Table 2 Abstraction table of articles on wildfire smoke exposure risk communication approaches

Randomized Controlled Trial

 Postma JM, et al. Promoting risk reduction among young adults with asthma during wildfire smoke: A feasibility study. 2022 [49].

The EPA Smoke Sense mobile application was adapted by technology partner, Urbanova. SSU: Participants recorded wildfire smoke observations, health symptoms, exposure reduction behaviors, and viewed information (air quality, next day forecast). SSU-Plus: Same features as SSU but had additional social features from Urbanova to maximize risk reduction (use of spirometer, mapping, and message board).

Three-arm, unblinded, randomized controlled trial. Recruiting via university student listservs. One control arm and two intervention arms using Smoke Sense Urbanova (SSU) and Smoke Sense Urbanova-Plus (SSU-Plus). Clinical outcomes were measured via Asthma Control Test (ACT) and forced expiratory volume (FEV). ACT measures five items (shortness of breath frequency, asthma symptoms, use of rescue medication, daily functioning, and asthma control) during self report recall. Higher scores indicate better control, with a score of over 19 indicating well-controlled asthma. FEV1 is the maximum amount of air that is subject can forcibly expel during the first–second following maximal inhalation. Decreases indicate airway narrowing. Spirometry measures FEV.

N = 67 young adults (18–26 years old) in Western US with self-reported asthma diagnosis by healthcare provider, own a smart phone, understand English. Most participants identified as female (79%), non-Hispanic (88%), and white (78%). Most reported using an iPhone (77%) versus Android (23%). At baseline the average ACT score was 20.4 (2.5), and the mean percent predicted FEV1 was 93.1% (17.8). Over half (55%) of participants reported being prescribed an asthma maintenance medication yet half reported ‘less than prescribed’ usage.

Smoke Sense app was usable and acceptable by intervention participants. In both intervention arms, Smoke Sense was primarily used to view the AQI and explore the fire and smoke map. Smoke Sense was less frequently used to report health symptoms, smoke observations and exposure reduction activities. And the least engagement was in learning risk reduction strategies. SSU-Plus had additional features, in order of frequency of use: spirometry, “Plus” menu of features (e.g., maps), and message board. 90% of SSU-Plus participants agreed that the app helped their asthma management, particularly with these features: air quality advisories, daily spirometry readings, and mapping feature, compared to 50% of SSU participants. SSU-Plus arm had significant increase in ACT at week 8 (Mean [SD]: 21.5 [2.3]) compared to baseline (20.0 [2.4]) (p = 0.0008), and significant decrease in percent predicted FEV1 at week 8 (88.6% [17.2]) compared to at baseline (94.9% [16.2]) (p = 0.0172). SSU arm had no difference in ACT at week 8 (21.0 [4.0]) from baseline (21.3 [2.1]) or in FEV1 at week 8 (95.6% [17.2]) from baseline (97.6% [14.6]) (all p > 0.05). Control arm had significant increase in ACT at week 8 (22.4 [1.9]) from week 0 (20.2 [3.7]) (p = 0.0320), but no change in percent predicted FEV1 at week 8 (92.9% [16.0]) compared to week 0 (88.4% [20.2]) (p > .05).

- increased use of certain features (e.g. peer message boards)

- weekly reminders to post messages

- graph spirometry readings over time

- more information about what spirometry readings meant would improve app engagement.

Quantitative Study

 Cao Y, et al. Is a picture worth a thousand words? Evaluating the effectiveness of maps for delivering wildfire warning information. 2016 [50].

Examination of the effectiveness of maps versus conventional text-based approaches for communicating spatial-related wildfire warning information in 3 fire-prone areas (urban, suburban and rural) of Western Australia related to residents’ abilities to use that information to predict fire spread and need for taking precautions. Experimental study based on understanding warning ‘effectiveness’ (warnings should appeal to general users, facilitate understanding and risk perception, assure information processing, and trigger appropriate responses) and effectiveness of maps for risk communication.

Online convenience survey recruiting via invitation postcards. Survey was conducted from Feb 2014 to Apr 2014 to quantitatively compare the communication outcomes of text-based messages versus various map designs and information elements (e.g., fire location, evacuation centers) within an experimental setting. Participants were tested on their abilities to extract key information about a simulated scenario of wildfire spread warnings from several different kinds of maps, and from text messages. They were also asked about their preference for maps versus text message information: i) accuracy of understanding, ii) risk perception, iii) response time, and iv) preference and ease of understanding.

N = 261 residents of 3 fire-prone areas (Kelmscott, Roleystone and Mundaring) in Western Australia. Sample was skewed towards residents who were middle aged (55% participants were 50–69 years old) and had higher education than general population of residents. 64% resided in place of residence for 10 or more years. 94% self-reported daily use of computers and 63% daily use of maps.

Accuracy of understanding, risk perception, response time, preference and perceived ease of understanding, and warning communications (map symbols, legends, text elements) were all measured by the survey. Significant differences (among different map designs: p < 0.05, p < 0.01, p < 0.005) were found between maps and text-based messages for understanding the wildfire risk situations. Map versions resulted in more accurate assessment of location, direction, and distance, and improved understanding of information than the text-based messages. Generally, when compared to text-based messages, maps stimulated better risk perception, but this varied between different communications. Response time also varied between maps and text-based messages, with some complex map designs resulting in a longer response time. Although maps did not always have better accuracy, risk perception, and response time, there was a preference for maps over text messages. Among 7 information elements, preference for text ranged from 1.2% to 37.1% of respondents, with majority of the respondents preferring maps or a combination of maps and text. Respondents reported better ease of understanding for all map designs except one (p < 0.005), compared to text. Warning communications included easily understood map symbols and colors, legends to explain color and size categories, and combining text elements with maps to improve descriptions.

- utilizing specific map designs depending on the information to be delivered

- incorporating visual maps and textual descriptions for enhanced understanding and interpretation

 Hano MC, et al. Knowing Your Audience: A Typology of Smoke Sense Participants to Inform Wildfire Smoke Health Risk Communication. 2020 [51].

Assessment of user responses to the EPA Smoke Sense app, which is an educational mobile app about wildfire smoke exposure risks (including in specific geographic areas) and protective measures. The study objectives were to assess how user demographics, health status, information needs and risk perceptions were associated with preparedness to adopt health behavioral recommendations. Study aimed to identify specific clusters of traits related to intention to act (or not) and provide recommendations for risk communication for specific clusters of traits. Study further assessed how such traits map to several common health behavioral models, especially the Precaution Adoption Process Model.

Investigators used multiple cluster analysis to identify perspective 5 trait clusters based on users’ responses from Sept 2018 to May 2019 about their health status, experiences with wildfire smoke, risk perception, self-efficacy, access to exposure-reducing resources, health information needs, and openness to health risk messages. Differences in these traits were examined based on user demographics, health status, activity level and engagement level (from least to most engagement: Users, Explorers, Observers and Learners) with the app. Traits were mapped onto the Precaution Adoption Process Model relative to adopting recommended behaviors. Cluster findings were used to recommend health risk messaging users in each of the 5 clusters.

N = 5,018 participants in the US who provided information on the “citizen science” section of the Smoke Sense app. Respondents were mostly white (76%), college-educated (69%). 52% were female and 72% were between 30 and 65 years old. 92% reported “good” to “excellent” health. 72% reported wildfire smoke issues where they lived and most agreed such smoke was a health problem and risk could be modified by behaviors to reduce exposure. Respondents varied in their perceptions of information needs about wildfire smoke risk information.

Study findings identified 5 major traits based on respondents’ attributes and perceptions. Traits were not strongly identified by demographics, individual engagement with the health risk tool varied across traits. Traits were mapped onto the Precaution Adoption Process Model which indicated where the 5 trait clusters fell relevant to people’s intention to adopt recommended behaviors: “Unengaged”, “Susceptible”, “Proactive”, “Cautious”, “Protectors”. Health conditions, risk perception, resources, information needs, self-efficacy, and receptiveness to health risk communication all played a role in differentiating five traits. There were significant differences in how different traits engaged with the app at deeper Observer (ANOVA F-ratio: F (4, 4,777) = 8.242, (p < 0.005)) and Learner (F (4, 4,777) = 9.587, (p < 0.005)) engagement levels. Notably, the Susceptible trait, followed by Cautious trait, had the highest engagement as Observers and Learners.

- tailoring risk messaging approaches for the 5 trait groups, including messaging for trait groups that may want more information, information about their smoke-relevant disease symptoms, information provided through health providers, etc.

- further study of how the traits that emerged in the study can be used to guide more targeted risk messaging strategies.

 Mott JA, et al. Wildland forest fire smoke: health effects and intervention evaluation, Hoopa, California, 1999. 2002 [52].

Interventions implemented by staff of the local medical center and other tribal organizations included distributing filtered and nonfiltered masks free of charge, vouchers for free hotel services in nearby towns to facilitate evacuation, portable high-efficiency particulate air (HEPA) cleaners, and several public service announcements (PSAs) through local media outlets. Due to limited resources, individuals who had adverse health effects and had preexisting conditions were prioritized for hotel vouchers and HEPA cleaners.

Retrospective quantitative study. A community survey of individuals treated at the reservation medical center and from randomly sampled households was completed. Individuals with preexisting cardiopulmonary conditions (coronary artery disease, asthma, chronic obstructive pulmonary disease, and other lung disease) were oversampled from the reservation medical center. And one randomly sampled person per household without any preexisting conditions was selected. The survey questions were about family demographics, intervention participation, and lower respiratory tract symptoms linked with forest fire smoke exposures before/during/after the large wildfire of 1999.

N = 289 residents of the Hoopa Valley National Indian Reservation, Humboldt County, California. Among subjects with preexisting conditions, 42.7% were male, 47.8% had household income at or below poverty level, 39.1% were less than 23 years old, 29.3% were 24–54 years old, and 31.5% were over 55 years old. Among subjects with no preexisting conditions, 49.2% were male, 55.3% had household income at or below poverty level, 35.4% were less than 23 years old, 52.8% were 24–54 years old, and 11.8% were over 55 years old.

Among evacuees of a wildfire smoke event, more than 80% of respondents correctly recalled a PSA without being shown a list of known PSAs, of these 66% took action to reduce smoke exposure as a result of hearing the PSA. Among those who could recall a PSA (n = 238), remaining indoors was the most common (78.6%, 95% CI: 73.4–83.8). Among those who took action because of a PSA (n = 157), “staying inside more often” was the most common action (83.4%, 95% CI: 77.6–89.2) taken. Respondents who recalled a PSA were less likely than those who could not recall a PSA to report worsening respiratory symptoms (OR = 0.25).

- PSAs via radio/telephone reduced respiratory symptoms to wildfire smoke

- PSAs combined with behavioral interventions are more effective

- prioritize persons with pre-existing cardiopulmonary conditions.

 Spano G, et al. Is Experience the Best Teacher? Knowledge, Perceptions, and Awareness of Wildfire Risk. 2021 [53].

Examine the impact of direct experience with wildfires on preparedness on the topic of wildfires among individuals. Assessing if there are group differences between fire-related employment or wildlife-urban interface proximity.

Cross-sectional design, single survey questionnaire administered throughout Italy. Questionnaire was disseminated through various channels, including social networks and personal and professional contacts snowball sampling. The general survey sample was grouped into two subsamples: (a) participants with direct wildfire experience and (b) participants without direct wildfire experience. Subsequently, the subsample of participants with direct experience was further classified as follows: (c) fire-related workers vs. (d) non-fire-related workers; and (e) WUI residents vs. (f) non-WUI residents.

N = 775 participants over the age of 18 and currently residents in Italy. Most respondents were from Apulia, (southern) Sardinia, Tuscany, (central) Lazio, and Lombardy (Northern Italy). Mean age was 37.4, and 50.7% female.

Individuals directly exposed to a wildfire, compared to those not exposed to a wildfire, have a significantly higher level of subjective knowledge of topic themes including wildfire definition (× 2 = 152.7, p < 0.001), self-perceived knowledge (× 2 = 17.2, p < 0.001), WUI definition (× 2 = 189.2, p < 0.001), and wildfire occurrence (× 2 = 23.4, p < 0.001), and significantly higher self-learning (× 2 = 0.3, p < 0.001) of the topic. There was no difference between the two groups on advanced knowledge, risk, and drivers/causes of wildfires. Participants with a direct experience of wildfires requested more information on the topic than those who had never been exposed to an event, which confirmed that knowledge and experience influence the number of sources from which individuals usually seek information.

- need for developing effective communication for high-risk groups, such as homeowners and fire-related workers, in order to effectively prepare them for threats and potential impacts of wildfires, and avoid the adverse health impacts of the exposure.

Mixed Methods Study

 Chapple DR, et al. Communicating bushfire risk in the Blue Mountains: a case study of the Fire Stories film. 2017 [54].

Risk communication case study of a documentary film that was made as a tool for learning about natural disasters for many residents in the Blue Mountains who have little or no experience with bushfire due to the transient nature of these residents. The film was locally produced and used an event-based approach, using a past bushfire event to provide a vicarious experience, to demonstrate fire risk to motivate residents to be prepared for bushfires and to increase risk awareness.

Participants watched the film in 2013 (either at the cinema screenings or over YouTube/DVD) and were recruited from film audience database and promotions in newspaper and on social media, participated in a survey which was conducted over 4 weeks in April–May 2015. The survey sought to understand 1) what did they do to increase safety after viewing the film until the end of 2015, 2) how did they respond to the 2013 bushfire season in terms of safety-enhancing activity, 3) whether and to what extent did the film and other factors contribute to safety-enhancing activity, and 4) what aspects of the film contributed most to the safety-enhancing activity.

N = 104 online completed questionnaires: 84 from the cinema screening and 20 from YouTube/DVD audience. Participants lived in Australia. 66% lived in the upper Blue Mountains, 87% were homeowners, 57% female, 52% were over 60 years old.

Between watching the film and the 2013 fires, the cinema audience respondents reported 257 activities with 153 of them being “new” activities for bushfire preparedness activities (e.g., seek more information about risks, speak with household and community about risk information). 85% of respondents rated the film elther very effective or effective in promoting community preparedness and resilience. 71% rated the element of the graphic map of the movement of the fire as the most frequent “high impact”. 60% of respondents reported thinking about the film “for a long time afterward”. 72% of additional bushfire safety activity between viewing the film and the 2013 fire event were related to the film, newspaper articles, and the information exposition. There was a broadening of spheres of concerns after the film and before the fire with respondents indicating new concerns for their street, neighborhood, and someone nearby who needs help with respondents commonly crediting viewing the documentary for changes. The use of the film helped respondents understand the effects of the devastating 1957 fires, the implications of living in fire-prone areas, and the new actions to prepare for bushfire season.

- use event-based approach for risk education to communicate, to enhance community bushfire safety efforts, to encourage social learning through storytelling,

 Sugerman DE, et al. Emergency Health Risk Communication During the 2007 San Diego Wildfires: Comprehension, Compliance, and Recall. 2012 [36].

Evaluation of a study of wildfire risk communication mass media messages in San Diego County disseminated during the Oct 2007 wildfire. Messages were disseminated by the San Diego County Health and Human Services Agency and by the American Heart and Lung Association about ways to reduce exposure to wildfire smoke and ash. Messages in English and Spanish contained both “non-technical” (simpler actions: stay inside, close window, etc.) and “technical” (required access to products (home filtration, N85 masks, etc.) and/or performing more complex actions) recommendations. Messages were disseminated via TV, radio, newspapers and the internet over a 3 week period.

Random digit dialing survey from a landline database (no cell). English and Spanish respondents only above 18 years old. Questions adapted by a prior survey and pretested. Variables included: socio-demographic, exposure to wildfire smoke/ash, unprompted recall of messages, prompted recall of messages, communication source of messages, reported understanding of messages, reported actions taken, and whether respondent used inhaler/oxygen or sought medical care.

N = 1,802 survey respondents (48% response rate; 93% cooperation rate (completion/eligible) in San Diego, CA. 50% were male, and most were middle aged (35–64 years), non-Hispanic White, educated past high school, employed full time, and spoke English as their primary language.

Most persons surveyed reported hearing fire-related health messages (87.9%) and nearly all (97.9%) understood the messages they heard. Univariate analysis showed that compliance with most to all messages was higher among persons who were female, spoke English as their primary language, were educated beyond high school, and earned incomes of $50,000 or more (p < .05). Respondents complied with most to all of the nontechnical health messages, including staying inside the home (58.7%), avoiding outdoor exercise (88.4%), keeping windows and doors closed (75.8%), and wetting ash before cleanup (75.6%). In contrast, few (< 5%) recalled hearing technical messages to place air conditioners on recirculate, use High-Efficiency Particulate Air filters, or use N-95 respirators during ash cleanup, and less than 10% of all respondents followed these specific recommendations. The authors found that nontechnical message recall, understanding, and compliance were high during the wildfires, and reported recall and compliance with technical messages that required buying supplies (N85 masks, HEPA filters, or carrying out more complicated actions) were much lower.

- limit the number of messages, explaining “technical” messages better

- making messages more accessible to vulnerable populations

- future disaster health communication should further explore barriers to recall and compliance with technical recommendations

Literature Reviews

 Fish JA, et al. Effectiveness of public health messaging and communication channels during smoke events: A rapid systematic review. 2017 [44].

Reviewed for these qualities

Communication channels: mass media, interpersonal, road signs, presentations at public events.

Effectiveness: awareness of public health messages, communication channel preferences, communication channel use, trust in communication sources, and compliance with advisories.

Rapid systematic review, search of 12 databases: MEDLINE, CINAHL, Current Content Connect, EMBASE, Science Direct, Scopus, PsycINFO, PsychARTICLES, JBI Databases of Systematic Reviews and Implementation Reports, Cochrane Database of Systematic Reviews, Evidence for Policy and Practice Information and Coordinating Centre/EPPI-Centre, the Campbell Library, and Database of Abstracts of Reviews of Effectiveness. Search terms included: “smoke event”, “bush fire”, “forest fire”, “wildfire”, “vegetation fire”, “wildland fire”, “prescribed fire”, “prescribed burn”; “public health messaging”, “risk communication”, “health warning”, “community warning”, “emergency warning”, “health information”. Study quality assessed based on effectiveness of communication channels for the general population, effectiveness of communication channels for at-risk populations, effectiveness of public health messages disseminated during smoke events. Articles from 2009 to 2016.

N = 10 articles from various geographic areas: North Rockies, South-Central US, Victoria Australia, Tasmania, NSW Australia, CA, Montana, Oregon, South Carolina, North Carolina. Review done by Australian research team.

Mass media, communication with healthcare workers, social networks from evidence are weak in terms of effectiveness of channels and sources of info, but indirect evidence shows these might be effective. Childcare, schools, retirement facilities are effective communication channels for specific age-defined groups. Neighbors and local residents were a trusted source of info. Preferred communication channels in US Australia: TV/radio/local paper, personal communication with authorities, social networks. Potential with texts, social media, mobile app, and twitter data, but insufficient evidence so far if communicating via internet is effective but may be useful depending on further research.

- use multiple communication channels

- leverage technologies such as social media and mobile apps

- identify and use most effective communication challenges to reach at-risk and vulnerable communities

- use simple language that is clear and specific about the situation with actions required

- conduct evaluations and studies to strengthen evidence base of messaging interventions

 Heaney E, et al. Efficacy of Communication Techniques and Health Outcomes of Bushfire Smoke Exposure: A Scoping Review. 2021 [35].

Papers covered media form and choice, smoke avoidance behaviors, advice on message style and content, communication limitations, bushfire smoke exposure, community medical/health seeking behaviors during smoke events, mortality and morbidity, psychological impacts, health consequences, health communication

Scoping review, search of PubMed database, ProQuest. Search terms included: “bushfire”, “prescribed burn”, “health messaging”, “wildfire”, “social media”. 20-year published time limit (Jan 1 2000 to Jun 1 2020).

N = 67 studies included: 20 Australia, 32 US, 8 Canada, 2 global, 1 of each: SE Asia, UK, Portugal, Sri Lanka, Belgium, China. Review done by Australian research team.

Communities used combo of traditional media sources (TV, radio, phone, newspaper) and non-traditional (Twitter and Facebook). Social media: real time dialogue between authorities and public, easy for messages to get amplified with sharing, builds trust when official accounts re-tweet or re-share posts. Still high utilization of traditional sources. Effective communication: guidance, timeframe, affected location, hazard/consequence, info source for those who want further info, guide public towards health promoting actions and safety during disaster, tailor info to local context to ensure info sources are credible/consistent, identify evacuation sites and any closed regions/roads. Messages: clear, specific, accurate, certain, and consistent, variety of languages and format, don’t inspire fear/panic in language, timely delivery, complete unbiased factual info, explain changes if any from prior messaging to keep consistent over time, encourage confidence, reduce confusion and distrust by ensuring consistent messaging.

- Communication should use both traditional and non-traditional media sources. Also use different media at different stages of disaster to tailor info delivered to each individual.

- Use effective communication and include pertinent info: actions/guidance, timeframes, affected location and evacuation areas, hazard, info source

- Messages: clear, specific, accurate, consistent, objective, build public confidence and trust

 Keegan SA, et al. Health protection messaging for populations susceptible to air pollution during landscape fire smoke events: an integrative review. 2021 [55].

Assess health protection messaging specific to smoke events, and risk communication contents surrounding bushfires and wildfires. Investigate which health protection message pathways and contents are appropriate for informing susceptible populations during smoke events.

Integrative review, searched Medline, Scopus, Embase, and CINAHL for peer-reviewed articles related to landscape fire air pollution and health risk communication. Search terms included: Bushfire, “bush fire”, wildfire, “wild fire”, “fire event”, “landscape fire”, “fire smoke”, “smoke pollution”, “air pollution”, “air quality”, “smoke event”; “Health protection message, “risk communication”, “public health message*”, “health warning”, “health alert”, “public health intervention”, “public health advi”, “health recommendation”. Also searched government websites, Google, Google Scholar, and websites for NGOs. Articles from 2010–2020, final search date was May 16, 2020.

NOTE: articles aimed at assessment of interventions or strategies were excluded.

N = 26 articles included in final review. 15 US, 4 Canada, 1 Canada-US, 5 Australia, 1 multi-national. Review done by Australian research team.

Regarding content, messages were either 1) emergency health alerts or 2) information campaigns. Clear, short, and non-technical advice was more likely to be recalled and complied. Traditional and modern communication channels were being used effectively. There is some evidence that health alerts are less likely to be received by persons over 75. There was sparse information on susceptible populations, which stated the need for messaging to begin ahead of bushfire season. Regarding delivery, this was essential to build community trust. Inconsistent messaging was confusing and lacked referencing of trusted sources. Scant and somewhat conflicting evidence on susceptible populations’ use of information. Television is preferred source of information but increasing preference for online and smartphone-based communication methods. Improved interagency collaboration and communication could help with better delivery.

- more detailed information on smoke and its health impacts for fire-prone communities, targeted messaging campaigns for communities who have high levels of exposure or fewer resources to take action

- more consistent, timely messaging and greater dissemination of information on AQI and protection strategies

- need for communications to start before bushfire seasons and advanced planning for messaging during smoke events, especially for susceptible populations

 Stieb DM, et al. Using maps to communicate environmental exposures and health risks: Review and best-practice recommendations. 2019 [37].

Examine the effectiveness or utility of maps as environmental exposure including wildfire smoke and health risk communication tools, addressed the issue from the perspective of health literacy, environmental health literacy, or public health ethics.

Librarian searched Embase (1974 to 2018 February 26) and MEDLINE (1946 to February 21, 2018) for commentaries, reviews, and primary studies about the evaluation of maps or map features as exposure and health risk communication tools. Search terms included: map, GIS, spatial analysis, pollution, environmental health, air, health behavior, health education, literacy, community, population, communication, public health, risk.

N = 37 total: 5 reviews/commentaries, 32 primary studies (18 US, 2 UK, 4 Australia, 2 Belgium, 3 Canada, 1 Austria/France/Germany, 2 Germany). The most common design was experimental (n = 9), followed by qualitative or mixed methods (n = 8) and cross-sectional surveys (n = 6). Half of the studies (n = 16) were conducted in the US. Review done by Canadian research team.

Map developers and users engage to define purposes, planning, and goals of maps, iterative process. Consider visual cognitive science (how maps are processed cognitively, images and purposeful info, color, position, texture, motion, patterns, objects). Consider risk perception differences between experts/non. Risk communications: convey spatial dimension of exposure, inform personal, community, or political decision-making to mitigate risk, improve knowledge, persuade behavior change, increase confidence in authorities, build capacity for stakeholder engagement. Communications with general public: maps improve understanding of risks and information, better comfort and accuracy of understanding, better risk perception, more meaning, and maps often more preferred

- understand the map developer’s societal role and mental model underlying map design

- define, understand and iteratively engage with map users

- define developers’ and users’ purposes

- inform map design using key theoretical constructs

- account for factors affecting risk perception

- adhere to risk communication principles and cartographic best practices

- avoid intentional and unintentional misrepresentation

- consider environmental justice and public health ethics implications

Qualitative Study

 Burns R, et al. From hypothetical scenario to tragic reality: A salutary lesson in risk communication and the Victorian 2009 bushfires. 2010 [33].

Four example messages were created for different time periods. First message for day of blaze has immediate danger and preventive action regarding the fire. Other three messages contain health warnings. Third and fourth message include a few sentences about smoke safety. Messages were disseminated via local papers, radio, TV, and internet.

Qualitative focus groups using a hypothetical bushfire scenario based on previous fires. This was a part of a larger study on additional disaster scenarios, participants recruited for this bushfire scenario were from larger study. Two messages for two time periods (day of blaze, two days later) were adapted from actual examples to test communications and timing. Researchers gave demographic and fire risk perception surveys then asked for comments on trustworthiness of message source.

N = 28 participants, Australia. Four focus groups: landcare members (including fire volunteers), community health workers, two ‘new parents’ groups. 4 male and 24 female, ages 20 through 60 and over, with and without young children, resident in town and semi-rural settings.

Those over 40 preferred radio and local papers, and emergency services. Those under 40 preferred TV and local papers, and state and council spokespersons. Greatest perceived source for all was family, friends, neighbors. For one message that mentions smoke was sent shortly after the fire - thoughts: alarming and anxiety-provoking, inadequate knowledge without clear action steps, seemed sensational and not reliable source, message was news not information. Other message that includes smoke and sent after the fire: more reassuring than concerning, threat to children was a “scare thing”, not enough information, no action steps.

- communicate via local places (post office, school)

- include information about location of fire, population message pertains to, phone number to evacuation center

- include safety of situation, phone number for regular updates, primary school should be information source

 Damon SA, et al. Public communication in unplanned biomass burning events. 2010 [56].

Verbal discussion of panel participants during conference: tools for communicating respiratory risk from biomass burning, information to know about a particular wildfire incident to communicate effectively with affected populations, disseminating information in timely and useful manner, information to know about the audiences at risk (e.g., health vulnerabilities, geographic location, demographics) to communicate effectively, what doesn’t work

90-min break-out session at a conference. Instruments/questions to discuss were about: basic tools for communicating respiratory risk, basic information to know about an incident and about at-risk populations to communicate effectively with affected populations, how to share information in timely manner, what doesn’t work. Used interview/discussion interview, hand-recorded notes, analyzed results afterwards

N = Number of participants not available. Montana, Georgia. Attendees of the 2007 International Biomass Smoke Health Effects conference cosponsored by University of Montana and CDC that attended this break-out session.

Through the lens of Health Belief Model and Stages of Change Model: communications must be “new” to prevent public from sliding back to earlier stage of change and from having tempered perception of risk, people need to believe they have sufficient self-efficacy and able to overcome barriers to taking action, must be notified appropriately when to take action and if any smoke conditions change. Have profiles on affected populations available to assess need for low-literacy, non-English, compliance with recommendations, status on Stages of Change continuum. Federal and state resources should be connected with local agencies, community-based organizations, emergency response, and media. Mobile air monitors should be given to public, “Stay inside” typical recommendation could use research on indoor air filters to be more scientifically grounded.

- materials must be “new” so public risk perception does not decrease after weathering previous years’ events

- convince public of personal relevance of threat and ability to take recommended protective actions

- research-based public messaging must be protective and non-alarming and be able to convince people of threat existence-their susceptibility-their ability to act

- channels: message blanketing (all local media outlets), local opinion leaders and organizations

 Dodd W, et al. Lived experience of a record wildfire season in the Northwest Territories, Canada. 2018 [57].

Explore the lived experience of individuals and communities affected by the 2014 wildfire season in the Northwest Territories of Canada, and to examine the impact of the wildfires and smoke on mental and emotional well-being, physical health, and livelihoods

Qualitative interviews were conducted during the period of October–December 2015 with residents of the four communities. A semi-structured interview guide was collaboratively developed by the research team. The audio portion from each video recording was manually transcribed. An iterative reflexive process was used to identify themes that emerged from the interviews. Coding was conducted within the QSR NVivo 11 software.

N = 30 participants (10 in Yellowknife; 7 in N’Dilo; 6 in Detah; 7 in Kakisa). Community members from four Subarctic communities residing in the Northwest Territories in Canada. The communities were Yellowkinfe, N’Dilo, Detah, and Kakisa. N’Dilo, Detah, and Kakisa are majority Dene (First Nation) communities while Yellowknife’s population includes both Indigenous (First Nations, Metis, Inuit) and non-Indigenous residents.

This paper is a qualitative paper about tribal experiences after summers with wildfires and shows the impacts that fires have had on mental health (personal and community isolation), decrease in physical health and activity due to the smoke limitations, and limits in cultural and “land-based” activities along with the decreased availability of country foods. These adverse experiences encouraged opportunities for community members to support and care for each other, and for several community-based initiatives to begin. A community-based program (FireSmart), initiated by the Government of Northwest Territories in Yellowknife and Kakisa, provided a workshop to encourage residents to remove brush and other flammable materials from around their houses.

- expand community-based initiatives to reach more people and ensure broad participation

- need for increased dialogue and education within their communities concerning the current impacts of climate change and how adaptation could mitigate these impacts

 Errett NA, et al. Building a Practice-Based Research Agenda for Wildfire Smoke and Health: A Report of the 2018 Washington Wildfire Smoke Risk Communication Stakeholder Synthesis Symposium. 2019 [58].

Stakeholder synthesis “World Café” meeting to define a practice-based agenda for wildfire smoke research, interventions and policies for Washington State.

Purposively-selected Washington state practitioners and academics with relevant professional responsibilities or expertise in wildfire smoke and health engaged in small group discussions using the “World Café Method” to identify practice-relevant research needs related to wildfire smoke and health. 8 Table (2 tables per population) were set up to discuss issues relevant to 4 affected populations: workers, at-risk, susceptible populations, rural, and urban/suburban populations. Notes from each discussion were coded and qualitatively analyzed using a content analysis approach.

N = 76 Washington State practitioners and academics from state and local public health, environmental, non-profit, policy and tribal groups and a few stakeholders from federal government.

Participants identified research needs that were grouped into the following research topics: exposure science, health risk research, risk communication research, behavior change and interventions research, and legal and policy research. Many needs were identified within those topical areas, including need for more precise data about wildfire exposure, short-term, long-term and cumulative effects of wildfire smoke exposure, assessment of risk perceptions and best communication messages/channels/messengers, prioritization of the most effective and feasible interventions for specific groups, research on effectiveness of interventions, and development of government policy regulations for worker protection, building codes, etc.

- additional research is needed to support risk assessment, risk communication, and risk management to protect communities and workers across Washington state from the growing threat of wildfire smoke

- encourage researchers, practitioners, and funders to use proposed research topics to inform future research-practice collaborations and policy- and practice-relevant research

 Hano MC, et al. Scaling Up: Citizen Science Engagement and Impacts Beyond the Individual. 2020 [34].

Explore the value of the EPA’s “citizen science” Smoke Sense app. The app provides educational information, has a forecasting feature for wildfire smoke, and integrates user input about wildfire locations, health symptoms (including subclinical and psychological, etc. symptoms. The study explored the perceived value of the app for individual, organizational and community use. The investigators provided background on the emerging use of “citizen science” projects to address complex health/social problems like wildfire smoke.

Qualitative study with semi-structured key informant interviews to explore the perceived value of the Smoke Sense app at individual, organizational and community levels. Recruited via purposive sampling strategy, drawing on list of organizations and individuals who had interacted with the Smoke Sense team as part of the citizen science project; inclusion criteria: (1) members of organizations in the health and environmental fields and thus part of the system that responds to smoke events; (2) individuals who engaged with Smoke Sense during its pilot year; and (3) individuals who worked in areas affected by smoke during the 2017 wildfire season. Inductive approach to inquiry beginning with case of interest and describe it with the purpose of synthesizing meaningful insights, data analysis used within-case inductive phenomenological thematic analysis

N = 8 respondents. Participants included three employees of public organizations at the local level, four at the state level, and one at the tribal level in the western United States, all of whose work intersects wildland fire and associated smoke, air quality, and public health.

This research investigated the motivations and expectations of organizational leaders across public-serving organizations who engage in citizen science projects regarding wildfire smoke and health. Results showed that participants significantly valued the app and its “citizen science” approach to collecting data, including novel information about wildfire smoke impacts from the public. Participants identified 3 motivations for using the app: 1. Individual: getting information to protect personal/family health; 2. organizational: using the app to support the work their organization does to get out information about wildfire smoke exposures, and to build relationships between federal, state, local, and tribal partners; 3. Community: support community efforts to protect sensitive groups. Participants thought Smoke Sense was an important decision-making tool at all these levels.

- further research needed about the app over time to gather more information about its use, value, and ways to improve it

- increase the impact of our citizen science research by supporting partners and scaling up the opportunities for engagement within projects from individuals to organizations, to advance the collective goal of addressing overarching complex social problems

 Marfori MT, et al. Public Health Messaging During Extreme Smoke Events: Are We Hitting the Mark? 2020 [38].

Examine the effectiveness of PH communications about smoke with participants wanting simple messages from official sites, details about short term and long term effects, and having concerns about the smoke on their health. Understand (1) the level of concern about the impacts of smoke on well-being, (2) how information about smoke and health was received and understood, (3) if public health information influenced individual actions and behavior, and (4) the acceptability of using portable HEPA cleaners for managing poor indoor air quality during the wildfires.

Qualitative research with semi-structured interviews with 24 households in Tasmania following a severe smoke episode in 2019). Recruited from news media, radio, newspaper, social media feeds of public health services and regional community groups, and direct invites to current users of smartphone app AirRater. Inductive analysis; interviews conducted 2–3 weeks after the fire

N = 24 residents from 24 different households in Huon Valley region of Tasmania. 18 of 24 were female, ages ranging from parents of young children to retirees, several with had a health condition affected by smoke such as asthma, and several were pregnant. At higher risk from exposure to air pollution from wildfire smoke and in the intended target group for public health messaging.

Participants almost universally commented about experiencing personal concerns, such as negative physical, social, and psychological impacts, from living in smoke-affected areas over a prolonged period but some were more concerned about the wildfires than about the smoke. Participants all recalled receiving different information for smoke and health from different sources: media releases, web-based information, social media posts from public health and emergency services, informal sources such as friends and colleagues, and information on social media reshared by individuals, community groups, and agencies. Participants had more distrust in messages the received from social media as compared to government-sponsored messages. Participants also stated that they searched more actively for information about the fires and smoke information was more incidental so some people might have missed the smoke information. Participants recalled simple and understandable main messages, and most would use portable HEPA filters again with one primary motive to avoid symptoms caused by smoke.

-use simple messages that can be verified by the government and repeat these messages for smoke exposure as the fire continues

- specify short-term and long-term health effects and more detailed information

- advice should be more timely and practical

- better information about AQI data to increase the public’s understanding

- specify which messaging was more important, since participants found it hard to decide between smoke or wildfire hazards

 Olsen CS, et al. Communicating About Smoke from Wildland Fire: Challenges and Opportunities for Managers. 2014 [59].

Communications recommendations focused on (1) importance of consistent and coordinated smoke and communication management, (2) prioritizing and strategies to reach current and extended audiences, and (3) develop personal relationships with members of the affected public.

Case-study with 45–90 min semi-structured interviews (mostly 1:1 key informant, some small group) among individuals in 4 research locations to provide diversity in geography, ecology, and social conditions, as well as communication strategies, partnerships in smoke management, and experiences and challenges

N = 60 participants. California, Montana, Oregon, South Carolina. Purposive sample of critical managers to make decisions on smoke management or as key stakeholder engaged in fire and smoke management discussion.

Most common strategies to communicate to public: radio announcements, websites, hotlines, public meetings. Challenges to communicating about smoke: uncertainty about effectiveness of communication strategies (how to effectively communicate, do messages reach as many people as intended), confusion from inconsistent messages from different agencies (unclear messaging sometimes contradictory, connection to regulations where agencies can prescribe burns but public cannot), internal priorities about importance of communicating with stakeholders (priorities between people in organizations, some push for public involvement, some want to do only what is minimally required), improve relationships with public

- making public communication an institutional priority

- coordinating messages across and within agencies

- utilize social networks and optimize resources

- piggybacking off of existing communication programs: media, signage like billboards

 Riden HE, et al. Wildfire Smoke Exposure: Awareness and Safety Responses in the Agricultural Workplace. 2020 [60].

Examination of wildfire smoke events and the impacts on health and safety for the agricultural industry workforce, the existing safety practices related to poor air quality from wildfire smoke, and the workplace social dynamics that impact safety.

Semi-structured interviews were conducted with agricultural employers and focus group discussions were held with farmworkers in three regions of California. Telephone interviews were conducted with employers. Focus groups were conducted with farmworkers in the 3 California regions. Telephone interviews were in English, 30–90 min, not recorded, detailed notes taken with keywords noted verbatim. Focus groups convened in-person, in Spanish conducted by a bicultural native Spanish speaker with extensive experience conducting focus groups with the agricultural community, 45–68 min, recorded and transcribed. Use of Atlas Ti, primary and secondary codes, themes, systematic analysis.

N = 16 agricultural employers (English) and 9 focus groups (N = 7–10 participants each) of farmworkers (Spanish-speaking) in 3 areas of California. Employers: San Joaquin Valley (n = 6), Imperial Valley (n = 4), and Salinas Valley (n = 6). Focus groups: convened in the San Joaquin Valley (n = 4), Imperial Valley (n = 2), and Salinas Valley (n = 3).

Qualitative study findings included: Agricultural employers had varying knowledge about and experience responding to poor air quality due to wildfire smoke. Respirators or masks were not mentioned as a potential protective measure when describing their safety practices. Farmworkers reported experiencing poor air quality due to wildfire smoke, although knowledge of safety precautions varied. Farmworkers reported employer and supervisors’ attitudes toward safety as having the greatest impact on the implementation of workplace safety measures. Employers and workers had variable knowledge about smoke risk and mitigation strategies. In particular, masks were not always raised as a protection for workers.

- assignment of responsibility for ensuring a safe work environment; knowledge of existing safety precautions and sources for learning about current air quality conditions

- tailor safety messaging about wildfire smoke, health promotion, and workplace safety strategies for employers and employees, recognizing roles and power imbalances in the occupational context

- need for wildfire smoke exposure resources (i.e., how to find your air quality index or protect your workers) to assist employers and supervisors in their compliance with new emergency safety regulation

 Thomas M, et al. Unpacking the Realities and Complexities of Sensemaking: Government Practitioners’ Experiences of Emergency Risk Communication. 2021 [61].

Exploration of lived experiences of government practitioners’ experiences of emergency risk communication related sensemaking during public health emergencies. Sensemaking is the concept in the fields of organizational studies and emergency response that is defined as a cognitive process that primarily takes place in an individual’s head and as a social process that is carried out and constructed through interaction.

Semi structured interviews to explore the lived experiences of the ERC professionals, specifically with questions about their experiences for smoke events that occurred in the last decade in Victoria, Australia. Focused on experiences during smoke events to evaluate sensemaking to positive outcomes. Recruited via purposeful sampling from the Environment Protection Authority Victoria and subsequent snowball recruitment based on interviewees’ recommendations of individuals at the EPA Victoria, Department of Health and Human Services, Country Fire Authority, Department of Environment, Land, Water and Planning, and Emergency Management Victoria. Conducted between June 2019-March 2020 face to face at the office of each participant. Approximately 60 min. Recorded and transcribed. Analyzed using thematic analysis.

N = 15 interviewees working as ERC professionals (ERC = Emergency Risk Communication). Victoria, Australia. 10 female, 5 male, several to 25+ years of experience in 8 with science expertise (public health, environmental health, air quality), 5 in communication (emergency, community engagement), 2 in operations (emergency and firefighting). Environment Protection Authority Victoria (n = 8), Department of Health and Human Services (n = 4), Country Fire Authority (n = 1), Department of Environment, Land, Water and Planning (n = 1), Emergency Management Victoria (n = 1)

Balancing achievable plausibility with expected accuracy - it is hard for practitioners to send out messages when they are first presented with information and they don’t want to provide inaccurate information. The pressure of making sense - practitioners feel pressure due to the time-critical aspect of the job as well as pressure from organizational and practical settings and have competing demands to meet. Professional expert identities and roles delimiting sense making - many interviewees expressed that their job titles and roles were important delimit areas for sense making, but it was hard if there were new roles like in the case of the health risk officer. Past lived experiences facilitate sensemaking - those with more lived experiences were viewed as more able to understand/assess the current situation. Personal relationships aiding collective sensemaking - working with the same group of stakeholders and decision makers over time was helpful in creating trust within the group for sensemaking.

- communication during the early stages of an emergency should outline what the government knows, what they do not know, and what they are doing to address uncertainty. Communication should explain that advice is based on the current information, and is subject to change as new information is available

- operations-based preparedness activities (e.g. field exercises, drills) to increase the knowledge of individuals involved, develop risk communication material, and build relationships between inter- and intra-organizational stakeholders

- preparedness activities should involve a variety of scenarios and simulate the pressure practitioners experience when sensemaking for real-world ERC, incorporate scientific information, and include risk communication language discussions to decide on consistent wording

 Van Deventer D, et al. Wildfire Smoke Risk Communication Efficacy: A Content Analysis of Washington State’s 2018 Statewide Smoke Event Public Health Messaging. 2020 [62].

Assessment of the types and content of the messages communicated by local authority figures, local government agencies, and media channels. Messages were assessed for information on health risk, personal interventions, administrative interventions, vulnerable populations, and trusted sources of information.

Content analysis of messaging examples of wildfire smoke risk information from local and state government organizations and mainstream media in 8 counties in Washington State in 2018. This was a purposive sample. Deductive, qualitative content analysis methods were used to create a codebook and to classify messages. The Extended Parallel Process Model (EPPM) risk communication framework was used.

N = 273 messages sampled from websites (either subpages or articles). 8 counties in Washington State (represents 57% of the state’s population) chosen based on historic wildfire smoke impact and representation of urban, rural, and suburban jurisdictions in 8 counties.

A total of 85 government sources and 188 media postings were analyzed. Of the 273 total messages, summary statistics were calculated. 71% of government sources and 66% of media sources contained information about the health risks of wildfire smoke, respectively. Of the messages containing risk information, 53% and 48% of government and media sources, respectively, contained information about personal interventions, whereas 33% of government sources and 20% of media sources contained information about administrative interventions. For government and media sources containing risk information and personal intervention information, the most common personal interventions were reducing activity (84% and 65%, respectively) and staying inside (78% and 83%, respectively). No confidence intervals were given. Authors reported that HEPA filtration was not mentioned by any site and evacuation was not mentioned on government sites. Overall, the majority of wildfire smoke messages from government sources communicated health risks in the 2018 Washington State wildfire smoke event. About half of the messages communicated information about personal interventions to reduce smoke exposure, suggesting that they contain messaging elements theorized by the EPPM to promote behavior change. However, the types of interventions suggested, and vulnerable populations addressed, varied widely across messages, indicating that government and news media would benefit from increased coordination of information about the health risks of smoke exposure, including approaches to exposure reduction, and/or risk communication tools, templates, and resources.

- need input from more diverse populations including unhoused populations (for instance finding a clean air space that is open for unhoused)

- need for people to be referred to trusted websites for additional information

- increase coordination among government and news media about health risks of smoke exposure