Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Nov 1.
Published in final edited form as: Indoor Air. 2021 Jul 7;31(6):2008–2019. doi: 10.1111/ina.12904

Household and behavioral determinants of indoor PM2.5 in a rural, solid fuel burning Native American community

Steven J Hadeed 1, Mary Kay O’Rourke 1, Robert A Canales 2,*, Lorencita Joshweseoma 3, Gregory Sehongva 3, Morris Paukgana 3, Emmanuel Gonzalez-Figueroa 1, Modhi Alshammari 1, Jefferey L Burgess 1, Robin B Harris 4
PMCID: PMC8530885  NIHMSID: NIHMS1719772  PMID: 34235761

Abstract

Indoor and outdoor concentrations of PM2.5 were measured for 24-hours during heating and non-heating seasons in a rural, solid fuel burning, Native American community. Household building characteristics were collected during the initial home sampling visit using technician walkthrough questionnaires, and behavioral factors were collected through questionnaires by interviewers. To identify seasonal behavioral factors and household characteristics associated with indoor PM2.5, data were analyzed separately by heating and non-heating seasons using multivariable regression. Concentrations of PM2.5 were significantly higher during the heating season (Indoor: 36.2 μg/m3; Outdoor: 22.1 μg/m3) compared to the non-heating season (Indoor: 14.6 μg/m3; Outdoor: 9.3 μg/m3). Heating season indoor PM2.5 was strongly associated with heating fuel type, housing type, indoor pests, use of a climate control unit, number of interior doors and indoor relative humidity. During the non-heating season, different behavioral and household characteristics were associated with indoor PM2.5 concentrations (indoor smoking and/or burning incense, opening doors and windows, area of surrounding environment, building size and height, and outdoor PM2.5). Homes heated with coal and/or wood, or a combination of coal and/or wood with electricity and/or natural gas had elevated indoor PM2.5 concentrations that exceeded both the EPA ambient standard (35 μg/m3) and the WHO guideline (25 μg/m3).

Keywords: solid fuel use, household air pollution, indoor PM2.5, household environmental risk factors, rural health

Introduction

Solid fuels (biomass, coal) are used for cooking, heating, and lighting by approximately 3 billion people globally (WHO, 2016). The reliance on these fuels results in an estimated 3.8 million premature deaths per year attributable to household air pollutants (HAPs) from combustion of these fuel sources (WHO, 2016). Fine particulate matter (PM2.5) is a hazardous air pollutant widely associated with adverse health outcomes such as stroke, respiratory disease, asthma, lung cancer and death (Apte et al., 2015; Bruce et al., 2000). People are exposed through inhalation to PM2.5 generated from the burning of solid fuels for heating and cooking. A large proportion of the global population relying on the use of these fuels reside in low-income communities of developing nations. Neighborhood and household level environmental monitoring is necessary to understand the health effects from PM2.5 exposures, and to guide targeted interventions and develop appropriate public health policy.

Despite the prevalent use of solid fuels in developing nations, recent studies (Bulkow et al., 2012; Rogalsky et al., 2014; Semmens et al., 2015) suggest that low income rural populations of the United States, and other affluent countries, experience similar exposures to HAPs. Rogalsky et al., (2014) estimate that more than 2.5 million households and 6.5 million people in the United States use solid fuel as a primary heating source. Exposure to HAPs is strongly associated with poverty, lack of access to cleaner fuels, and improper stove maintenance. In northern and western regions of the United States, solid fuel use is highly prevalent in rural communities with high poverty levels. Solid fuel use and poverty rates are highest in the Southwestern United States, specifically the Four Corners region of Arizona, Utah, Colorado and New Mexico (Rogalsky et al., 2014). Native American communities are disproportionately affected by air pollution and are at greater risk for exposures and adverse health effects due to the economic disadvantages and health disparities that exist on tribal lands (U.S. EPA, 2015).

Monitoring of indoor air pollutants in tribal communities in the United States is limited. Robin et al. (1996) examined the association between indoor concentrations of PM10 and respiratory infection in children living in solid fuel burning homes on the Navajo Nation. In wood heated homes (n=3) that cooked with electricity or gas, median concentrations of PM10 were 100.9 μg/m3 (GM=57.4), whereas homes that cooked and heated with only wood (n=4) had median indoor concentrations of 85.6 μg/m3 (GM=62.8). Homes using electricity or gas for heating and cooking had the lowest indoor concentrations (median=22.2 μg/m3, GM=18.8). In homes cooking with wood, acute lower respiratory infection in children was five times higher than children living in gas or electric homes. Quantifying and characterizing PM2.5 concentration in coal burning homes on the Navajo Nation was most recently assessed in 19 homes and reported an average 24-hour PM2.5 concentration of 37.9 μg/m3, with mean daily concentrations as high as 109 μg/m3, exceeding the EPA 24-hour ambient standard of 35 μg/m3 (Bunnell et al., 2010). Although these two studies were conducted in a different tribal community, the Navajo Nation and Hopi Tribe border one another, use coal originating from the same coal mine, and experience similar socioeconomic constraints.

Relatively few studies have assessed rural, indoor air quality in the United States, where populations continue to experience greater health disparities and disease burden. The prevalent use of solid fuel in rural households highlights the need for quantifying indoor concentrations of PM2.5 in these communities. The Hopi Tribe of rural Northern Arizona is among those affected. A large proportion of Hopi households report burning coal (37%) and wood (33%) for heating (U.S. Census Bureau, 1995). Alarmingly, roughly half of youth on Hopi lands live in poverty (U.S. Census Bureau, 2010). A 2012 health survey reported 24% of participants experience asthma attacks (Joshweseoma, 2015). No prior environmental air monitoring study has been undertaken with the Hopi Tribe. The goal of this study is to determine behavioral and household characteristics associated with indoor PM2.5 concentrations during heating and non-heating seasons in households of a rural, solid fuel burning, Native American community.

Methods

Study Population & Setting

The Hopi are a Native American Tribe of Pueblo People in rural, Northern Arizona who live in 12 villages within the 6,557 square kilometers of Hopi land. The area contains 7,185 Hopi residing in 2,081 households with a median income of $34,016 (US) per year (U.S. Census Bureau, 2010). The Hopi Tribal lands are situated on the Colorado Plateau in northeastern Arizona with an average elevation of 7200 feet. It also experiences wide variability in temperature and precipitation with substantial variability between winter and summer wet seasons and dry intervening seasons (Crimmins et al., 2015). Until 2020, members of the Hopi Tribe were provided free coal from a large nearby mining operation, which served as a major heating source. Although coal was widely available, it was used exclusively for seasonal heating purposes, whereas electricity and liquid propane gas were used primarily for cooking.

Hopi homes were classified into three categories: Traditional, modern or modular/mobile and building materials varied among these home types. Traditional homes were frequently built of rock and plastered with mud mortar and included packed dirt floors and roofing of traditional tuuma (dirt over brush or sticks). Some of these homes were built prior to 1,100 AD based on archaeological records of the area (Carson et al., 2020, Whiteley, 1988). Most modern homes had concrete slab floors with frame or block walls and asphalt roofs. Modular/mobile homes were constructed in a factory and had air flowing under the structure, a metal exterior, coated with a sprayed insulation and finished with paneling and ceiling tile. Hopi homes could have elements of each building type and were categorized by the percent of the dominant material. Some homes lacked electricity, and running water depending on age and location. In general, heating stoves at Hopi were located in the main living area where people spend most of their waking hours.

There are no standard floor plans for Hopi homes and there are no building code requirements. People build and modify their homes as they choose. The Hopi Environmental Health Project (HEHP) is a community based participatory research (CBPR) project conducted with Hopi partners and with oversight by a Community Advisory Board (CAB). The CAB prohibited photos of homes or people to protect privacy.

Recruitment & Selection of Participants

Potential households were identified spatially using Google Earth with 222 potential housing structures randomly selected and approached for recruitment. Of the 222, 126 households were eligible and 76 households were enrolled in the study. The enrolled households were distributed across Hopi land in proportion of the population of each village relative to the tribal population. Inclusion criteria were that members of the Hopi Tribe lived in the household and someone 18 years or older (i.e., head of household) agreed to individual and household monitoring. All participants provided a signed consent form. The protocols and consenting documents were approved by the University of Arizona Institutional Review Board. In addition, the project was approved by the Hopi Tribal Council. The CAB reviewed all procedures and all local research staff were members of the Hopi Tribe and living at Hopi.

Field Work Training & Data Monitoring

The Hopi field office was located more than six hours from the university research staff. The HEHP university research faculty and staff provided required and regular training, both in-person and virtual, for all elements of the study including: use of the air quality monitors, maintenance and calibration of the monitors, recruitment and consenting of study participants, individual measurements and interviewing techniques. As required by the CAB and agreements with the Hopi Tribe, all personally identifiable information was retained at Hopi and not returned to the university research team. Only de-identified measurements were returned to the university for data entry and analysis. Data and equipment coordinators regularly checked with the field teams and visits were conducted to provide oversight and additional training.

Air Quality Monitoring

Attempts were made to sample homes twice for 24-hours, once during the heating season, and again during the non-heating season. Longer sampling times were proposed, however the CAB viewed sampling for longer periods as too intrusive. Of the 76 homes sampled, 70 were sampled during the heating season and 66 during the non-heating season; 60 households completed both winter and summer sampling.

Indoor and outdoor concentrations of PM2.5 were collected each season using real-time area monitors (pDR-1500, ThermoFisher) set at 1-minute logging intervals for 24-hours. Samplers operated concurrently. Monitors were calibrated by field technicians prior to placement at each home and in accordance with the instruction manual. Further, all project monitoring equipment was returned for annual factory calibration. Unfortunately, the field implementation office was located 6 hours away from the laboratory and lacked the resources for gravimetric assessment of filter weights in the field. Each area monitor simultaneously measured air temperature and relative humidity. Monitors were placed approximately 1.5 meters off the ground to represent the breathing zone of inhabitants. The indoor air monitor was placed in the living room where household occupants spent most of their time and the outdoor monitor was placed near the main entrance of the household. Combustion gases (NO, NO2, SO2, CO) were measured using a real-time passive monitor (GrayWolf AdvancedSense Pro), co-located with the indoor PM2.5 monitor and set at the same 1-minute logging interval. The GrayWolf sampler was returned to the manufacturer for annual factory calibration.

Home & Activity Assessments

Household building characteristics were assessed using questionnaires collected during the initial home sampling visit. Important residential characteristics including primary heating fuel type (solid fuel: coal and/or wood; EG: electric and/or gas; combination: coal and/or wood with electric and/or gas) and housing type (modern, traditional, mobile) were noted. The technician also recorded the size of the house, building and flooring material, number of rooms, number of doors and windows. Other factors reported were presence of pets inside the home, dustiness of the interior environment, condition of the roof, visible crack or gaps in the walls or ceiling, noticeable signs of mold or pests, the condition and number of household appliances, and ventilation devices (fans, air conditioning). Characteristics of the exterior home environment were collected; they included categorizing the area surrounding the home (natural, cleared, actively used for livestock/gardening), the outside parking area (paved, gravel, unpaved), and the surrounding roadways leading to the residence (paved or unpaved).

Primary respondents also were interviewed using culturally appropriate and population specific both pre and post-sampling questionnaires for each of the season visits. These questions were developed and modified under the guidance of the CAB. Questions were designed to identify specific indoor and outdoor behavioral activities practiced by household occupants that may be associated with indoor PM2.5. Participants were asked general questions regarding cooking, cleaning, burning of incense, smoking, craft-activity (art or traditional cooking), other traditional practices, and whether solid fuel or air circulating devices were used during the sampling period. More specific activity questionnaires were determined by the CAB to be too burdensome and were not used.

Statistical Data Analysis

Mean imputation was used for homes with partial or incomplete real-time monitoring data. A detailed explanation of imputation methods in this context has been discussed in a previous paper (Hadeed, et al., 2020). Indoor PM2.5 concentration was averaged over 24-hours and was naturally log-transformed. Variables collected from questionnaires were entered as categorical, binary, or nominal variables. Frequency tables were used to assess the distribution of responses for categorical variables, and cells with low counts were collapsed into appropriate categories. Categorical variables were analyzed with appropriate reference categories assigned.

Univariable linear regression was used to identify household and behavioral explanatory variables that were significantly associated with indoor PM2.5 and to assess the magnitude of their association. To ensure important variables were not excluded from potential selection into multivariable regression, a relaxed significance level (p≤0.20) was used to screen candidate variables. Collinearity among explanatory variables was tested prior to development of a multivariable model. Two explanatory variables highly correlated (r≥0.70) resulted in exclusion of one of the two variables. To identify the seasonal behavioral factors and household characteristics associated with indoor concentrations of PM2.5, separate analyses were conducted for heating and non-heating sample periods. Multivariable regression was used to determine the household and behavioral factors associated with indoor PM2.5 concentrations for each season. Variable selection was carried out using backwards elimination with a cutoff of 10% (p≥0.10).

Results

Summary of Housing Characteristics

Household building characteristics are summarized in the supplemental material (Table S1). A large proportion of households were considered as modern construction homes (59%; n=45), followed by traditional (24%; n=18) and modular/mobile (17%; n=13) homes. Homes had an average size of 93.7 m2 and 3–4 rooms. Three heating fuel types were common during the winter heating season, EG as fuel (28%; n=21), solid fuel of coal and/or wood (38%; n=29), and “combination” fuels of both EG and solid fuel (34%; n=26). Most homes in this rural community were surrounded by natural undisturbed vegetation (41%) or were cleared of vegetation (42%).

Reporting of the dustiness of the interior environment by the primary field technician indicated more than one-third of homes were fairly dusty (37%) and 13% were extremely dusty, with the remainder of homes being relatively dust free. This variable was dichotomized due to low cell count in extremely dusty homes. Concrete and wood were dominant flooring components in these homes, 34% and 49%, respectively; however, 12% of homes had packed earth floors. Twenty percent of homes had greater than 20% of their interior floor covered with carpeting or rugs. Additional housing characteristics are summarized in supplemental Table S1 and include the number of interior and exterior doors and windows, the height of the home, and surface of the parking area outside the home.

Household Air Quality by Season

Of the 70 homes sampled during the heating season, two homes had missing indoor air monitoring data and one household was excluded (participant limitation). During the heating season, the 24-hour indoor mean PM2.5 concentration in these 67 homes was 36.2 μg/m3 and the mean outdoor concentration was 22.1 μg/m3 (Table 1). Of the 66 homes sampled in the non-heating season, 5 homes had missing indoor air monitoring data and an additional 5 homes had missing outdoor monitoring data. Indoor non-heating 24-hour mean concentration of PM2.5 in these 61 homes was 14.6 μg/m3 and the mean outdoor 24-hour concentration was 9.3 μg/m3 (Table 1). Indoor concentrations of combustion gases (NO, NO2, CO, SO2) were measured in ppm. Most homes had concentrations below the instrument detection level. As a result, seasonally averaged values are in fractions of ppb. As expected, values are lower in the non-heating season than during the heating season (Table 1).

Table 1:

Concentrations of Air Pollutants by Sampling Season

Heating Non-Heating
Variable Mean SD Mean SD
Indoor PM2.5 (μg/m3) 36.2 45.0 14.6 17.0
Outdoor PM2.5 (μg/m3) 22.1 40.2 9.3 9.8
Indoor NO (ppb) 0.089 0.140 0.0056 0.023
Indoor NO2 (ppb) 0.089 0.099 0.0860 0.067
Indoor CO (ppb) 3.100 5.900 0.5900 0.890
Indoor SO2 (ppb) 0.030 0.086 0.0160 0.035
Open in a new tab

Indoor Activities Performed by Season

During the heating season, 25% of participants reported they engaged in craft-activities (art/cooking) during the 24-hour sampling period. Seven homes (10.4%) reported smoking or burning incense indoors, and participation in the removal of ash generated from heating devices was reported in over 50% of households. Half of participants reported cleaning their floors or having their pets indoors during a sampling period. A central or wall mounted climate control unit was used in 16% of homes, and about one-third of participants reported seeing an indoor pest within the past 24-hours (Table 2). During the non-heating sampling, over 80% of homes reported opening doors and windows for ventilation, and over 25% of participants engaged in craft-activities indoors. Smoking and burning of incense indoors was reported in 8 homes (13.1%) and more than 50% of homes reported seeing a pest indoors during the 24-hour period (Table 2).

Table 2:

Mean (24-hour) Indoor PM2.5 Concentrations (μg/m3) by Household Characteristics and Behavioral Factors for Each Sampling Season.

Heating Non-Heating
Variable Mean SD Freq (%) Mean SD Freq (%)
Area of Home (m2)
< 55.5 31.4 37.1 15 (22.4) 21.6 24.4 16 (26.2)
55.5 – 69.2 51.1 55.2 17 (25.4) 17.9 17.7 16 (26.2)
69.3 – 130.0 32.3 49.3 19 (28.4) 10.5 10.5 15 (24.6)
130.1 – 504.0 29.7 33.2 16 (23.9) 7.0 3.9 14 (23.0)
Indoor Smoking or Incense
No 33.0 43.6 60 (89.6) 12.5 15.3 53 (86.9)
Yes 64.0 50.2 7 (10.4) 28.0 22.6 8 (13.1)
Parking Area
Paved-Gravel 27.8 47.3 22 (32.8) 12.6 9.9 16 (26.2)
Unpaved 40.4 43.7 45 (67.2) 15.2 19.0 45 (73.8)
Surrounding Area of Home
Natural Undisturbed 24.7 27.5 25 (37.3) 11.3 11.7 26 (42.6)
Cleared 40.1 46.8 29 (43.3) 20.4 22.4 25 (41.0)
Actively Used 51.2 65.8 12 (17.9) 9.0 7.1 9 (14.8)
Missing 32.4 0.0 1 (1.5) 3.0 0.0 1 (1.6)
Pests Found in House
No Pest 31.6 40.5 46 (68.7) 12.7 15.2 27 (44.3)
Tick-Roach-Insect 32.4 60.1 11 (16.4) 17.8 20.1 27 (44.3)
Rodent 61.6 42.1 10 (14.9) 9.0 7.4 7 (11.5)
Dustiness of Interior Observed by Technician
Not Dusty 25.6 31.4 35 (52.2) 12.8 17.5 30 (49.2)
Dusty 47.8 54.4 32 (47.8) 16.6 16.8 30 (49.2)
Missing 3.6 0.0 1 (1.6)
Open Doors/Windows
No 32.0 43.1 35 (52.2) 31.6 28.4 10 (16.4)
Yes 40.9 47.2 32 (47.8) 11.2 11.6 51 (83.6)
Floor Cleaning During Sampling
No 43.3 54.6 34 (50.7) 13.9 16.6 27 (44.3)
Yes 28.9 31.4 33 (49.3) 15.1 17.6 34 (55.7)
Any Craft Activities During Sampling
No 38.6 49.6 50 (74.6) 13.5 16.6 44 (72.1)
Yes 29.4 27.5 17 (25.4) 17.4 18.2 17 (27.9)
Climate Control Device
No 40.0 46.9 56 (83.6) 15.0 16.8 44 (72.1)
Yes 17.1 28.1 11 (16.4) 13.3 18.1 17 (27.9)
Ash Removal During Sampling
No 31.8 44.8 30 (44.8) 14.2 17.2 59 (96.7)
Yes 39.8 45.4 37 (55.2) 23.6 11.5 2 (3.3)
Pets Indoors During Sampling
No 37.7 48.1 29 (43.3) 14.7 18.3 27 (44.3)
Yes 35.1 43.1 38 (56.7) 14.4 16.2 34 (55.7)
Open in a new tab

Indoor PM2.5 Across Household & Behavioral Factors by Season

During the heating season, the more modern style homes had lower indoor mean concentrations (27.8 μg/m3) compared to traditional (53.2 μg/m3) and mobile (46.1 μg/m3) homes (Figure 1). Homes heated with solid fuel had mean indoor 24-hour concentrations of 45.2 μg/m3, whereas combination and EG fueled homes had indoor levels of 39.2 μg/m3 and 21.2 μg/m3, respectively. A substantial reduction in indoor PM2.5 was observed between heating and non-heating season by housing and heating fuel type (Figure 1).

Figure 1:

Figure 1:

Open in a new tab

Mean Concentrations of PM2.5 across Fuel & Housing Type by Season

Average indoor PM2.5 concentrations were greater during the heating season compared with the non-heating season. Differences in household characteristics, behaviors and activities by residents (Table 2) may be related to the seasonal increases. The primary difference across seasons was the use of heating fuel type. Homes heated exclusively with coal and wood had indoor concentrations that exceeded the EPA 24-hour ambient PM2.5 standard of 35μg/m3 and the WHO recommendation of 25μg/m3. Even homes that used a combination of EG and solid fuel during the heating season had indoor concentrations that exceeded both the EPA standard and WHO guideline. Interestingly, ambient PM2.5 concentrations recorded outside of coal and wood burning homes were higher (30.0 μg/m3) compared to EG fueled homes (9.2 μg/m3), suggesting the potential effect household burning of these fuels can have on local ambient air quality. Activities such as indoor smoking or burning of incense also resulted in elevated indoor concentrations (64.0 μg/m3) during the heating season, whereas homes that used climate control units (fans and thermostats) had much lower indoor concentrations (17.1 μg/m3) than homes without climate control units (40.0 μg/m3).

Seasonal differences in household and behavioral factors associated with indoor PM2.5 concentrations were observed. Indoor PM2.5 was strongly associated with housing type, heating fuel type, presence of indoor pests, use of a climate control unit, number of interior doors and indoor relative humidity during the winter heating season. Presence of a pest indoors may be an indicator of indoor cleanliness, structural integrity, and other factors associated with indoor air pollutants. Compared to modern homes, mobile homes had elevated indoor PM2.5 concentrations, which may reflect the expansion and contraction of building material during the day when fuel use for heating is greatest. There is a positive association between indoor relative humidity and PM2.5. Perhaps hygroscopic growth of ultrafine particles or aggregation of particles is responsible. Winkler (1988) states that particles < 0.1 μm and those above 1 μm grow hygroscopically with similar patterns. He et al. (2019) state that humidity can serve a bridge that aggregates particulate matter. Negative associations between climate control devices (fans and thermostats) could be a function of increased air circulation, exchange rate and filtration of air, whereas the number of interior doors could serve as a barrier for compartmentalizing the living room from highly polluted rooms (such as the kitchen).

During the non-heating season, indoor PM2.5 concentrations gradually declined as the size of the home increased (Table 2). For example, homes <55.5 m2 had concentrations of 21.6 μg/m3 compared to homes sized 55.6 m2 – 69.2 m2 (17.9 μg/m3), 69.3 m2 – 130 m2 (10.5 μg/m3), and homes over 130 m2 (7.0 μg/m3). In homes that burned incense or had a smoker, indoor concentrations were 28.0 μg/m3 compared to 12.5 μg/m3 in homes without smoking or incense. Other behavioral and household characteristics were associated with indoor PM2.5 concentrations (opening doors and windows, area of surrounding environment, and outdoor PM2.5). Elevated indoor PM2.5 concentrations among homes cleared of ground cover may be explained by fugitive dust being suspended during high wind events and making entry into the home. To reduce windblown dust from entering homes, villages and household owners may consider opting for natural vegetation that anchors soil or avoiding clearing vegetation around their homes. Paving roads and driveways may also assist in reducing indoor dust. Opening of doors and windows, as well as the size and height of the home, were negatively associated with indoor PM2.5, possibly due to dilution of indoor air with outside clean air and increased volume of the home.

Univariable Analysis: Heating Season

Candidate variables for inclusion into multivariable regression were identified from univariable regression (Table 3). Many other variables, including the combustion gases of Table 1, were evaluated, and they were not significantly associated with PM2.5, so they were not included in the modeling. Log-natural (LN) indoor PM2.5 was significantly associated (p<0.05) with various behavioral and household variables during the winter heating season, that included heating fuel type, indoor smoking or burning of incense, presence of pests, housing type, parking area of the home, dustiness of interior environment indoor, use of a climate control unit, use of kitchen or bathroom ventilation, presence of carpet or rug, number of interior doors, area of home, and indoor relative humidity.

Table 3:

Factors Associated with PM2.5 Concentrations Measured in Heating Season - Univariable and Multivariable Regression Models

Univariable Regression Multivariable Regression
LN-PM2.5 Coef. Percent Change p-value Coef. Percent Change p-value
Household Characteristics
Area of Home −0.0003 −0.03 0.06 NS
Number of Interior Doors −0.16 −15.0 0.01 −0.135 −12.6 0.014
Number of Exterior Doors −0.28 −24.2 0.10 NS
>20% Carpet/Rugs (Yes vs No) −0.66 −48.1 0.05 NS
Height of Home −0.05 −5.3 0.12 NS
Home (Traditional vs Modern) 0.75 110.7 0.03 0.219 24.5 0.464
Home (Mobile vs Modern) 0.33 39.5 0.39 0.822 127.4 0.02
Pest (Insect vs no pest) −0.50 −39.1 0.17 −0.136 −12.8 0.644
Pest Rodent vs no pest) 0.86 137.2 0.02 0.646 90.7 0.05
Dusty (Yes vs No)* 0.73 107.5 0.01 NS
Behavioral
Any Kitchen or Bathroom Vent Use (Yes vs No)* −0.96 −61.7 0.00 NS
Climate Control Device (Yes vs No) −1.07 −65.8 0.00 −0.612 −45.8 0.06
Fuel Category (Combination EG-CW vs EG) 0.81 125.3 0.01 0.543 72.2 0.073
Fuel Category (Solid fuel CW vs EG) 0.99 170.0 0.00 0.642 90.1 0.033
Smoke or Incense (Yes vs No) 0.94 156.9 0.02 NS
Ash Removed (Yes vs No) 0.45 56.7 0.08 NS
Environmental
Indoor Relative Humidity 0.04 3.7 0.03 0.031 3.1 0.029
Outdoor PM2.5 0.006 0.6 0.11 NS
Parking Area (Unpaved vs Paved-Gravel)* 0.74 109.0 0.01 NS
Open in a new tab
*

Excluded from multivariable regression due to collinearity (Dusty correlated with Climate Control Device, Fuel Category, and Ash Removed; Any Kitchen or Bathroom Vent Use correlated with Housing Type and Climate Control Device; Parking Area correlated with Housing Type, Climate Control Device and Fuel Category)

NS= Not Selected

For example, compared to EG heated homes, indoor PM2.5 concentrations were 170% higher in homes heated with solid fuels, and 125.3% higher in homes using a combination of EG and solid fuel. Additionally, use of a climate control unit was associated with a 65.8% reduction in indoor PM2.5 compared to homes without climate control units. Dustiness of the interior environment was positively associated with indoor concentrations. Compared to non-dusty homes, indoor concentrations of PM2.5 were 107.5% higher in dustier homes (Table 3). Compared to modern housing types, indoor concentrations were 110.7% and 39.5% higher in traditional and mobile homes, respectively.

Other household and behavioral variables moderately associated (p≤0.1) with LN-indoor PM2.5 during the winter heating season included removal of ash, number of exterior doors, height of the home, and outdoor PM2.5. Indoor concentrations were 56.7% higher in homes that removed ash generated from heating fuels during the 24-hour sampling period (Table 3). Additionally, for every 1-unit increase in the number of exterior doors, indoor PM2.5 decreased by 24.2%.

Dustiness of interior environment, use of kitchen or bathroom vents, and parking area surfaces were identified as collinear with other explanatory variables and were removed from inclusion into variable selection. Dustiness correlated with climate control unit, fuel category, and ash removal. Use of kitchen or bathroom vent was correlated with housing type and use of a climate control unit, whereas parking area was correlated with housing type, climate control unit and heating fuel type.

Multivariable Analysis: Heating Season

Results of the multivariable regression modeling (R2=0.45) included six important household and behavioral factors associated with indoor PM2.5 concentrations during the heating season in 65 homes with complete monitoring and questionnaire data (Table 3). Housing type, heating fuel type, presence of pests, and indoor relative humidity were all positively associated with indoor concentrations, whereas use of a climate control device and number of interior doors were negatively associated with indoor levels.

After adjusting for behavioral and household characteristics, compared to homes heated with EG fuel, indoor PM2.5 concentrations were 90.1% higher in homes burning solid fuels and 72.2% higher in homes heated with a combination of EG and solid fuel sources. Homes reporting the presence of a rodent during the 24-hour sampling period had 90.7% higher indoor concentrations compared to homes with no pests. Additionally, mobile homes had 127.4% higher indoor concentration of PM2.5 compared to modern homes. Indoor relative humidity was significantly associated with indoor heating PM concentrations. For every 1-percent increase in indoor relative humidity, indoor PM2.5 concentrations increased by 3.1%. Homes using a climate control device had 45.8% lower indoor concentrations compared to homes that did not use a climate control device, and concentrations decreased by 12.6% for every 1-unit increase in the number of interior doors.

Univariable Analysis: Non-Heating Season

Univariable regression identified 7 household and behavioral factors that were significantly associated with indoor PM2.5 concentrations in the non-heating season and were candidates for inclusion into multivariable regression (Table 4). Opening of doors and windows, the number of exterior windows, and the area and height of the home were negatively associated with indoor PM2.5 concentrations. For example, opening of doors and windows for ventilation was associated with a 61.2% decrease in indoor concentrations. Similarly, indoor PM2.5 decreased by 4.2% for every 1 foot (0.3048 m) increase in the height of the house.

Table 4:

Factors Associated with PM2.5 Concentrations Measured for Non-Heating Season - Univariable and Multivariable Regression Models

Univariable Regression Multivariable Regression
LN-PM2.5 Coef. Percent Change p-value Coef. Percent Change p-value
Household Characteristics
Area of Home −0.0004 −0.04 0.004 −0.00023 −0.02 0.042
Number of Exterior Windows −0.054 −5.3 0.079 NS
Height of Home −0.043 −4.2 0.140 −0.048 −4.7 0.038
Behavioral
Open Doors (Yes vs No) −0.946 −61.2 0.001 −0.947 −61.2 0.001
Smoke or Incense (Yes vs No) 0.862 136.7 0.009 0.783 118.8 0.004
Environmental
Outdoor PM2.5 0.035 3.5 0.003 0.026 2.6 0.005
Area of Home (Cleared-Scraped vs Natural Undisturbed) 0.493 63.7 0.045 0.361 43.5 0.059
Area of Home (Actively Used vs Natural Undisturbed) −0.092 −8.8 0.782 −0.121 −11.4 0.661
Open in a new tab

NS= Not Selected

Outdoor PM2.5 was positively associated with indoor concentrations. For every 1 μg/m3 increase in outdoor PM2.5, indoor PM2.5 increased by 3.5%. Indoor concentrations were 136.7% higher in homes that reported indoor smoking or burning of incense. The surrounding environment in which homes were located was also associated with indoor PM2.5. Compared to homes surrounded by natural vegetation, homes cleared of natural vegetation had 63.7% higher indoor concentrations (Table 4).

Multivariable Analysis: Non-Heating Season

Six of the seven candidate variables were selected into multivariable regression yielding a model with significant performance (R2=0.60). After adjusting for behavioral and household factors, there is an apparent increase in indoor PM2.5 as outdoor concentrations increase (Table 4). Indoor smoking and burning of incense remained the greatest contributor to indoor concentrations (118.8%). Compared to homes surrounded by natural vegetation, indoor PM2.5 concentrations were 43.5% higher in homes cleared of natural vegetation, despite a modest reduction in significance (p=0.059). Similar to univariable analysis, the area and height of homes were both negatively associated with indoor concentrations. Significant reduction of indoor PM2.5 was observed in homes opening doors and windows for ventilation (61.2%), compared to homes that kept doors and windows closed.

Discussion

Air pollution remains a significant public health hazard, especially indoor air pollutants from the burning of solid fuels. Relatively few studies assess rural ambient and indoor air quality in the United States, where populations continue to experience greater health disparities and disease burden (Rogalsky et al., 2014; Semmens et al., 2015). Our study explored this gap by quantifying household indoor concentrations of PM2.5 in a rural, solid fuel burning, Native American community in the United States, and examined household and behavioral factors associated with indoor concentrations during heating and non-heating seasons using each home as its own control.

We found elevated indoor PM2.5 in communities burning solid fuel for heating. Similar studies addressing use of solid fuel for both heating and cooking have reported greater concentrations. High indoor concentrations of PM2.5 were observed in 53 open fire wood burning homes in an indigenous rural Mexican community (Zuk et al., 2007). Pre-intervention 48-hour PM2.5 concentrations were measured 1 m from a woodstove (693 μg/m3), in kitchens 1.5 m from the woodstove (658 μg/m3), and outside the homes (94 μg/m3). Our study sampled homes with closed stoves, while the study by Zuk et al., (2007) evaluated homes using open fires which contributed to significantly higher indoor concentrations.

In rural China, Hu et al., (2014) measured 24-hour indoor PM2.5 in homes burning “smoky” bituminous coal (185μg/m3), “smokeless” anthracite coal (104 μg/m3), mixed coal (164 μg/m3), wood (393 μg/m3), plant material (417 μg/m3), and mixed fuel (210 μg/m3). Regardless of stove ventilation type, indoor PM2.5 concentrations were well in excess of the WHO indoor guideline and EPA ambient standards. These concentrations far exceeded indoor levels observed in our study, and may be due to differences in stove design, inadequate venting, coal type, and burning duration by participants. Regardless, solid fuel clearly contributed to increased concentrations of PM2.5 indoors.

Seasonal and household characteristics associated with indoor PM2.5 concentrations were examined in 2258 households in a low income urban community in Dhaka, Bangladesh (Gurley et al., 2013). Average 24-hour indoor PM2.5 concentration inside homes fueled only with biomass had the highest indoor concentrations (308 μg/m3). A reduction in 24-hour average concentration was observed among homes burning a combination of EG and biomass fuel (193 μg/m3) and homes cooking with only EG fuel (165 μg/m3). These values were much greater than those we found on Hopi land. Indoor concentrations were found to be greater during the winter season (225 μg/m3 increase) in households with 1 or fewer external windows or doors. This finding was similar to ours. Gurley et al. (2013) found that reduced ventilation results in higher winter concentrations. Air infiltration resulted in a 22 μg/m3 reduction in PM2.5 for every 1 unit increase in the number of external windows or doors. Homes in our study may be “tighter”, resulting in an increase of indoor PM2.5 among homes burning coal and wood.

Pokhrel et al., (2015) identified similar effects of biomass and solid fuel use on indoor PM2.5 concentration in Nepal. Twenty-four hour concentrations of PM2.5 were measured in kitchens of 824 homes of four primary cooking fuel types. Average concentrations of PM2.5 were 656 μg/m3 in biomass households (n=218), 169 μg/m3 in kerosene burning homes (n=187), 101 μg/m3 in propane gas burning homes (n=238), and 80 μg/m3 in electric powered homes (n=181). Indoor PM2.5 concentrations exhibited strong seasonal affects, with higher concentrations observed during winter months in biomass burning homes. Similar to the Hopi community, biomass may be burned more frequently in the winter, and windows and doors kept shut. However, the authors also suggested higher winter indoor concentrations may occur when brick kilns are in high operation (Pokhrel et al., 2015). Although the location of monitoring and fuel source differed from our study, similar increases in PM2.5 were observed inside and outside of coal and wood heated homes, demonstrating the adverse effects of these fuels on local ambient and indoor air quality.

Differences in the magnitude of indoor concentration between our study and other studies conducted outside the United States are due to differences in housing characteristics (including ventilation), ambient/local conditions, stove design, and most importantly fuel types (Naeher et al., 2007; Semmens et al., 2015; Hu et al., 2014). Prevalent use of low temperature combustion of biomass is one reason for significantly greater concentrations of indoor air pollutants observed in these studies. Furthermore, solid fuel was used only for seasonal heating in our study, as opposed to daily use for both heating and cooking among other international studies.

Many household studies conducted in the United States rely on self-reported use of wood stoves and other solid fuels (Morris et al., 1990; Browning et al., 1990; Honicky et al., 1991; Noonan & Ward, 2007; Noonan et al., 2012; Ware et al., 2014), and report inconsistent associations between woodstove use and respiratory health outcomes. Our study used trained Hopi Tribal technicians to collect all information and examine woodstoves. Few studies (Sexton et al., 1984; Leaderer et al., 1994; Ward et al., 2008; Ward et al., 2011; Ward et al., 2017; Semmens et al., 2015; Noonan et al., 2012; Robin et al., 1996; Bunnell et al., 2010; Weichenthal et al., 2013) in the United States have quantified indoor concentrations and exposures to PM2.5 in solid fuel heated homes, and the potential health effects of these exposures remains largely unknown among populations at the greatest risk. Sexton et al. (1984) quantified indoor concentrations of PM10 in rural Vermont (n=163) homes and reported 24-hour average concentrations of 24 μg/m3 in wood burning homes (n=116) compared to 18 μg/m3 in non-wood burning homes (n=28). We collected PM2.5 and report higher particulate concentrations.

Leaderer et al., (1994) measured weekly outdoor and indoor PM2.5 concentrations in wood-gas burning and non-wood burning suburban communities in Suffolk and Onondaga County, New York. In Suffolk County, weekly geometric mean outdoor concentration (n=19) of PM2.5 was 16.9 μg/m3, and indoor concentration in wood-gas burning homes (n=15) was 18.1 μg/m3 and 17.3 μg/m3 in non-burning homes (n=30). In Onondaga County, weekly geometric mean outdoor PM2.5 concentration (n=36) was 15.8 μg/m3, and indoor concentration in wood-gas burning (n=16) was 19.1 μg/m3 and 14.1 μg/m3 in non-burning (n=45) homes (Leaderer et al., 1994). Indoor concentrations in these studies were much lower compared to our study, and may be due to differences in household construction and environments (suburban vs rural), sampling duration and methodology, supplementing wood heating with gas heating, or possibly due to differences in demographic and socioeconomic factors in these communities. The authors note that the elemental composition of indoor PM2.5 in homes burning wood-gas was similar to profiles in non-burning homes, suggesting appliances in these homes were well maintained and did not leak PM2.5 into the indoor environment, which may be a function of household income.

More recently, indoor concentrations of PM2.5 were measured in a composite study of 96 rural wood heated homes in the Northwestern United States (Montana, Idaho, Alaska) as part of a larger randomized trial. During two winter pre-intervention sampling periods, concentration of PM2.5 inside wood heated homes was 28.8 μg/m3 and 29.1 μg/m3 during the first and second winter sampling periods, respectively (Semmens et al., 2015). Thirty percent of homes had concentrations of PM2.5 that exceeded the EPA standard and WHO guideline. Similar to our study, indoor relative humidity and housing type were associated with increased indoor PM2.5 concentrations. Semmens et al., (2015) found that every 1% increase in indoor relative humidity was associated with a 2% increase in indoor PM2.5, and compared to mobile homes and apartments, modern homes had 36% lower indoor PM2.5 concentrations. An important finding was the inverse relationship between household income and indoor PM2.5 concentrations, which may be attributed to properly installed and maintained stoves in higher income households. A similar relationship may exist in our findings, as low income households are unable to afford modern, well maintained, stoves when heating with solid fuels.

Heating stove changeout may be a viable approach for mitigating indoor air pollution inside solid fuel heated homes in rural and tribal communities. Sixteen wood burning homes in Libby, Montana had pre-intervention 24-hour average indoor PM2.5 concentrations of 51.2 μg/m3 and post-intervention concentrations of 15.0 μg/m3 (Ward et al., 2008). The efficacy of woodstove changeout interventions was further demonstrated in the rural Nez Perce tribal community in Idaho. Pre-stove intervention median 24-hour indoor concentrations of PM2.5 was 39.2 μg/m3 in 16 wood heated households. Post-intervention, indoor concentrations dropped to 19.0 μg/m3, suggesting the potential benefits of woodstove changeout interventions on reducing indoor air pollutants (Ward et al., 2011).

Exposure to HAPs is strongly linked with poverty (Rogalsky et al., 2014). Few studies have been completed in rural low-income solid fuel burning communities of the United States (Robin et al., 1996; Bunnell et al., 2010; Ward et al., 2011). Each of these studies conducted in tribal communities report similar results showing increased concentrations in solid fuel heated households. The Hopi Tribe, much like other rural solid fuel burning communities in the United States, has a high rate of poverty. Over one quarter (28%) of households make less than $20,000 per year (U.S. Census Bureau, 2010). Poverty rate on the Hopi Lands (35%) is more than double the poverty rate in the state of Arizona (15%) (U.S. Census Bureau, 2010). Similar to findings from Semmens et al., (2015), indoor concentrations of air pollutants may be inversely associated with household income. Household concentrations of PM2.5 reported in this study may be similar across other low-income solid fuel burning communities in the United States.

There are several limitations to our study. First, predictor variables and activities performed during the sampling period were self-reported to interviewers by participants and are prone to potential recall bias and thus misclassification error. Second, fuel heating type is unlikely to be the only contributing source of indoor PM2.5, and other behavioral sources (home furniture, frequency and duration of cooking/cleaning, social gathering, and traditional practices) may not have been accounted for in our analysis as the home was measured only one time per season. Additionally, the amount of fuel used during the sampling period and the number of times participants stoked and fed the fire was not measured. Such activities may have contributed to elevated indoor PM2.5. Further, the pDR-1500 monitors were not calibrated with gravimetric measurements in the Hopi homes. Instead, they were calibrated by the manufacturer using test dust that might differ from local, house dust.

Despite these limitations, this study makes strides in quantifying indoor concentrations of air pollutants in a rural, Native American tribal community where heating with solid fuels is widely practiced. Fuel use and behavioral factors were examined in relationship to indoor concentrations of PM2.5. After adjusting for these factors, we were able to quantify the magnitude of their association with PM2.5. These variables enabled us to gauge the representativeness of households in our study across Hopi Lands and communities that share similar behavioral and housing characteristics. We were able to demonstrate a clear seasonal difference in factors associated with indoor PM2.5. Information gathered from this study can inform the Tribe about household environmental hazards present within their community and help guide policy development and targeted interventions for reducing exposures to environmental contaminants.

Conclusion

Millions of households in the United States relying on solid fuels for heating, yet research on respiratory health effects from HAPs in low income rural communities continues to remain largely ignored. Native Americans experience greater respiratory morbidity; exposure to indoor air pollutants may be a contributing factor. We observed a seasonal difference in behavioral and household characteristics associated with indoor PM2.5 in Hopi households. Homes heated with solid fuels, or a combination of solid and cleaner burning fuels, had indoor PM2.5 concentrations that exceeded the EPA 24-hour ambient standard and the WHO guideline during the winter heating season. These results are similar to other studies across the globe that show indoor use of solid fuels without adequate ventilation is significantly associated with indoor air pollution. Occupants residing in these households may be exposed to hazardous levels of indoor air pollutants and subject to acute and chronic adverse health effects. Results from this study will help to address the current gap in the literature by highlighting the environmental hazards associated with solid fuel use in rural, low-income communities within the United States. Further effort is needed to quantify environmental concentrations of air pollutants and respiratory health outcomes in underserved rural and tribal communities throughout the United States.

Supplementary Material

1

Practical Implications.

Indoor air pollution associated with solid fuels has been researched extensively in developing nations, yet it has been largely ignored in low-income, rural, solid fuel burning communities in the United States and other developed countries.

As found in other studies across the globe, homes heated with solid fuels had significantly elevated indoor concentrations of PM2.5.

Knowledge of housing characteristics associated with exposures can inform residents and housing authorities of important factors that contribute to indoor air pollution, and help guide policy development and targeted interventions, thereby reducing exposures to environmental contaminants.

Behavioral factors that contributed to elevated indoor PM2.5, may be modified to reduce household occupants’ exposures to hazardous pollutants.

These findings call attention to the potential hazards facing many low-income, rural households in the United States that rely on solid fuels for heating.

Acknowledgments

This work was funded by NIEHS: P50ES026089, P30 ES006694, T32 ES007091, and EPA: R836151. “Research reported in this publication was supported by the National Institute of Environmental Health Sciences of the National Institutes of Health under Award Number P50ES026089. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.” and “This publication was developed under Assistance Agreements No. 836151 awarded by the U.S. Environmental Protection Agency to The University of Arizona. It has not been formally reviewed by EPA. The views expressed in this document are solely those of the authors and do not necessarily reflect those of the Agency. EPA does not endorse any products or commercial services mentioned in this publication.”

We express our deepest appreciation to the Hopi Tribe for considering the project and to all project participants and their families for welcoming us into their homes. We are extremely grateful to our Hopi community partners, members of the Hopi Community Advisory Board, and the incredible Community Health Representatives and Field Technicians that made this project possible (Delores Ami, Joyce Hamilton, Elyse Monroe, Rose Namoki, Beatrice Norton, Madeline Sahneyah, Gayl Honanie, Alfonso Mahkewa, Sehoy Talaswaima, Philbert Poseyesva, Del Marino).

Finally, we cannot begin to express our thanks and gratitude to all the undergraduate and graduate students on the HEHP team for the hard work and commitment to the success of this project (Connor Kelly, Jonathan Blohm, Julie Mattes, Joel Chavez, Graciela Olivas, Liora Fiksel, Paulina Colombo, Amanda Bates, Veronica Lugo, McKenna Nelson, Adam Carl).

Footnotes

Conflict of Interest Statement

The authors have no conflicts of interest to declare.

References

  1. Apte J, Marshall J, Cohen A, & Brauer M (2015). Addressing global mortality from ambient PM2.5. Environmental Science & Technology, 49(13), 8057–8066. doi: 10.1021/acs.est.5b01236 [DOI] [PubMed] [Google Scholar]
  2. Browning KG, Koenig JQ, Checkoway H, Larson TV, & Pierson WE (1990). A questionnaire study of respiratory health in areas of high and low ambient wood smoke pollution. Pediatric asthma, allergy & immunology, 4(3), 183–191 [Google Scholar]
  3. Bruce N, Perez-Padilla R, & Albalak R (2000). Indoor air pollution in developing countries: A major environmental and public health challenge. Bulletin of the World Health Organization, 78(9), 1078–1092. [PMC free article] [PubMed] [Google Scholar]
  4. Bulkow LR, Singleton RJ, DeByle C, Miernyk K, Redding G, Hummel KB, … Hennessy TW (2012). Risk factors for hospitalization with lower respiratory tract infections in children in rural alaska. Pediatrics, 129(5), e1220–e1227. doi: 10.1542/peds.2011-1943 [DOI] [PubMed] [Google Scholar]
  5. Bunnell JE, Garcia LV, Furst JM, Lerch H, Olea RA, Suitt SE, & Kolker A (2010). Navajo coal combustion and respiratory health near shiprock, new mexico. Journal of Environmental and Public Health, 2010, 1–14. doi: 10.1155/2010/260525 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Carson RT, Hanemann WM & Whittington D The Existence Value of a Distinctive Native American Culture: Survival of the Hopi Reservation. Environ Resource Econ 75, 931–951 (2020). 10.1007/s10640-020-00412-5 [DOI] [Google Scholar]
  7. Crimmins MA, Ferguson DB, Weiss JL, and Faulstich H. 2015. Hopi Climate: An Overview to Support Drought Monitoring and Management. Tucson, AZ: Climate Assessment for the Southwest. [Google Scholar]
  8. Gurley ES, Salje H, Homaira N, Ram PK, Haque R, Petri WA Jr, Bresee J, Moss WJ, Luby SP, Breysse P, & Azziz-Baumgartner E (2013). Seasonal concentrations and determinants of indoor particulate matter in a low-income community in Dhaka, Bangladesh. Environmental research, 121, 11–16. 10.1016/j.envres.2012.10.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Hadeed SJ, O’Rourke MK, Burgess JL, Harris RB, & Canales RA (2020). Imputation methods for addressing missing data in short-term monitoring of air pollutants. Science of The Total Environment, 730, 139140. 10.1016/j.scitotenv.2020.139140 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. He Y, Gu Z, Lu W, Zhang L, Okuda T, Fujioka K, Luo H, Yu CW (2019). Atmospheric humidity and particle charging state on agglomeration of aerosol particles. Atmospheric Environment, 197:141–149. [Google Scholar]
  11. Honicky RE, & Osborne JS 3rd (1991). Respiratory effects of wood heat: clinical observations and epidemiologic assessment. Environmental health perspectives, 95, 105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Hu Wei, et al. “Personal and indoor PM2. 5 exposure from burning solid fuels in vented and unvented stoves in a rural region of China with a high incidence of lung cancer.” Environmental science & technology 48.15 (2014): 8456–8464. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Joshweseoma L (2015). 2012 Hopi On-Reservation Health Survey. Annual Arizona Tribal Collaborative Conference. Retrieved from http://www.azdhs.gov/documents/prevention/health-systems-development/cancer-prevention-control/tribal-collaborative-conference/day-one-brfss-data-panel-hopi-lori-joshweseoma.pdf [Google Scholar]
  14. Leaderer BP, Koutrakis P, Briggs SL, & Rizzuto J (1994). The mass concentration and elemental composition of indoor aerosols in Suffolk and Onondaga Counties, New York. Indoor Air, 4(1), 23–34 [Google Scholar]
  15. Morris K, Morganlander M, Coulehan JL, Gahagen S, & Arena VC (1990). Wood-burning stoves and lower respiratory tract infection in american indian children. American Journal of Diseases of Children, 144(1), 105–108. doi: 10.1001/archpedi.1990.02150250117047 [DOI] [PubMed] [Google Scholar]
  16. Naeher LP, Brauer M, Lipsett M, Zelikoff JT, Simpson CD, Koenig JQ et al.. 2007. Woodsmoke health effects: a review. Inhal Toxicol19(1):67–10617127644. [DOI] [PubMed] [Google Scholar]
  17. Noonan CW, & Ward TJ (2007). Environmental tobacco smoke, woodstove heating and risk of asthma symptoms. Journal of Asthma, 44(9), 735–738. [DOI] [PubMed] [Google Scholar]
  18. 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(6), 492–500. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Pokhrel AK, Bates MN, Acharya J, Valentiner-Branth P, Chandyo RK, Shrestha PS, … & Smith KR (2015). PM2. 5 in household kitchens of Bhaktapur, Nepal, using four different cooking fuels. Atmospheric Environment, 113, 159–168. [Google Scholar]
  20. Robin LF, Lees PS, Winget M, Steinhoff M, Moulton LH, Santosham M, & Correa A (1996). Wood-burning stoves and lower respiratory illnesses in Navajo children. The Pediatric infectious disease journal, 15(10), 859–865. Chicago: [DOI] [PubMed] [Google Scholar]
  21. Rogalsky DK, Mendola P, Metts TA, & Martin 2, William J (2014). Estimating the number of low-income americans exposed to household air pollution from burning solid fuels. Environmental Health Perspectives, 122(8), 806. doi: 10.1289/ehp.1306709 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Semmens EO, Noonan CW, Allen RW, Weiler EC, & Ward TJ (2015). Indoor particulate matter in rural, wood stove heated homes. Environmental Research, 138, 93–100. 10.1016/j.envres.2015.02.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Sexton K, Spengler JD, & Treitman RD (1984). Effects of residential wood combustion on indoor air quality: a case study in Waterbury, Vermont. Atmospheric Environment (1967), 18(7), 1371–1383 [Google Scholar]
  24. U.S. Census Bureau (1995). Housing of American Indians on Reservations- Equipment and Fuels. Statistical Brief. [Google Scholar]
  25. U.S. Census Bureau (2010). Demographic Analysis of the Hopi Tribe Using 2010 Census and 2010 American Community Survey Estimates. Arizona Rural Policy Institute In Governor’s Office on Tribal Relations U.S. Census Bureau. [Google Scholar]
  26. U.S. Environmental Protection Agency. (2015). Tribal Agenda. In Office of Air & Radiation (OAR). Retrieved from https://www.epa.gov/sites/production/files/2016-02/documents/oar_tribal_air_agenda_final_2015.pdf [Google Scholar]
  27. Ward TJ, Semmens EO, Weiler E, Harrar S, & Noonan CW (2017). Efficacy of interventions targeting household air pollution from residential wood stoves. Journal of Exposure Science and Environmental Epidemiology, 27(1), 64–71. 10.1038/jes.2015.73 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Ward T, & Noonan C (2008). Results of a residential indoor PM2. 5 sampling program before and after a woodstove changeout. Indoor air, 18(5), 408–415 [DOI] [PubMed] [Google Scholar]
  29. 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. Science of the total environment, 409(4), 664–670 [DOI] [PubMed] [Google Scholar]
  30. Ware DN, Lewis J, Hopkins S, Boyer B, Montrose L, Noonan CW, … & Ward TJ (2014). Household reporting of childhood respiratory health and air pollution in rural Alaska Native communities. International journal of circumpolar health, 73 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Weichenthal S, Mallach G, Kulka R, Black A, Wheeler A, You H, … & Sharp D (2013). A randomized double-blind crossover study of indoor air filtration and acute changes in cardiorespiratory health in a First Nations community. Indoor Air, 23(3), 175–184. [DOI] [PubMed] [Google Scholar]
  32. Whiteley PN (1988). Deliberate acts: Changing Hopi culture through the Oraibi split. The University of Arizona Press, Tucson. pp 373. [Google Scholar]
  33. Winkler P (1988). The growth of atmospheric aerosol particles with relative humidity. Physica Scripta, 37:223–230. [Google Scholar]
  34. World Health Organization. (2016). Household air pollution and health. WHO Fact Sheet. [Google Scholar]
  35. Zuk M, Rojas L, Blanco S, Serrano P, Cruz J, Angeles F, Tzintzun G, Armendariz C, Edwards RD, Johnson M, Riojas-Rodriguez H and Masera O (2007) The impact of improved wood-burning stoves on fine particulate matter concentrations in rural Mexican homes, J Exposure Sci Environ Epidemiol, 17, 224–232. [DOI] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

1
NIHMS1719772-supplement-1.pdf (199.7KB, pdf)

RESOURCES