Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 May 1.
Published in final edited form as: Air Qual Atmos Health. 2024 Jan 9;17(5):967–978. doi: 10.1007/s11869-023-01492-0

Randomized trial in rural Native American homes heated with wood stoves: results from the EldersAIR study

Curtis W Noonan a,b, Ethan S Walker a,b, Erin O Semmens a,b, Annie Belcourt b, Johna Boulafentis c, Crissy Garcia d, Jon Graham a, Nolan Hoskie f, Eugenia Quintana f, Julie Simpson c, Paul Smith a,b, Howard L Teasley Jr e, Desirae Ware a,b, Emily Weiler a,b, Tony J Ward a,b
PMCID: PMC11446504  NIHMSID: NIHMS1995441  PMID: 39363883

Abstract

Residential wood burning has both practical and traditional value among many indigenous communities of the United States Mountain West, although household biomass burning also results in emissions that are harmful to health. In a household-level three-arm placebo-controlled randomized trial we tested the efficacy of portable filtration units and education interventions on improving pulmonary function and blood pressure measures among elder participants that use wood stoves for residential heating. A total of 143 participants were assigned to the Education (n=49), Filter (n=47), and Control (n=47) arms. Blood pressure and spirometry measures were collected multiple times during a per-intervention winter period and during a follow-up post-intervention winter period. Despite strong PM2.5 exposure reduction results with the Filter arm (50% lower compared to Control arm), neither this intervention nor the Education intervention translated to improvements in the selected health measures among this population with a mixture of chronic conditions. Intention to treat analysis failed to demonstrate evidence that either of the intervention arms had beneficial effects on the blood pressure or the spirometry measures. Post-hoc evaluation of effect modification for blood pressure and spirometry outcomes did not reveal any interaction influence on the outcomes according to sex, residential smoking, chronic disease history and study area.

Keywords: Biomass, rural, indigenous, blood pressure, spirometry

Introduction

Approximately 13 million US households burn wood fuel for heating (U.S. Energy Information Agency 2015). Residential wood burning is often the most economically and culturally preferred form of home heating in rural Native American communities. Fire is also included among the oral traditions of many indigenous peoples (Walters et al. 2020). For example, conversations with one tribal community revealed that tribal members historically served as stewards of fire and that fire was intricately connected to their oral histories of creation. Thus, residential wood burning has both practical and traditional value among many indigenous communities of the rural Mountain West.

Wood burning also generates biomass smoke, known to be associated with poor indoor air quality and corresponding health risks. The most well studied indoor smoke exposure settings have been in low income countries where biomass burning is commonly used for cookstoves. In addition to observations of adverse childhood respiratory health in these settings, respiratory and cardiovascular health impacts to adults are of concern. Several studies have shown blood pressure elevations among adults exposed to high concentrations of biomass smoke from cookstoves (Li et al. 2020). Women exposed to biomass smoke from cookstoves were also shown to be at higher risk of chronic bronchitis and COPD (Po, FitzGerald and Carlsten 2011). Household air pollution exposures in lower income countries, typically via biomass-fueled cookstoves, are greater than those observed in higher income country settings where woodstoves are used for home heating. Nevertheless, homes that use woodstoves for heat often have indoor fine particulate matter concentrations (PM2.5; airborne particles ≤ 2.5 μm in aerodynamic diameter) that approach or exceed existing health-based standards. Although the US does not have an indoor air quality standard for PM2.5, the U.S. Environmental Protection Agency (EPA) health-based 24-hour standard for ambient PM2.5 is 35 μg/m3. Although the EPA does not have corresponding indoor air quality guidelines, the World Health Organization 24-hour standard of 15 μg/m3 is applicable to both ambient and indoor environments. Daily average indoor PM2.5 concentrations of 20 to 50 μg/m3 have previously been measured in wood stove households across rural areas of the US, including in Native American communities (Noonan et al. 2012, Semmens et al. 2015, Ward and Noonan 2008, Singleton et al. 2017, Walker et al. 2021, Walker et al. 2022b, Walker et al. 2022a).

As with the effort to address household air pollution from cooking, active efforts are underway to identify evidenced-based and implementable strategies for reducing indoor PM exposures in settings where woodstoves are used for heating. Several (Wheeler et al. 2014, Allen et al. 2011, Hart et al. 2011, Cheek et al. 2020, Ward et al. 2017, Walker et al. 2022b) but not all (Walker et al. 2022a) studies have found that portable air filters reduce indoor PM2.5 in homes heated with wood stoves. Limited evidence also points to the importance of training to achieve maximum combustion efficiency and lower emissions when using high quality woodstoves (Ward et al. 2011). Educational messaging focused on best-burn practices is promoted by the EPA as a strategy to mitigate emissions from residential wood stoves (U.S. Environmental Protection Agency 2021). However, development and testing of educational strategies is limited; no evidence from randomized trials indicates that such practices effectively reduce indoor PM2.5 in wood stove households (Walker et al. 2022b). Further assessment of both air filtration and educational interventions in wood stove households is needed to inform low-cost strategies for reducing indoor exposures to wood smoke. Moreover, layering community-level approaches over household-level interventions to improve indoor air quality may improve overall efficacy, but as of yet this is a completely untested strategy.

In the Residential Wood Smoke Interventions for Improving Health in Native American Populations (EldersAIR) study, our goal was to evaluate strategies to improve indoor air quality and health outcomes among rural, Native American senior and elder households that used woodstoves as a primary source of residential heating. The EldersAIR project was part of a consortium of studies under the Intervention Research to Improve Native American Health (IRINAH) umbrella of NIH-funded studies. The guiding principles of these studies were to engage community partners in developing and testing multi-level interventions to improve health outcomes in Native American communities. The community-level intervention was a wood distribution program led by tribal agencies with a goal of procuring, processing, storing and delivering dry, high-quality wood fuel to study participants and other community elders (Walker et al. 2022a). The household-level intervention reported here was a randomized three-arm placebo-controlled trial focused on improving pulmonary function and blood pressure measures among participants. We have previously reported on efficacy of the air filtration arm in reducing indoor PM2.5 concentrations by more than 50% (Walker et al. 2022a). Here we report the a priori intention to treat analysis and the post-hoc exposure-response analysis for the pulmonary function and blood pressure measures.

Methods

Study design

EldersAIR took place in communities located within the Navajo Nation (NN) and Nez Perce reservations in the Western US. The EldersAIR household intervention study was a three-arm randomized, pre-post, placebo-controlled trial among rural Native American adults aged 50 years and older living in households that used wood stoves as a primary heating source. Consistent with input from our community partners, participants were not excluded according to smoking history or history of respiratory or cardiovascular disease. Participants were enrolled across multiple cohorts (enrollment years 2014–2018). Each participant was followed during two consecutive winter periods: a pre-intervention baseline (Winter 1) and a post-intervention follow-up period (Winter 2). Each winter sampling period corresponded to the home heating season, November through March. Health and exposure measures were collected at two 48-hour visits during both winter periods, with each visit separated by at least three weeks.

Recruitment, eligibility criteria, and informed consent

Specific chapters or locales within the NN and Nez Perce Tribe (NPT) territories were selected by community partners. Households with a resident 50 years of age or older that used a wood stove as a primary heating source were considered for inclusion in the study. Our recruitment goal was 63 households per study area, or 21 households per treatment arm within each study area, for a total of 126 households, or 42 households per treatment arm. The target sample size was based on a combination of previous literature, power calculations, and feasibility following recommendations from community partners. Power calculations were based on pulmonary function measures with two group comparisons at 42 participants per treatment yielding 80% power to detect mean differences of 18.4 ml for FEV1, 24.9 ml for FVC and 0.31% for FEV1/FVC%. Recruitment strategies varied by study area. For the NN community, recruitment occurred through chapter houses, word of mouth, and through other community event forums. NPT households were recruited from an existing NPT Senior Wood Delivery Program that was administered by the NPT Forestry and Fire Management Division with household eligibility determined by the Social Services Division. Participants provided written informed consent prior to participation. Participants were also compensated for the time they spent performing study tasks and reimbursed for the cost of electricity to use the air filtration units. Following the study, all households received the educational tools and the air filtration units used in the Education and Filtration study arms. The EldersAIR study was approved by the University of Montana IRB, the Navajo Nation Human Research Review Board, and the Nez Perce Tribal Executive Committee. The EldersAIR trial is registered at ClinicalTrials.gov under trial number NCT02240069.

Randomization and treatment arms

Treatment assignment occurred following the Winter 1 baseline measures and was stratified on cohort (year of enrollment) nested within study area (NN and NPT). Randomization (EOS) took place within blocks of three households in each cohort/study area stratum. Intervention arms were similar to those used in previous randomized trials of wood stove households (Noonan et al. 2019, Walker et al. 2022b). Briefly, the Education arm participants received a community-adapted educational intervention that included methods for identifying low moisture content fuel and achieving optimal combustion efficiency, both considered to be desirable strategies for minimizing smoke emissions. Tools provided to Education arm participants included a moisture meter to measure fuel moisture, fire starters to facilitate the rapid lighting of the fire, and a wood stove thermometer to ensure that the wood stove was burning within the optimal temperature range. Additionally, participants watched short, culturally-relevant, digital stories that were developed with input from community stakeholders (Walters et al. 2020). These short videos stressed the most important factors related to best-burn practices, including using low-moisture fuels and burning their wood stove at the optimal temperatures.

The Filtration and Placebo treatment arms used similar approaches. Filtration arm households received one of several portable air filtration models (Filtrete FAP03 and FAP02, 3M Company, USA; Winix 5500 and 5300, Winix America Inc, USA). The Filtrete units were chosen based on its ideal combination of price, availability, clean air delivery rate (CADR), and ratings for larger room sizes. However, production of the Filtrete unit was discontinued prior to completion of the EldersAIR study, so the Winix models were selected based on comparable price and CADR relative to the Filtrete unit. Placebo households were provided with one of the above portable air filtration models but with no filter inside the unit. In both Filtration and Placebo households, the unit was placed in the same room as the wood stove and participants were instructed to leave the unit running continuously on the high setting. Study field personnel assessed the air filtration units during household visits to ensure they were turned on, running at the desired setting, and filters were replaced as needed (i.e., when the “change filter” light came on).

Health measures

Field personnel collected participant measures of spirometry, blood pressure, heart rate and oxygen saturation at each household visit. Prior to each sampling season, field personnel were trained, or re-trained, on the collection of health measures. For breathing measures, we used an EasyOne Diagnostic portable spirometer (Ndd Medical, Andover, MA) with disposable spirette breathing tubes. Field personnel were trained by a pulmonologist (Yawn et al.) or a respiratory therapist on the use of the spirometer as well as best practices for coaching participants to assist in capturing accurate readings. Participants completed a lung function screening questionnaire, and the health measure visit was rescheduled if the participant had any contraindications for completion of the spirometry measures, including unstable cardiac status, recent surgeries, or recent respiratory infections. Participants were asked to avoid alcohol, tobacco and inhaled medications for a minimum of four hours prior to the scheduled spirometry. Field personnel demonstrated the spirometry maneuver for participants prior to coaching them through correct measures. Participants had to complete three acceptable maneuvers per the instrument readout with a limit of nine attempts, or participant fatigue, to achieve the three acceptable maneuvers. Participants rested a minimum of 30 seconds between attempts. A pulmonologist (Yawn et al.) confirmed acceptable measures based on the instrument’s overall session quality grade of A, B or C and visual inspection of the flow-volume loop. Reproducibility criteria were based on recommendations of less than 100 ml variance. Percent predicted values for FVC and FEV1 were generated by the EasyOne instrument based on age, sex and height inputs (NHANES III). For blood pressure measures, we used the automated e-sphyg 2 (American Diagnostic Corporation, Hauppauge, NY). Following a 5-minute rest period with the participant in a seated position, an inflatable cuff was placed on the participant’s upper arm over the brachial artery. Field staff recorded systolic blood pressure, diastolic blood pressure, and heart rate following the blood pressure measurement. For blood oxygen saturation and pulse rate, we used the CMS50DL (Contec, Qinhuangdao, China).

Exposure assessment

Measures of indoor PM2.5 concentrations for the EldersAIR study were described previously (Walker 2022). Briefly, at each visit we measured PM2.5 mass concentrations over 48-hour sampling periods at 60-second time intervals using light-scattering aerosol monitors (DustTrak 8530, TSI, USA). DustTrak instruments were placed 1 to 1.5 meters above ground level in the same room as the wood stove. A rigorous Quality Assurance / Quality Control program was followed during the air sampling measurements. Prior to field deployment, DustTraks were collocated with a certified BAM 1020 Continuous Particulate Monitor (Met One Instruments, Inc, USA) to assess performance and were factory calibrated as necessary. Instruments were cleaned and zero calibrated according to manufacturer standards prior to each sampling event. We previously developed a DustTrak wood smoke correction factor by collocating these instruments with a certified BAM 1020 Continuous Particulate Monitor (Met One Instruments, Inc, USA) reference monitor (McNamara, Noonan and Ward 2011). Before data analysis began, the correction factor of 1.65 was applied to all DustTrak PM2.5 measurements.

DustTrak data were assessed for quality by checking descriptive statistics (n, minimum [min], mean, standard deviation [sd], median, maximum [max]), gaps in sampling or missing observations, and by visually inspecting a time-series plot of the PM2.5 concentrations following each sampling session. Data from 45 DustTrak sampling sessions (9.6%) were excluded from analysis due to apparent instrument malfunction that resulted in extremely high minimum values (n=4), a large proportion of negative values (n=4), or sampling sessions less than 80% of the expected duration (n=37).

Covariates

Field personnel administered questionnaires to participants at the beginning of each winter season to collect information on demographic and household characteristics. Demographic characteristics included participant age, sex, race, ethnicity, education level, and household income. Household characteristics included number of levels in the home, size of the home (square meters), number of pets in the home, age of the home, and whether or not any residents smoked inside or outside the home. Each participant also completed a health history questionnaire to report on current or past chronic disease history. We used a Kill A Watt device (P3 International Corporation, USA) in each household to measure power usage by the air filtration units; measures of compliance were based on percent expected kilowatt usage compared to laboratory tests for each filter unit/setting.

The overall quality of the wood stoves was assessed by participant-reported age of the stoves and whether or not the stove was an EPA certified model. We also implemented a wood stove quality grading method that has previously been utilized during wood stove studies in our group (Walker et al. 2021). An expert, independent wood stove consultant who was blinded to study site and intervention assignment assessed photos of a given household’s wood stove, stovepipe, chimney, and wood storage. Each stove was assigned a grade of high-, medium-, or low-quality based on the wood supply, the stove and chimney system, and the maintenance of the stove. In addition, participants self-reported wood stove practices during a typical winter season, including primary wood collection method (i.e., purchase, harvest, or a local wood delivery program), the length of time they allowed the wood to dry prior to burning, and time since the chimney was last cleaned. Participants self-reported stove use during the PM2.5 sampling periods compared to a typical winter season (i.e., no burning, light burning, average burning, heavy burning). We also assessed stove use by placing temperature logging devices (iButton DS1921G, Thermochron, Australia or LogTag UTRIX-16, OnSolution, Australia) near the wood stoves to record temperature at 20-minute intervals during each winter of observation. Wood moisture content was measured during household visits using a pin-type moisture meter (MMD4E, General Tools & Instruments LLC, USA).

Statistical analysis

Analysis was conducted using R version 4.0.4 (The R Foundation for Statistical Computing, Austria). We calculated descriptive statistics for continuous variables (n, mean, sd, min, median, max) and categorical variables (n, percentage of total) across all participants and separately for each treatment arm and study area.

We conducted the primary analysis with linear mixed models using the lme4 package (Bates et al. 2015). In the intent-to-treat (ITT) framework, with a given spirometry or blood pressure variable as the outcome of interest, we included a 3-level fixed effect for treatment as the primary explanatory variable (levels = Placebo arm, Education arm, Filtration arm), a continuous fixed term for mean baseline (Winter 1) response measure, and a nested random effect to account for repeated measures (i.e., participant:home:cohort:area). The ITT models utilized the study’s randomization and were not adjusted for potential confounding variables in the primary analyses. Sensitivity analyses were conducted that added potential confounding variables into the ITT models to assess the effectiveness of the randomization in controlling for confounding. Model assumptions were evaluated in all analysis frameworks.

We conducted exposure-response (ER) analyses to evaluate associations between indoor PM2.5 measures and outcome measures outside of the RCT framework. The ER analyses utilized a linear mixed model fit by restricted maximum likelihood (REML) with blood pressure or spirometry measures as the outcome and mean 48-hour indoor PM2.5 concentration as the exposure of interest. As with the ITT models, ER models included a nested random term (i.e., participant:home:cohort:area) to account for repeated measures in the analysis as well as area-level factors such as the wood yards. Because the ER models did not utilize the study’s randomization to help control for confounding, covariates were included in the model to adjust for potential confounders. Confounders were identified a priori through previous literature and by assessing independent associations with the exposure and outcomes of interest. Final ER models were adjusted for age, gender, income, education, household resident smoking status, history of cardiovascular disease, and history of respiratory disease. We included a natural cubic spline term for date of household visit with four degrees of freedom to account for potential confounding by time over the course of the trial. Model estimates are expressed as change in health outcome per interquartile range (IQR; 32 μg/m3) increase in mean 48-hour indoor PM2.5 concentration.

We also evaluated potential modification of the effect of the treatments within each of the model frameworks by including interaction terms in the primary models by a priori selected factors of sex, presence of a smoking in the household, history of chronic disease, and study site. We assessed significance of the interaction terms using Type II Wald Chi-square tests. Model assumptions were evaluated in all analysis frameworks.

Results

A total of 149 participants were initially enrolled in the EldersAIR study with six participants dropping out prior to randomization, leaving 143 participants assigned to the Education (n=49), Filter (n=47), and Control (n=47) arms (Fig. 1). Of the 143 randomized participants, post-intervention (Winter 2) measures were collected for 131 participants. Table 1 shows participant demographics and cohort entry year by treatment arm. A majority of the participants identified as female (70%), American Indian or Alaska Native (AI/AN) (97%), and not Hispanic (97%). Participants were 70 years of age on average and the majority reported having at least a high school education. Study homes were generally single-level (89%), and 61% of participants reported a household income of less than $20,000 per year. One third of the households had a resident who smoked tobacco products, and 8% said the resident(s) smoked while indoors. In general, participant and household characteristics were balanced across treatment arms.

Figure 1:

Figure 1:

Open in a new tab

EldersAIR Participant Recruitment, Enrollment, and Retention

Table 1:

Participant and household characteristics at baseline (Winter 1)

All Participants (total n=149) Placebo (total n=47) Filtration (total n=47) Education (total n=49)
Participant sex
 Female, n (%) 105 (70) 31 (66) 33 (70) 36 (73)
 Male, n (%) 44 (30) 16 (34) 14 (30) 13 (27)
Participant race
 AI/AN, n (%) 144 (97) 47 (100) 43 (91) 48 (98)
 Asian, n (%) 1 (1) 0 (0) 1 (2) 0 (0)
 More than 1 race, n (%) 4 (3) 0 (0) 3 (6) 1 (2)
Participant ethnicity
 Hispanic, n (%) 5 (3) 1 (2) 1 (2) 3 (6)
 Not Hispanic, n (%) 144 (97) 46 (98) 46 (98) 46 (94)
Participant education
 < High school, n (%) 27 (18) 11 (23) 5 (11) 11 (22)
 High school, n (%) 46 (31) 13 (28) 16 (34) 13 (27)
 Some college, n (%) 52 (35) 17 (36) 18 (38) 15 (31)
 College degree, n (%) 24 (16) 6 (13) 8 (17) 10 (20)
Participant age (years)
 n 148 46 47 49
 mean (sd) 69.6 (9.2) 68.9 (8.7) 69.8 (9.5) 69.8 (9.3)
 min, median, max 51, 69, 98 52, 66, 93 55, 70, 98 51, 69, 89
Participant Cohort (1st Winter of Participation)
 2014–2015, n (%) 12 (8) 4 (9) 4 (9) 4 (8)
 2015–2016, n (%) 25 (17) 8 (17) 8 (17) 9 (18)
 2016–2017, n (%) 38 (26) 11 (23) 13 (28) 13 (27)
 2017–2018, n (%) 43 (29) 14 (30) 13 (28) 13 (27)
 2018–2019, n (%) 31 (21) 10 (21) 9 (19) 10 (20)
All Households (total n=144) Placebo (total n=45) Filtration (total n=46) Education (total n=47)
Household income
 < $20,000 88 (61) 27 (60) 26 (57) 30 (64)
 $20,000 to $39,000 37 (26) 10 (22) 13 (28) 13 (28)
 $40,000 + 19 (13) 8 (18) 7 (15) 4 (9)
Resident smokes
 Yes, n (%) 48 (33) 17 (38) 14 (30) 16 (34)
 No, n (%) 95 (66) 28 (62) 32 (70) 30 (64)
Resident smokes inside
 Yes, n (%) 12 (8) 2 (4) 4 (9) 6 (13)
 No, n (%) 122 (85) 41 (91) 36 (78) 40 (85)
Age of stove
 < 6 years, n (%) 26 (18) 10 (22) 7 (15) 7 (15)
 6 to 10 years, n (%) 27 (19) 9 (20) 11 (24) 7 (15)
 11 to 15 years, n (%) 11 (8) 3 (7) 4 (9) 4 (9)
 16 + years, n (%) 80 (56) 23 (51) 24 (52) 29 (62)
EPA certified stove
 Yes, n (%) 18 (13) 7 (16) 5 (11) 6 (13)
 No, n (%) 47 (33) 14 (31) 18 (39) 15 (32)
 Do not know, n (%) 79 (55) 24 (53) 23 (50) 26 (55)
Open in a new tab

n = number of homes or observations; sd = standard deviation; min = minimum; max = maximum; AI/AN = American Indian/Alaska Native; EPA = United States Environmental Protection Agency

Baseline health measures, averaged within Winter 1, are presented in Table 2. Based on field blood pressure reading, 39% (n=58) of participants had hypertension defined as systolic blood pressure above 140 mg/dL and diastolic blood pressure above 90 mg/dL (Qaseem et al. 2017). Baseline spirometry measures indicated 47% (n=70) with FVC < 80 percent predicted, 34% (n=51) with FEV1 < 80 percent predicted, and 10% (n=15) with FEV1/FVC < 0.7. Considering the baseline spirometry measures together, 27% (n=40) of participants had indication of restrictive impairment and 7% (n=11) of participants had indication of obstructive impairment (Fabbri et al. 2004). Results from the health history questionnaire show self-reporting of chronic health conditions among participants. The most commonly reported conditions were diabetes (46%), hypertension (42%), and cardiovascular disease (56%). Among those reporting asthma, COPD, chronic bronchitis or emphysema (n=29), 14% (n=4) had corresponding baseline spirometry measures indicative of obstructive impairment. Among those reporting hypertension (n=62), 87% (n=54) had a corresponding baseline blood pressure measurement consistent with hypertension.

Table 2:

Baseline health measures and reported disease history (Winter 1)

All Participants (total n=149) Placebo (total n=47) Filtration (total n=47) Education (total n=49)
Body mass index (kg/m 2 )
 n 140 44 45 46
 mean (sd) 29 (6) 30 (6) 30 (5) 29 (5)
 min, median, max 19, 28, 48 21, 28, 48 21, 30, 45 19, 28, 40
Systolic blood pressure (mmHg)
 n 135 43 45 42
 mean (sd) 137 (19) 137 (20) 143 (19) 132 (18)
 min, median, max 86, 135, 200 96, 134, 200 99, 142, 186 86, 129, 168
Diastolic blood pressure (mmHg)
 n 135 43 45 42
 mean (sd) 77 (12) 77 (14) 79 (10) 74 (11)
 min, median, max 51, 77, 143 58, 75, 143 51, 80, 101 51, 73, 101
Heart rate (beats per minute)
 n 138 43 46 44
 mean (sd) 74 (10) 74 (11) 73 (9) 73 (11)
 min, median, max 48, 73, 101 50, 76, 96 58, 72, 95 48, 72, 101
Oxygen saturation (%)
 n 138 43 46 44
 mean (sd) 95 (3) 95 (3) 95 (3) 95 (3)
 min, median, max 87, 95, 99 88, 95, 99 90, 95, 99 87, 95, 98
Forced vital capacity (FVC; liters)
 n 138 44 44 45
 mean (sd) 2.6 (0.8) 2.5 (0.9) 2.6 (0.7) 2.6 (0.8)
 min, median, max 0.4, 2.5, 5.0 0.4, 2.4, 5.0 0.6, 2.4, 4.4 1.3, 2.6, 4.3
Forced expiratory volume in 1 second (FEV1; liters)
 n 138 44 44 45
 mean (sd) 2.0 (0.7) 2.0 (0.8) 2.1 (0.6) 2.1 (0.7)
 min, median, max 0.3, 2.0, 4.8 0.3, 2.0, 4.8 0.5, 2.0, 3.5 0.9, 2.2, 3.6
FEV1/FVC ratio
 n 138 44 44 45
 mean (sd) 0.79 (0.08) 0.79 (0.08) 0.80 (0.07) 0.79 (0.08)
 min, median, max 0.52, 0.80, 1.00 0.52, 0.80, 0.95 0.63, 0.81, 0.96 0.56, 0.80, 1.00
Health History, n (%)
 Asthma 25 (17) 7 (15) 8 (17) 10 (20)
 Chronic Bronchitis 7 (5) 2 (4) 3 (6) 2 (4)
 Cancer 18 (12) 4 (9) 11 (23) 3 (6)
 CAD 6 (4) 5 (11) 0 (0) 1 (2)
 CHF 4 (3) 2 (4) 2 (0) 0 (0)
 COPD 7 (5) 1 (2) 4 (8) 2 (4)
 Diabetes 69 (46) 25 (53 20 (43) 21 (43)
 Emphysema 0 (0) 0 (0) 0 (0) 0 (0)
 Hay fever 18 (12) 8 (17) 4 (9) 6 (12)
 Heart attack 10 (7) 1 (2) 4 (9) 5 (12)
 Hypertension 62 (42) 17 (36) 28 (60) 15 (31)
 Stroke 9 (6) 1 (2) 5 (11) 3 (6)
 Any CVD 84 (56) 24 (51) 32 (68) 24 (49)
 Atherosclerotic CVD 21 (14) 5 (11) 9 (19) 7 (14)
 Any respiratory disease 62 (42) 16 (34) 22 (47) 22 (45)
 Any condition 124 (83) 37 (79) 43 (91) 39 (80)
Tobacco use, n (%)
 Cigarettes current 23 (15) 8 (17) 6 (13) 8 (16)
 Cigarettes ever 69 (46) 21 (45) 23 (49) 21 (43)
 Cigars/pipes ever 8 (5) 2 (4) 4 (9) 2 (4)
Tobacco (cigarette) pack years
 n 67 20 23 20
 mean (sd) 13 (16) 15 (16) 13 (16) 11 (18)
 min, median, max 0, 5, 72 0, 12, 55 0, 5, 54 0, 2, 72
Open in a new tab

n = number of participants or observations; sd = standard deviation; min = minimum; max = maximum

Changes in blood pressure and spirometry measures for Education and Filter arms relative to Placebo arm are shown in Table 3. The ITT model estimates and 95% confidence intervals (CIs) were imprecise for each measure, and there was no evidence that either of the intervention arms had beneficial effects on either the blood pressure or the spirometry measures. Consideration of different modeling assumptions in sensitivity analysis did not reveal any discrepancy with the ITT results (Fig. 2). Post-hoc evaluation of effect modification for blood pressure and spirometry outcomes did not reveal any interaction influence on the outcomes in the hypothesized direction according to sex, residential smoking, chronic disease history and study area (Table S1).

Table 3:

Primary results, intent-to-treat framework

Estimate (95% CI) P-value
Systolic Blood Pressure
 Placebo treatment (n = 75) Reference
 Education treatment (n = 74) −3.46 (−11.45, 4.54) 0.39
 Filtration treatment (n = 76) 0.25 (−7.73, 8.24) 0.95
Diastolic Blood Pressure
 Placebo treatment (n = 75) Reference
 Education treatment (n = 74) −1.00 (−5.42, 3.42) 0.65
 Filtration treatment (n = 76) −0.58 (−4.97, 3.81) 0.79
Forced Vital Capacity
 Placebo treatment (n = 31) Reference
 Education treatment (n = 30) −0.09 (−0.24, 0.07) 0.28
 Filtration treatment (n = 40) 0.01 (−0.14, 0.16) 0.89
Forced Expiratory Volume in 1 Second
 Placebo treatment (n = 31) Reference
 Education treatment (n = 30) −0.11 (−0.24, 0.02) 0.10
 Filtration treatment (n = 40) 0.01 (−0.11, 0.13) 0.90
Forced Expiratory Volume in 1 Second / Forced Vital Capacity Ratio
 Placebo treatment (n = 31) Reference
 Education treatment (n = 30) −0.02 (−0.05, 0.00) 0.06
 Filtration treatment (n = 40) 0.00 (−0.03, 0.02) 0.71
Open in a new tab

Models adjusted for baseline (Winter 1) outcome; model includes nested random term: participant:home:cohort:area. CI = confidence interval

Figure 2: Estimates and 95% confidence intervals for primary results with sensitivity analyses for Intent-to-treat framework.

Figure 2:

Open in a new tab

BP = blood pressure; FVC = forced vital capacity; FEV1 = forced expiratory volume in 1 second. Primary: Models adjusted for baseline (Winter 1) outcome; model includes nested random term: participant:home:cohort:area. Confounders: Primary model plus fixed terms for education, resident smoking status, gender, stove grade, age, body mass index. No Home Term: Primary model with no random term for household. All Sessions: for lung function outcomes, models include all sessions measured in the field (primary analysis included only “acceptable” sessions; a pulmonologist confirmed acceptable measures based on the instrument’s overall session quality grade of A, B or C and visual inspection of the flow-volume loop).

The ER analysis also failed to show any consistent association for systolic and diastolic blood pressure, FVC, FEV1, or FVC/FEV1 ratio (Table 4). Post-hoc evaluation of effect modification in the ER analysis did not reveal any interaction influence on the outcomes in the hypothesized direction according to sex, residential smoking, chronic disease history and study area (Table S1).

Table 4:

Primary results, exposure-response framework

Change in health outcome per IQR increase (32 μg/m3) in indoor PM2.5 Estimate (95% CI)
Systolic Blood Pressure (n = 404) −0.46 (−1.60, 0.69)
Diastolic Blood Pressure (n = 404) −0.28 (−0.99, 0.44)
Forced Vital Capacity (n = 281) −0.01 (−0.03, 0.02)
Forced Expiratory Volume in 1 Second (n = 281) 0.02 (0.00, 0.04)
Forced Expiratory Volume in 1 Second / Forced Vital Capacity Ratio (n = 281) 0.01 (0.00, 0.01)
Open in a new tab

Models adjusted for age, gender, income, education, household resident smoking status, history of cardiovascular disease, history of respiratory disease; model includes nested random term: participant:home:cohort:area and a spline term for date of household visit (4 degrees of freedom). PM2.5 = fine particulate matter; CI = confidence interval; IQR = interquartile range

Discussion

Interest in reducing smoke concentrations in woodstove households is motivated by the epidemiological evidence regarding ambient PM exposures and health effects, as well as the adverse respiratory and cardiovascular health impacts attributed to higher exposures to household air pollution in lower income countries (GBD 2019 Risk Factors Collaborators 2020). In both European and US settings, chronic respiratory conditions were more common among adults living in woodstove homes (Sood et al. 2010, Orozco-Levi et al. 2006), and reductions in pulmonary function measures were observed among people with chronic respiratory conditions living in woodstove homes (Sood et al. 2010, White et al. 2022). Small experimental biomass exposure studies have shown some evidence of elevated inflammatory markers, but evidence of associations with cardiovascular function is limited (Sigsgaard et al. 2015). One intervention study in healthy adults living in woodstove homes found that improved indoor air quality was associated with decreases in c-reactive protein (Allen et al. 2011), although associations with endothelial function were not consistently observed (Kajbafzadeh et al. 2015).

Reported here is one of the few household intervention randomized trials conducted among persons living in homes that use woodstove for domestic heating. Innovations in this study included the introduction of an education intervention arm to promote best burn practices and the first randomized trial to assess household air pollution and health measures among rural elder indigenous populations in the US. We did not observe any ITT impacts to blood pressure or pulmonary function measures in our primary analyses. These findings point to the difficulties and uncertainties of conducting such trials in field settings, including intervention fidelity, external impacts to exposures, and assessment of highly variable and behavior-dependent pre-clinical outcome measures.

First, both intervention arms require substantial compliance and learning behaviors, and deviations to intervention fidelity can impact the intended goal of improving indoor air quality. As with the few previous long-term studies of household air filter interventions (Noonan et al. 2017, Chuang et al. 2017, Hansel et al. 2022), the air filter intervention proved to be efficacious in reducing mean indoor PM2.5 in homes assigned to this arm with a pre-post mean percent PM2.5 change (Noonan et al.) of −50.5% (−27.8% to −66.1%) (Walker et al. 2022a). This finding is similar to another pre-post randomized trial in rural and indigenous populations (Ward et al. 2017), and a recent review found that portable air filtration units reduced indoor PM2.5 by 23% to 92% across a variety of indoor settings and study designs (Cheek et al. 2020). The effectiveness of portable air filters is subject to variation in residents’ compliance with filter usage and recommended settings, but we have shown that PM2.5 reductions are robust to moderate departures from full compliance (Ward et al. 2017).

The efficacy of the education arm is far more vulnerable to poor adoption of new behaviors, use of provided tools, or deviations from best burn practices. We previously reported that this treatment arm did not significantly reduce indoor PM2.5, with mean change (Noonan et al.) of −2.9% (+40.0% to −66.1%) (Walker et al. 2022a). We adapted the education delivery and messaging to each community based upon best practices promoted by the EPA and other agencies (US Environmental Protection Agency 2021). Also, it is likely that the effectiveness of such strategies will vary substantially according to different home characteristics and, most importantly, the quality of woodstove. We noted substantial variation in wood stove quality in this study with over half of woodstoves greater than 15 years old and of poor to medium grade quality.

A second factor that could have impacted the ITT findings are other sources of PM exposure that were not targeted by the interventions. Potential exposure to tobacco smoke could have been the most important non-woodstove exposure source. Based on recommendations from our community partners we did not exclude study participation where a household resident smoked tobacco. One-third of participating households included a resident who smoked. Only 8% of homes reported indoor smoking, but environmental tobacco smoke should still be considered an important exposure factor that could impact participants’ blood pressure and pulmonary function measures. Although there was a significant interaction between intervention and household smoking in SBP analyses, we generally did not observe strong evidence of effect modification by this factor in post-hoc analyses.

Finally, it is possible that the subclinical health outcomes measured in this study were not as sensitive to year-to-year changes in indoor air quality or were not measured frequently enough to best characterize blood pressure and breathing function status within winter periods. Blood pressure and pulmonary function were selected for this study based upon prior findings in other settings with high household air pollution as well as the feasibility and repeatability of capturing these measures in the field with trained personnel. Several studies in lower income countries have shown that women using traditional biomass fueled cookstoves had higher blood pressure measures compared to women using non-biomass fuels or improved cookstoves (Li et al. 2020). Intervention studies in such settings have shown some evidence of improved blood pressure following implementation of improved or cleaner-burning cookstoves, although randomized trials have been few in number and results have been inconsistent overall (Kumar et al. 2021, Ye et al. 2022). A recent meta-analysis of air filter intervention studies showed overall reductions in blood pressure with significant reductions observed for systolic, but not diastolic, measures (Faridi et al. 2023). These summary results were less robust when restricted to certain study characteristics that were present in our study, including study location (North America/Europe compared to Asia) and study population (mixed health compared to healthy only). Among studies focused on portable air filter interventions in elderly populations, non-significant mean changes in blood pressure were in both directions (Morishita et al. 2018, Guo et al. 2021, Shao et al. 2017, Bräuner et al. 2008). . All of these studies were small and of short duration, several days to two weeks. Our study is distinct from earlier studies in that we monitored changes over a several months period following intervention, however it is also possible that detection of changes in blood pressure would require multiple years of observation for this parameter.

Several studies in lower income countries have evaluated household air pollution exposure and association with lung function in children (Aithal et al. 2021), but intervention studies in these settings have not shown robust benefits to lung function (Smith-Sivertsen et al. 2009, Romieu et al. 2009). Cross-sectional studies of adults have generally shown ambient PM exposure to be associated with lower FEV1 and FVC measures among adults (Götschi et al. 2008, Hou et al. 2020, Yang et al. 2021, Rice et al. 2015), and a recent study showed similar findings specifically among elderly adults (Chen et al. 2019). Household intervention studies among elderly adults are limited, but one case-crossover study of a two-week household air filter intervention in Beijing observed no change in lung function measures (Shao et al. 2017).

Based on input from our community partners we did not exclude any participants according to baseline health status. However, the health measures used in this study and the response of these measures to intervention may have been dependent on baseline health status. For example, a recent analysis from the Agricultural Lung Health Study found that use of wood burning for residential heating was associated with lower FEV1 and FVC among participants with asthma but not among participants that did not have asthma (White 2022). By contrast, associations between PM2.5 exposure estimates and pulmonary function measures were not more robust among Framingham cohort participants with prevalent chronic lung conditions (Rice et al. 2015). The Beijing study (Shao et al. 2017) showed some suggestion of a positive effect among elderly adults without COPD compared to those with COPD, but as with our study the sub-group sample sizes were too small to discern robust effects.

Overall, despite promising exposure reduction results from at least one of the two intervention arms we did not observe beneficial changes to blood pressure or lung function measures. Wood stove interventions that are sustainable and effective are still needed, especially within rural and underserved communities that rely on wood burning as the primary source of home heating.

Supplementary Material

Table S1

Acknowledgements

The authors thank the many families that chose to participate in the EldersAIR study. The project also benefited from the strong support of community members and local research assistants from all study locations.

Funding

The study is funded by the National Institute of Environmental Health Sciences (NIEHS, 1R01ES022583). Intervention design was co-developed for the KidsAIR study (NIEHS, 1R01ES022649). Support also provided by Center for Population Health Research (NIGMS, 1P20GM130418).

Footnotes

Competing Interests

The authors declare no competing interests.

Statements and Declarations

Ethics approval

The EldersAIR study was approved by the University of Montana IRB, the Navajo Nation Human Research Review Board, and the Nez Perce Tribal Executive Committee

Consent to participate

Participants provided written informed consent prior to participation in the study. Participants were compensated for the time they spent performing study tasks and reimbursed for the cost of electricity when using the air filtration units within their homes.

Consent for publication

The authors declare that they agree with the publication of this paper in this journal.

Data availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

References

  1. Aithal SS, Gill S, Satia I, Tyagi SK, Bolton CE & Kurmi OP (2021) The Effects of Household Air Pollution (HAP) on Lung Function in Children: A Systematic Review. Int J Environ Res Public Health, 18. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Allen RW, Carlsten C, Karlen B, Leckie S, van Eeden S, Vedal S, Wong I. & Brauer M. (2011) An air filter intervention study of endothelial function among healthy adults in a woodsmoke-impacted community. Am J Respir Crit Care Med, 183, 1222–30. [DOI] [PubMed] [Google Scholar]
  3. Bates D, Maechler M, Bolker B. & Walker S. (2015) Fitting Linear Mixed-Effects Models Using lme4. Journal of Statistical Software, 67(1), 1–48. [Google Scholar]
  4. Bräuner EV, Forchhammer L, Møller P, Barregard L, Gunnarsen L, Afshari A, Wåhlin P, Glasius M, Dragsted LO, Basu S, Raaschou-Nielsen O. & Loft S. (2008) Indoor particles affect vascular function in the aged: an air filtration-based intervention study. Am J Respir Crit Care Med, 177, 419–25. [DOI] [PubMed] [Google Scholar]
  5. Cheek E, Guercio V, Shrubsole C. & Dimitroulopoulou S. (2021) Portable air purification: Review of impacts on indoor air quality and health. Sci Total Environ, 766:142585. [DOI] [PubMed] [Google Scholar]
  6. Chen CH, Wu CD, Chiang HC, Chu D, Lee KY, Lin WY, Yeh JI, Tsai KW & Guo YL (2019) The effects of fine and coarse particulate matter on lung function among the elderly. Sci Rep, 9, 14790. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Chuang HC, Ho KF, Lin LY, Chang TY, Hong GB, Ma CM, Liu IJ & Chuang KJ (2017) Long-term indoor air conditioner filtration and cardiovascular health: A randomized crossover intervention study. Environ Int, 106, 91–96. [DOI] [PubMed] [Google Scholar]
  8. Fabbri L, Pauwels RA, Hurd SS & Committee GS (2004) Global Strategy for the Diagnosis, Management, and Prevention of Chronic Obstructive Pulmonary Disease: GOLD Executive Summary updated 2003. COPD, 1, 105–41; discussion 103–4. [DOI] [PubMed] [Google Scholar]
  9. Faridi S, Allen RW, Brook RD, Yousefian F, Hassanvand MS & Carlsten C. (2023) An updated systematic review and meta-analysis on portable air cleaners and blood pressure: Recommendations for users and manufacturers. Ecotoxicol Environ Saf, 263, 115227. [DOI] [PubMed] [Google Scholar]
  10. GBD 2019 Risk Factors Collaborators (2020) Global burden of 87 risk factors in 204 countries and territories, 1990–2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet, 396, 1223–1249. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Guo M, Du C, Li B, Yao R, Tang Y, Jiang Y, Liu H, Su H, Zhou Y, Wang L, Yang X, Zhou M. & Yu W. (2021) Reducing particulates in indoor air can improve the circulation and cardiorespiratory health of old people: A randomized, double-blind crossover trial of air filtration. Sci Total Environ, 798, 149248. [DOI] [PubMed] [Google Scholar]
  12. Götschi T, Heinrich J, Sunyer J. & Künzli N. (2008) Long-term effects of ambient air pollution on lung function: a review. Epidemiology, 19, 690–701. [DOI] [PubMed] [Google Scholar]
  13. Hansel NN, Putcha N, Woo H, Peng R, Diette GB, Fawzy A, Wise RA, Romero K, Davis MF, Rule AM, Eakin MN, Breysse PN, McCormack MC & Koehler K. (2022) Randomized Clinical Trial of Air Cleaners to Improve Indoor Air Quality and Chronic Obstructive Pulmonary Disease Health: Results of the CLEAN AIR Study. Am J Respir Crit Care Med, 205, 421–430. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Hart JF, Ward TJ, Spear TM, Rossi RJ, Holland NN & Loushin BG (2011) Evaluating the effectiveness of a commercial portable air purifier in homes with wood burning stoves: a preliminary study. J Environ Public Health, 2011, 324809. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Hou D, Ge Y, Chen C, Tan Q, Chen R, Yang Y, Li L, Wang J, Ye M, Li C, Meng X, Kan H, Cai J. & Song Y. (2020) Associations of long-term exposure to ambient fine particulate matter and nitrogen dioxide with lung function: A cross-sectional study in China. Environ Int, 144, 105977. [DOI] [PubMed] [Google Scholar]
  16. Kajbafzadeh M, Brauer M, Karlen B, Carlsten C, van Eeden S. & Allen RW (2015) The impacts of traffic-related and woodsmoke particulate matter on measures of cardiovascular health: a HEPA filter intervention study. Occup Environ Med, 72, 394–400. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Kumar N, Phillip E, Cooper H, Davis M, Langevin J, Clifford M. & Stanistreet D. (2021) Do improved biomass cookstove interventions improve indoor air quality and blood pressure? A systematic review and meta-analysis. Environ Pollut, 290, 117997. [DOI] [PubMed] [Google Scholar]
  18. Li L, Yang A, He X, Liu J, Ma Y, Niu J. & Luo B. (2020) Indoor air pollution from solid fuels and hypertension: A systematic review and meta-analysis. Environ Pollut, 259, 113914. [DOI] [PubMed] [Google Scholar]
  19. McNamara ML, Noonan CW & Ward TJ (2011) Correction Factor for Continuous Monitoring of Wood Smoke Fine Particulate Matter. Aerosol and Air Quality Research, 11, 316–323. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Morishita M, Adar SD, D’Souza J, Ziemba RA, Bard RL, Spino C. & Brook RD (2018) Effect of Portable Air Filtration Systems on Personal Exposure to Fine Particulate Matter and Blood Pressure Among Residents in a Low-Income Senior Facility: A Randomized Clinical Trial. JAMA Intern Med, 178, 1350–1357. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Noonan CW, Navidi W, Sheppard L, Palmer CP, Bergauff M, Hooper K. & Ward TJ (2012) Residential indoor PM2.5 in wood stove homes: follow-up of the Libby changeout program. Indoor Air, 22, 492–500. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Noonan CW, Semmens EO, Smith P, Harrar SW, Montrose L, Weiler E, McNamara M. & Ward TJ (2017) Randomized Trial of Interventions to Improve Childhood Asthma in Homes with Wood-burning Stoves. Environ Health Perspect, 125, 097010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Noonan CW, Semmens EO, Ware D, Smith P, Boyer BB, Erdei E, Hopkins SE, Lewis J. & Ward TJ (2019) Wood stove interventions and child respiratory infections in rural communities: KidsAir rationale and methods. Contemp Clin Trials, 89, 105909. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Noonan CW, White MC, Thurman D. & Wong LY (2005) Temporal and geographic variation in United States motor neuron disease mortality, 1969–1998. Neurology, 64, 1215–1221. [DOI] [PubMed] [Google Scholar]
  25. Orozco-Levi M, Garcia-Aymerich J, Villar J, Ramirez-Sarmiento A, Anto JM & Gea J. (2006) Wood smoke exposure and risk of chronic obstructive pulmonary disease. Eur Respir J, 27, 542–6. [DOI] [PubMed] [Google Scholar]
  26. Po JY, FitzGerald JM & Carlsten C. (2011) Respiratory disease associated with solid biomass fuel exposure in rural women and children: systematic review and meta-analysis. Thorax, 66, 232–9. [DOI] [PubMed] [Google Scholar]
  27. Qaseem A, Wilt TJ, Rich R, Humphrey LL, Frost J, Forciea MA, Fitterman N, Barry MJ, Horwitch CA, Iorio A, McLean RM & C. G. C. o. t. A. C. o. P. a. t. C. o. H. o. t. P. a. S. o. t. A. A. o. F. Physicians (2017) Pharmacologic Treatment of Hypertension in Adults Aged 60 Years or Older to Higher Versus Lower Blood Pressure Targets: A Clinical Practice Guideline From the American College of Physicians and the American Academy of Family Physicians. Ann Intern Med, 166, 430–437. [DOI] [PubMed] [Google Scholar]
  28. Rice MB, Ljungman PL, Wilker EH, Dorans KS, Gold DR, Schwartz J, Koutrakis P, Washko GR, O’Connor GT & Mittleman MA (2015) Long-term exposure to traffic emissions and fine particulate matter and lung function decline in the Framingham heart study. Am J Respir Crit Care Med, 191, 656–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Romieu I, Riojas-Rodríguez H, Marrón-Mares AT, Schilmann A, Perez-Padilla R. & Masera O. (2009) Improved biomass stove intervention in rural Mexico: impact on the respiratory health of women. Am J Respir Crit Care Med, 180, 649–56. [DOI] [PubMed] [Google Scholar]
  30. Semmens EO, Noonan CW, Allen RW, Weiler EC & Ward TJ (2015) Indoor particulate matter in rural, wood stove heated homes. Environ Res, 138, 93–100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Shao D, Du Y, Liu S, Brunekreef B, Meliefste K, Zhao Q, Chen J, Song X, Wang M, Wang J, Xu H, Wu R, Wang T, Feng B, Lung CS, Wang X, He B. & Huang W. (2017) Cardiorespiratory responses of air filtration: A randomized crossover intervention trial in seniors living in Beijing: Beijing Indoor Air Purifier StudY, BIAPSY. Sci Total Environ, 603–604, 541–549. [DOI] [PubMed] [Google Scholar]
  32. Sigsgaard T, Forsberg B, Annesi-Maesano I, Blomberg A, Bolling A, Boman C, Bonlokke J, Brauer M, Bruce N, Heroux ME, Hirvonen MR, Kelly F, Kunzli N, Lundback B, Moshammer H, Noonan C, Pagels J, Sallsten G, Sculier JP & Brunekreef B. (2015) Health impacts of anthropogenic biomass burning in the developed world. Eur Respir J, 46, 1577–88. [DOI] [PubMed] [Google Scholar]
  33. Singleton R, Salkoski AJ, Bulkow L, Fish C, Dobson J, Albertson L, Skarada J, Kovesi T, McDonald C, Hennessy TW & Ritter T. (2017) Housing characteristics and indoor air quality in households of Alaska Native children with chronic lung conditions. Indoor Air, 27, 478–486. [DOI] [PubMed] [Google Scholar]
  34. Smith-Sivertsen T, Díaz E, Pope D, Lie RT, Díaz A, McCracken J, Bakke P, Arana B, Smith KR & Bruce N. (2009) Effect of reducing indoor air pollution on women’s respiratory symptoms and lung function: the RESPIRE Randomized Trial, Guatemala. Am J Epidemiol, 170, 211–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Sood A, Petersen H, Blanchette CM, Meek P, Picchi MA, Belinsky SA & Tesfaigzi Y. (2010) Wood smoke exposure and gene promoter methylation are associated with increased risk for COPD in smokers. American journal of respiratory and critical care medicine, 182, 1098–104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. U.S. Energy Information Agency. 2015. Residential Energy Consumption Survey. 2015 RECS Survey Data. [Google Scholar]
  37. U.S. Environmental Protection Agency (2021) Best Wood-Burning Practices. https://www.epa.gov/burnwise/best-wood-burning-practices (last accessed January 26, 2023).
  38. US Environmental Protection Agency (2021) Best Wood-Burning Practices. https://www.epa.gov/burnwise/best-wood-burning-practices (last accessed.
  39. Walker ES, Noonan CW, Belcourt A, Boulafentis J, Garcia C, Graham J, Hoskie N, Quintana E, Semmens EO, Simpson J, Smith P, Teasley H, Ware D, Weiler E. & Ward TJ (2022a) Efficacy of air filtration and education interventions on fine particulate matter among rural Native American homes heated with wood stoves: Results from the EldersAIR randomized trial. Sci Total Environ, 843, 157029. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Walker ES, Noonan CW, Semmens EO, Ware D, Smith P, Boyer BB, Erdei E, Hopkins SE, Lewis J, Belcourt A. & Ward TJ (2021) Indoor fine particulate matter and demographic, household, and wood stove characteristics among rural US homes heated with wood fuel. Indoor Air, 31, 1109–1124. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Walker ES, Semmens EO, Belcourt A, Boyer BB, Erdei E, Graham J, Hopkins SE, Lewis JL, Smith PG, Ware D, Weiler E, Ward TJ & Noonan CW (2022b) Efficacy of Air Filtration and Education Interventions on Indoor Fine Particulate Matter and Child Lower Respiratory Tract Infections among Rural U.S. Homes Heated with Wood Stoves: Results from the KidsAIR Randomized Trial. Environ Health Perspect, 130, 47002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Walters KL, Johnson-Jennings M, Stroud S, Rasmus S, Charles B, John S, Allen J, Kaholokula JK, Look MA, de Silva M, Lowe J, Baldwin JA, Lawrence G, Brooks J, Noonan CW, Belcourt A, Quintana E, Semmens EO & Boulafentis J. (2020) Growing from Our Roots: Strategies for Developing Culturally Grounded Health Promotion Interventions in American Indian, Alaska Native, and Native Hawaiian Communities. Prev Sci, 21, 54–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Ward T, Boulafentis J, Simpson J, Hester C, Moliga T, Warden K. & Noonan C. (2011) Lessons learned from a woodstove changeout on the Nez Perce Reservation. Sci Total Environ, 409, 664–70. [DOI] [PubMed] [Google Scholar]
  44. Ward T. & Noonan C. (2008) Results of a residential indoor PM(2.5) sampling program before and after a woodstove changeout. Indoor Air, 18, 408–415. [DOI] [PubMed] [Google Scholar]
  45. Ward TJ, Semmens EO, Weiler E, Harrar S. & Noonan CW (2017) Efficacy of interventions targeting household air pollution from residential wood stoves. J Expo Sci Environ Epidemiol, 27, 64–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Wheeler AJ, Gibson MD, MacNeill M, Ward TJ, Wallace LA, Kuchta J, Seaboyer M, Dabek-Zlotorzynska E, Guernsey JR & Stieb DM (2014) Impacts of Air Cleaners on Indoor Air Quality in Residences Impacted by Wood Smoke. Environmental Science & Technology, 48, 12157–12163. [DOI] [PubMed] [Google Scholar]
  47. White JD, Wyss AB, Hoang TT, Lee M, Richards M, Parks CG, Beane-Freeman LE, Hankinson JL, Umbach DM & London SJ (2022) Residential Wood Burning and Pulmonary Function in the Agricultural Lung Health Study. Environ Health Perspect, 130, 87008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Yang T, Chen R, Gu X, Xu J, Yang L, Zhao J, Zhang X, Bai C, Kang J, Ran P, Shen H, Wen F, Huang K, Chen Y, Sun T, Shan G, Lin Y, Wu S, Zhu J, Wang R, Shi Z, Xu Y, Ye X, Song Y, Wang Q, Zhou Y, Ding L, Yao W, Guo Y, Xiao F, Lu Y, Peng X, Zhang B, Xiao D, Wang Z, Zhang H, Bu X, An L, Zhang S, Cao Z, Zhan Q, Yang Y, Liang L, Cao B, Dai H, van Donkelaar A, Martin RV, Wu T, He J, Kan H, Wang C. & Group CPHS (2021) Association of fine particulate matter air pollution and its constituents with lung function: The China Pulmonary Health study. Environ Int, 156, 106707. [DOI] [PubMed] [Google Scholar]
  49. Yawn BP, Duvall K, Peabody J, Albers F, Iqbal A, Paden H, Zubek VB & Wadland WC (2014) The impact of screening tools on diagnosis of chronic obstructive pulmonary disease in primary care. Am J Prev Med, 47, 563–75. [DOI] [PubMed] [Google Scholar]
  50. Ye W, Steenland K, Quinn A, Liao J, Balakrishnan K, Rosa G, Ndagijimana F, Ntivuguruzwa JD, Thompson LM, McCracken JP, Diaz-Artiga A, Rosenthal JP, Papageorghiou A, Davila-Roman VG, Pillarisetti A, Johnson M, Wang J, Nicolaou L, Checkley W, Peel JL, Clasen TF & Household I. Air Pollution Intervention Network trial (2022) Effects of a Liquefied Petroleum Gas Stove Intervention on Gestational Blood Pressure: Intention-to-Treat and Exposure-Response Findings From the HAPIN Trial. Hypertension, 79, 1887–1898. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Table S1
NIHMS1995441-supplement-Table_S1.docx (30.8KB, docx)

Data Availability Statement

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

RESOURCES