A qualitative investigation of Hispanic construction worker perspectives on factors impacting worksite safety and risk

In the United States, Hispanic construction workers have had consistently higher work-related death rates compared to non-Hispanic construction workers [1]. In 2009, Hispanic workers experienced the highest rate of work-related fatal injuries at 3.7 incidents per 100,000 full time equivalent workers, compared to 3.4 for whites and 3.0 for blacks [2]. Dong's study [3] of more than 7,000 construction workers' medical records found that Hispanics were nearly 30% more likely to have medical conditions due to work-related injuries than white, non-Hispanics, after controlling for occupation, gender, age, and education. Their injuries were also more likely to be severe enough to cause lost workdays. Hispanic construction workers number almost 3 million and hold 30% of non-management jobs in construction [4]. These workers generally have limited-English language ability and lower education levels than other construction workers. They also are more likely to hold the more hazardous jobs within the construction industry. Despite a declining trend in work-related deaths overall, among Hispanic workers such fatalities continued to rise until the crash of the construction industry in 2009 [5].

Several investigators have examined, or have attempted to explain, the reasons for the disparity in injury rates. Language barriers, cultural differences, lack of safety training, economic disadvantage, lack of construction experience, and relegation to the most dangerous jobs within construction have all been suggested as reasons behind the rates [6–10]. Dong and Platner [1, 3, 11] note that Hispanics are over-represented in jobs with the more hazardous conditions laborers, helpers, and roofers face, and that "these dangerous occupations typically offer fewer opportunities for worker control of the work process, less on-the-job training, less job security, and lower average wages." One of the most prominent of the reports on Hispanic workers' greater risk cited "inadequate knowledge and control of recognized safety hazards and inadequate training and supervision of workers, often exacerbated by different languages and literacy levels of workers" as contributory factors based on federal and state investigations of the deaths of 200 Hispanic workers [7].

Language barriers, i.e., communication challenges resulting from lack of a common language between employers and employees, are often cited among the reasons for the higher rates and/or as a key factor to address in order to ameliorate the disparity. Menendez and Havea [8], in response to the growth in the proportion of construction fatalities among Hispanic construction workers, noted that Hispanic construction workers may need more language and literacy-appropriate safety training and information about their rights. Pransky et al. [9] hypothesized that the increased risk was potentially attributable to several factors, including: limited economic and political resources, language and cultural barriers, relegation to the most hazardous jobs, language difficulties, and workplace discrimination that may result in inadequate safety. However, in their study among non-agricultural Hispanic workers, they did not find that injured workers were less likely to speak English, to have received safety training, or to have received safety training in Spanish than non-injured Hispanic workers.

Goodrum and Dai [10] attributed higher Hispanic injury rates to the limited effectiveness of traditional training techniques due to Hispanic construction workers' lower education levels and more limited ability to understand English language and American culture. However, a language only becomes a barrier to safety training when training is provided. McGlothlin's survey [12] of Hispanic construction workers in New Orleans found that 57% had never received any safety training. O'Connor [13] assessed the adequacy of safety training provided to young Latino immigrant construction workers (14-21 years old) and found that, although the workers frequently performed tasks that posed significant injury risks, one quarter of respondents received no safety training at all, and approximately another quarter received less than an hour of training.

Forst et al's hospital-based interviews [14] with injured Hispanic workers (more than half of whom worked in construction) found that the injured workers provided the following explanations for their injuries: unsafe working conditions, including not understanding or being trained about the hazards, problems with personal protective equipment (not provided or defective), being overworked, and carelessness on their own part. Another hospital-based interview study found that Hispanic construction workers were more likely than blacks or whites to report that "nothing" could have prevented their injuries [15].

To better understand and address these issues, we conducted an exploratory qualitative research study as part of a larger community-based participatory research project. The project, Protección en Construcción: The Lawrence Latino Safety Partnership, is a National Institute for Occupational Safety and Health-funded study conducted by the University of Massachusetts Lowell, Laborers International Union of North America Local 175, the City of Lawrence, MA Mayor's Health Task Force and Lawrence Community Connections, a health and community development non-profit organization. Focus groups were conducted with both union and non-union workers to explore their perceptions of work-related hazards and potential factors associated with the disproportionate number of fatal and non-fatal accidents and exposures experienced by Latino construction workers (we use the terms "Latino" and "Hispanic" interchangeably). An additional aim of this research was to identify potential intervention elements. Finally, the intention of this work was to cultivate relationships with workers and to build community networks in preparation for the development of an intervention to reduce hazards. We have found focus groups to be successful in research aimed at building and strengthening community relationships (Sprague Martinez, L.S., Freeman, E., Peréa, F.C., From engagement to action: Assessing community readiness for disparities mobilization. Submitted manuscript).

Two focus groups of three consecutive sessions for Hispanic construction workers - one for union and the other non-union workers - were held in April-May 2009. Participants came three nights in a row for two hours each night. This study used a modified convenience sample to select participants. A list of Spanish-speaking members of the labor union project partner was provided to University research staff who randomly selected workers from the list and invited 12 to participate. Eight workers participated in the union focus group at the Local's hall. Non-union Hispanic construction workers were recruited by the project's Hispanic community health outreach staff, the "promotores," who live and work in the Lawrence, MA community. The promotores identified potential participants through their community networks (e.g., neighborhoods, churches, family friends) and invited 12 to participate. Ten Hispanic construction workers participated in the non-union focus groups at a local senior center. All participants were male and were born in either Puerto Rico or the Dominican Republic. Focus group participants were given $75 for participating in the study over the three nights.

Participants authorized our recording of the sessions and agreed to participate through an oral consent process approved by the University of Massachusetts Lowell Institutional Review Board. All focus group discussions occurred in Spanish without English translation. One interviewer led the discussion from an interview guide that had been developed through our community-based participatory research process. That process included solicitation of desired themes from the project partners and other community members. The university researchers drafted the questions and translated them from English to Spanish, and these were again reviewed and revised by project members. The draft questions were piloted with three non-union construction members who considered the value and clarity of the questions. We again revised the interview guide, and it was approved by the project steering committee. In addition to the discussion leaders, a bilingual note taker participated in the focus groups, and served as the translator of the transcripts.

Our goal was to create an environment where questions could be answered honestly and deeply. Many comments and themes were repeated over the course of the three nights, despite the changing of the questions night to night. The first night's questions asked about falls they had experienced or knew about and causative factors. We probed further on the role of supervisors and contractors in health and safety on site, equipment and training. The second night's questions focused on prevention and safety practices and programs and how we could encourage participation in our intervention program. On the third night we asked about the role of others in making positive change, such as spouses, community organizations, and the United States Occupational Safety and Health Administration (OSHA).

Focus group recordings were transcribed in Spanish, and subsequently translated by the same person, who had also served as the note taker. No names were associated with comments, but the interviewer's comments were identified and letters were assigned to distinguish speakers. These translated conversations were input into QSR NVivo 9^® and coded based on the first author's impression of emergent themes, the themes of the interview guide, and the main research questions that concerned contributing factors to the cause of accidents and unsafe conditions on construction worksites. These theme codes were: why accidents happen, training, retaliation, speaking up, respect for workers, construction equipment, personal protective equipment, speed versus safety, supervisors, language issues, necessity for work, and responsibility for safety. Many quotes were multiply coded, reflecting the over-layering of the themes and experience of the participants. Quotes were selected that summarized other comments, were representations of a poignant sentiment, or described an uncommon view. The quotes shown below were edited from the English transcripts for grammar and spelling in English and written English.

The analysis of the focus groups series that we held with Hispanic construction workers points to a tough safety climate on many of their worksites. Supervisor pressure to work fast and workers' necessity for work and fear of retaliation if they speak up about unsafe conditions were discussed by participants as the primary reasons behind Hispanic construction workers' disparate rates of injury on the job. Lack of training and language and cultural differences may play a role in contributing to these conditions, but were not prominently identified by the workers we interviewed. Although workers with limited English language skills may not be inclined to view their own limitation as a reason for accidents, workers did identify their responsibility for hazardous conditions, in particular by not "complaining." We interpret the findings above to suggest an analytical construct that Hispanic construction workers are more likely to face hazardous conditions that might lead to injury, because Hispanic construction workers, more than other workers, 1) face greater production pressures in a safety-challenged construction work environment, 2) are subject to disrespectful attitudes and intimidation, and 3) are more likely to tolerate unsafe conditions.

There are several potential limitations of this study. These data represent the views of a few individuals at a specific time and place. Limited qualitative data are not, as a rule, generalizable to larger populations. It is especially important to note that Hispanics cannot be characterized as a singular or uniform cultural, linguistic, economic or other social group. Additional limitations of the study include: lack of comparison of the data with alternative perspectives, possible errors of translation from Spanish to English, lack of verbal cues for interpretation, possible lingual and cultural misinterpretation, and unreflected bias of the interpreter of the results. We do believe, however, that these data and analysis are representative of the experiences of some Hispanic construction workers in the Northeast United States circa 2009.

There has been little exploration of the impact of a negative psychosocial environment on the rate of injuries in construction, although a recent study of workers in a variety of industries found association between job threat, harassment, and pressure, and higher rate of occupational injuries [16]. Zohar and others [17] have attributed lower occupational injury rates to better "safety climates" which include such constructs as demonstrated management commitment to safety and opportunities for meaningful employee participation in safety programs. Gillen et al. [18] found numerous challenges in evaluating relationships between safety climate and other psychosocial factors and injury severity in their study of injured construction workers. Indeed, the dynamic and overlapping potential factors affecting safety and organizational climate, hazardous conditions and workers' safety-related knowledge and actions, would prove challenging to evaluate in the construction sector.

Qualitative investigations of social phenomena are useful for unearthing the "why" behind what quantitative investigations have revealed to be the "what" - in this case, injury disparities in construction. Focus group findings' generalizability is limited, because the results are based on data from few non-randomly selected participants. Nevertheless, our results resonate with the findings of other investigators. Menzel et al. [19] conducted focus groups among union and non-union Latino construction workers in Nevada to elucidate their perceptions of the sources of risk in their jobs. They found many similar themes to those that arose in our study. The salient concepts of the origins of risk that they discuss include language barriers, lack of trade skills, and "traditional Latino values." The workers in their focus groups also identified economic pressure to work quickly due piecework pay or production deadlines, lack of appropriate tools and safety equipment, lack of or inadequate work tool and safety training, economic competition with other workers for scarce jobs, lack of sufficient skills in the chosen construction trade, exploitation, and immigration status as factors that contributed to risk and injury. Robertson's focus group study [20] of Latino construction workers exposed to noise identified the influence of ''traditional Latino values of collectivism, family focus, proscribed [sic] gender roles, and respect for authority figures'' on these workers' hearing protection behavior. Robertson's study also found that "participants agreed that Latino workers needed jobs and were willing to work in unsafe situations."

In some cases, these and other authors suggest that "traditional Latino values" such as respect for authority, hard work and prioritization of family interests over self-interest, play a role in either explaining Hispanics' greater risk, or can be taken advantage of to promote self-protective behaviors. However, given the context of the work environment described by the workers who participated in our focus groups, we question the explanatory role of "cultural" factors. Even if cultural factors play a role in reinforcing workers' response to pressure, removing the intimidation might also remove the influence of the traditional cultural value of "respect for authority." Palinkas and Arciniega [21] suggest that "traditional Latino values" held by an individual are unlikely to pose a barrier to economic advancement. They suggest that cultural norms change when confronted with new contexts in which the costs of maintaining those norms rises and the benefits of changing increase.

In introducing her paper "Hazardous Constructions of Latino Immigrants in the Construction Industry: The Case of a Post-Katrina New Orleans," Trujillo-Pagan [22] writes that many scholars and practitioners focus on "linguistically- and culturally-appropriate training and personal protective equipment and reproduce an emphasis on the worker rather than the employer." Her interviews with Latino construction workers in New Orleans found that "Latino workers believe they are racialized as 'hard workers' and that this racialization accounts for workplace discrimination. Specifically, Latino workers believed they were given more dangerous and risky work assignments because employers knew Latino workers would do the work... They assumed their ability to withstand a tough working situation was part of what they were paid for.." Additionally, the workers that she interviewed reported that "They had not received any training or information, and were unlikely to request this information from their employer because they did not believe their employer knew any more than they did. They were also concerned about their employer perceiving them as weak, unwilling to work, or as someone who hace problemas, i.e. makes problems; a "troublemaker."

How public health investigators and policy authorities interpret the causes of the disparate Latino risk for injury in construction will influence the design of interventions to address the problem. At the Latino Health and Safety Summit in April 2010 United States Secretary of Labor Linda Solis suggested that worker training was a central strategy to reduce Latino risk, and announced that OSHA would begin enforcing in earnest the OSHA policy that worker training must be provided in a way that is understandable to workers, i.e., in their language [23]. Programs that focus on worker training in Spanish, however, may not adequately address additional causative factors, such as the greater peril Hispanic workers may face in construction workplaces, and the challenges even well-trained and informed workers face in changing those conditions. (It is interesting to note that language barriers are hypothesized to be a barrier to safety, far more often than as a barrier to construction.) Interventions focused on changing hazardous conditions, rather than "hazardous traditional workers," would have the added benefit of addressing poor conditions for all workers, regardless of ethnicity or their personal values.

We are unaware of any interventions designed to shift contractor practices in order to lessen the risks faced by Hispanic workers on their worksites. However, some worker training programs have been designed specifically to address the needs of Hispanic workers [12, 24]. In New Jersey, researchers developed and delivered a hazard awareness training program targeting Latino immigrant day laborers, many of whom were working in construction. Both qualitative and quantitative pre- and post-training assessments demonstrated increased knowledge and "self-protective actions" among trainees. The authors conclude that "participatory, peer-lead training...may have a positive effect...but major changes would require employer engagement" [25].

These focus group findings suggest that Hispanic construction workers in the United States perceive a tough safety climate and hazardous conditions. Rather than focus on the workers as the target of change, we designed an intervention for contractors employing Hispanic construction workers. The intervention program, Leaders in Safe Construction (LISC), includes multiple steps and coaching for contractors to enhance their ability to assure safety for all workers. LISC includes a supervisor training component aimed at improving their management skills, and increasing their understanding and respect for their workers. More details on this program can be found at http://www.lawrenceworksafe.org. We believe that widespread and effective alteration in the high injury rates experienced by immigrant Hispanics working in construction in the United States, may be affected by enhanced safety climates and reduced hazards. While Spanish-language training for workers is necessary, it may not translate into improved conditions. Indeed, the key may be better enforcement of workplace standards and targeted training of those with the power over these conditions: construction employers.