8

Interventions and Risk Communication

The workshop’s final session featured three presentations on risk communication. George Gray of Milken Institute School of Public Health at the George Washington University discussed the challenge of communicating the risk of indoor PM exposure. William Hallman of Rutgers University talked about the lessons learned from communicating about other health risks. Lee Ann Kahlor spoke about what could be learned from a benchmark study to gauge public understanding and information seeking related to indoor PM risk. An open discussion moderated by William Nazaroff concluded the workshop.

THE CHALLENGE OF COMMUNICATING INDOOR PM RISK¹

Two important unanswered questions regarding indoor PM are what to tell people about the associated risks and how to tell them, George Gray said. “This is particularly important because persuasion is the only thing we can do, as there are no regulatory avenues into the indoor environment,” he explained. “What we have to do is help people understand the risks and make decisions.” This challenge is compounded, he added, by the lack of data definitely linking indoor PM exposures to specific health risks.

___________________

¹ This section is based on the presentation by George Gray, a professor of environmental and occupational health at Milken Institute School of Public Health at the George Washington University, and the statements are not endorsed or verified by the National Academies of Sciences, Engineering, and Medicine.

One of the first things the scientific community needs to understand with regard to explaining risk, Gray said, is how the public feels about risk because that attitude plays an important role in the way the public receives information (IOM, 2001). Attitude is shaped by a number of factors, including the magnitude of the risk, whether the risk outcome is fatal or not fatal, if the risk arises from a voluntary or involuntary behavior, if it is controllable or uncontrollable, if it is manmade or natural, whether it is observable or not, and if there is any benefit associated with the risk (see Figure 8-1). These factors, Gray said, have been identified over the past 20 years or so from research in the field of cognitive psychology (Fischhoff et al., 1993; Slovic et al., 2004). As an example of how benefits can affect decisions about risk, he observed that people choose to drive even though they know it is inherently risky because there is a tangible benefit.

Gray believes that getting the public to grasp the risk of indoor PM exposure is going to be difficult because many of the most important sources are familiar, voluntary, and linked to benefits. Cooking, for example, is something that people do every day, and it produces food they enjoy eating. Burning candles or incense is voluntary and makes the house smell good or may be part of a cultural or religious experience that is important to an individual.

There are, however, perception factors that might help with communicating risk. Children, for example, could be at risk as could a developing fetus, and those risks are controllable, sometimes through relatively simple and even inexpensive actions. Some of the important sources of adverse exposures are already areas of at least some concern among the public, such as secondhand tobacco smoke and outdoor PM. A 20-year-old study (Slovic et al., 1995) found that 80 percent of the Canadian public—as well as 80 percent of toxicologists—said that indoor air quality posed a slight, moderate, or high risk, suggesting that the public (at least in Canada) is aware that there may be a health risk associated with exposure to indoor pollutants.

Among the other challenges that the scientific community will face in conveying the risk of indoor PM to the public is the need to understand the magnitude of the risk before deciding how much to communicate about it, Gray said. “Is this something that everybody needs to know about?” he asked, noting the limited bandwidth that people have for thinking about risks and taking action to mitigate them. He suggested that comparing a risk to the risk associated with environmental tobacco smoke might be an approach that would get the public’s attention, while comparing the risk of indoor PM exposure to the risks of cooking on a gas stove would lead the public to downplay the risks of indoor PM exposure.

It will be important too, Gray said, to communicate the relative contributions of different sources, particularly indoor versus outdoor, and how much of the exposure to these sources is under the individual’s control.

“One of the things that people want to know if you are going to talk to them about risk is, What can I do?” he said. “Don’t just tell me to be scared—tell me that there is something I can do.” He added that mitigation approaches that require changes in personal behavior are hard to sell, either because people benefit from those behaviors or do not feel the risk is big enough to warrant changing those behaviors. Another challenge, Gray said, will involve sorting out the risk tradeoffs, such as ventilation versus energy use, and explaining those to the public in a way that does not cause them to take actions that actually increase risk.

There are examples of success from which the indoor PM field can learn and borrow. Indoor radon is a close analog, Gray said, and it is an area where effective risk communication has over time changed perception and the actions that the public takes to mitigate that risk. One difference is that radon exposure has no known benefits associated with it. Secondhand smoke is an area in which there has been substantial behavioral change over the past 20 years, and there are similarities to indoor PM with regard to the risk to children and the voluntary nature of the activity. There is, however, a different level of awareness, concern, and dread associated with tobacco smoke, Gray said. The public has become more aware of the risks associated with indoor pesticide use, and risk communication there has focused on how to minimize the risks from exposure, particularly to children. Indoor pesticide use is voluntary and does have benefits, making it analogous to indoor PM, but it also comes with the connotation that, by default, synthetic pesticides are bad. There may also be lessons that can be applied from the evidence presented in various studies of indoor air quality in the developing world relating largely to indoor activities such as cooking with wood and other forms of biomass.

In closing, Gray said that if the field is to effectively communicate the risk of indoor PM, it will need the research disciplines represented at this workshop to produce data on the magnitude of the risks from exposure to indoor PM. But success will also depend on developing a better understanding of how the public thinks and feels about these risks, he said. “Maybe we can borrow information from similar situations to help guide us think about how we can be effective in communicating this information if and when we decide it is the thing to do.”

EMPOWERING PEOPLE TO REDUCE INDOOR PM EXPOSURES²

Risk communication is integral to the risk analysis process, William Hallman said, and it also involves more than just determining what to tell people. “It is also about asking the right questions, which requires being very clear about our particular goals,” he explained. Citing a 1989 report from the National Research Council, Hallman listed three essential goals for risk communication: education, advocacy or persuasion, and fostering partnerships for decision making. All three of these goals, he added, are implicit outcomes with regard to indoor PM.

Education aims to provide information and context so that people can choose what they believe is the right action to take in the face of risk. Advocacy or persuasion, in contrast, tries to change beliefs, attitudes, and behaviors in order to convince people to adopt a particular position: to take or not take an action. The goal of fostering partnerships, Hallman said, is to collect and discuss information with stakeholders in a way that leads to better collective decisions. He stressed the importance of involving stakeholders, manufacturers, and regulators early in this process to get their perspectives on how to deal with the issues related to indoor PM. Hallman added that simply alarming people is ethically problematic. “We have to tell them not just what the problem is, but what to do about the problem,” he said.

Interventions, Hallman said, can be focused on changing technology, behavior, policy, or regulations. With regard to technology, it is important to determine who has the responsibility and resources to make the necessary technology choices, an issue that was raised in earlier discussions concerning disadvantaged communities and individuals. One solution for these communities—where the ability to get property owners to make changes to HVAC systems is limited, for example—might be to provide incentives or tax rebates for the purchase of portable HEPA filtration units.

The invisibility of PM matters, Hallman said, and it represents a key issue in communicating the risk of indoor PM exposure. One approach to discussing the risks, he said, is to use how people feel when other impose risks on them. “You can harness strong feelings to try to get fairly rapid changes in policies and regulations, and that is exactly what happened with secondhand smoke,” he said. In contrast, when individuals are responsible for their own exposures, it is easy to ignore them, particularly when the problem is invisible. Radon is the prototypical example of the latter, and Hallman described the work that he and his colleagues did with EPA in the

___________________

² This section is based on the presentation by William Hallman, a professor and the chair of the Department of Human Ecology at Rutgers, the State University of New Jersey, and the statements are not endorsed or verified by the National Academies of Sciences, Engineering, and Medicine.

early 1990s trying without much success to persuade people to spend $7 on a carbon canister to test their homes for radon.

When a threat is invisible, people look for proxy indicators, such as smoke detectors and carbon monoxide detectors. In the area of food safety, which Hallman has studied, people use their sense of smell and taste as indicators, or else they look for visual indicators, such as mold. The problem with this particular mental model of food spoilage is that the bacteria that cause spoilage are not necessarily pathogenic. Hallman has found that the public perceives that a clean-looking home is a germ-free home (Hallman, 2008) and that the presence of mold, dust, filth, and bad smells are good indicators of germs being present. “People are using proxies of things they can see to represent things they cannot see,” Hallman said, “so one question we need to ask is what are the proxies that people will use or are using for indoor PM,” given that PM may have no taste or smell.

Hallman noted that while the size of PM is an important concern to the scientific community, he believes that this will be of little concern to the public and that it is not the place to start a communication effort. He also said that the lack of a feedback loop matters and that it can be difficult for people to make the link between an invisible cause and a later, visible effect. He again used food safety as an example. People do not recognize the symptoms of foodborne illness, and few people believe they have made themselves or other sick because the most common mental model is that eating contaminated food will make one sick immediately. “As a result, they do not connect their poor food safety practices with ultimate illness,” Hallman said. “I suspect we have a similar issue here, although we need the research to show that.”

Hallman said if there was one key message he wanted to convey at this workshop, it was that mental models matter. While it is important to know what people know about an issue and what they want to know about an issue, it is more important to know how they think about an issue. Certainly, he said, answering the questions that people are most interested in makes it more likely they will be willing to hear more of the information the scientific community wants to convey. “But I would also argue as a psychologist,” he said, “that it is important to understand how people think about an issue, how they construct a ‘big picture’ or world view about an issue.” In that context, Hallman said, the scientific community needs to realize how much most Americans overestimate their knowledge and understanding of science and technology. “Only 20 percent of Americans rate their understanding of science as poor, but ask them any science question and most will get it wrong.” In fact, Hallman added, Americans often exhibit false fluency, where they may have the right vocabulary but not the right constructs. The lesson here, he said, is that when conducting social science research, people in the field needs to

be careful to ask the right questions phrased in a way that will yield both qualitative and quantitative data on how people think about indoor PM, not just the facts they know about it.

The issue with mental models, Hallman said, is that people try to make sense of the world by creating meaning based on whatever information they have available and filling any gaps with misinformation or by inventing information. This is a particular problem in the world of the Internet and social media, he said. “A quick search on indoor PM issues suggests there is already bad information available on indoor PM.” Hallman also explained that people use analogies and metaphors to organize their mental models, and he recounted some work that he did for CDC on how the public thinks the immune system works. Without any prompting, the most common explanation people gave used a military metaphor with “good guy” blood cells engaging in a war against “bad guy” germs or cancer. Then, when asked how vaccines work, the most common answer was that a vaccine adds more troops to the germ fighters. “It is like the cavalry riding in, which makes absolute sense, except that it is absolutely wrong,” Hallman said, adding that his response as a risk communicator was to stick with the military metaphor but explain that vaccination helps the germ fighters recognize the enemy and respond more quickly to them. “We need to identify the metaphors we can use that will get into people’s heads relatively quickly rather than trying to give them a lecture on PM_2.5,” he said.

Acknowledging that he lacked data because no mental modeling research appears to exist regarding indoor PM_2.5, Hallman said that his guess is that the mental models the public might use to understand indoor PM would include the following:

• “Bad air” triggers asthma attacks, and since nobody in my family has asthma, I do not have a problem.
• “Air pollution” is “outside,” and I keep my windows closed, so I do not have a problem.
• “Air pollution” is caused by cars, factories, and power plants that produce “smoke.”
• Sources of “indoor air pollution” have to do with combustion and smoke, and because nobody in my house smokes, I do not have a problem.
• Combustion is associated with open flames and not electricity, and I have an electric stove, so I do not have a problem.
• My cooking produces pleasant aromas, not harmful PM, so I do not have a problem.
• Candles produce small flames and so are not a significant source of PM, so I do not have a problem.

• Water “washes things” and is clean, so my water is not a source of PM.
• Air “fresheners” clean the air, so I do not have a problem.
• “Particulate matter” means “dust,” and I vacuum twice per week, so I do not have a problem.
• Why is a filter rated 16 better than a filter rated 1? Isn’t being number 1 the best?

Hallman also noted the importance of advertising claims to the formation of mental models and the need to look at the information being passed to consumers through advertisements for air cleaning products and services. He said that a quick search yields bad information associated with these products, so the field will need to think about how to counter this misinformation and do so in a way that provides useful holistic advice. Simply warning people does not work, Hallman said, but telling them what to do and helping them do it by providing practical, effective, and affordable advice can work. Consistency and dependability will be key to avoiding the problem that the field of nutrition now has—i.e., as a result of ever-changing dietary recommendations, many Americans lack trust in nutritional advice. Hallman stressed the need to engage stakeholders—including partner organizations involved in science, health, and engineering, as well as appliance manufacturers and professionals in the building trades—to develop consistent, credible messages for consumers.

Hallman concluded his presentation with a list of what people will likely want to know about indoor PM:

• What causes indoor air pollution?
• Am I affected, how will I know, and who else is vulnerable?
• What are the immediate and long-term consequences?
• Does this explain my health issues?
• Can I do anything about it? Do I know what to do? Do I have what I need? Can I do it by myself? Where should I start?
• Who is responsible for the problem, and can it be prevented?
• Who will solve the problem, how long will it take, how effective will the solution be, who will pay for it, and how expensive will it be?
• How will I know that the problem has been solved, and can I trust that it has been solved?

Hallman said that not having all of the answers to these questions does not preclude engaging in risk communication today, particularly given that this conversation about exposure to particulate matter in indoor air is already occurring. His prescription for communicating information in the

face of uncertainty is to acknowledge that uncertainty and say, “This is what we know and do not know now. This is why it matters, this is what we are doing to become more certain, and this is what you can do while we continue to work on this particular problem.”

WHAT COULD BE LEARNED FROM A BENCHMARK STUDY³

In the workshop’s final presentation, Lee Ann Kahlor discussed the desirability of gathering baseline data to jump start a risk communication research agenda focused on PM. Such data are needed, she said, because little is known about how the public perceives indoor environmental issues broadly and the issues of PM in particular. “We need a current snapshot of PM risk, knowledge, and attitudes but also data to give us direction about how to make information more compelling, especially to audiences that are uniquely vulnerable,” Kahlor said.

Communication, she continued, is about telling compelling stories to a variety of audiences that need to hear those stories. When thinking about stories and audiences, the research community must remind itself often that it is a privileged group and that while only 29 percent of adults 25 years of age and older have a 4-year college degree, the other 71 percent can make important decisions about their health and their family’s health on a moment’s notice when necessary. It is also important to remember that 43 percent of Americans live in rental housing, and therefore are limited in the actions they can take, and that 15 percent live in poverty, which means they cannot afford most of the remediation approaches described at this workshop.

Each of the many audiences for information on PM needs to be reached uniquely, in ways that respond to its members’ resources and values and that take into account the competing threats to their health, safety, and quality of life, Kahlor said. Mass communications research can identify the psychological and social psychological factors unique to each audience, and crafting narratives with these factors in mind will help develop a version of the story that resonates in important ways with that unique audience. This kind of research can also identify which of the many information channels that exist are best suited for conveying stories to specific audiences to achieve a desired effect, which can be a change in attitude, knowledge, or behavior. Kahlor said that one aspect of developing a strategic communication plan is understanding what the information source says and with what intention. “So we as information creators want to be very honest about

___________________

³ This section is based on the presentation by Lee Ann Kahlor, associate professor in the Stan Richards School of Advertising at The University of Texas at Austin, and the statements are not endorsed or verified by the National Academies of Sciences, Engineering, and Medicine.

our intention,” Kahlor said. “Why do we think we have this story to tell, and why do we think it going to resonate with our audience?” Being self-conscious about intentions is critical, she said.

From the presentations she had heard, Kahlor said she had developed a list of research questions that could be part of a baseline study. These included the following:

• Who is vulnerable to PM risks?
• What information do they have about PM risks, and what more do they want to know?
• What risk messages about PM would be compelling for what audiences?
• Who would be credible sources for messages about PM?
• What partnerships would be useful in getting messages out?

To answer these questions, Kahlor recommended using a risk-information-seeking framework, one that focuses on people helping themselves to information outside of a formal education setting. She noted that people do not simply absorb the information that the scientific community generates, but rather, they must be engaged in such a way that they process the information and make it meaningful in a way that makes sense personally. “We need to think about what motivates people to seek information,” Kahlor said. Such motivations might include social norms in terms of what others in their social circle and community expect them to know about topic; a sense of risk; existing knowledge that provides a mental model through which new information flows; how much knowledge people think they need; beliefs about seeking information and whether useful information exists for them to use; and beliefs about the ability to get and process information. Kahlor has organized these motivations in a model she calls the Planned Risk Information Seeking Model (Kahlor, 2010), illustrated in Figure 8-2. One motivation not included in this model, she said, is avoidance, which refers to actively avoiding information. Avoidance can be driven by both negative affect and a desire to maintain hope and feel happy.

Communicating risk is a complicated task, Kahlor said, because it is affected by the fact that understanding is difficult when the audience lacks a knowledge of basic science. “You are not just explaining your science, you are explaining all of science so that you can provide some scaffolding with which to process your information,” she explained. In that context, it is important to decide what a person really needs to know in order to hear a message and to make meaning from it. Other concepts of interest include vulnerability indicators, such as age, life stage, overall health, community, and place; trusted sources for health risk information; comparative social,

economic, safety, and health risks in the environment; and existing risky behaviors.

Kahlor said that in her opinion the best approach for getting data to explore these questions and concepts would be to start with a survey that produces generalizable data from a cross-section of Americans, with a special effort to include vulnerable communities in the sample. For example, because children are vulnerable and parents are natural information seekers when it comes to their children, the survey might include a subset of parents of children up to age 12. Residents of disadvantaged communities, identified by zip code or county, and the elderly could both be oversampled in order to provide a sense of some of the challenges associated with getting information into those communities.

To get a sense of how long it would take to conduct such a survey and how much it would cost, Kahlor contacted a private company that conducts probability-based online surveys for academia, government, and nonprofit organizations. A 15-minute survey of 2,500 individuals, includ-

ing 500 parents of children up to age 12 and 500 residents of designated disadvantaged communities—as a starting point—would take 2 months and cost $65,000. This sum, she noted, did not include conducting the survey in Spanish or sampling elderly and retired persons, which she thought would be a good idea.

Kahlor observed that the data from this survey would provide information on the social norms that would help or hinder getting information out to audiences. It would also reveal patterns in perceived risk and related worry and existing knowledge versus how much knowledge individuals think they need. The data would help researchers understand beliefs about the information available and attitudes toward that information, the ability to get and process information, and whether such information will be accessible. The data would also identify trusted sources for conveying information and the preferred channels for receiving information. It will be important to distribute the data through multiple channels in order to reach as many disciplines as possible, and this will require presenting the data in multiple venues, including journals and conferences.

In summary, Kahlor said, once the data are available to illuminate who the audiences for information on PM are and what information they need, the challenge will be to focus on the stories that need to be told and on tailoring those stories in ways that are meaningful to these audiences. Partnerships will be critical to delivering those stories in a way that resonate with place, time, and most critically, the people themselves.

DISCUSSION

To start off the final discussion, William Nazaroff asked the panelists if their messages to the workshop would change if the audiences for information were expanded beyond the general public to include architects, members of the building trades, manufacturers of mitigation equipment, professional associations, health care providers, and policy makers. Kahlor said she thought the answer would be no, that stories would just be tailored for each of these unique audiences. Gray added that, given the influence that standards can have on the design of buildings and the products that go into them, the standards community should be included as an important audience. Hallman disagreed, however, saying that he believes talking to these other audiences requires different conversations, not different messages, and directing those conversations requires answering different questions.

William Fisk asked if, given the invisibility of particles, communication should revolve around sources and approaches to mitigation and the tangible actions that people can take. Gray said he thought that was a good idea, but added that there will still be the need to get people to care about why they need to make an investment in time and money or change some

aspect of their behavior. That, Gray said, is why the big challenge is making the invisible visible, at least in a mental model. Howard Kipen asked if getting hard data on the health effects of exposure to indoor PM—as opposed to extrapolating from outdoor PM data—would help address this challenge, and Hallman said that the answer was yes, that those data for indoor PM are essential.

Brett Singer asked the panelists to comment on whether the precautionary principle—the idea that knowing that something could be harmful would warrant action if there is an easy way to reduce risk—would be a useful framework in the absence of hard data. Kahlor responded that the precautionary principle can be helpful, but the danger is that the media will convert uncertainty into certainty. The argument, she said, has to be put forward with care so that it gets to the intended audience in the right form. Hallman suggested that this was the approach the field was taking today. “I think the assumption around the room is that exposure to PM is not a good thing at any level and we should try to reduce it,” he said. “Unfortunately, in the United States we do not regulate on the precautionary principle very often, and when we do it is called government overreach.” As a result, he said, the goal should be to reduce exposures but not based on the precautionary principle.

Gray noted that relying on the precautionary principle could have the effect of increasing disparities because people with resources would be more likely to take action than those whose resources are limited. Hallman agreed with Gray’s assessment and said an additional complication is that those without the resources to take action are often left feeling guilty about not being able to protect their children and their families.

Brent Stephens asked about the challenge of communicating the type of relative risks that were discussed in the prior session, and Hallman responded that people often have a difficult time with very large and very small numbers. The shorthand that the technical community uses to communicate within its members serves to alienate the public, he said, and so trying to get the public to think like scientists and understand scientific language and concepts will not work.

With regard to changing personal behavior, Marc Weisskopf asked whether the message that PM could be influencing an individual’s energy level might be a good secondary message, given that this is an effect that people would experience every day and that does not take years to develop. Kahlor replied by reiterating the need to picture what people would do with that kind of information, and Weisskopf acknowledged that concern but countered that there may be easy and inexpensive actions—using the kitchen hood or opening the apartment windows at the appropriate time—that would provide some benefit. “It would seem that people would act on something if it was more likely to affect something they experience

every day,” he said. Gray cautioned that it is important that the message be credible and that the subsequent action needs to produce a meaningful benefit that will not be overwhelmed by all of the other things going on in someone’s life. “That would hurt credibility,” he said.

Hallman cautioned that once people are convinced that indoor PM is a problem and learn there are multiple sources, they will pick and choose which ones to address. One lesson from nutrition research, he said, is that people form an indulgence mental model in which somehow the apple in their pocket cancels out the big breakfast they ate. With PM, people could run their kitchen fans and then feel better about burning candles, he suggested.

In response to an online question about how to improve compliance for air cleaner and range hood use, Hallman said that the answer is to make them automatic, to make “on” the default condition and to make them quiet. Vito Ilacqua from EPA agreed with Hallman and suggested that the way to get such automatic systems into homes would be to start with the early adopters, the people willing to pay a higher price to get a market started. As an analogy, Ilacqua noted that thermostats were a luxury item at one time but are now standard in every home and apartment. Hallman said that the idea had potential and suggested linking automatic systems to the “Internet of things” that could provide remote sensor readings of airborne PM to provide the necessary input. “These things are possible, and they are not necessarily all that expensive,” Hallman said. “With research dollars, it could happen.”

Along those same lines, Barbara Turpin asked if outfitting homes with PM monitors that display real-time PM levels would change people’s behaviors, similar to the way that cars are now outfitted with fuel economy readouts. Gray and Hallman both said they thought that was a good idea, and Hallman added that providing people with social norm data could also help motivate behavior change, though it could also create anxiety.

Paula Olsiewski wondered if it was possible to frame the interventions discussed at this workshop in terms of creating a healthier home. Kahlor said she thought that could be a useful framework because it is general and not tied to a specific risk factor. The key will be the rest of the message—what an individual can do to create that healthier home. Gray agreed with the importance of the “What next?” part of that message and added that some of those what-next steps will be things that some stakeholders who hold a different point of view will not want to have happen. Hallman added that there is a science of science communication and methods for testing these kinds of messages before putting them out to the world, and he said that this is the kind of research needed once the baseline data are available. “We need programmatic research in this area and not just a couple

of studies,” he said. He noted, too, that EPA was the initial leader in the federal government on risk communication.

Richard Corsi remarked that when he gives talks to the general public about indoor PM, members of the public pay attention when he talks about candles, and they then inquire where they can get more information. Corsi said that the EPA and California Air Resources Board websites have good information on them about indoor PM, and he asked the panelists if those sources of information are important and if the public perceives the importance of getting that kind of information. Kahlor said that when experts direct individuals to websites, they can be good sources of information, but in some settings, such as a doctor’s office, a well-designed pamphlet could be a more useful information source. Hallman added that when people want health-related information, they typically go to the Internet these days, and so it would behoove the experts to check what Wikipedia and WebMD have to say about indoor PM to make sure they are providing good information. Most members of the public, he said, are not likely to think of going to the EPA website as their first source of information. He also suggested that this community could create a website with an intuitive name that would naturally draw people to good information.

Gediminas Mainelis asked if the panelists had any thoughts about how to communicate information in a way that will not lead to people altering their behavior in a way that negates any improvements that might make, such as putting ionizers and filters in their homes and then smoking inside. Hallman said risk homeostasis theory (Wilde, 1998) addresses this kind of behavior. David Rowson from EPA then asked the panelists for ideas on how to conduct non-regulatory risk communication around an issue that involves promoting, encouraging, and suggesting behavior change rather than requiring it and enforcing compliance. Hallman replied that the argument can be framed in terms of a gain—health will improve if indoor PM levels are reduced—or a loss and that social norms can be used to encourage positive behaviors. He added, though, that it is important to be realistic about what risk communication alone can accomplish. “The idea that we are going to encourage 90 percent of the public to do something based on the information we provide is unrealistic, which is why regulations are needed,” he said. Offering an example that he said was relevant to indoor PM, he noted that getting New Jersey homeowners to test for radon only happened when stakeholders insisted that radon tests be done before people could get mortgages.

To end the discussion, Terry Brennan noted that when he was building single-family residences in the 1980s, he would violate manufacturer’s warranties by wiring thermostats into the range hood so that they would turn on automatically when someone was cooking, and he installed high-efficiency filters in the HVAC systems. “That was easy for me to do as the

builder because I had control over what went into the building, and it was an easy sell to the homeowner.” Since then, he said, he has worked with builders who put up 5,000 houses per year. Making changes at that scale would represent an important step, but doing so would require making changes in standards, which he said is an uphill process, given that for every group involved in the discussions that wants change there is another group that does not. The key in that case, Hallman said, is to convince the policy makers that the public wants these changes, which is where the data from the study Kahlor proposed could help.