Results of a Community-University Partnership to Reduce Deadly Hazards in Hardwood Floor Finishing

Results of a Community-University Partnership to Reduce Deadly Hazards in Hardwood Floor FinishingLenore S. Azaroff • Hoa Mai Nguyen • Tuan Do • Rebecca Gore • Marcy Goldstein-GelbSpringer Science+Business Media, LLC 2011Abstract A community-university partnership used community-based participatory research (CBPR) to design, implement, and evaluate a multi-cultural public health campaign to eliminate flammable products and reduce use of products high in volatile organic compounds (VOCs) in hardwood floor finishing in Massachusetts. Leading par¬ticipants were Vietnamese-American organizations and businesses. Following the public health campaign, a multi¬lingual survey of self-reported experiences with fires, product use, exposure to outreach activities, and changes made, was conducted with floor finishers. One hundred nine floor finishers responded. Over 40% reported fires at their companies’ jobs, mostly caused by lacquer sealers. Over one third had heard radio or TV shows about health and safety in floor finishing, and over half reported making changes as a result of outreach. Exposure to various out¬reach activities was associated with reducing use of flam-mable products, increasing use of low-VOC products, and greater knowledge about product flammability. However, most respondents still reported using flammable products. Outreach led by community partners reached large pro¬portions of floor finishers, was associated with use of safer products, and adds to recent work on CBPR with immigrant workers. Continued use of flammable products supports the belief that an enforceable ban was ultimately necessary to eradicate them.L. S. Azaroff (&). T. Do. R. Gore Work Environment Department, University of Massachusetts Lowell, One University Avenue, Lowell, MA 01854, USA e-mail: Lenore_Azaroff@uml.eduH. M. Nguyen Vietnamese-American Initiative for Development, Dorchester, MA, USAM. Goldstein-Gelb Massachusetts Coalition for Occupational Safety and Health, Dorchester, MA, USAKeywords Community-based participatory research. Immigrant workers. Environmental justice. Toxics use reduction. Occupational healthIntroductionProtecting worker health and safety in small businesses has long been an important challenge [1] that has become more urgent with the growth in low-wage, contingent, and informal employment. Increasing numbers of immigrants, in particular, hold non-standard jobs in hazardous indus¬tries, where forces that promote safety, such as government enforcement, labor union representation, and industry standards, are rare [2–6].One fruitful approach to improving such work envi¬ronments is community-based participatory research (CBPR). CBPR involves community, labor, and academic partners in jointly defining problems and solutions and assessing effectiveness [7]. A series of CBPR projects has been funded by the National Institute for Occupational Safety and Health (NIOSH), the Environmental Protection Agency, and the National Institute for Environmental Health Sciences (NIEHS) through the Environmental Jus¬tice Partnerships for Communication [8] and Partnerships for Environmental Public Health (PEPH) programs [9].Small firms often lack information about products, depend on suppliers for education, and choose controls based on custom [10, 11]. CBPR projects have sought to address these obstacles by reducing exposures to toxic substances in print shops [12], car repair [13], painting [14], and house cleaning [15]. The Dorchester Occupa¬tional Health Initiative (DOHI), funded by NIEHS through the Partnerships for Communication and PEPH programs, addressed hazardous products in hardwood floor finishing.DOHI focused on the Dorchester neighborhood of Boston, where most floor finishers are from Vietnam [16]. Partners were the Vietnamese-American Initiative for Development (Viet-AID); Massachusetts Coalition for Occupational Safety and Health (MassCOSH); New Ecol¬ogy, Inc. (NEI); the University of Massachusetts Lowell (UML); and Dorchester House, Codman Square, and Bowdoin Street Health Centers.Hazards of Floor FinishingDOHI identified little information about floor finishing in the scientific literature and found that floor finishers were too busy and suspicious of competitors to share their knowledge in planned focus groups. Two Boston-area house fires that killed three Vietnamese floor finishers and badly burned three others in 2004 and 2005 [17–19] then created a ‘‘teachable moment’’ that moved DOHI to con¬vene experts from industry, labor, and environmental pro¬tection who explained the following:• Vietnamese-American finishers often applied a first coat of an inexpensive, quick-drying product called ‘‘lacquer sealer.’’• Lacquer sealers are extremely flammable, with flash points as low as 9F and high volatility [20, 21].• Lacquer sealers are not designed for floor finishing and result in a poor quality finish. However, they initially produce an attractive amber color, require little skill to apply, and disguise errors in sanding.• Lacquers, as well as wood dust, rags wet with solvents, and a few other flammable products frequently combust.• A broad range of high quality, non-flammable products, both solvent-borne (‘‘oil-based’’) and water-borne, are available.• The sickening vapors from solvent-borne floor finishes can be avoided by using water-borne products and very-low-VOC solvent-borne products.• Proper application of water-borne products can produce a superior finish.The cause of fires perceived to be most urgent and most feasibly eliminated was lacquer sealers, which had started the two fatal fires [19]. (Lacquers are defined as ‘‘clear or pigmented wood finishes, including clear lacquer sanding sealers, formulated with nitrocellulose or synthetic resins to dry by evaporation without chemical reaction.’’ [22])Many floor finishing products contain volatile organic compounds (VOCs). Breathing VOCs can cause acute (feeling dizzy or ‘‘drunk,’’ headache, respiratory irritation)and long-term (neurological impairment, liver and kidney damage) health effects [23]. In addition, floor finishes are important sources of VOCs that contribute to ground level ozone [24–26]. Some finishes (including some low in VOCs [12]) also contain known carcinogens, asthmagens, or reproductive toxins, but several low-VOC products do not [27].Respirable wood dust causes respiratory irritation, der¬matitis, and cancer, and dusts from certain types of wood cause asthma [28]. Sanding painted floors can expose workers to lead dust. Sanding some floor finishing products from the 1950s and 60s can cause significant exposures to PCBs [29]. Dustless sanding systems, which typically cost several thousand dollars, are designed to dramatically reduce dust.Flammable or toxic dusts and vapors are particular risks for smokers, and about 30% of Vietnamese immigrants smoke. They are also three times more likely than other U.S. residents to not visit a doctor due to lack of money [30].Public Health Campaign to Prevent Hazards in Wood Floor FinishingDOHI learned that floor finishing products and practices were determined by a complex system (Figure 1). Partners launched a campaign targeting several levels.Societal and Industry LevelsMassCOSH recruited several product distributors, floor finishers, environmental experts, labor representatives, and community leaders to form the Massachusetts Floor Fin¬ishing Safety Task Force. Members agreed on the need toFig. 1 Simplified ecological model of factors affecting wood floor finishers’ products and practices, based on DOHI findings arrow indicates direction of influence exertedeliminate lacquers to prevent fires in floor finishing and explored a range of policy alternatives.All floor product distributors in Massachusetts known to the Task Force voluntarily halted sales of lacquer sealers to floor finishers for several months. Floor finisher demands then led one to resume lacquer sales, forcing the others to follow. This demonstrated the need for enforceable rules.The Task Force then moved to a legislative approach. One bill was filed with the Massachusetts legislature to require certification and safety training of floor finishers, and a second to prohibit the sale and use of flammable products in commercial wood floor finishing [31–34].The Task Force also proposed a regulation banning flammable floor finishing products to the Massachusetts Board of Fire Prevention Regulations. Based on feedback from the Board, the Task Force submitted an alternative regulation to require safety measures and permits for jobs using flammable products.Task Force members testified before the legislature and the Fire Prevention Board and visited legislators several times [35]. In June 2010, the Board passed the regulation requiring permitting and safety measures [36, 37]. In July, the legislature passed a bill banning the sale and use of lacquer sealers for wood floor finishing [38, 39].DOHI disseminated information to consumers via fact sheets, media releases, and presentations to homebuyer clas¬ses. Viet-AID collaborated with the Massachusetts Depart¬ment of Public Health (MDPH) in translating a Fatality Assessment and Control Evaluation (FACE) Alert describing the fatal fires and safety recommendations, and translating and designing it to make it appealing to Vietnamese floor finishers [40, 41]. MDPH sent over 3,000 copies of the FACE Alert to audiences including insurance agencies, building inspectors, property managers, and landlord associations. The Massa¬chusetts Department of Fire Services distributed copies to fire chiefs and to institutions affiliated with the Massachusetts Burn Injury Reporting System.The Task Force developed a Floor Finishing Safety Checklist that was distributed to floor finishing companies, product distributors, and product manufacturers [42].NEI identified quality as customers’ main concern [43]. In response, MassCOSH and NEI encouraged several agencies and organizations involved in building and maintenance to use non-flammable, low-VOC products to obtain high quality finished floors.Company and Individual LevelsViet-AID staff promoting free floor finishing health and safety trainings in Vietnamese discovered low awareness of the fatal fires and little perceived need for safety education. Almost no one attended DOHI’s first advertised trainings, which were canceled. Viet-AID staff then conducted aninformal assessment of contractors’ interests and learned that they most needed more business. Viet-AID placed adver-tisements and columns in three major Vietnamese-language newspapers, six announcements on popular Vietnamese-language radio stations, and three discussions on the local Vietnamese public affairs television program with infor¬mation about health and safety and advertising trainings on ‘‘Making Your Floor Finishing Business More Profitable.’’Industry experts in the Task Force designed training materials and curricula. Viet-AID identified one of just three certified master floor finishers in Massachusetts, a bilingual Vietnamese immigrant and talented instructor who volunteered as a trainer.Several aspects of the training were designed to over¬come barriers raised by floor finishers. When participants argued that water-borne products were too expensive, instructors guided them through analyses of time and labor costs of jobs using faster-drying water-borne products versus solvent-borne products. They explored the ‘‘high road’’ to competition, providing superior products and services that can enable contractors to charge more while building a good reputation and loyal customer base. In response to concerns that adopting unfamiliar products might harm quality and customer satisfaction, the sessions included hands-on practice in use of water-borne products, expert instructors, and trainers’ offers to travel to job sites with free technical assistance. Workers desiring additional practice attended advanced sessions at a corporate facility. Trainings helped dispel common beliefs, based on older formulations of water-borne products and incorrect appli¬cation of current ones, that they produce an inferior finish.Training participants viewed causes of floor finishing fires, discussed flash points, reviewed material safety data sheets (MSDSs), and received the FACE Alert.By 2010, 146 floor finishers had completed 5 h of training in use of environmentally preferable products.Viet-AID recruited participants to a planned Safer Floor Finishers Network. Network members must commit to give up lacquers and offer water-borne products as an option. The Massachusetts Toxics Use Reduction Institute (TURI) and Viet-AID designed websites marketing Network members.Assessment of ResultsThe main goal of the public health campaign was to reduce fires caused by lacquer sealers. Secondary goals included reducing health risks by promoting water-borne products. With no known data source on fires or product use, we relied on a survey of experiences with fires, products, exposure to outreach activities, changes made, and asso¬ciations between changes and exposure to these activities [44]. This was conducted prior to the banning of lacquers.MethodsQuestionnaire DesignThe interviewer-administered questionnaire included items about characteristics of workers surveyed and their busi¬nesses and types and frequencies of products used. Open-ended questions asked for reasons that respondents did or did not use certain products.Questions asked about exposure to outreach activities and other influences and whether the respondent had made chan¬ges as a result. Descriptions of such changes were requested.Additional questions asked about knowing of fires that had happened on other companies’ jobs or at jobs done by one’s own company; knowing others who feel dizzy or drunk when using floor finishing products or feeling such sensations oneself; having heard of the recent fatal fires; and familiarity with flash points.The questionnaire was translated into Vietnamese and Spanish and back-translated. It was field tested with four Vietnamese floor finishers and wording improved.Population SelectionViet-AID staff identified 418 businesses performing hard¬wood floor finishing listed in telephone and business directories in mainland central and eastern Massachusetts. Researchers attempted to call or visit each fourth business listed alphabetically. Most were no longer in operation, not present at listed addresses, or non-responsive. Researchers were able to interview only three people this way.Interviewers then set up survey stations at three floor supply stores, visited coffee shops where floor finishers gathered, and identified finishers through friends and family. In one city, an interviewer failed to contact all 10 floor finishing contractors listed in directories, but recruited 10 different respondents through personal connections. In addition, 16 respondents were recruited because they had participated in a Viet-AID training. Interviews were con¬ducted between June 2009 and March 2010.A gift card was offered for participation. Each potential respondent was read an informed consent form in his or her preferred language. No personal identifying information was collected. All procedures were approved by the Insti¬tutional Review Board for the Protection of Human Sub¬jects at UML.Data AnalysisData were entered into a database in Microsoft Access. Open-ended responses were categorized by inspection. Associations between predictors and outcomes were expressed as prevalence ratios calculated by log-binomialmodels [45] using proc genmod in SAS version 9.2 (SAS Institute, Inc., Cary, North Carolina).ResultsRespondents (Table 1)One hundred nine finishers participated. About three quarters spoke Vietnamese. The great majority (83%) worked for small companies, with a maximum of five employees when busiest.Knowledge and Experiences (Table 1)Of 98 respondents, 64 said they knew others who had had fires on the job. Of 91 who answered questions about fires at their own companies’ jobs, 39 (43%) said that they knew of such fires. With the option of reporting multiple causes, 33 reported lacquers as causes of fires, 16 named wood dust, and 10 named used rags.Two thirds of respondents said that they were aware of the recent fatal firesFifty-two of 104 respondents said they would be able to find a product’s flash point. Forty named correct locations for locating a flash point, such as the material safety data sheet (MSDS) or product label.Twenty-four of 103 respondents reported that they sometimes felt dizzy or drunk when using floor finishing products, and 16 said they knew others who feel this.Products (Table 2)Two thirds of respondents reported using lacquer sealers for at least some jobs, though nearly half said they used them for ‘‘just a few’’ jobs. The most common reason given for using lacquers was that they look nice (37% of 89); the most common reason given for limiting their use was the risk of fire or explosion (30%).Most (85%) reported using water-based products for at least some jobs, though two thirds said they used them for ‘‘just a few’’ jobs. The greatest number (23% of 90) cited customer requests as a reason for using them, while 10% cited customer requests as a reason for not using them. Seventeen percent said they used water-based products because they are safer or non-flammable and 13% because they are easy or quick to use, while nine percent reported avoiding them because they are difficult or slow to use. Ten percent gave quality or durability as reasons for using them, while 13% gave poor quality or appearance as a reason for not using them.One third of respondents reported using dustless systems for at least some jobs. Of the others, most said that these systems are too large or expensive or their current practices controlled dust adequately.Table 1 Floor Finishers Surveyed, n = 109Number (%)Business characteristicsProximity to Boston, n = 106 respondents Close 83 (78) Far (Worcester or Essex Counties) 23 (22)Speak Vietnamese, n = 109 Yes 82 (75) No 27 (25)Other Employees Speak Vietnamese, n = 108 Yes 82 (76) No 24 (22) Don’t know 2 (2)Years Experience in Floor Finishing, n = 96 1–5 years 37 (39) 6–10 years 34 (35) 11–15 years 10 (10) 16 ? years 15 (16)Number Employees when Busiest, n = 9712 (12) 1–5 69 (71) 6–10 12 (12) 11–15 4 (4)Experience with firesKnow Other Finishers Who Have Had Fires, n = 98 Yes 64 (65) No 29 (30) Don’t know 5 (5)Know of Fires on Own Company’s Jobs, n = 91 Yes 39 (43) Caused by lacquers 33 (36) Caused by used rags 10 (11) Caused by wood dust 16 (18) Other cause 2 (2) No 51 (56) Don’t know 1 (1)Heard of Fatal Fires, n = 105 Yes 70 (67) No 35 (33)Reasons for not using Expensive 10 (11) Customers don’t want 9 (10) Difficult or slow to use 8 (9) Looks bad 6 (7) Poor quality 6 (7) Lack of familiarity 4 (4)Reasons for using Customers request 21 (23) Safer or non-flammable 15 (17) Easy or quick to use 12 (13) Durable/good quality 9 (10) Avoid health effects or odors 3 (3)Exposure to Outreach Activities (Table 3)Main outreach activities targeting floor finishers were radio programs, TV programs, and newspaper articles. For each of these, 41–46% of 79 Vietnamese-speaking respondents reported having seen or heard this outreach.Table 2 Products used by floor finishers surveyed, n = 109Reasons for using or not using water-based products, n = 90Dustless systemsUse dustless systems, n = 100 No jobs 67 (67) Just a few jobs 14 (14) Most jobs 8 (8) All jobs 11 (11)Twenty-two Vietnamese-speaking respondents had par¬ticipated in a Viet-AID training. Thirty respondents had seen the FACE Alert. Of these, 15 had not attended Viet-AID trainings. Twenty-one reported seeing English news¬paper articles with floor finishing safety information.Sixty-three percent of Vietnamese speakers and 26% of non-Vietnamese speakers reported hearing information fromTable 3 Floor Finishers’ Exposure to Outreach Activities or Input, n = 106other floor finishers about health and safety. Twenty-three percent of Vietnamese speakers and 48% of non-Vietnam-ese-speakers reported that customers had asked them to use or to avoid products because of health or safety.Reported ChangesFifty-five of 95 respondents reported that they had made a change as a result of outreach activities or other input. Fifteen said that they had reduced use of lacquers, and 14 that they had stopped using lacquers. Fourteen reported increasing use of water-based products. Seven said they stopped smoking near the job site, and two that they avoided turning light switches on and off when applying flammable products. Fifteen reported changes that had not been promoted in the outreach activities: opening windows, using personal pro¬tective equipment, or just ‘‘being careful.’’Factors Associated with Products and Practices (Table 4)Knowing of a fire at a job done by one’s own company and speaking Vietnamese were independently associated with several outcomes and predictors. Unless otherwise noted, models were adjusted for these factors. To identify pre¬dictors of changes among people who knew of fires at their companies’ jobs, as well as those who did not, results were stratified by this factor. Associations reported below were statistically significant at P 0.10.Factors Associated with ProductsVietnamese speakers were more than three times as likely as others to report using lacquers for at least some jobs (unadjusted PR 3.2, 95% CI 1.6, 6.4). To investigate fac¬tors associated with limiting lacquer use, the responses ‘‘don’t use’’ and use ‘‘for just a few jobs’’ were combined. Vietnamese speakers were less likely to use lacquers for noor just a few jobs (PR 0.80, 95% CI 0.64, 0.99). Two factors were associated with limited lacquer use for all respondents: attending a Viet-AID training (PR 1.5, 95% CI 1.0, 2.3) and citing durability or quality as a reason for using water-based products (PR 1.4, 95% CI 0.94, 2.0). A factor associated with reduced prevalence of limited lac¬quer use was saying that a reason for using lacquers is that they look nice (PR 0.48, 95% CI 0.31, 0.77). Among respondents who knew of fires at their companies’ jobs, seeing the FACE Alert was associated with this outcome (PR 1.6, 95% CI 0.99, 2.7).Factors associated with using water-based products for all or most jobs were citing speed or ease of use as a reason for using water-based products (PR 4.2, 95% CI 1.4, 13) and citing customer requests as a reason for using lacquer sealers (PR 3.8, 95% CI 1.1, 13). Hearing customer requests regarding products was associated with a reduced prevalence of using water-based products for all or most jobs (PR 0.30, 95% CI 0.093, 0.95). Among respondents who knew of fires at their companies’ jobs, citing durability and quality as a reason for using water-based products was the only factor associated with using them for all or most jobs (PR 3.8, 95% CI 1.0, 14). For others, both hearing from customers about products (PR 0.20, 95% CI 0.050, 0.83) and hearing from other finishers about health and safety (PR 0.23, 95% CI 0.050, 1.1) were associated with reduced prevalences of using water-based products for all or most jobs.Factors Associated with Reported ChangesRespondents whose companies had had a fire (unadjusted PR 1.8, 95% CI 1.2, 2.6) and Vietnamese speakers (unadjusted PR 2.3, 95% CI 1.2, 4.6) were more likely than others to report making a change.For all respondents, factors associated with making any change were hearing radio programs (PR 1.8, 95% CI 1.1,Table 4 Associations between floor finishers’ experiences/knowledge/attitudes and main outcomes prevalence ratios (95% confidence intervals)1232.7), hearing from other finishers (PR 1.8, 95% CI 1.0, 3.3), and seeing any articles (PR 1.5, 95% CI 0.93, 2.5).Among those whose companies had had a fire, no pre¬dictors were associated with making a change. Among others, predictors associated with making changes were hearing from other finishers (PR 5.2, 95% CI 1.2, 20), correctly identifying locations of flash points (PR 3.4, 95% CI 1.2, 10), hearing from customers (PR 3.2, 95% CI 1.1, 9.3), seeing the FACE Alert (PR 2.3, 95% CI 1.2, 4.2), hearing radio programs (PR 1.9, 95% CI 1.1, 3.3), and seeing any articles (PR 1.7, 95% 0.96, 3.2).Respondents whose companies had had a fire were more likely to report either reducing or stopping use of lacquers (unadjusted PR 2.9, 95% CI 1.2, 6.7). For all respondents, a factor associated with reducing or stopping use of lacquer sealers was seeing Vietnamese articles (PR 2.5, 95% CI 1.1, 5.7). Among those whose companies had had a fire, factors associated with reducing or stopping lacquer use were attending a Viet-AID training (PR 1.9, 95% CI 0.93, 3.9) and seeing Vietnamese articles (PR 3.0, 95% CI 1.0, 8.5). For others, no factors were significantly associated with reducing or stopping lacquer use.Factors associated with adding water-based products were seeing the FACE Alert (PR 6.2, 95% CI 1.9, 20), citing avoidance of fires as a reason for using water-based products (PR 11, 95% CI 1.5, 80), and attending a Viet-AID training (PR 3.8, 95% CI 1.3, 11). A factor associated with reduced prevalence of adding water-based products was the belief that lacquers look nice as a reason for using them (PR 0.26, 95% CI 0.062, 1.1).Among respondents who knew of fires at their compa¬nies’ jobs, no factor was associated with increasing use of water-based products. Among others, predictors of adding water-based products were citing avoidance of fires as a reason for using water-based products (PR 7.6, 95% CI 1.1, 53), seeing the FACE Alert (PR 7.0, 95% CI 1.7, 28), and citing customers’ requests as a reason to use water-based products (PR 2.6, 95% CI 0.89, 7.3).Predictors that covaried were included in models toge¬ther. When adjusted for knowing of a fire at a job done by one’s company, the best model for adding water-based products included both attending a Viet-AID training (PR 3.1, 95% CI 1.2, 8.2) and citing safety as a reason for using water-based products (PR 8.5, 95% CI 1.1, 63).Factors Associated with Knowledge and AttitudesFactors associated with respondents saying that they could locate product flash points were hearing from other fin¬ishers (PR 2.5, 95% CI 1.3, 5.2), attending a Viet-AID training (PR 2.2, 95% CI 1.1, 4.3), seeing the FACE Alert (PR 1.9, 95% CI 1.2, 2.9), sometimes feeling drunk or dizzy when using floor finishing products (PR 1.7, 95% CI1.2, 2.6), and hearing customers request products (PR 1.5, 95% CI 0.96, 2.4). For those whose companies had had a fire, these were seeing any articles (PR 3.8, 95% CI 1.3, 11), seeing the FACE Alert (PR 3.4, 95% CI 1.4, 8.2), hearing from other finishers (PR 2.5, 95% PI 0.87, 7.0), citing avoidance of fires as a reason for using water-based products (PR 2.4, 95% CI 1.0, 5.9), feeling drunk or dizzy (PR 2.2, 95% CI 1.1, 4.3), and seeing Vietnamese articles (PR 2.1, 95% CI 0.97, 4.7). For others, they were hearing from other finishers (PR 2.5, 95% CI 0.97, 6.5), and hearing from customers (PR 2.4, 95% CI 1.1, 5.3).Exposures to certain outreach activities were associated with avoiding lacquers because they cause fires or explo¬sions. Among Vietnamese speakers, this was attending a Viet-AID training (PR 4.1, 95% CI 2.1, 7.9). Among all respondents, the PR for this belief was 2.6 for those who had seen the FACE Alert (95% CI 1.4, 4.7). Attending the Viet-AID training was associated with a reduced preva¬lence of stating that they used lacquer because it looks nice (PR 0.094, 95% CI 0.014, 0.64).Exposures to certain outreach activities were associated with respondents reporting that they used water-based products because they are safer. These were attending a Viet-AID training (PR 2.3, 95% CI 1.5, 3.6), and seeing the FACE Alert (PR 2.2, 95% CI 1.2, 3.5).Some experiences were associated with reporting that they used water-based products because they are durable or good quality. Among Vietnamese speakers, the PR for stating this belief was 5.0 for those who had seen Viet¬namese articles (95% CI 1.1, 22). Among all respondents, the PR was 10 for those who had seen any articles (95% CI 1.3, 80), 8.6 for those who had seen TV programs (95% CI 1.9, 39), 7.5 for those who had seen the FACE Alert (95% CI 1.7, 34), and 3.3 for those who had attended a Viet-AID training (95% CI 0.99, 11).DiscussionCommunity, labor, government, and industry partners defined the source of a deadly problem, developed feasible solutions, identified influences on product choices, and designed corresponding outreach and policy initiatives. This fit the classic definition of a public health communi¬cation campaign: ‘‘… an integrated series of communica¬tion activities, using multiple operations and channels, aimed at populations or large target audiences usually of long duration purpose.’’ [46] We do not propose that information alone motivated or enabled floor finishers to change products. Thus, associations between exposure to outreach and changes do not imply simple causation, but help describe the process of change.Limitations of the StudyWithout other data sources, this research relied on a survey, which suffers from reliance on self-reports with no vali¬dation. Researchers attempted and failed to execute a sur¬vey sampling procedure. Instead, interviewers found floor finishers where they gathered and used personal networks. This is consistent with other community-based projects, in which, for example, indigenous community educators surveyed farm workers at homes, labor camps, and com¬munity centers [47], or Vietnamese-American residents interviewed members of personal networks about their occupational health [48].The lack of an accurate list of businesses precluded the enumeration of eligible respondents. We estimate the number of floor finishers in the target area at 500–1,500, meaning we sampled between 22% and 7%. There is no apparent reason to believe that respondents differed sys¬tematically from non-respondents. Therefore, we believe that the survey is meaningfully representative of hardwood floor finishers in eastern and central Massachusetts. The exception was the 16 respondents who were included because they had participated in Viet-AID trainings, so might differ in some ways from the underlying population and were more likely to have seen FACE Alerts, which were distributed at the trainings.Uncovering Elusive DataStrikingly, 43% of respondents admitted knowledge of fires at jobs performed by their companies, mostly caused by lacquers. This finding supported anecdotal information from Task Force members about fires as a serious and frequent problem in the industry. Researchers were not able to locate documentation of such fires in other data sources, and news media typically cover only the few tragic events involving loss of life or massive property damage.The association between speaking Vietnamese and using lacquer sealers was also consistent with statements by Task Force members that these were especially marketed to Vietnamese immigrants. Results indicating that many people chose to use lacquers because they look nice also matched anecdotal information from the Task Force.Effects of the Public Health CampaignOutreach activities reached large proportions of the target population, with about one third of respondents reporting exposure to the FACE Alert or TV shows, radio, or articles with information on health and safety. More than half reported making changes as a result of such input, and more than a quarter reported reducing or stopping use of lacquers. Several associations between changes andoutreach suggest causality. For example, both seeing the FACE Alert and attending Viet-AID trainings were strongly associated with adding water-based products. (The FACE Alert did not promote the use of water-based products, only non-flammable products generally, but some respondents reported using water-based products because they are not flammable.)Based on the model of fires creating a ‘‘teachable moment’’ in which floor finishers might be open to alter¬natives, we examined knowing of a fire at a job done by one’s company as an effect modifier of associations between exposure to outreach activities and changes. Associations between seeing the FACE Alert and using lacquer for all or just a few jobs, and between attending Viet-AID trainings and stopping or reducing lacquer use, were in fact larger for those who knew of fires at their companies’ jobs. Some other associations were much lar¬ger between exposure to certain inputs and making changes for those who did not know of such fires, an interesting result inconsistent with our model.Results reinforced suggestions by customers and floor finishers that messages about products should be framed in terms of quality and business advantages, a concept seen in other projects [49]. The perception that water-based prod¬ucts are quick and easy to use, or durable and good quality, were strongly associated with using water-based products for all or most jobs. However, the perception that water-based products are safer or non-flammable was also strongly predictive of increasing their use.Outreach activities may have been especially effective at fostering the perception that water-based products pro¬duce a good quality finish, with people who reported exposure to any one of several outreach activities several times more likely than others to state this as a reason for using them.Respondents reporting exposure to any of several outreach activities were also much more likely to express key concepts such as risk of fire and explosions as a reason for avoiding lacquers and correct locations of product flash points.Perhaps most importantly, results demonstrate that vol¬untary measures such as product withdrawal by some vendors or educational initiatives were insufficient to halt the use of a hazardous and unnecessary product. Two thirds of respondents to the survey, administered before the leg¬islative ban, reported using lacquer sealers for at least some jobs. This reflects the consensus among Task Force mem¬bers that legislation was needed to enable businesses to refuse to sell or use a product that is temptingly cheap and convenient.This project adds to the growing body of literature highlighting the potential of synthesizing the expertise of community, labor, industry, and researchers to prevent hazards affecting vulnerable workers in small businesses.Acknowledgements The authors would like to thank Lap Le, Van Lan Truong, Marcela Villarroel, and Nisha Narvekar, who per¬sisted in conducting the survey in challenging environments. We are grateful to Ed Connelly and Rachel Nania of New Ecology, Inc. for their expert communications with consumers and developers. Lee Weber kindly opened her businesses to field researchers. Cora Roe¬lofs and David Kriebel provided invaluable input into study design and presentation of results. This work would not have been possible without the determination and time provided by Karen van Unen, Stephen M. Tringale, and the Dorchester House Multi-Service Center and Codman Square Health Center in obtaining support and over¬coming countless administrative obstacles. This research was funded by the National Institute for Environmental Health Sciences grants 3R25ES012585 and 3R25ES012585-04S1.References1. Rosenstock, L., Olenec, C., & Wagner, G. (1998). The National Occupational Research Agenda: a model of broad stakeholder input into priority setting. American Journal of Public Health, 88(3), 353–356.2. Kalleberg, A. L. (2009). Precarious work, insecure workers: employment relations in transition. American Sociological Review, 74(1), 1–22.3. Smith, B. E., & Winders, J. (2008). ‘We’re here to stay’: Eco¬nomic restructuring, Latino migration and place-making in the US South. Transactions of the Institute of British Geographers, 33(1), 60–72.4. DeFilippis, J., Martin, N., Bernhardt, A., & McGrath, S. (2009). On the character and organization of work in the cities of the US. Urban Geography, 30(1), 63–90.5. Azaroff, L. S., Lax, M. B., Levenstein, C., & Wegman, D. H. (2004). Wounding the messenger: The new economy makes occupational health indicators too good to be true. International Journal of Health Services, 34(2), 271–303.6. Gong, F., Baron, S., Ayala, L., Stock, L., McDevitt, S., & Hea¬ney, C. (2009). The role for community-based participatory research in formulating policy initiatives: promoting safety and health for in-home care workers and their consumers. American Journal of Public Health, 99(Suppl 3), S531–S538.7. Cook, W. K. (2008). Integrating research and action: a systematic review of community-based participatory research to address health disparities in environmental and occupational health in the United States. Journal of Epidemiology and Community Health, 62(8), 668–676.8. Baron, S., Sinclair, R., Payne-Sturges, D., Phelps, J., Zenick, H., Collman, G. W., et al. (2009). Partnerships for environmental and occupational justice: contributions to research, capacity and public health. American Journal of Public Health, 99(S3), S517– S525.9. Birnbaum, L. S., Zenick, H., & Branche, C. M. (2009). Envi¬ronmental justice: A continuing commitment to an evolving concept. American Journal of Public Health, 99(Suppl 3), S487– S489.10. Schulte, P. A., Okun, A., Stephenson, C. M., Colligan, M., Ahlers, H., Gjessing, C., et al. (2003). Information dissemination and use: critical components in occupational safety and health. American Journal of Industrial Medicine, 44(5), 515–531.11. Oldershaw, P. (2001). Control banding–a practical approach to judging control methods for chemicals. Journal of Preventive Medicine, 9(4), 52–58.12. Sutton, P., Wolf, K., & Quint, J. (2009). Implementing safer alternatives to lithographic cleanup solvents to protect the healthof workers and the environment. Journal of Occupational and Environmental Hygiene, 6(3), 174–187.13. Bra˚tveit, M., Hollund, B. E., & Moen, B. E. (2004). Reduced exposure to organic solvents by use of water-based paint systems in car repair shops. International Archives of Occupational and Environmental Health, 77(1), 31–38.14. Materna, B. L., Harrington, D., Scholz, P., Payne, S. F., Stubbs,H. A., Hipkins, K., et al. (2002). Results of an intervention to improve lead safety among painting contractors and their employees. American Journal of Industrial Medicine, 41(2), 119–130.15. Gute, D. M., Siqueira, E., Goldberg, J. S., Galva˜o, H., Chianelli, M., & Pirie, A. (2009). The Vida Verde Women’s Co-Op: Bra¬zilian immigrants organizing to promote environmental and social justice. American Journal of Public Health, 99(Suppl 3), S495–S498.16. Azaroff, L. S., Doan, T., Nguyen, H., Goldstein-Gelb, M., Fraser-Cook, M., & Kota, S. (2006). Protecting workers and residents from wood floor-finishing hazards. New Solutions, 16(2), 119–138.17. Ellement, J., Smalley, S. (2004). Another floor sander dies; eyeing of regulations urged. Boston Globe September 4, 2004; Third Edition, A1.18. Zaremba, J. (2005) Fumes ignite, killing Hull man: sealing agent blamed for fire at D Street house where floor refinisher, 43, died. The Patriot Ledger July 04, 2005.19. NIOSH FACE report Floor Fatality Assessment and Control Evaluation (FACE) Program Sander Dies When Wood Floor Refinish Product Ignites Massachusetts Case Report: 05-MA-044 Release Date: March 31, 2006 Available at www.cdc.gov/niosh/ face/stateface/ma/05ma044.html. Accessed December 8, 2010.20. Condie, R. G. (1991). Solvents can be deadly, so choose alter¬natives. The Toronto Star November 17, 1991; E1.21.Woodfloordoctor.com. Lacquer finish floor fires. Available at www. woodfloordoctor.com/_product_reviews/articles/lacquerfinishfloor fires.shtml, accessed September 7, 2005.22. South Coast Air Quality Management District (AQMD). Rule 1113. Architectural Coatings. Amended July 13, 2007. Available at http://www.aqmd.gov/rules/reg/reg11/r1113.pdf. Acces¬sed September 18, 2010.23. Environmental Protection Agency. An introduction to indoor air quality (IAQ): Volatile organic compounds. available at www.epa.gov/iaq/voc.html#Health Effects. Accessed November 12, 2010.24. South Coast Air Quality Management District (AQMD). Volatile Organic Compounds (VOC) Emissions from Paint and Archi¬tectural Coatings. Available at www.aqmd.gov/prdas/coatings/ vocs.htm. Accessed September 18, 2010.25. Avery, R. J. (2006). Reactivity-based VOC control for solvent products: More efficient ozone reduction strategies. Environ¬mental Science and Technology, 40(16), 4845–4850.26. South Coast Air Quality Management District (AQMD). (2010). Final 2007 Air Quality Management Plan. Diamond Bar, CA: June 2007. Available at http://www.aqmd.gov/aqmp/07aqmp/ index.html, accessed September 18, 2010.27. Green Seal. (2005). Wood finishes and stains. Green Seal Choose Green Report, February 2005. Available at www.wbdg.org/ccb/ GREEN/REPORTS/cgrwoodfinish.pdf. Accessed December 8, 2010.28. National Institute for Occupational Safety and Health. (2005). Wood dust. In NIOSH pocket guide to chemical hazards. NIOSH Publication Number 2005-149, p. 335; 2005.29. Rudel, R. A., & Perovich, L. J. (2009). Endocrine disrupting chemicals in indoor and outdoor air. Atmospheric Environment, 43(1), 170–181.30. Centers for Disease Control and Prevention (CDC). (2004). Health status of Cambodians and Vietnamese–selected commu¬nities, United States, 2001–2002. MMWR Morb Mortal Wkly Rep, vol. 53(33), 760–765.31. Mishra. R. (2005). Sanders’ deaths yet to bring changes: Safety proposals sit for months. Boston Globe September 2, 2005.32. Ross, C. (2006) More safety urged in floor finishing bill. Boston Herald May 17, 2006.33. Nichols, R. (2006). Panel targets floor finishing: seeks regula¬tions, safer sealants. Boston Globe.May 17, 2006.34. Occupational Hazards. (2006). Mass.: Deaths of floor finishing workers spark safety movement. Occupational Hazards online June 7, 2006. Available at www.occupationalhazards.com/ Classes/Articles/ArticleDraw_P.aspx. Accessed July 3, 2008.35. Reardon, N. (2009). Ban on flammable floor sealers considered. The Patriot Ledger (Quincy, MA) July 10, 2009, p. 2.36. Gloucester Daily Times, Mass. (2010). New fire mandates target floor finishing, flammables. Gloucester Daily Times (Massachu¬setts) June 12, 2010.37. Code of Massachusetts Regulations 57 CMR: Board of Fire Prevention Regulations 10.15 Floor Finishing or Refinishing.38. Reardon, N. (2010). New statute outlaws flammable floor fin¬ishes. The Patriot Ledger (Quincy, MA) July 20, 2010.39. Massachusetts General Laws Chapter 94 Section 329 Sale, pos¬session, or use of lacquer sealer prohibited; penalties. Text of section added 2010.40. Occupational Health Surveillance Program Massachusetts Department of Public Health, Office of the State Fire Marshal Department of Fire Services. (2006). Wood floor sanders killed when floor finishing product catches fire. Boston, Massachusetts: Massachusetts Department of Public Health.41. Occupational Health Surveillance Program Massachusetts Department of Public Health, Office of the State Fire MarshalDepartment of Fire Services. (2006). Nhung Nguoi Da´nh Bo´ng Sa`n Nha` Bang Go Da˜ Bi Chet Khi San Pham Da´nh Bo´ng Sa`n Nha` Boc Cha´y-O Tieu Bang. Boston, Massachusetts: Massachusetts Department of Public Health.42. Massachusetts Floor Finishing Safety Task Force. (2006). Mas¬sachusetts wood floor finishing daily safety checklist. Dorchester, Massachusetts: Massachusetts Coalition for Occupational Safety and Health.43. Nania, R. (2009). The safer floor finishing initiative. Master’s thesis. Boston: Emerson College.44. Barbeau, E. M., Li, Y., Calderon, P., Hartman, C., Quinn, M., Markkanen, P., et al. (2006). Results of a union-based smoking cessation intervention for apprentice iron workers (United States). Cancer Causes and Control, 17(1), 53–61.45. Deddens, J. A., & Petersen, M. R. (2008). Approaches for esti¬mating prevalence ratios. Occupational and Environmental Medicine, 65, 501–506.46. Flay, B. R, & Burton, D. (1990). Effective mass communication strategies for health campaigns. In C. Atkin, & L. Wallack (Eds.), Mass communication and public health: Complexities and con¬flicts (p. 130). Newbury Park, CA: Sage (129–146).47. Samples, J., Bergstad, E. A., Ventura, S., Sanchez, V., Farquhar,S. A., & Shadbeh, N. (2009). Pesticide exposure and occupational safety training of indigenous farmworkers in Oregon. American Journal of Public Health, 99(Suppl 3), S581–S584.48. Roelofs, C., Azaroff, L. S., Holcroft, C., Nguyen, H., & Doan, T. (2008). Results from a community-based occupational health survey of Vietnamese-American nail salon workers. Journal of Immigrant and Minority Health, 10(4), 353–361.49. May, J., Hawkes, L., Jones, A., Burdick, P., Ginley, B., Santiago, B., et al. (2008). Evaluation of a community-based effort to reduce blueberry harvesting injury. American Journal of Indus¬trial Medicine, 51(4), 307–315.