reduction of 50% in
personal carbon monoxide (CO) concentrations was
associated with a lower incidence of physician-diagnosed
pneumonia (RR = 0.82, 95% CI 0.70–0.98) but the
intention to treat analysis did not show significant
results. The authors attribute the difference in results
between the intention-to-treat and exposure-response
analyses to less exposure misclassification in the latter,
i.e., the intention-to-treat analysis does not account for
subject-specific exposures to HAP. Moreover, larger
expected reductions may lower the risk of
physiciandiagnosed pneumonia even further. Thus, recent
intervention efforts are shifting towards stoves that use
cleaner fuels [
53, 54, 59–62
Current research has demonstrated that liquefied
petroleum gas (LPG) fuels can significantly reduce HAP
when compared to biomass fuels. However, there have
been few studies that have reported on the relationship
between direct measures of HAP and health outcomes
using LPG stoves as an intervention to replace biomass
]. Studies in Guatemala and Bangladesh
demonstrated PM reductions with LPG stove use that ranged
from 45 to 90%, although these studies were
crosssectional and enrolled a small number of participants
53, 54, 61
Additionally, few studies have linked LPG stove use to
health outcomes [
8, 59, 62–66
], and even fewer have
incorporated longitudinal follow-up of HAP and health [
Using retrospective, longitudinal data, a study in China
concluded that women cooking with cleaner fuels,
including LPG, were more likely to report better health and had
a lower probability of chronic and acute diseases when
compared to women cooking with biomass fuels [
cross-sectional study in a rural village near Mexico City
found that women cooking with biomass fuels had more
respiratory symptoms and higher kitchen PM10
concentrations, compared to those cooking with LPG [
crosssectional study of 760 women in rural India found that
43% of those using biomass fuels had abnormal peak
expiratory flow (PEF) readings when compared to 23% of
those using LPG [
]. Cataracts have also been associated
with the use of biomass fuels for cooking [
], and this also
appears to be reduced with LPG. A case-control study
along the Nepal-India border found that women using
traditional cookstoves were 90% more likely to develop
cataracts than women using a combination of biogas,
LPG, and kerosene (OR = 1.90, 95% CI 1.00–3.61) [
This study, however, did not consider the effects of LPG
stove use alone, and included kerosene [
], which has
been classified by the WHO as a dangerous fuel source
]. Overall, there is a lack of longitudinal research
studies examining the health effects of switching from
biomass fuel to LPG for cooking.
A major barrier to assessing the effect of LPG fuel use on
HAP and health is stove stacking, in which families
continue using their traditional biomass stoves after receiving
an LPG stove. Since just 1 h per week of traditional stove
use raises PM2.5 concentrations above WHO guidelines,
even a small amount of stove stacking can negate many of
the benefits of clean cooking technologies [
research has shown that people who use LPG stoves like
the speed of cooking, the cleanliness, and ease of use, but
that the cost of fuel is frequently a barrier to adoption [
Additional barriers to exclusive adoption of LPG stoves
include fears that LPG stoves are unsafe, lack of knowledge
about their proper operation, a desire for more burners,
and changes in the taste of food cooked on LPG stoves
]. However, previous research on the adoption of LPG
stoves has focused on situations where LPG must be
purchased. To date, there is no research on factors that
might hinder or facilitate adoption once cost is removed as
an obstacle [
]. Thus, it is necessary to gain an in-depth
understanding of factors that motivate families to adopt
clean fuels and facilitate the exclusive use necessary to
achieve the potential health benefits.
Previous research has rarely examined other
consequences of stove interventions such as changes in diet
and quality of life. Anderman et al. observed that
households provided with methane biogas stoves had more
diverse diets (measured by number of food groups) than
households using a firewood stove [
little is known about how LPG stoves might affect dietary
patterns. One of the few studies that has looked at
quality of life and biomass burning found that respiratory
quality of life improved after installation of ventilated
cookstoves in rural Bolivia [
]. To date, no intervention
has assessed how a gas stove intervention affects
There is a lack of studies with sufficient sample size and
longitudinal follow-up to determine the health effects of
LPG fuel use when compared to biomass fuel use.
Additionally, few studies have collected high-quality
environmental measurements of both kitchen and personal air
quality. Furthermore, many studies have struggled to
reduce HAP concentrations due to stove stacking. This
study aims to address these gaps by conducting a
longitudinal assessment of clinical outcomes, and personal
exposure and kitchen HAP concentrations over 2 years, while
explicitly investigating and incorporating behavior change
into study activities. Understanding the factors influencing
LPG adoption will enable us to better motivate exclusive
LPG adoption and better quantify the potential impact of
LPG stoves on exposure to HAP and health. Because no
field intervention trial of LPG fuel use has convincingly
demonstrated that a reduction of HAP decreases morbidity
or mortality, policy-makers have been reluctant to invest in
its more widespread adoption. This trial could provide
evidence needed to justify investment in LPG as an effective
public health intervention.
This study is a randomized, field intervention trial
testing the efficacy of LPG stove use and fuel distribution,
compared to traditional, open-fire stove use, as a
strategy to reduce HAP and improve cardiopulmonary
outcomes in the rural, high-altitude setting of Puno, Peru.
LPG stove use will be monitored and compared to
standard cooking practices to determine the relative
effect of LPG adoption on HAP concentrations and
subsequent improvements in cardiopulmonary outcomes over
a 1-year period. As a secondary objective, in the second
year of follow-up, we will measure intervention
effectiveness by characterizing the sustainability of LPG among
participants in the intervention arm and initial adoption
of LPG among those in the control arm (Fig. 1).
Study setting and population
The trial will take place in rural areas surrounding the city
of Puno in southeastern Peru (Fig. 2), on the shore of Lake
Titicaca at 3825 m above sea level. Puno is the capital and
largest city of the Puno Province which, in 2007, had a
population of 230,000 inhabitants [
]. Study participants
live in rural communities in the province of Puno, where
biomass-burning, open-fire stoves are used for cooking
]. Median distance to nearest house between all
households in the study area is 101 m with an interquartile range
of 56 to 189 m (Fig. 3). Median nearest distance to the
highway for households in the study area is 1701 m with an
interquartile range of 893 to 3103 m. Only 4% of
households are within 100 m from the road.
We seek to enroll 180 women, 90 in both the
intervention and control arms. One woman per household will be
enrolled. To be eligible for the trial, women must be aged
25–64 years, be the primary cook, be a full-time resident in
their current location for at least 6 months, be capable of
understanding study procedures, providing informed
consent, and respond to questionnaires, use biomass fuels daily
for cooking, and have a cooking area separate from their
sleeping area. The latter criterion was added to exclude
households that use biomass fuel stoves to heat their living
space. Women will be excluded from participating if they
plan to move from the area within 1 year, have
hypertension (taking antihypertension medications, or systolic blood
pressure ≥ 140 mmHg or diastolic pressure ≥ 90 mmHg) or
a diagnosis of COPD (post-bronchodilator FEV1/FVC
below the lower limit of normal of a reference population),
smoke cigarettes daily, are pregnant or have plans to
become pregnant in the next year, and have active pulmonary
tuberculosis or are taking antituberculosis medications for
pulmonary tuberculosis. Women in these communities
speak Aymara (local language) and Spanish. Field staff
members who will collect data from participants are native
Spanish speakers, and at least one member of the field team
who speaks Aymara will attend each visit.
Participants will be provided with an LPG stove, with
three burners, that connects to an external gas tank
(Fig. 4). Formative research showed that three-burner
stoves were preferred by the community over the
commonly available two-burner stoves. Stoves will be
purchased from a local manufacturer (Industrias SURGES,
Juliaca, Peru). Current open-fire stoves will not be
replaced or destroyed by study personnel. LPG fuel will be
locally purchased and delivered to intervention
participants’ homes for the first year of the study.
Control participants will receive vouchers to pick free fuel
from distributors for the second year of the study. Before
receiving an LPG stove, all intervention participants will
attend a community meeting where they will observe a
cooking demonstration and receive behavioral messages
based on formative research to promote exclusive LPG
stove use. As part of the cooking demonstration,
participants will receive safety information and training on how
to correctly operate and maintain the LPG stove. Correct
and exclusive use of the LPG stove will be reinforced
during the approximately bi-monthly gas delivery visits to the
Each participant who is enrolled in the study will receive a
set of baseline pre-intervention assessments and multiple
follow-up assessments over 24 months. The 90 participants
assigned to the intervention arm will receive a free LPG
stove and free fuel delivery to their homes approximately
twice monthly during the first year to ensure a steady
supply (Fig. 5). Rate of gas use will be tracked by a field staff
member. Participants will be assigned to the intervention
or control arms with a 1:1 ratio using random permuted
block sizes of 2, 4, and 6 [
]. The randomization schedule
was created in R [
] by an investigator who will not be
involved in the recruitment or baseline interviews of
participants. Study-arm allocation will be masked from field staff
in envelopes until baseline measurements are complete. A
field staff member will then open the envelopes.
The study will be conducted over 3 years, with
staggered enrollment over the first year. Our field team will
start recruitment by meeting with the leaders of each
community to explain the study, obtain approval to
approach community residents for participation, and
schedule community meetings before enrollment. Our
team will use previously collected census data to identify
potentially eligible households. At the beginning of the
study, we will randomly determine the order in which
communities and subsectors within communities will be
visited. Each month for the first year, we will randomly
select households within subsectors from the census list
to obtain informed consent and assess eligibility. We
aim to enroll 15 households each month.
A nurse will be present during enrollment and clinical
evaluations to thoroughly explain the trial, the procedures,
and answer questions. We will obtain verbal informed
consent (i.e., waiver of documentation of consent) and we will
leave a written copy of the Consent Form signed by study
staff with each participant. Any serious adverse events will
be identified by the study team who will generate electronic
reports within 24 h after the event. Participants would be
directed towards the nearest emergency health facility for
appropriate medical care. However, since this is a low-risk
study, we do not expect any serious adverse events.
After obtaining informed consent, all participants will be
asked to answer questions about their sociodemographics,
medical history and current clinical symptoms. Personal
exposure and kitchen 48-h PM2.5 and CO concentrations
will be measured for each participant at baseline and
longitudinally during the first year. We will monitor nitrogen
dioxide (NO2) in a subset of 100 participants (50 in both
the intervention and control arms). Stove use will be
Time point (months)
assessed by continuously monitoring the temperature of
LPG and traditional stoves. Health outcomes will be
measured at baseline and repeatedly at different intervals in
the first year (Fig. 6). Primary outcomes include respiratory
symptoms, spirometry, blood pressure, endothelial function
and carotid intima-media thickness, and quality of life.
Secondary outcomes include dietary salt intake and other
macronutrients, inflammatory markers of HAP in urine
and blood, and exhaled CO.
At the end of the 12-month intervention period,
control-arm participants will receive a free LPG stove
and vouchers to cover a 1-year supply of fuel.
Participants will be followed for a second year to determine
patterns of sustained use (intervention arm) or initial
adoption (control arm), and their effects on HAP and
health outcomes (Fig. 7).
The study will be implemented in partnership by
Asociación Benéfica PRISMA (A.B. PRISMA) in Lima, Peru;
Universidad Peruana Cayetano Heredia in Lima, Peru;
and Johns Hopkins University, Baltimore, MD, USA.
Technical support will be provided by experts from
Emory University in Atlanta, GA, USA and Washington
University in Saint Louis, MO, USA. This study was
approved by the Institutional Review Boards of A.B.
PRISMA and Universidad Peruana Cayetano Heredia in
Fig. 7 Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) Figure data collection schedule for the second year
Lima, Peru, and the Johns Hopkins Bloomberg School of
Public Health in Baltimore, MD, USA.
We based sample size calculations on preliminary data
obtained by our group from prior studies [
] and feasibility
intervention trials on the association between HAP and
blood pressure, PEF and St. George’s Respiratory
Questionnaire (SGRQ) (Table 1). To measure a 5-mmHg (SD = 10)
lowering of systolic blood pressure (SBP), a 20-L/min/m2
(SD = 40) improvement in height-adjusted PEF, and a
10-point (SD = 20) higher SGRQ result between
intervention and control arms with 80 to 90% power
and 95% confidence, we need to enroll at least 63 to 85
participants in each arm. These estimations are consistent with
previously observed differences in health outcomes [
]. Thus, we aim to enroll 180 women (90 in each arm) to
account for potential dropouts and loss to follow-up.
Stove adoption and social behavioral components
After enrollment, we will invite intervention arm
participants to a group meeting that will include a cooking
demonstration with local ingredients and recipes,
developed based on formative research. This session will
include information about using an LPG stove safely and
present motivational messaging to encourage
participants to use LPG exclusively.
We will conduct longitudinal, qualitative, in-depth
interviews to explore factors related to LPG adoption. Three
rounds of interviews will be conducted. First, we will
interview 10 participants who use their LPG stove for ≥ 80% of
cooking events and 10 participants who use their LPG stove
for < 80% of cooking events at 1–3 months post
intervention. We will explore differences in barriers, motivators,
practices, and preferences between these two types of LPG
stove users to identify strategies for promoting exclusive
adoption. The same 20 households will then be asked to
participate in a second interview, after owning their LPG
stove and receiving free fuel for 10–12 months, to
determine how their perceptions and stove-use practices may
have changed over time. Finally, we will return to these 20
households for a third interview, at 15–17 months post
intervention, to understand factors influencing sustained
use or abandonment of LPG stoves when fuel is no longer
provided for free. Additional interviews may be conducted
to ensure adequate representation of exclusive and
nonexclusive LPG stove users.
We will also conduct quarterly behavioral
questionnaires to collect information on observed and
selfreported use of all household stoves, reasons for stove
use, time spent cooking and collecting biomass fuels,
and fuel expenditures. In intervention households, we
will also collect information on the likes and dislikes of
the LPG stove, stove maintenance and repair behaviors,
and overall opinions of LPG stoves. These visits will be
used to reinforce exclusive use of LPG through behavioral
messaging and address any problems or concerns with the
gas stove. We will continuously measure stove use during
the 2 years on both LPG and traditional stoves using the
Digit-TL temperature monitor (LabJack Corporation,
Lakewood, CO, USA).
We will measure 48-h personal exposure and kitchen
concentrations of PM2.5 with the ECM Monitor (RTI
Inc., Research Triangle Park, NC, USA) and CO with the
EL-USB-CO data logger (Lascar Electronics, Erie, PA,
USA) at each environmental assessment visit. Both the
ECM and EL-USB-CO are light-weight monitors and
can be easily worn by participants without disrupting
daily activities (Fig. 8). The ECM has a light-scattering
laser for real-time assessment of PM2.5, and a 0.3-L/min
pump that will be continuously on for 48 h to
gravimetrically collect PM2.5 in a 15-mm diameter filter. We will
calibrate the ECM pumps daily with a TSI 4100
flowmeter (TSI Incorporated, 500 Cardigan Road, Shoreview,
MN, USA). We will use 15-mm Teflon filters with a 2-μm
membrane (Measurement Technology Laboratories LLC,
Minneapolis, MN, USA). All filters will be pre- and
postweighed in a humidity- and temperature-controlled room
using a XP2U microbalance (Mettler Toledo, Columbus,
OH, USA) located in the Department of Environmental
Health and Engineering of the Bloomberg School of Public
Health at Johns Hopkins University. We will conduct direct
readings of PM2.5 every 3 out of 10 s and time-weight
average these values into 1-min intervals, and conduct direct
readings of CO in 1-min intervals. We will ask participants
to wear the ECM and CO monitors near the breathing
zone in the pocket of an apron provided to them. The
aprons, which are similar to those commonly used by
women in our setting, were selected as the most feasible
and acceptable method for personal exposure sampling
during formative research (Fig. 9). Participants will be
encouraged to wear the aprons during awake hours and
colocate them near their beds when sleeping.
We will measure kitchen concentrations of PM2.5 and
CO using the same instruments described above. Monitors
will be placed approximately 1 m from the combustion
zone, at 1.5 m of height from the floor, and at least 1 m
away from doors and windows, when possible. We will also
measure personal exposure and kitchen
concentrations of NO2 in a subsample of households using the
Aeroqual 500 series monitors (Aeroqual Limited,
Auckland, New Zealand). Passive NO2 samplers from
Ogawa (Ogawa USA, Pompano Beach, FL, USA) will
measure personal exposure to NO2 in a subsample of
20 women. Passive NO2 samples will be analyzed
using standard methods [
Kitchen samples of PM2.5 will include 10% blanks and
10% duplicates, and all reported concentrations will be
blank-corrected. Kitchen CO samples will include 10%
duplicates. Calibration checks will be performed every 3
months for the CO monitors using a chamber to test the
devices with clean air and two, different, known CO
concentrations. Staff will also record general
characteristics of the kitchen including wall materials, and the
presence of windows and doors.
Ambient particulate matter pollution (pDR-1000, Thermo
Fisher Scientific, Wharton, MA, USA) and meteorology
(Vantage VUE, Davis Instruments, Hayward, CA, USA) will
be monitored in a central location during the study period.
Pulmonary health outcomes
We will measure respiratory symptoms with the SGRQ.
The SGRQ is a standardized, self-completed
questionnaire for measuring impaired health and perceived
wellbeing in individuals with chronic airway disease [
This questionnaire is easy to use, gives a comparative
measurement of respiratory health between populations,
and can be used to quantify changes in respiratory
health following interventions [
]. It has been
translated and validated for use in Spanish.
We will measure exhaled CO using the Micro CO
Meter (Micro Direct, Lewiston, ME, USA) as an additional
measure of compliance. We will take two exhaled CO
measurements and average them to obtain the final value.
The Micro CO Meter will be calibrated monthly with a
20-ppm CO concentration gas cylinder connected directly
to the device.
We will use the EasyOn PC handheld spirometer (ndd
Medical Technologies Inc., Zurich, Switzerland) to assess
lung function [
]. The EasyOn PC has an ultrasonic
flow reader that is not affected by air density and is suitable
for use at high altitudes. We will check spirometer
calibration weekly using a 3-L syringe (Hans Rudolph Inc., Kansas
City, MO, USA). If a spirometer reads more than 3.5%
above or below 3 L , it will not be used in the field and
will be repaired or replaced. All team members involved in
administering tests will be trained to comply with standard
]. Regular supervision and feedback will take
place via a centralized quality-control program [
will record forced vital capacity (FVC), forced expiratory
volume in 1 s (FEV1), peak expiratory flow (PEF), and store
individual flow-volume curves for quality-control
assessment and further analysis. Using a salbutamol inhaler
approved for use by the Peruvian General Directorate of
Medication, Drugs, and Supplies, we will administer two
puffs (100 mcg/puff) via a spacer and repeat spirometry
after 15 min for reversibility testing, i.e., an improvement
of > 12% or > 0.2 L in baseline forced expiratory volumes.
We will measure oxyhemoglobin saturation with a Rad
5v pulse oximeter (Masimo, Irvine, CA, USA). We will
take two oxyhemoglobin saturation measurements and
average them to obtain the final value.
Cardiovascular health outcomes
SBP and diastolic blood pressure (DBP) will be measured
in triplicate in 5-min intervals and in the sitting position
using an automatic blood pressure monitor OMRON
HEM-780 (Omron, Tokyo, Japan). At enrollment, we
will determine the arm with the highest SBP and will
use that arm thereafter for all measurements. We will
average the second and third values to determine final
SBP and DBP.
cIMT is a noninvasive, surrogate marker of
atherosclerosis and has also been shown to predict future
cardiovascular events [
]. Increased thickness of
the carotid artery is associated with increased risk for
stroke, myocardial infarction, and other adverse
cardiovascular outcomes as previously shown by our group and
33, 83, 84
]. We will follow standard guidelines for
cIMT assessment [
]. A high-frequency portable
ultrasound (M-Turbo, Sonosite Inc., Bothell, WA, USA) with a
linear transducer (6–13 MHz) will be used to visualize the
carotid vessel. Special attention will be paid to obtain vessel
interface and carotid plaques, wall thickness of the distal
common carotid artery, and Doppler velocity. The cardiac
cycle will be tracked with an electrocardiogram (EKG)
monitor built into the ultrasound [
] for cIMT assessment.
We will evaluate subjects in the morning after an overnight
fast. cIMT images will be saved in DICOM format on a
local server, and then uploaded to a secure Cloud server
(Ultralinq, New York, NY, USA) for transfer to the USA.
Endothelial function will be measured using brachial
artery reactivity testing (BART) [
] and peripheral arterial
tonometry (PAT) [
], which will be administered
simultaneously. BART and PAT are non-invasive methods for
assessing endothelial function and are associated with future
CVD risk [
]. A higher BART score, indicating
increased flow-mediated dilation (FMD) in the brachial
artery, is associated with better cardiovascular health. A
systematic review of 23 studies and 14,753 participants
found that a 1% higher FMD was associated with an 8%
lower risk of a future CVD event [
]. We will use portable
ultrasound (M-Turbo, Sonosite, Bothell, WA, USA) with a
high-frequency linear probe (6–13 MHz) to visualize the
brachial artery in longitudinal view following standard
]. We will measure the diameter of the vessel
at end diastole before and after distal arm ischemia. We will
use a blood pressure cuff immediately distal to the
antecubital fossa and inflate the cuff to 240 mmHg to obtain
sufficient occlusion for distal arm ischemia. We will maintain
cuff inflation for 5 min and measure post-ischemia vessel
diameter at 30, 60, 90, 120, and 150 s after cuff deflation
and calculate the percentage change in diameter from
baseline to each post-ischemia measurement. Time of dilation
related to cardiac cycle will be tracked with an EKG
monitor built into the ultrasound [
]. BART images will be
saved in DICOM format on a local server, and then
uploaded to a secure Cloud server (Ultralinq, New York,
NY, USA) for transfer to the USA.
Image analysis for BART and cIMT will be done at
Washington University. All studies will be analyzed by
two sonographers who will be blinded. The primary
reader will evaluate all tests from the study, while the
secondary reader (expert) will evaluate 10% of the tests
at random for quality control. Intra- and inter-observer
intra-class correlation coefficients at the Washington
University Core Laboratory for BART and CIMT
measurements are greater than 0.91.
An EndoPAT (Itamar, Caesarea, Israel) will be used for
PAT. Finger pulse-wave amplitude will be recorded with
the EndoPAT probe placed on the index finger of the
same arm for BART. A second EndoPAT probe will be
placed on the contralateral index finger. A reactive
hyperemia index will be recorded and normalized. The
EndoPAT device documents the percentage change
between pre- and post-ischemia tests. EndoPAT data will
be saved in a local server for FTP transfer. Data analysis
will be done at Johns Hopkins.
Anthropometry and nutrition outcomes
Each participant will have weight and height (standing
and sitting) measured at baseline and at 12 months post
intervention using a standardized protocol. Dietary
intake will be assessed through 24-h recalls. Participants
will describe what they consumed both individually and
in shared meals. Whenever possible, food will be
weighed using a Henkel Max 5000-g d-1G scale (HV
Digital Eirl, Lima, Peru). Otherwise, participants will be
asked to identify portion sizes visually using a standardized
book developed by A.B. PRISMA [
]. This book contains
to-scale pictures of common food items, plate-ware and
corresponding weight in grams. Salt intake will be assessed
in a subset of 100 participants with a 24-h urine test.
Participants will be provided with a receptacle to urinate for a
24-h period. The samples will be mixed and aliquoted
before shipment to a clinical laboratory (Medlab, Arequipa,
Peru) for analysis. The sodium measured in urine will be
used to validate salt intake recorded in the 24-h recalls.
We will measure quality of life with the RAND 36-Item
Short Form Health Survey (SF-36) [
] and the EuroQol
five dimensions questionnaire (EQ5D) [
]. We will use
these quality-of-life indicators to calculate differences in
quality-adjusted life years between the intervention and
control arms. The validated Spanish versions of both the
SF-36 and the E5QD are being used in this study.
Urine and blood markers
Women will be provided with a 500-mL urine collection
cup and instructed to urinate briefly to waste before
collecting the remainder of the urine void in the cup.
The time of urine collection and time of previous urine
void (if known) will be recorded and the total volume
collected will be estimated and recorded. The urine will
be transferred in its entirety to 3-oz Qorpak bottles and
labeled. The labeled bottles will be secured in freezer
boxes and will be stored in a cooler with ice packs until
they can be transported to a − 20 °C freezer.
Blood will be collected as dried blood spots (DBS). A
finger from the non-dominant hand will be swabbed
with a sterile alcohol wipe. The sterile lancet will be used
to puncture the skin and the initial drop of blood will be
wiped away with an alcohol swab. Blood from the finger
will be allowed to drip onto five standard spots on a
Guthrie DBS card. The finger will be squeezed until a
large drop of blood appears on the finger and the drop
will be quickly applied to an unfilled spot on the card.
Each spot will contain approximately 100 μL blood so
the total blood collection will be less than 1 mL. Cards will
be labeled and dried at room temperature (20–25 °C) on
drying rack for 10 h without the use of external heat or
fan. Dried cards will be placed in individually labeled
ziptop bags into which a desiccant pouch and a humidity
indicator card will be placed. Urine and DBS samples will
be blinded to assignment before biomarker analysis at
Emory University (Table 2).
Most clinical biomarkers will be analyzed in duplicate
through immunoassay methods with the MesoScale MSD
Mutiplexer Clinical Analyzer. We will perform one
reagent blank per run, three quality-control materials per
run, and eight calibration samples per plate. For mass
spectrometry-based methods we will use Agilent 7000
Triple quadrupole MS/MS with a gas chromatograph and
electron impact and chemical ionization capabilities. We
will perform calibrations, reagent blanks, and matrix
blanks every 25 samples.
We will conduct intention-to-treat analyses of
cardiovascular and pulmonary health endpoints (primary outcomes)
during the first year of the intervention. We will calculate
48-h personal exposure and kitchen PM2.5 and CO
concentrations, and compare differences in these concentrations
between treatment arms. We will use linear mixed-effects
models to examine the effect of the intervention on
subject-specific trajectories of cardiovascular (SBP, DBP,
metrics of FMD, and cIMT) and pulmonary (SGRQ score,
PEF, and FEV1) endpoints. As a sensitivity analysis, we will
evaluate exposure reduction and clinical outcome
relationships. Specifically, we will ignore random allocation and
instead we will examine if there is an exposure-response
relationship between reduction in pollutant concentrations
and change in clinical outcomes. For this analysis, we will
use linear mixed-effects models to measure the
subjectspecific changes in clinical outcomes in relation to
reductions of personal exposure and kitchen concentrations.
Since this type of analysis breaks randomization, we will
adjust for any potential confounders.
Stove-use monitoring data will be summarized to
determine the number of cooking events and cooking
P53 tumor-associated antigen antibodies (p53 TAA antibodies)a
Lung and other cancer biomarker
Volatile organic chemicals (mercapturate metabolites)a
Polycyclic aromatic hydrocarbons (1-OH pyrene and 1-, and 2-naphthols) Carcinogen exposure biomarker
Short term tobacco smoke biomarker
Tobacco smoke biomarker
Carcinogen exposure biomarker
duration per day for each participant in LPG and traditional
stoves. Stove-use monitoring will also be useful in
exposure-response analyses described above.
Secondary analyses will focus on effectiveness
outcomes during the second year. Specifically, we plan to
calculate the cumulative incidence of abandonment of
LPG stoves and percentage use of LPG stoves in the
intervention arm, adoption of LPG stoves in the control
arm, and factors affecting these statistics. When
participants decide to drop out, field staff will complete a
questionnaire to document potential biases related to missing
data. We will incorporate recommended methods from
the academic literature and sensitivity analysis to properly
account for missing data [
Qualitative data will be analyzed inductively, coding
transcripts with emerging themes using ATLAS.ti
(ATLAS.ti, Berlin, Germany). Information from coded
quotes will be synthesized and compared within and
Data management and quality assurance
Questionnaires and other field forms will be collected
on tablets using the Research Electronic Data Capture
software (REDCap, Vanderbilt University Medical
Center, Nashville, TN, USA) [
]. Images collected
from BART, PAT, and cIMT will be uploaded to a secure
Cloud server (Ultralinq Health Care Solutions, New
York, NY, USA). Twenty-four-hour food recalls will be
collected on paper forms, combined with nutrition facts
from official Peruvian Ministry of Health documents
], and entered into a database. Qualitative
interviews will be audio-recorded, transcribed by local field
staff into Spanish, and then translated into English. Data
processing and storage will be centralized at a server in
Puno, Peru. Personal information will be maintained
confidential before, during, and after the trial.
The study team will be in constant contact with the
study site team with weekly scheduled meetings and
additional communication as needed for quality control.
All data that is collected will be de-identified and
uploaded onto Cloud servers for real-time access. As
study procedures are completed, the study team will
have real-time oversight of all activities.
Publications on the results and analysis of the primary
outcomes and main objectives will be prioritized over
those on secondary outcomes and objectives. We will
follow the Consolidated Standards of Reporting Trials
(CONSORT) 2010 guidelines when reporting the main
results of the trial [
]. Paper topics and authorship will
be discussed with the respective team leaders of each
component and the principal investigator. In
accordance with NIH data-sharing policy, we will submit our
de-identified, limited dataset to the National Heart,
Lung and Blood Institute Biologic Specimen and Data
Repository after publication [
LPG stoves have been recently proposed as a strategy to
achieve WHO air-quality guidelines in LMICs, but there
are limited studies, much less studies with repeated
measurements, which measure the effects of LPG stove use
on HAP and adult cardiopulmonary outcomes. Similarly,
there is little information on potential outcomes if
sustained LPG stove adoption is achieved. To address these
gaps, we plan to conduct an LPG field intervention trial
with extensive behavioral support and repeated HAP
and health outcome assessments to understand how
LPG adoption could reduce disease burden in LMICs.
We will measure stove use, personal exposure and
kitchen pollutant concentrations, and cardiopulmonary
outcomes repeatedly during 2 years. We will monitor
biomarkers of exposure and nutritional changes.
Longitudinal measurements over 2 years will allow us to
identify short-term health benefits of the intervention.
Randomization will provide confidence in the
comparability between intervention and control arms. If this
trial demonstrates that LPG stoves reduce HAP and
improve health, it would provide critical evidence that
shifting from biomass to LPG fuel stoves is an effective
way to reduce the burden of cardiopulmonary-related
illnesses and death.
One of the challenges in cookstove interventions is the
difficulty of ensuring exclusive LPG stove use. Study
participants may continue to use their traditional open-fire
stoves for various reasons. First, these open-fire stoves
will not be removed from the home and participants will
be free to use either stove. To address this challenge, we
will incorporate regular visits to reinforce the exclusive
use of LPG stoves. Second, participants may not like the
taste of food or may believe that they cannot cook
traditional dishes like in an open-fire stove. Therefore, we
have incorporated cooking demonstrations to show
participants that it is possible to prepare traditional
dishes with an LPG stove that taste similar to meals
cooked on an open-fire stove. Third, participants may
have safety concerns that may deter the use of LPG
stoves, but this will also be addressed through education
during our field visits. Fourth, achieving a consistent
supply of LPG fuel to our study participants may be
challenging. To address this potential concern, we have
incorporated strategies to monitor LPG use and deliver
fuel tanks based on usage rates.
The second year of our study will provide valuable
information on continued use and factors related to
abandonment or adoption of LPG stoves. These results
will inform scale-up and implementation of future LPG
programs. An additional challenge is the lack of outdoor
air-quality measurements in our study area to control for
external sources of pollutants that could affect HAP,
including neighbors with biomass-burning stoves. To characterize
ambient air pollution in our study area, air quality will be
monitored in a central location. However, given the large
distances between houses in our rural setting,
crosscontamination between households is less likely.
While we hypothesize that LPG stoves will reduce HAP
and improve cardiopulmonary outcomes, there are other
sources of energy, such as electricity, which are likely to
produce even greater benefits through a more consistent
supply and lower emissions. However, electric stoves are
not currently appropriate for many communities around
the world because of the lack of access to electricity.
Moreover, LPG has become more widely available with a
supply infrastructure to meet demands in many countries.
In Peru, the communities in our study area have
experience with LPG stoves and there are existing government
initiatives to make LPG more affordable.
The trial has been registered at www.clinicaltrials.gov
(NCT02994680). Trial has not started enrollment at the
time of manuscript submission. Enrollment is expected to
begin on 18 January 2017. This protocol paper for this trial
complies with the SPIRIT (Standard Protocol Items:
Recommendations for Interventional Trials) [
and the World Health Organization Trial Registration
] (Figs. 6 and 7, Additional files 1 and 2).
Additional file 1: SPIRIT Checklist. (PDF 682 kb)
Additional file 2: World Health Organization Trial Registration Data Set
Items. (PDF 45 kb)
A.B. PRISMA: Asociación Benéfica PRISMA; BART: Brachial artery reactivity
testing; cIMT: Carotid intima-media thickness; CO: Carbon monoxide;
COPD: Chronic obstructive pulmonary disease; CVD: Cardiovascular disease;
DBP: Diastolic blood pressure; DBS: Dried blood spots;
EKG: Electrocardiogram; EQ5D: EuroQol five dimensions questionnaire;
FEV1: Forced expiratory volume at 1 s; FMD: Flow-mediated dilation;
FVC: Forced vital capacity; HAP: Household air pollution; IRB: Institutional
Review Board; LMIC: Low- and middle-income countries; LPG: Liquefied
petroleum gas; NIH: National Institute of Health; NO2: Nitrogen dioxide;
PAT: Peripheral arterial tonometry; PEF: Peak expiratory flow; PM: Particulate
matter; PM2.5: PM less than 2.5 μm in aerodynamic diameter; SBP: Systolic
blood pressure; SF-36: 36-Item Short Form Health Survey; SGRQ: St. George’s
Respiratory Questionnaire; SPIRIT: Standard Protocol Items:
Recommendations for Interventional Trials; WHO: World Health Organization
USA), and Joshua Rosenthal (Fogarty International Center, National Institutes
of Health, Bethesda, MD, USA) for their help with this study.
Financial support was received from the Global Environmental and
Occupational Health, Fogarty International Center, United States National
Institutes of Health (1U2RTW010114-01); the Global Alliance for Clean
Cookstoves of the United Nations Foundation (UNF 16-80), and the Johns
Hopkins Center for Global Health. The content is solely the responsibility of
the authors and does not necessarily represent the official views of these
Availability of data and materials
WC, VD, KK, SH, KS, and GG conceived the original study design. KK, SH, and
WC provided expert guidance to the design and implementation of the
study. MF, DG, JK, CM, KW, MM, EF, KL, and SN were responsible for pilot
samples, protocol design and study implementation. KS and GG provided
expert guidance in the design and logistics of the project. MF, KW, DG, JK,
and WC led the writing of the manuscript. WC had ultimate oversight over
the design of this trial. All authors contributed to the development of the
study design and the writing of the manuscript. All authors read and
approved the final manuscript.
Ethics approval and consent to participate
The trial received approval by Johns Hopkins School of Public Health
Institutional Review Board (IRB00007128), A.B. PRISMA Ethical Institutional
Committee (CE2402.16), and Universidad Peruana Cayetano Heredia
Institutional Review Board (SIDISI 66780). We will be requesting verbal
consent to participate from all participants at the time of enrollment. Any
amendments will undergo ethical review board approval at all institutions
Consent for publication
Written informed consent was obtained from the participant for publication
of their individual details and accompanying images in this manuscript. The
Consent Form is held by the authors and is available for review by the
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.
1. Bruce N , Perez-Padilla R , Albalak R . Indoor air pollution in developing countries: a major environmental and public health challenge . Bull World Health Organ . 2000 ; 78 : 1078 - 92 .
2. Smith KR , Bruce N , Balakrishnan K , Adair-Rohani H , Balmes J , Chafe Z , et al. 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 . 2014 ; 35 : 185 - 206 .
3. World Health Organization. Global Health Risks: mortality and burden of disease attributable to selected major risks . Geneva: WHO Press; 2009 . http://www.who.int/healthinfo/global_burden_disease/ GlobalHealthRisks_ report_full.pdf, Accessed 21 Jul 2016 .
4. World Health Organization. Burden of disease from household air pollution for 2012 . 2014 . http://www.who.int/phe/health_topics/outdoorair/ databases/FINAL_HAP_ AAP_BoD_24March2014 . pdf. Accessed 21 Jul 2016 .
5. Dennis R , Maldonado D , Norman S , Baena E , Martinez G . Woodsmoke exposure and risk for obstructive airways disease among women . Chest . 1996 ; 109 : 115 - 9 .
6. Menezes AM , Victora CG , Rigatto M. Prevalence and risk factors for chronic bronchitis in Pelotas, RS, Brazil: a population-based study . Thorax . 1994 ; 49 : 1217 - 21 .
7. Orozco-Levi M , Garcia-Aymerich J , Villar J , Ramírez-Sarmiento A , Antó JM , Gea J . Wood smoke exposure and risk of chronic obstructive pulmonary disease . Eur Respir J . 2006 ; 27 : 542 - 6 .
8. Shrestha IL , Shrestha SL . Indoor air pollution from biomass fuels and respiratory health of the exposed population in Nepalese households . Int J Occup Environ Health . 2005 ; 11 : 150 - 60 .
9. Rinne ST , Rodas EJ , Bender BS , Rinne ML , Simpson JM , Galer-Unti R , et al. Relationship of pulmonary function among women and children to indoor air pollution from biomass use in rural Ecuador . Respir Med . 2006 ; 100 : 1208 - 15 .
10. Ramírez-Venegas A , Sansores RH , Pérez-Padilla R , Regalado J , Velázquez A , Sánchez C , et al. Survival of patients with chronic obstructive pulmonary disease due to biomass smoke and tobacco . Am J Respir Crit Care Med . 2006 ; 173 : 393 - 7 .
11. Akhtar T , Ullah Z , Khan MH , Nazli R . Chronic bronchitis in women using solid biomass fuel in rural Peshawar, Pakistan . Chest. 2007 ; 132 : 1472 - 5 .
12. Liu S , Zhou Y , Wang X , Wang D , Lu J , Zheng J , et al. Biomass fuels are the probable risk factor for chronic obstructive pulmonary disease in rural South China . Thorax . 2007 ; 62 : 889 - 97 .
13. Pérez-Padilla R , Regalado J , Vedal S , Paré P , Chapela R , Sansores R , et al. Exposure to biomass smoke and chronic airway disease in Mexican women. A case-control study . Am J Respir Crit Care Med . 1996 ; 154 ( 3 Pt 1 ): 701 - 6 .
14. Liu Q , Sasco AJ , Riboli E , Hu MX . Indoor air pollution and lung cancer in Guangzhou, People's Republic of China . Am J Epidemiol . 1993 ; 137 : 145 - 54 .
15. Ko YC , Lee CH , Chen MJ , Huang CC , Chang WY , Lin HJ , et al. Risk factors for primary lung cancer among non-smoking women in Taiwan . Int J Epidemiol . 1997 ; 26 : 24 - 31 .
16. Smith K , McCracken J , Weber M , Hubbard A , Jenny A , Thompson L , et al. Effect of reduction in household air pollution on childhood pneumonia in Guatemala (RESPIRE): a randomised controlled trial . Lancet . 2011 ; 378 : 1717 - 26 .
17. Bautista LE , Correa A , Baumgartner J , Breysse P , Matanoski GM . Indoor charcoal smoke and acute respiratory infections in young children in the Dominican Republic . Am J Epidemiol . 2009 ; 169 : 572 - 80 .
18. Mishra V . Indoor air pollution from biomass combustion and acute respiratory illness in preschool age children in Zimbabwe . Int J Epidemiol . 2003 ; 32 : 847 - 53 .
19. Fullerton D , Bruce N , Gordon S. Indoor air pollution from biomass fuel smoke is a major health concern in the developing world . Trans R Soc Trop Med Hyg . 2008 ; 102 : 843 - 51 .
20. Miller KA , Siscovick DS , Sheppard L , Shepherd K , Sullivan JH , Anderson GL , et al. Long-term exposure to air pollution and incidence of cardiovascular events in women . N Engl J Med . 2007 ; 356 : 447 - 58 .
21. Boy E , Bruce N , Delgado H . Birth weight and exposure to kitchen wood smoke during pregnancy in rural Guatemala . Environ Health Perspect . 2002 ; 110 : 109 - 14 .
22. Yucra S , Tapia V , Steenland K , Naeher LP , Gonzales GF . Association between biofuel exposure and adverse birth outcomes at high altitudes in Peru: a matched case-control study . Int J Occup Environ Health . 2011 ; 17 : 307 - 13 .
23. World Health Organization. Burning opportunity: clean household energy for health, sustainable development, and wellbeing of women and children . Geneva: WHO Press; 2016 . http://apps.who.int/iris/bitstream/10665/204717/ 1/9789241565233_eng.pdf.
24. Mendis S , Armstrong T , Bettcher D , Branca F , Lauer J , Mace C , et al. The WHO Global Status Report on Noncommunicable Diseases 2014 . 2014 . http://apps.who.int/iris/bitstream/10665/148114/1/9789241564854_eng. pdf?ua=1. Accessed 26 Jul 2016 .
25. Institute of Medicine (US) Committee on Preventing the Global Epidemic of Cardiovascular Disease: Meeting the Challenges in Developing Countries . Promoting cardiovascular health in the developing world: a critical challenge to achieve global health . Washington (DC): National Academies Press (US); 2010 . http://www.ncbi.nlm.nih.gov/books/NBK45693/. Accessed 29 Aug 2016 .
26. NH UnitedHealth. Global response to non-communicable disease . BMJ . 2011 ; 342 : d3823 .
27. Brook RD , Franklin B , Cascio W , Hong Y , Howard G , Luepker R , et al. Air pollution and cardiovascular disease: a statement for healthcare professionals from the expert panel on population and prevention science of the American Heart Association . Circulation. 2004 ; 109 : 2655 - 71 .
28. Pope CA , Burnett RT , Thurston GD , Thun MJ , Callee EE , Krewski D , et al. Cardiovascular mortality and long-term exposure to particulate air pollution: epidemiological evidence of general pathophysiological pathways of disease . Circulation . 2004 ; 109 : 71 - 7 .
29. Pope C , Burnett R , Krewski D , Jerrett M , Shi Y , Calle E , et al. Cardiovascular mortality and exposure to airborne fine particulate matter and cigarette smoke: shape of the exposure-response relationship . Circulation . 2009 ; 120 : 941 - 8 .
30. Brook RD , Brook JR , Urch B , Vincent R , Rajagopalan S , Silverman F. Inhalation of fine particulate air pollution and ozone causes acute arterial vasoconstriction in healthy adults . Circulation . 2002 ; 105 : 1534 - 6 .
31. O 'Neill MS , Veves A , Zanobetti A , Sarnat JA , Gold DR , Economides PA , et al. Diabetes enhances vulnerability to particulate air pollution-associated impairment in vascular reactivity and endothelial function . Circulation . 2005 ; 111 : 2913 - 20 .
32. Rundell K , Hoffman J , Caviston R , Bulbulian R , Hollenbach A . Inhalation of ultrafine and fine particulate matter disrupts systemic vascular function . Inhal Toxicol . 2007 ; 19 : 133 - 40 .
33. Painschab MS , Davila-Roman VG , Gilman RH , Vasquez-Villar AD , Pollard SL , Wise RA , et al. Chronic exposure to biomass fuel is associated with increased carotid artery intima-media thickness and a higher prevalence of atherosclerotic plaque . Heart . 2013 ; 99 : 984 - 91 .
34. Brook RD , Rajagopalan S , Pope C , Brook JR , Bhatnagar A , Diez-Roux AV , et al. Particulate matter air pollution and cardiovascular disease: an update to the scientific statement from the American Heart Association . Circulation. 2010 ; 121 : 2331 - 78 .
35. Burnett RT , Pope CA , Ezzati M , Olives C , Lim SS , Mehta S , et al. An integrated risk function for estimating the global burden of disease attributable to ambient fine particulate matter exposure . Environ Health Perspect . 2014 ; 122 : 397 - 403 .
36. McCracken JP , Smith KR , Díaz A , Mittleman MA , Schwartz J . Chimney stove intervention to reduce long-term wood smoke exposure lowers blood pressure among Guatemalan women . Environ Health Perspect . 2007 ; 115 : 996 - 1001 .
37. Baumgartner J , Schauer JJ , Ezzati M , Lu L , Cheng C , Patz JA , et al. Indoor air pollution and blood pressure in adult women living in rural China . Environ Health Perspect . 2011 ; 119 : 1390 - 5 .
38. Baumgartner J , Zhang Y , Schauer JJ , Ezzati M , Patz JA , Bautista LE . Household air pollution and children's blood pressure . Epidemiology . 2012 ; 23 : 641 - 2 .
39. Baumgartner J , Zhang Y , Schauer JJ , Huang W , Wang Y , Ezzati M. Highway proximity and black carbon from cookstoves as a risk factor for higher blood pressure in rural China . Proc Natl Acad Sci U S A . 2014 ; 111 : 13229 - 34 .
40. Wilker EH , Ljungman PL , Rice MB , Kloog I , Schwartz J , Gold DR , et al. Relation of long-term exposure to air pollution to brachial artery flowmediated dilation and reactive hyperemia . Am J Cardiol . 2014 ; 113 : 2057 - 63 .
41. Briet M , Collin C , Laurent S , Tan A , Azizi M , Agharazii M , et al. Endothelial function and chronic exposure to air pollution in normal male subjects . Hypertension . 2007 ; 50 : 970 - 6 .
42. Menezes AM , Perez-Padilla R , Jardim JB , Muiño A , Lopez MV , Valdivia G , et al. Chronic obstructive pulmonary disease in five Latin American cities (the PLATINO study): a prevalence study . Lancet . 2005 ; 366 : 1875 - 81 .
43. Buist AS , McBurnie MA , Vollmer WM , Gillespie S , Burney P , Mannino DM , et al. International variation in the prevalence of COPD (The BOLD Study): a population-based prevalence study . Lancet . 2007 ; 370 : 741 - 50 .
44. Smith-Silverstein T , Diaz E , Pope D , Lie RT , Diaz A , McCracken J , et al. Effect of reducing indoor air pollution on women's respiratory symptoms and lung function: the RESPIRE randomized trial , Guatemala. Am J Epidemiol . 2009 ; 170 : 211 - 20 .
45. Mannino DM , Braman S. The epidemiology and economics of chronic obstructive pulmonary disease . Proc Am Thorac Soc . 2007 ; 4 : 502 - 6 .
46. Kurmi OP , Semple S , Simkhada P , Smith WCS , Ayres JG . COPD and chronic bronchitis risk of indoor air pollution from solid fuel: a systematic review and meta-analysis . Thorax . 2010 ; 65 : 221 - 8 .
47. Gordon SB , Bruce NG , Grigg J , Hibberd PL , Kurmi OP , Lam KH , et al. Respiratory risks from household air pollution in low and middle income countries . Lancet Respir Med . 2014 ; 2 : 823 - 60 .
48. Jaganath D , Miranda JJ , Gilman RH , Wise RA , Diette GB , Miele CH , et al. Prevalence of chronic obstructive pulmonary disease and variation in risk factors across four geographically diverse resource-limited settings in Peru . Respir Res . 2015 ; 16 : 40 .
49. Miele C , Jaganath D , Miranda J , Bernabé-Ortiz A , Gilman R , Johnson C , et al. Urbanization and daily exposure to biomass fuel smoke both contribute to chronic bronchitis risk in a population with low prevalence of daily tobacco smoking . COPD J Chron Obstruct Pulmon Dis . 2015 ; 13 : 1 - 10 .
50. Romieu I , Riojas-Rodríguez H , Marrón-Mares A , Schilmann A , Perez-Padilla R , Masera O . Improved biomass stove intervention in rural Mexico: impact on the respiratory health of women . Am J Respir Crit Care Med . 2009 ; 180 : 649 - 56 .
51. Ezzati M , Kammen DM . Quantifying the effects of exposure to indoor air pollution from biomass combustion on acute respiratory infections in developing countries . Environ Health Perspect . 2001 ; 109 : 481 - 8 .
52. Thomas E , Wickramasinghe K , Mendis S , Roberts N , Foster C . Improved stove interventions to reduce household air pollution in low and middle income countries: a descriptive systematic review . BMC Public Health . 2015 ; 15 : 650 .
53. Albalak R , Bruce N , McCracken JP , Smith KR , De Gallardo T. Indoor respirable particulate matter concentrations from an open fire, improved cookstove, and LPG/open fire combination in a rural Guatemalan community . Environ Sci Technol . 2001 ; 35 : 2650 - 5 .
54. Naeher LP , Leaderer BP , Smith KR . Particulate matter and carbon monoxide in highland Guatemala: indoor and outdoor levels from traditional and improved wood stoves and gas stoves . Indoor Air . 2000 ; 10 : 200 - 5 .
55. Jary HR , Kachidiku J , Banda H , Kapanga M , Doyle JV , Banda E , et al. Feasibility of conducting a randomised controlled trial of a cookstove intervention in rural Malawi . Int J Tuberc Lung Dis . 18 : 240 - 7 .
56. World Bank. Household cookstoves, environment, health, and climate change: a new look at an old problem . 2011 . http://documents.worldbank. org/curated/en/732691468177236006/pdf/ 632170WP0House00Box0361508B0PUBLIC0.pdf.
57. Balmes J , Pope D , Dherani M , Zhang J , Duan X , Bates M , et al. WHO guidelines for indoor air quality: household fuel combustion. Review 4: Health effects of household air pollution (HAP) exposure . 2014 . http://www. who.int/indoorair/guidelines/hhfc/evidence/en/. Accessed 20 Jul 2016 .
58. Mortimer K , Ndamala CB , Naunje AW , Malava J , Katundu C , Weston W , et al. A cleaner burning biomass-fuelled cookstove intervention to prevent pneumonia in children under 5 years old in rural Malawi (the Cooking and Pneumonia Study): a cluster randomised controlled trial . Lancet . 2016 . doi: 10 .1016/S0140- 6736 ( 16 ) 32507 - 7 .
59. Nie P , Sousa-Poza A , Xue J . Fuel for life: domestic cooking fuels and women's health in rural China . Int J Environ Res Public Health . 2016 ; 13 : 810 .
60. Anderman TL , DeFries RS , Wood SA , Remans R , Ahuja R , Ulla SE . Biogas cook stoves for healthy and sustainable diets? A case study in Southern India . Nutr Environ Sustain . 2015 ; 28 .
61. Begum BA , Paul SK , Dildar Hossain M , Biswas SK , Hopke PK . Indoor air pollution from particulate matter emissions in different households in rural areas of Bangladesh . Build Environ . 2009 ; 44 : 898 - 903 .
62. Sukhsohale ND , Narlawar UW , Phatak MS , Agrawal SB , Ughade SN . Effect of indoor air pollution during cooking on peak expiratory flow rate and its association with exposure index in rural women . Indian J Physiol Pharmacol . 2013 ; 57 : 184 - 8 .
63. Pokhrel AK , Smith KR , Khalakdina A , Deuja A , Bates MN . Case-control study of indoor cooking smoke exposure and cataract in Nepal and India . Int J Epidemiol . 2005 ; 34 : 702 - 8 .
64. Shen G , Chen Y , Du W , Lin N , Wang X , Cheng H , et al. Exposure and size distribution of nitrated and oxygenated polycyclic aromatic hydrocarbons among the population using different household fuels . Environ Pollut. doi: 10 .1016/j.envpol. 2016 . 07 .002.
65. Adetona O , Li Z , Sjödin A , Romanoff LC , Aguilar-Villalobos M , Needham LL , et al. Biomonitoring of polycyclic aromatic hydrocarbon exposure in pregnant women in Trujillo, Peru-Comparison of different fuel types used for cooking . Environ Int . 2013 ; 53 : 1 - 8 .
66. Titcombe ME , Simcik M. Personal and indoor exposure to PM2.5 and polycyclic aromatic hydrocarbons in the southern highlands of Tanzania: a pilot-scale study . Environ Monit Assess . 2011 ; 180 : 461 - 76 .
67. Regalado J , Perez-Padilla R , Sansores R , Páramo Ramirez J , Brauer M , Paré P , et al. The effect of biomass burning on respiratory symptoms and lung function in rural Mexican women . Am J Respir Crit Care Med . 2006 ; 174 : 901 - 5 .
68. Johnson MA , Chiang RA . Quantitative guidance for stove usage and performance to achieve health and environmental targets . Environ Health Perspect . 2015 . doi: 10 .1289/ehp.1408681.
69. Puzzolo E , Pope D , Stanistreet D , Rehfuess EA , Bruce NG . Clean fuels for resource-poor settings: a systematic review of barriers and enablers to adoption and sustained use . Environ Res . 2016 ; 146 : 218 - 34 .
70. Alexander D , Linnes JC , Bolton S , Larson T. Ventilated cookstoves associated with improvements in respiratory health-related quality of life in rural Bolivia . J Public Health . 2014 ; 36 : 460 - 6 .
71. Instituto Nacional de Estadistica e Informatica de Peru. Censos Nacionales 2007 : XI de Población y VI de Vivienda-Cuadros Estadísticos . Instituto Nacional de Estadistica e Informatica de Peru. 2007 . http://censos.inei.gob. pe/cpv2007/tabulados/. Accessed 23 Aug 2016 .
72. Pollard SL , Williams DL , Breysse PN , Baron PA , Grajeda LM , Gilman RH , et al. A cross-sectional study of determinants of indoor environmental exposures in households with and without chronic exposure to biomass fuel smoke . Environ Health . 2014 ; 13 : 21 .
73. Meinert , CL , Tonascia, S. Clinical trials: design, conduct, and analysis , 1st ed. New York: Oxford University Press; 1986 .
74. R Core Team. R: a language and environment for statistical computing . Vienna: R Foundation for Statistical Computing; 2016 . https://www.r-project. org/. Accessed 18 Nov 2016 .
75. Burroughs Peña M , Romero KM , Velazquez EJ , Davila-Roman VG , Gilman RH , Wise RA , et al. Relationship between daily exposure to biomass fuel smoke and blood pressure in high-altitude Peru . Hypertension . 2015 ; 65 : 1134 - 40 .
76. Camp PG , Ramirez-Venegas A , Sansores RH , Alva LF , McDougall JE , Sin DD , et al. COPD phenotypes in biomass smoke- versus tobacco smoke-exposed Mexican women . Eur Respir J . 2014 ; 43 : 725 - 34 .
77. McCracken JP , Wellenius GA , Bloomfield GS , Brook RD , Tolunay HE , Dockery DW , et al. Household air pollution from solid fuel use: evidence for links to CVD . Glob Heart . 2012 ; 7 : 223 - 34 .
78. Jones PW , Quirk FH , Baveystock CM , Littlejohns P. A self-complete measure of health status for chronic airflow limitation: The St . George's Respiratory Questionnaire . Am Rev Respir Dis . 1992 ; 145 : 1321 - 7 .
79. Jones PW . St. George's Respiratory Questionnaire: MCID . COPD: J Chron Obstruct Pulmon Dis . 2005 ; 2 : 75 - 9 .
80. Miller MR , Hankinson J , Brusasco V , Burgos F , Casaburi R , Coates A , et al. Standardisation of spirometry . Eur Respir J . 2005 ; 26 : 319 - 38 .
81. Ferguson GT , Enright PL , Buist AS , Higgins MW . Office spirometry for lung health assessment in adults: a consensus statement from the National Lung Health Education Program . Chest . 2000 ; 117 : 1146 - 61 .
82. Cobble M , Bale B. Carotid intima -media thickness: knowledge and application to everyday practice . Postgrad Med . 2010 ; 122 : 10 - 8 .
83. Lorenz MW , Von Kegler S , Steinmetz H , Markus HS , Sitzer M. Carotid intimamedia thickening indicates a higher vascular risk across a wide age range: prospective data from the Carotid Atherosclerosis Progression Study (CAPS) . Stroke . 2005 ; 37 : 87 - 92 .
84. Bots ML , Hoes AW , Koudstaal PJ , Hofman A , Grobbee DE . Common carotid intima-media thickness and risk of stroke and myocardial infarction: the Rotterdam Study . Circulation . 1997 ; 96 : 1432 - 7 .
85. Stein JH , Korcarz CE , Hurst RT , Lonn E , Kendall CB , Mohler ER , et al. Use of carotid ultrasound to identify subclinical vascular disease and evaluate cardiovascular disease risk: a consensus statement from the American Society of Echocardiography Carotid Intima-Media Thickness Task Force Endorsed by the Society for Vascular Medicine . J Am Soc Echocardiogr . 21 ; 93 - 111 .
86. Sharma K , Blaha MJ , Blumenthal RS , Musunuru K. Clinical and research applications of carotid intima-media thickness . Am J Cardiol . 2009 ; 103 : 1316 - 20 .
87. The National Heart, Lung, and Blood Institute. Atherosclerosis Risk in Communities Study Protocol Manual 6A: Ultrasound Assessment: Scanning Procedures . 1997 . http://www.cscc.unc.edu/aric/visit/Ultrasound_ Assessment-_ Scanning_Procedures.4_6a.pdf.
88. Schäberle W. Ultrasonography in vascular diagnosis: a therapy-oriented textbook and atlas . Springer Berlin Heidelberg; 2005 .
89. Mattace-Raso FUS , van der Cammen TJM , Hofman A , van Popele NM , Bos ML , Schalekamp MADH , et al. Arterial stiffness and risk of coronary heart disease and stroke . Circulation . 2006 ; 113 : 657 - 63 .
90. Hamburg NM , Palmisano J , Larson MG , Sullivan LM , Lehman BT , Vasan RS , et al. Relation of brachial and digital measures of vascular function in the community: The Framingham Heart Study . Hypertension . 2011 ; 57 : 390 - 6 .
91. Bruno RM , Gori T , Ghiadoni L . Endothelial function testing and cardiovascular disease: focus on peripheral arterial tonometry . Vasc Health Risk Manag . 2014 ; 10 : 577 - 84 .
92. Gokce N , Keaney JF , Hunter LM , Watkins MT , Nedeljkovic ZS , Menzoian JO , et al. Predictive value of noninvasively determined endothelial dysfunction for long-term cardiovascular events in patients with peripheral vascular disease . J Am Coll Cardiol . 2003 ; 41 : 1769 - 75 .
93. Muiesan ML , Salvetti M , Paini A , Monteduro C , Galbassini G , Poisa P , et al. Prognostic role of flow-mediated dilatation of the brachial artery in hypertensive patients . J Hypertens . 2008 ; 26 : 1612 - 8 .
94. Ras RT , Streppel MT , Draijer R , Zock PL . Flow-mediated dilation and cardiovascular risk prediction: a systematic review with meta-analysis . Int J Cardiol . 2013 ; 168 : 344 - 51 .
95. Corretti MC , Anderson TJ , Benjamin EJ , Celermajer D , Charbonneau F , Creager MA , et al. Guidelines for the ultrasound assessment of endothelialdependent flow-mediated vasodilation of the brachial artery: a report of the International Brachial Artery Reactivity Task Force . J Am Coll Cardiol . 2002 ; 39 : 257 - 65 .
96. Prisma ONG . Salud y Saneamiento. Prisma. http://www.prisma.org.pe/ proyectos/salud-y-saneamiento/. Accessed 17 Aug 2016 .
97. 36 - Item Short Form Survey from the RAND Medical Outcomes Study . https://www.rand.org/health/surveys_tools/mos/36-item-short-form. html. Accessed 26 Sep 2017 .
98. EQ-5D instruments - EQ-5D. https://euroqol.org/eq-5d-instruments/. Accessed 26 Sep 2017 .
99. Little RJ , D'Agostino R , Cohen ML , Dickersin K , Emerson SS , Farrar JT , et al. The prevention and treatment of missing data in clinical trials . N Engl J Med . 2012 ; 367 : 1355 - 60 .
100. Harris PA , Taylor R , Thielke R , Payne J , Gonzalez N , Conde JG . Research electronic data capture (REDCap)-A metadata-driven methodology and workflow process for providing translational research informatics support . J Biomed Inform . 2009 ; 42 : 377 - 81 .
101. Dominguez Curi CH , Aviles Arias DA , Satalaya Perez A. Tablas auxialiares para la formulacion y evaluacion de regimenes alimentarios . 2014 . http:// www.ins.gob.pe/repositorioaps/0/5/jer/doc_tec_norm/TAFERA_1_ compressed. pdf. Accessed 17 Aug 2016 .
102. Dominguez Curi CH , Aviles Arias DA , Satalaya Perez A . Tabla de factores de conversion de peso de alimentos cocidos a crudos. 2014 . http://www.ins. gob.pe/repositorioaps/0/5/jer/doc_tec_norm/TAFERA_2_compressed. pdf. Accessed 17 Aug 2016 .
103. Consort-Welcome to the CONSORT website . http://www.consort-statement. org/. Accessed 2 Aug 2017 .
104. About BioLINCC. https://biolincc.nhlbi.nih.gov/about/. Accessed 3 Aug 2017 .
105. Chan A-W , Tetzlaff JM , Altman DG , Laupacis A , Gøtzsche PC , Krleža-Jerić K , et al. SPIRIT 2013 Statement: Defining standard protocol items for clinical trials . Ann Intern Med . 2013 ; 158 : 200 .
106. WHO | WHO data set . http://www.who.int/ictrp/network/trds/en/. Accessed 10 Aug 2017 .