5
Health Effects of Exposure to Indoor Particulate Matter
The second day of the workshop focused on health, metrics, and assessment, and in his opening remarks Richard Corsi noted that there is overwhelming scientific evidence that increases in levels of outdoor fine particulate matter (PM2.5) are associated with a range of short-term and chronic health effects, including asthma exacerbation, acute and chronic bronchitis, heart attack, increased susceptibility to respiratory infections, and premature death. Moreover, the burden of these health effects is greater for underserved and marginalized communities, something he asked the participants to keep in mind when listening to the presentations and discussions. He then posed three questions that the day’s presentations would address:
- What do we know about the health effects of exposure to indoor fine particulate matter?
- How do we practically measure indoor fine particulate matter?
- What do the measurements mean?
The day’s first session featured three presentations by physicians on how exposure to indoor particulate matter can affect human health. Howard Kipen (Rutgers University School of Public Health) addressed the overall health burden associated with exposure to PM2.5, and Meredith McCormack (Johns Hopkins School of Medicine) discussed the link between indoor PM2.5 exposure and pulmonary disease, as well as disparities in economically challenged communities. Stephanie Holm (University of California, San Francisco) talked about the health effects of exposure to wildfire smoke
and other ambient air pollutants and how certain building characteristics might mitigate those adverse health effects. Elizabeth Matsui and Linda McCauley (Emory University’s Nell Hodgson Woodruff School of Nursing) comoderated an open discussion following the three presentations.
THE OVERALL (MOSTLY CARDIOVASCULAR) HEALTH BURDEN OF INDOOR PM2.5 EXPOSURE
Howard Kipen outlined his presentation by saying that he would talk about what he perceived to be the confusion between the health effects of indoor and outdoor air pollution—a confusion that results from the fact that outdoor particulate matter penetrates indoors. He began by displaying data from the Global Burden of Disease study showing that exposure to household air pollution from solid fuels (which occurs primarily in less and least developed countries) and ambient particulate matter pollution are two of the 10 top causes of morbidity, expressed as disability-adjusted life years (Figure 5-1) (Lim et al., 2012). The biggest components of the morbidity associated with ambient particulate matter exposure are cardiovascular disability, chronic and acute respiratory disease, and infection, and these and other health effects (such as developmental issues, blood clotting, and metabolic problems) associated with particulate matter in general and PM2.5 in particular are ubiquitous (Figure 5-2) (Thurston et al., 2017). He explained that because it is not always possible to study these clinical events, particularly when looking at their relationship to changes in air pollution over time, researchers tend to use biomarkers as proxies for both clinical events and exposures.
Taking an example from the PM10 literature (he asserted that PM2.5 would be similar) Kipen showed the results of an analysis of administrative datasets generated from death certificates attributing mortality to respiratory disease, cardiovascular disease, and other causes. That analysis showed that the estimated percentage change in day-to-day death rates associated with each 10 microgram per cubic meter (μg/m3) increase in PM10 is around 4 percent for respiratory disease, 2 percent for cardiovascular disease, and 1 percent for all causes of mortality. Kipen said that the takeaway message is that, while we know that the strength of the association between air pollution and respiratory disease is stronger, the vast burden of morbidity from air pollution is actually due to increased cardiovascular disease, largely because cardiovascular disease is so much more prevalent in adults living in the developed world.
It is likely that the mechanisms that link particulate matter and cardiovascular disease involve inflammation and oxidative stress that arise when particulate matter interacts with lung tissue and triggers physiological
events that can exacerbate atherosclerosis and, over time, lead to a heart attack (Figure 5-3) (Brook, 2008). Ultrafine particles that cross from the alveoli into the blood stream can also cause blood vessels to constrict and increase blood pressure.
Before the 2008 Beijing Olympics, Kipen participated in a study called the Health Effects of Air Pollution Reduction Trial. This trial involving healthy 20-year-olds living in Beijing tested the hypothesis that there would be biomarkers of systemic inflammation, endothelial dysfunction, increased coagulation, and autonomic dysfunction, as well as direct measures of oxidative stress, that would improve significantly during the period in which the Chinese government promised to reduce air pollution and that would then return to baseline in the post-Olympic period (Huang et al., 2012; Rich et al., 2012; Zhang et al., 2013).
During the “clean air” period, levels of all the pollutants measured, including PM2.5, dropped significantly except for ozone, which naturally increases when nitric oxide levels fall. At the same time, biomarkers of pulmonary inflammation and oxidative stress, systemic inflammation and oxidative stress, asthma, coagulation, and autonomic tone, which affects heart rate and blood pressure, all improved. Unfortunately, when pollution levels increased once the Olympics ended, the biomarkers returned to where they
were before the clean air period. Kipen noted that this study did not measure heart attacks, atherosclerosis, or other clinical disease, but rather the biochemistry and cell biology that underlie those chronic diseases. “Most importantly, we saw that all of these changes were reversible in healthy young people,” said Kipen. “So all of the health outcomes associated with exposure to PM2.5, whether indoors or outdoors, might be reversible.”
In another study, Kipen and his collaborators used a device that measures pressure inside the heart as a marker of heart failure to see if there was physiologic change associated with a 10 μg/m3 change in ambient particulate matter. On days when there were increases in hospital admissions related to heart failure associated with increased ambient particulate matter, there was a corresponding increase in the pressure in the right ventricle of the heart. In essence, this work identified a biomarker for heart failure (Rich et al., 2008; Wellenius et al., 2006).
Epidemiology of Exposure to Indoor Air
Switching gears, Kipen said that the epidemiology of indoor exposure to air pollutants—including PM2.5 and other constituents—has been done but isn’t generally acknowledged because people are largely exposed to “outdoor” pollution when they are indoors, as the speakers in the workshop’s first session described. “It turns out that indoor exposures to PM2.5 probably account for 40 to 60 percent of total mortality attributed to PM2.5 because even though we are measuring the pollution outdoors, that is not where we are,” said Kipen. Based on that idea, he calculated that 71 percent of the average adult’s dose of PM2.5 comes from particles that were outdoors and infiltrated the indoor space.
After reviewing 16 studies that used an indoor air cleaner to lower household levels of PM2.5, Kipen found that the air cleaners lowered indoor PM2.5 levels between 40 percent and 82 percent but did not eliminate them. These studies also found that reduced indoor PM2.5 levels correlated with reductions in systolic blood pressure and increases in peak blood flow, as well as nonsignificant reductions in endothelial function and inflammatory markers. While the potential for this type of research exists, one problem is that different researchers report tracking different biomarkers, making comparisons and meta-analyses less useful than they could be, he noted.
Kipen then discussed a study from 2005 in which his team exposed (with Institutional Review Board approval) 130 healthy, nonsmoking women in a test chamber to a mixture of VOCs and ozone similar to what might be found in so-called sick buildings (Laumbach et al., 2005). While the number of ultrafine particles rose significantly and then fell as they coalesced and formed PM2.5 over time, there were no significant differences in nasal or upper airway symptoms or markers of nasal inflammation. Kipen concluded that VOCs and their oxidation products may not cause acute nasal effects at low concentrations, a finding similar to one in which laboratory rats were exposed to limonene and ozone (Sunil et al., 2007). The animal study did find some subclinical inflammatory changes in the lung that might be important in long-term exposure to indoor PM2.5.
A more recent study found that household air cleaners can decrease the oxidative potential of indoor PM2.5 when the data were normalized by volume, but not by mass (Brehmer et al., 2020). However, personal air monitor data failed to show any improvement in oxidative potential associated with air cleaning. Kipen noted that most of these studies have involved small numbers of individuals, which could explain the variable results.
On a final note, Kipen mentioned that he and his collaborators were now examining SARS-CoV-2 and how the virus may be associated with particles inside homes.
PULMONARY DISEASE ASSOCIATED WITH FINE PARTICULATE MATTER EXPOSURE IN INDOOR ENVIRONMENTS AND DISPARITIES IN ECONOMICALLY CHALLENGED COMMUNITIES
One of the unique characteristics of PM2.5, said Meredith McCormack, is its ability to penetrate deep into the lungs and reach all the way to the alveoli, and the smallest components of PM2.5, the ultrafine particles, can even translocate into the blood stream. As Kipen noted, it is possible to extrapolate from what is known about outdoor air quality to draw some inferences about the respiratory health effects of indoor PM2.5, particularly on lung development, said McCormack.
The human lung continues to develop from birth until the early 20s, and during that time the alveoli increase exponentially and are susceptible to the health effects of air pollution, as has been shown with research on outdoor air exposures. In fact, studies have demonstrated that exposure to PM2.5 affects the trajectory of lung development (Horak et al., 2002; Lee et al., 2010) and is associated with an accelerated decline in lung function with age (Doiron et al., 2019; Guo et al., 2018), McCormack explained. Conversely, improvements in air quality can improve optimal lung growth and attenuate the decline associated with normal aging.
Lung function is known to affect the risk of developing lung disease, and low lung function during early life is a risk factor for developing asthma during childhood and chronic obstructive pulmonary disease (COPD) later in life. In a study McCormack and her collaborators conducted in Baltimore City—where many people live in row houses with shared walls and in close proximity to a road—results showed that indoor PM2.5 levels in children’s bedrooms1 were higher than outdoor levels and exceeded the annual outdoor limit for exposure set by the Environmental Protection Agency in three-quarters of the homes studied (McCormack et al., 2009). When compared to the results obtained from suburban homes, it was clear that inner-city children were exposed to much higher levels of indoor PM2.5. Moreover, there was a consistently positive association between elevated PM2.5 levels and asthma, other respiratory issues, and the need for children to use rescue medication (Figure 5-4).
In addition to spending time at home, children spend a good portion of their days in school environments—6–10 hours each school day. Few studies, however, have directly assessed school environmental conditions, in part because of the complexity of indoor air quality studies, said McCormack. She and her collaborators in Baltimore (Majd et al., 2019), as well as another group in Boston (Phipatanakul et al., 2011), have studied
___________________
1 In-home levels were measured in the subject child’s bedroom because that was the location where the child was expected to spend a substantial portion of their time indoors.
air quality in inner-city schools. Both groups, said McCormack, found that the average air quality was within what she termed “reasonable limits” in the 30 schools they each studied, with an average PM2.5 level of about 5 μg/m3. The Boston study, she added, found that outdoor factors contributed to indoor air quality, as did activities in the indoor spaces.
In Baltimore, about half the schools studied subsequently underwent a major renovation, and McCormack’s team found that improvement in school facilities was associated with improvement in air quality and reduction in PM2.5 concentrations. “So even though the baseline level was within a reasonable range, that modernization of school facilities led to even better air quality,” said McCormack.
Indoor Air Pollution and COPD
Poor indoor air quality is a recognized contributor to COPD, particularly in low- and middle-income countries where indoor cooking using biomass fuels produces elevated levels of indoor PM2.5. McCormack noted, though, that in many parts of the United States solid fuels (wood, coal, coke) are the primary heating source, accounting for more than 2.5 million households and 6.5 million people overall (Figure 5-5) (Rogalsky et al., 2014). Some 500,000 to 600,000 people who use solid fuels as their primary heating source live in households below the federal poverty line.
McCormack noted that several studies have demonstrated a relationship between solid fuel use for heating or cooking and COPD, as well as respiratory disease, even among people who have never smoked tobacco (Barry et al., 2010; Raju et al., 2019a,b). Other studies have found that even in homes where PM2.5 levels were fairly low, around 11 μg/m3, indoor PM2.5 was associated with increases in severe COPD exacerbations, respiratory symptoms, and rescue medication use among former smokers (Hansel et al., 2013). During warm months, there was a synergistic effect of indoor heat and PM2.5 exposure that increased COPD exacerbations, respiratory symptoms, and rescue inhaler use (McCormack et al., 2016).
Susceptibility Factors and Interventions
To try to understand whether there are distinct individual factors that might increase susceptibility to the adverse health effects of PM2.5, McCormack collaborated in a study led by her colleague Elizabeth Matsui in which they looked at the association between asthma-related symptoms and weight in 150 children. The results showed that overweight and obese children had increased nocturnal and exercise-related symptoms and that there was a synergistic effect of body mass and PM2.5 levels on asthma symptoms (Lu
et al., 2013). The investigators found a similar trend when they examined the effect of exposure to secondhand smoke on asthma symptoms.
More recently, McCormack and her collaborators were able to measure tidal volume2 in children with asthma and found an association between increasing body mass and increasing tidal volume. Modeling results showed that body mass was associated with increased particulate matter deposition rate. “Potentially, this is a characteristic important among children with obesity,” said McCormack, who added that she has seen the same association in adults with COPD (McCormack et al., 2015), suggesting that obese individuals have a greater adverse response to particulate pollution.
Interventions can reduce PM2.5 levels. One study, for example, found that an air cleaner intervention in homes with an adult smoker reduced indoor PM2.5 levels by 50 percent and improved symptoms and increased the number of symptom-free days by 14 to 18 percent in children with asthma (Butz et al., 2011). McCormack noted that a similar effect size was seen in a trial with adult former smokers who have COPD (Sampson and Holgate, 1998).
___________________
2 Tidal volume is the amount of air that moves in or out of the lungs with each respiratory cycle.
In conclusion, McCormack noted four primary implications from the studies she spoke about:
- Indoor particulate pollution is a contributor to health disparities.
- Indoor particulate pollution is associated with increased asthma and COPD morbidity.
- Obesity may increase susceptibility to particulate pollution health effects.
- Improving indoor air quality represents a therapeutic target to modify disease activity.
WILDFIRE SMOKE AND OTHER AMBIENT AIR POLLUTION COME INDOORS: HEALTH EFFECTS AND THE BUILDING CHARACTERISTICS THAT MITIGATE THEM
Wildfire events can expose large populations to smoke, over regions that extend far beyond the fire, said Stephanie Holm. “Even if you are not located close to a wildfire smoke event, you can still be affected by smoke from that event,” said Holm. She noted that, while seeing smoke is a cause for concern, visibility alone isn’t a reliable indicator of potential risk because conditions such as fog and rain can affect it. “So just because you can’t see smoke, if the air quality index is bad you should be concerned,” she explained. She added that, though a common perception is that California has the biggest concerns about wildfires, the National Oceanographic and Atmospheric Administration estimates that the risk of very large fires in the mid-21st century, compared to the end of the 20th century, extends to large portions of the United States (Figure 5-6).
As Kipen and others had noted throughout the workshop sessions, roughly half of the outdoor PM2.5 comes indoors (Azimi and Stephens, 2020). Moreover, in terms of an individual’s total exposure to PM2.5, roughly 53 percent is outdoor PM2.5 that people are exposed to in their home or other buildings. Holm observed that many air pollution epidemiologic studies use residential location as a proxy for exposure to outdoor ambient air pollution, although some of that exposure is indoors. For example, a 2017 Medicare study with 460 million person-years of follow-up found that a 10 μg/m3 increase in PM2.5 was associated with an approximately 7 percent increase in mortality (Di et al., 2017). The Children’s Health Study in Los Angeles, which followed three cohorts involving more than 2000 adolescents, found that a 12.6 μg/m3 decrease in PM2.5 was associated with a significant increase in lung function (Gauderman et al., 2015). Conversely, the Imaging Dementia—Evidence for Amyloid Scanning (IDEAS) Study found that a 10 μg/m3 increase in PM2.5 increased the odds
of finding amyloid (which is thought to play a role in Alzheimer’s disease) during a positron emission tomography brain scan (Iaccarino et al., 2021).
After reviewing how infiltration, mechanical ventilation, and natural ventilation determine how much outside PM2.5 comes inside, Holm noted that while it might be tempting to think that a solution is to close up these three avenues into a home, doing so could lead to exposure problems from indoor-generated PM2.5 (as well as moisture and mold). The key to addressing this problem is filtration using a filter with a minimum efficiency reporting value (MERV) rating of 13 or higher for central heating and air conditioning systems. For portable air cleaning systems, the unit needs a high enough clean air delivery rate to account for the size of the home and should not produce ions or ozone.
Interventions and Health Effects
Studies have shown that weatherization that decreases infiltration, combined with ventilation and education on intervention strategies, can
improve asthma symptoms and overall self-reported health (Breysse et al., 2011, 2014). In contrast, decreasing infiltration without improving ventilation or filtration can actually increase indoor PM2.5 exposure (Yuan et al., 2018) and might increase the rate of severe asthma events (Fabian et al., 2014). In schools, improved ventilation has been associated with fewer absences (Mendell et al., 2013) and improvements in math and language performance (Wargocki and Wyon, 2007). “When thinking about improving exposure to PM2.5 from outdoor sources, ventilation needs to be accompanied by filtration,” Holm added, but she cautioned that ion generation technologies may negate the benefits of filtration (Allen and Barn, 2020) and suggested sticking to mechanical air filtration.
Several studies have shown that improved filtration yields health benefits. For example, it is associated with an increase in birth weight among full-term infants (Barn et al., 2018). In children, improved filtration has been associated with a decrease in asthma morbidity (Martenies and Batterman, 2018), but does not decrease airborne nicotine exposure if there are smokers in the house, which may explain why filtration has fewer benefits in such households (Gold et al., 2017).
A few studies have looked at interventions to reduce exposure to wildfire-generated PM2.5. A study in Australia, for example, showed that tightening the building envelope reduced peak PM2.5 levels by as much as 76 percent (Reisen et al., 2019), and a study in Southern California found that running a home’s heating and air conditioning system produced marked reductions in indoor wildfire-generated PM2.5, particularly when running continuously with filters rated MERV-9 or higher (Fisk and Chan, 2017). Studies have also found portable air cleaners to produce substantial decreases in indoor PM2.5 during wildfire events (Barn et al., 2008; Henderson et al., 2005; Stauffer et al., 2020). Holm noted, however, that total PM2.5 infiltration may be higher during wildfire smoke events as a result of filters getting saturated and because there may be more ultrafine particulate matter that filtration captures less efficiently (Mendoza et al., 2021).
With regard to the effect of interventions on health outcomes during wildfire events, a modeling study found that MERV-13 filtration and a portable air cleaner installed in every home in Southern California during the 2003 wildfire season could have prevented more than half of the hospital admissions due to respiratory issues (Fisk and Chan, 2017). Another study found that filtration was the only intervention that benefited health among residents of the Hoopa Valley National Indian Reservation in northwestern California (Mott et al., 2002); mask wearing and leaving the area did not confer the same health benefits.
Holm closed by noting that the Western States Pediatric Environmental Health Specialty Unit, where she is codirector, has produced a number of educational resources on reducing wildfire smoke exposures for children.3
DISCUSSION
Matsui opened the discussion by asking the speakers to identify important research gaps that need to be filled. Kipen replied that because biomarkers may not always be suitable for studying health effects, there is a need to identify markers related to “real” endpoints that have never been measured. He noted that the National Institute of Environmental Health Sciences may have some requests for application coming soon in this area. McCormack noted that indoor air studies can be challenging because of the intensity of the study protocol, which leads to smaller sample sizes that make it difficult to identify multisystem effects that may be downstream from pollution exposure. She would like to see studies that look at more than one outcome in at-risk populations and longitudinal studies that measure downstream consequences of exposure to particulate matter. Holm agreed with all of these suggestions and added that she is concerned that there is not sufficient characterization of the difference in exposure profiles for indoor versus outdoor particulate matter and the effects on health in the broader population. She noted that there are gaps in understanding the effects of short-term spikes in PM2.5 exposure as opposed to levels that are averaged out over time. The recent availability of low-cost exposure monitoring technologies may help.
Comoderator Linda McCauley, impressed with the data on intrusion of outdoor particulate matter into the indoor environment, said that she would like to see public health messages reflecting such data. She noted that the vast majority of public health messages today concern when it is safe for outdoor play and recreation. Many people, however, do not believe they have an indoor air quality issue. Holm agreed with McCauley, noting that the natural response in schools is to keep children inside when the outdoor air quality is poor without considering the school’s indoor air quality. She reported that Washington State is looking at what the messaging should be regarding indoor and outdoor levels, and that the EPA and several Pediatric Environmental Health Specialty Units4 are collaborating on a series of
___________________
3 These resources are available at https://wspehsu.ucsf.edu/projects/wildfires-and-childrens-health-2/.
4 The Pediatric Environmental Health Specialty Units are a national network of experts in the prevention, diagnosis, management, and treatment of health issues that arise from environmental exposures from preconception through adolescence. Additional information is available at https://www.pehsu.net/.
workshops on how to message during wildfire smoke events considering indoor versus outdoor exposures.
McCormack added that the EPA’s Tools for Schools5 and its anti-bus-idling campaign are good examples of a focus on the contribution of outdoor sources to indoor PM2.5 levels and making actionable recommendations. She also suggested that the COVID-19 pandemic has provided a unique opportunity to educate the public about the importance of indoor air quality. Kipen observed that an additional consideration was the possibility that residences that were tightly sealed to prevent infiltration could be subject to increases in exposure to other indoor contaminants, which may result in their own health problems.
Responding to a question about ultrafine particulate matter, McCormack noted that new technologies are now making it possible to study levels and health effects of ultrafine particulate matter and enhancing the ability of air cleaners to reduce exposure to it.
Jeff Siegel noted that studies on interventions have not focused on how variations in their performance—for example, in how an air cleaner performs under different conditions or over time—affect the data being generated. McCormack agreed that this is a challenge, and Kipen added that indoor studies have also been too short to truly judge the effectiveness of the interventions being tested.
Marina Vance asked the panelists for their thoughts on what a framework to assess the health effects from indoor sources of particulate matter would look like and if there would ever be a place for a guideline for indoor PM2.5. Holm endorsed such a guideline, and maybe even a standard; from her perspective of working with schools, she finds it hard to advise them on steps to take because there is no guideline or standard for indoor PM2.5 exposure. McCormack agreed and supported conducting the research needed to develop a standard and put one in place. Schools, she said, might be a good place to start that work. McCauley followed up by asking if systematic monitoring of indoor levels might be a part of such an effort and McCormack replied that that would be useful and that newer technologies made it possible.
Both McCormack and Holm noted there is an environmental justice component to this, given that public health officials during a smoke event may tell schools to send their students home, but that assumes that the indoor air quality at home will be better than at the school. “That is especially unlikely to be true for our lower-resourced families,” said Holm, “so we are exacerbating preexisting disparities for those children.” In her opinion, the way to improve access to good-quality indoor air for all children is to ensure that schools have clean indoor air.
___________________
This page intentionally left blank.