Language Accommodations in Workers' Compensation: Comparing Ontario and Quebec.

Workers who experience language barriers are more likely to get injured or sick because of their work and have poorer claim and return-to-work outcomes compared to other workers. To better understand the systemic factors that shape access to compensation in contexts of language barriers, we compared language accommodation policies and practices in the Quebec and Ontario workers' compensation systems. We uncovered gaps limiting access to professional interpreters in both provinces, although gaps were more pronounced in Quebec where workers were responsible for the cost of interpreters. We argue that simply improving the linguistic competence of workers' compensation systems is not sufficient to tackle access barriers and must be accompanied by efforts to address the root causes of social and economic inequities for workers who experience language barriers.

Introduction

This article compares how Ontario and Quebec’s workers’ compensation systems have accommodated the increasing language diversity that has accompanied growing immigration. Evidence indicates that injured workers who experience language barriers have poorer claim and return-to-work outcomes compared to other workers.[2,3] Recently, an internal audit of the Ontario Workplace Safety and Insurance Board (WSIB) found that just 27% of workers who completed a WSIB work transition program and experienced language barriers found employment, with that proportion being zero percent for some job categories. These outcomes have largely been explained by workers’ difficulties filling out forms, understanding letters, communicating with providers, following procedures, respecting deadlines, and self-advocating.[5-7] However, in emphasizing how workers accommodate to the system rather than how the system accommodates workers’ needs, the existing research risks framing difficulties as an individual, as opposed to systemic, deficit. Accordingly, little information is available on language policies and practices within workers’ compensation systems, particularly in relation to how these policies and practices are influenced by laws and discourse around language and diversity.

The importance of accommodating the language needs of injured workers is increased when considering that language barriers have been associated with elevated rates of occupational injuries, illnesses, and deaths.[9-11] This is due to the over-representation of workers who experience language barriers in hazardous work environments such as those that involve physically demanding labor, as well as to the increased risk they may face within jobs because of challenges around training and other health and safety communication.[13-15] Despite obstacles to reporting,[16,17] injured workers who experience language barriers are more likely than their fluent counterparts to file a workers' compensation claim,[18-20] perhaps due to the severity of their injuries or their lack of alternative income supports. However, the inability to access workers’ compensation benefits and services and to return to work promptly, safely, and sustainably can have devastating financial, family, and health consequences for workers, all of which risk amplifying existing socioeconomic and health inequities.

In previous publications, we revealed how the lack of consideration of language barriers in return-to-work policies and practices, such as in the determination of suitable employment, negatively shaped the experiences and outcomes for workers.[22,23] In this article, we examine language accommodations policies and practices at the Ontario Workplace Safety and Insurance Board (WSIB) and the Quebec Commission des Normes, de l’Équité, de la Santé et de la Sécurité du travail (CNESST). Language accommodation policies and practices are defined as those intended to address language barriers between injured workers and the compensation system and its representatives, such as pertaining to interpretation (verbal) and translation (written).

Methods

We conducted qualitative interviews with injured workers and key informants in Ontario and Quebec over a 2-year period (2016-2017). Most of the interviews were conducted in the Greater Toronto Area and Montreal Metropolitan Area. In Ontario, we interviewed 14 workers and 24 key informants, and in Quebec, 13 workers and 8 key informants. We recruited workers who self-reported difficulties communicating verbally, or in writing, in their province's dominant language (English in Ontario and French in Quebec), and who experienced an injury or illness because of their work. While we did not screen for workers who had filed a claim (since we were interested in documenting challenges to reporting), we advertised our study as looking to document experiences with workers’ compensation. To ensure our analysis of policies and practices was as current and comprehensive as possible, we did not include workers whose claims were closed more than 3 years prior to the interview.

Workers were recruited through social media, online ads, clinics, unions, community organizations, and by leafleting in various community locations. We translated our recruitment material into several languages representing the constituencies of the community organizations that assisted with recruitment; however, we interviewed workers in any language. We offered all workers the assistance of a professional interpreter, but most chose to communicate with researchers without one (19/27). All workers were provided with and signed an English or French consent form that was also verbally explained by the interviewer prior to the interview. When an interpreter was present, he or she orally translated the consent form for workers using a script that repeatedly checked for understanding. In the interview, workers were asked about their pre- and post-injury trajectories, interactions with various stakeholders (eg, employers, healthcare practitioners, compensation staff, etc.), and recommendations to address barriers.

The key informants we recruited had specialized knowledge about the workers’ compensation process in Ontario or Quebec and specifically about the challenges facing individuals who experience language barriers. We recruited key informants through our networks and snowball sampling. Interviews with key informants were conducted in person or over the phone, and all signed a consent form that was sent to them prior to the interview. Key informants were asked about how they perceived, evaluated, and addressed language barriers within their contexts. They were also asked about larger trends, and about strategies that could tackle barriers.

All interviews were semi-structured and lasted from one and a half to 2 h. They were recorded and transcribed, and data were managed with NVivo software. Interview segments were coded, and the codes were grouped into themes through an analysis process that spanned and iteratively informed data collection. Ethics approval for this study was obtained from the McMaster University Research Ethics Board (project # 2016 154). All names are pseudonyms and details have been changed as appropriate to protect the anonymity of participants.

Results

Participant Characteristics

Characteristics of the injured workers interviewed are presented in Table 1. Women represented almost half of the worker participants (13 of 27). A large majority (20 of 27) was over the age of 40 years and had been living in Canada for more than 10 years at the time of the interview (19 of 27). In Ontario, the plurality of workers was from Asia and spoke Mandarin as first language, whereas in Quebec, the plurality was from Latin America and spoke Spanish. The workers had varying levels of spoken and written proficiency in English or French. A vast majority had attained a level of education at the college level or higher (20 of 27), with three having pursued training or obtained degrees in Canada. At the time of injury, workers were employed in factories (10 of 27), the low-wage service sector (6 of 27), construction (4 of 27), or in other manual jobs such as in warehouses and recycling plants (5 of 27). A few (5 of 27) reported working for temporary agencies, in seasonal work, or having irregular hours. All had experienced a workplace injury or illness, with the injury or illnesses occurring between 1 month and 12 years prior to the interview (median of 5 years), and 22 had filed a workers’ compensation claim. Of those who filed a claim, fourteen had their claim accepted and received some benefits or services as a result. Others (8 of 22) had their claim denied, three of whom had appealed and were waiting for a decision at the time of interview.

Table 1.

Characteristics of Injured Worker Participants.

Characteristics		Ontario (n = 14)	Quebec (n = 13)
Sex	Male	8	6
	Female	6	7
Age (years)	40 or less	2	5
	41 to 60	10	6
	60 +	2	2
Years spent in Canada	5 years or less	1	1
	6 to 10	1	5
	11 to 25	11	3
	More than 25	1	4
Region of origin	Asia	8	1
	Middle East and North Africa	2	4
	Caribbean	2
	Europe	2	1
	Latin America		6
	Canada		1
First language	Mandarin	7
	Arabic	2	1
	Caribbean Patois	1
	Korean	1
	Portuguese	1	1
	Spanish	1	5
	Turkish	1	2
	English		1
	Russian		1
	Cambodian		1
	Persian		1
Education/training	High school or less	3	4
	College or professional training	3	4
	Bachelor	5	4
	Masters	2	1
	Unknown	1
Occupation	Service sector	2	4
	Factory	4	6
	Construction	3	1
	Other manual (maintenance, warehouse)	3	2
	Technical	2

The characteristics of the 32 key informants are presented in Table 2. They included healthcare providers (eg, family physicians, psychiatrists, rehabilitation specialists, etc.), union representatives, and worker advisers (eg, community legal workers). In Ontario, we additionally conducted interviews with WSIB staff and with representatives from two independent agencies of the Ontario Ministry of Labour which offer advice, education, and representation in matters of workers’ compensation, the Office of the Worker Adviser (OWA) and Office of the Employer Adviser (OEA). In Quebec, we were not able to interview CNESST staff and there is no equivalent of the OEA and OWA.

Table 2.

Characteristics of Key Informants.

	Ontario (n = 24)	Quebec (n = 8)
Worker advisers	9	5
Healthcare providers	4	1
Employer advisers	1
Workers’ compensation staff	8
Union representatives	2	2

Language Accommodations in Ontario and Quebec

The WSIB and CNESST, like most workers’ compensation systems in Canada, report employing formal interpretation and translation services as well as specialized services, staff, or tools to respond to language needs. However, our study documented widely different policies and practices with regards to these language accommodations. In Ontario, where there is no official language but where English is the language of everyday life, the French Language Services Act guarantees access to provincial governmental services in French in areas with significant numbers of Franco-Ontarians, but no such requirement exists for other languages. Despite the lack of a legislative requirement or formal internal policy, the WSIB accommodated language needs by offering free external interpretation and translation services to staff and workers upon request. Our results indicate, however, that these services were not always requested when needed, with several factors contributing to gaps.

Workers tended to underestimate their need for language services; on the other hand, those who felt they needed language services were typically unaware they could request it. As a result, staff perception of language barriers dictated the involvement of language services:

… At the beginning … I didn't even know there is an interpreter available … At the beginning you know they speak very fast and I don't really understand and in the end, they found a translator for me … They don't understand me. That's why they got the interpreter. (Harry - Injured worker, speaking through an interpreter)

However, no systematic process was in place for staff to determine language needs. Doctors, employers, and workers were able to report the worker's language needs on claim forms, but this information was not recorded by the WSIB's automated claim registration system. Instead, language needs were generally determined informally by staff during phone conversations with workers, a practice that failed to capture less evident barriers. Additionally, when language needs were known, this information was inconsistently shared with the various staff and other stakeholders who encountered injured workers across the claim and return-to-work process, some of whom, like independent healthcare providers, also lacked access to professional language services. While WSIB staff could make available interpreters for workers’ meetings with other stakeholders, this practice was occasional and varied among staff. In addition, staff was sometimes reluctant to request language services due to logistical difficulties, and workers sometimes refused it, believing they had sufficient English language skills to navigate the compensation system. As a result, WSIB staff and other stakeholders, at times, communicated with workers in English or through informal interpreters. Informal interpreters often consisted of workers’ family members, friends, or sometimes supervisors or employers; this raised concerns among some key informants about accuracy, confidentiality, consent, and objectivity.

Importantly, the presence of a formal interpreter did not always avert miscommunication. Quality issues were reported by workers, WSIB staff, and other stakeholders who noted that some interpreters lacked professionalism, objectivity, familiarity with workers’ compensation terminology, or even the required language skills. Harry, for instance, was eventually provided with interpretation services but commented that “Sometimes the translator I feel is good and sometimes I think my English level is same as theirs” (speaking through an interpreter). No process was in place to evaluate the effectiveness of language services from the point of view of the worker, though staff feedback could be relayed to the provider.

The WSIB provided few other language accommodations. While a worker guide in 10 languages was available on the WSIB website, the website and forms were only available in English and French. Workers were able to complete forms over the phone with the assistance of a WSIB-provided interpreter, or fill them out in their preferred language, but the workers in our study, as well as some worker advisers, were unaware of these options. Difficulties with forms sometimes resulted in inaccuracies, some of which were intentionally introduced by employers and supervisors who purported to help workers:

I have a couple worker that because they don't speak English, they don't write English … the employer help them to fill the form 6 … So, the story is not come from the worker it's come from the employer … Then the case will get into trouble because … employer's very smart. They say “okay I’ll help you to fill the form. All you have to do is just sign it”. (Raelyn - Community legal worker)

In Quebec, the Charter of the French Language establishes French as the official language of government, education, commerce, and the workplace. In line with the Charter and other relevant legislation, the CNESST establishes its language policy at least every 5 years. The stated aim of the policy at the time of our study, and still in effect, is to “promote the use and quality of French at the [CNESST]” and “favor French unilingualism in the activities of the latter” (p. 3, our translation). The policy establishes that the CNESST should communicate first and foremost in French, though staff is authorized to communicate in another language if the situation warrants it. Communication with employers or businesses, however, must be exclusively in French. On its website, the CNESST clarified the conditions under which an interpreter would be provided: “The CNESST offers its services in French and English. If the worker cannot express himself or herself in these languages, he or she must pay for the cost of the interpreter he or she needs. If it is the CNESST who needs an interpreter, it will request it and pay” (our translation). The systemic emphasis on worker responsibility for addressing language barriers was further reflected in the attitude of CNESST staff who, according to some key informants, expressed beliefs that workers should learn and speak French.

Our interviews indicate that CNESST staff sometimes ignored requests for English communication and rarely called upon professional interpretation or translation for other languages. Workers described struggling to pay for interpreters due to their low or lack of income. For example, Aleyna, an injured garment worker, explained:

I didn't understand of course and each time I had to bring someone with me by paying the person. Even if this person was busy and worked, he would stop and come with me to help me in this process … Obviously it had an impact on me to have to pay each time for an interpreter, but maybe there are resources, but I don't know. (Speaking through an interpreter, translated from French)

In the absence of professional language services, communication took place in French or was assisted by informal interpreters. At times, CNESST staff with the required language skills communicated with workers directly in their language. Each of these practices was problematized by key informants as failing to protect workers’ interests:

The directive is to speak French. And secondly, when they assign a case for example of an Arabic worker to an agent who is also Arabic, often it will be more to attempt to trap him, to harm him or to tell him “Look you complain that you don't speak French and that your diplomas are not recognized, but look at me, I made it. Get off your ass!” (Jacques - community legal worker, translated from French)

Communication gaps also existed with other stakeholders such as family doctors who, as in Ontario, lacked access to professional language services. Our results indicate that the CNESST did not make interpreters available for workers’ meetings with other stakeholders. At the time of writing, it provided a link for stakeholders on its website for a pay-for-service bank of interpreters.

The CNESST provided few other language accommodations. A Spanish-language pamphlet on health and safety rights was available on its website, as well as links to various integration programs for newcomers including French language classes. The website was only partially translated into English and forms were only available in French and English. No option was available to fill out forms with the help of a CNESST-provided interpreter or in a language other than French or English. As in Ontario, difficulties with forms provided opportunities for inaccuracies to be introduced, either inadvertently by workers or their helpers, or intentionally by employers or supervisors seeking to undermine their claims.

Many workers in Ontario and Quebec navigated the claim process without being able to fully communicate and understand the information in phone calls, appointments, letters, forms, and other documents. Gaps in language accommodations, which were wider in Quebec, resulted in both provinces in mistakes, omissions, misunderstandings, and inconsistencies which resulted in delays, complexities, and attacks on credibility that postponed or prevented diagnosis, compensation, treatment, rehabilitation, recovery, and return to work. The risks of failing to properly accommodate workers’ language needs were illustrated by Luc, a union representative from Quebec:

Recently I have a worker who gets helped by his wife … He speaks a little bit of French, but like I said very, very basic. He has an appointment at the Bureau d’Évaluation Médicale … He can't speak himself … and he needs his wife, except his wife just had surgery and she can't leave the house …. So the employer sends a letter to the [CNESST] asking to cut his income replacement benefits because he doesn't show up at the appointment … The CNESST agent could not that day call the worker directly to say “Sir, why did you miss your appointment with the BEM?” Because the guy will say “What? I not understand”. So the agent gets fed up and calls me. But if there's no me in the portrait as representative, well the agent gets fed up and says “He's at home, I cut him off.” (Translated from French)

Factors Amplifying Language Barriers

Language barriers were amplified by several characteristics common to both compensation systems. First, the compensation process was adversarial, as worker and doctor reports were typically challenged by employers and compensation staff, and as workers were required to continually self-advocate. For example, in Ontario, Evelyn described how the lack of an independent investigation by the WSIB put her in the position of having to contest her employer's description of her job: “I had to say listen, that job description does not reflect on the real job … But what WSIB failed to do is do a proper investigation of the actually injury. Come into the location and examine what happened. So it's just their word and my word.”

Communication was also frequently described as antagonistic, with workers feeling they were treated with disrespect, aggression, or suspicion. It was also depicted as typically brief, largely phone-based, and one-sided. Namely, workers described difficulties reaching staff, as well as a lack of receptiveness to their inputs, questions, or concerns. At the same time, decisions were often communicated in writing and without justification, and required workers wishing to challenge them to respond within strict timelines. These issues were made more difficult by the complexity of the terminology and bureaucratic nature of the compensation and appeals process. As the claim progressed, the potential for difficulties was multiplied by the involvement of an increasing number of stakeholders. Since stakeholders were sometimes poorly coordinated and/or experienced turnover, workers reported having to relay messages among them and retell their stories again and again. Taken together, these dynamics resulted in a breakdown of trust which impeded effective communication, even when language services were employed.

Discussion

Our study contributes to the literature on language-related difficulties navigating workers’ compensation[5-7] by documenting the language accommodation policies and practices of the Ontario and Quebec systems. Our study stands in contrast with existing health and safety research which largely posits the source of racialized inequities as intrinsic to workers due to vaguely defined linguistic and cultural differences. The prevailing characterization of language barriers as a deficit of individuals carries the risk of blaming the victims for their injuries and illnesses, and for their difficulties accessing care, compensation, and sustainable return to work. Our study counteracts this perspective by bringing to the forefront the workers’ compensation policies and practices that shape access in contexts of language barriers. As such, we documented different approaches to language accommodations by the WSIB and CNESST. In Ontario, professional interpretation and translation were available to staff and workers but there were gaps in how language needs were identified and addressed. On the other hand, in Quebec, policies and practices promoted workers’ ‘responsibilization’ for addressing language barriers. Particularly concerning is the fact that workers had to pay for their own interpreters, noting that workers who experience language barriers tend to be employed in low-wage manual or service jobs and earn a fraction of the income of the general population.

The CNESST policies and practices we documented reflect the laws and politics of language and diversity in Quebec. In contrast to the policy of multiculturalism which establishes Canada as a cultural mosaic, Quebec has instead adopted interculturalism as its approach to diversity. Rooted in efforts to protect the French language and culture, which represent a minority in Canada, interculturalism encourages exchange between newcomers of diverse backgrounds and the white francophone majority, in the French language and with a focus on integration. Recently, motivated by data showing a decline in the use of French at home and in the workplace,[28,29] the Quebec government proposed important reforms to the Charter of the French Language, including measures to further entrench French as the exclusive language of government. While provisions exist for communication in other languages where there is a concern for health and public safety, the weighting of these concerns appeared absent in CNESST policies and practices which downloaded responsibility for securing interpretation and translation to economically and socially vulnerable injured workers. It has been argued, in the context of the Quebec Tribunal Administratif du Travail (CNESST appeals tribunal), that the absence of free interpretation services may in fact constitute discrimination under federal and provincial human rights legislation.

Our analysis demonstrates that identifying language needs systematically and providing information and services in different languages at no cost to staff and workers, are crucial to ensuring that injured workers who experience language barriers recover, are compensated, and return to work. Indeed, due to shifting demographics in the Global North, there have been calls to increase the cultural and linguistic competence of service providers and organizations to ensure services are responsive to the needs of diverse communities. At the WSIB, this approach was further reflected in the establishment, beginning in 2017, of specialized teams of return-to-work specialists and case managers in various locations to handle claims involving language barriers. The teams developed and implemented best practices for interacting with workers who experience language barriers; however, as our study concluded, caseloads for members of the specialized teams remained unchanged, impeding the implementation of best practice, and no changes to policy had been implemented to address systemic barriers (eg inflexible deadlines). Accordingly, by focusing on means of communication rather than mechanisms of power, competence essentializes and decontextualizes inequities and therefore fails to address the root causes of marginalization.

Our findings indicate that language accommodations alone are not sufficient to ensure equity in claim experiences and outcomes, in support of previous research that showed that injured workers experience English language proficiency dependent recovery barriers despite access to interpreter services. Namely, our analyses documented systems that lacked the opportunity for workers to have their voices heard, even when their words were interpreted and translated—systems that are adversarial, antagonistic, depersonalized, rigid, unreceptive, one-sided, complex, bureaucratic, and poorly coordinated. While such systemic problems negatively impact workers irrespective of language, they may be amplified in contexts of language barriers. Previous analyses of our data also show that specific adjudication and return-to-work policies and practices that inadequately take language into account contributed to poor experiences and outcomes. For example, participants in both provinces reported that the identification of a suitable occupation often overestimated the worker’s language skills or underestimated the language requirements of the job. It will be important to consider how proposed reforms to the health and safety system in Quebec, which have been argued will lead to greater restrictions on workers’ access to compensation, might magnify or create new barriers for injured workers in general and those who experience language barriers in particular. At the same time, it will be important for analyses to consider the role of precarious work in increasing risk and delaying or denying compensation for workers who experience language barriers, as well as that of intersecting factors such as literacy in shaping access to compensation.

While workers navigated different paths and benefited from different supports in Ontario and Quebec, gaps in language accommodations existed in both provinces and ultimately all workers we spoke to who filed a claim experienced communication difficulties with the workers’ compensation system. This is especially striking as most workers we interviewed were able to conduct the interview in French or English yet found it difficult to navigate compensation. These difficulties led to mistakes, omissions, misunderstandings, and inconsistencies which resulted in delays, complexities, and attacks on credibility, negatively impacting access to benefits and services. Workers typically faced long-term health problems, failed to return to their preinjury employment, and had difficulties finding and maintaining new work. They reported finding themselves under- or unemployed, with many having to rely, reluctantly, on family members and/or government supports. Given immigrant workers constitute almost half of compensation cases on the Island of Montreal, the failure to properly protect and compensate them, due to systematic exclusion based on language, has important public health and social equity implications. These inequities may also be gendered, as data indicates that a higher proportion of immigrant women than immigrant men lack proficiency in French or English, and that women are less likely to make language gains over time.

Our study is limited by the fact that compensation policies and practices often change, and it is possible that some of the descriptions included in this article do not reflect current experiences or conditions. For example, none of the workers in our study had experience with the specialized teams at the WSIB. In addition, given that we were not able to interview CNESST representatives, the descriptions of Quebec policies and practices were obtained exclusively from the experiences of claimants and other key informants and may not reflect the full range of practices within the CNESST. Our findings may also be influenced by our recruitment strategy, which relied heavily on worker advocacy organizations, and which may therefore reflect more complex cases. Lastly, most of the workers in our sample had been in Canada for more than a decade, with some for more than 25 years. A number of the challenges described, such as employment precarity, are more prevalent among newcomers, and it is possible that our study failed to capture those experiences. At the same time, our results dispel assumptions that language barriers are associated with newcomer status and are, as such, a transient condition part of a normal process of integration.

Conclusion

Despite limitations, our study contributes to the literature on inequities in access to workers’ compensation by untangling the role of language accommodation policies and practices. Ours counteracts the prevailing characterization of language barriers as an individual rather than a systemic deficit, which carries the risk of victim blaming. Our results support the recommendation for workers’ compensation systems to ensure that access to benefits and services is irrespective of one's proficiency in the dominant language, and to consider, in the development and application of policies and practices, the broader social and economic context that shapes the experiences of workers who experience language barriers, including the precariousness of employment. In addition to the devastating physical, emotional, financial, and social impacts for injured workers, the failure to accommodate language needs and consider the root causes of marginalization will amplify existing inequities and result in undue costs for society.