Article Text

Original research
Prevalence and correlates of workplace violence: descriptive results from the National Transgender Discrimination Survey
Free
  1. Idia Enogieru1,
  2. Ashley L. Blewitt-Golsch2,
  3. Lauren J. Hart3,
  4. Sara LeGrand2,
  5. Kathryn Whetten2,
  6. Truls Ostbye2,3,
  7. Candice Y. Johnson3
  1. 1 Department of Health Policy and Management, The University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA
  2. 2 Duke Global Health Institute, Duke University, Durham, North Carolina, USA
  3. 3 Department of Family Medicine and Community Health, Duke University, Durham, North Carolina, USA
  1. Correspondence to Dr Candice Y. Johnson, Family Medicine and Community Health, Duke University, Durham, North Carolina, USA; candice.y.johnson{at}duke.edu

Abstract

Objective To describe the lifetime prevalence of workplace harassment, physical violence and sexual assault against transgender and non-binary workers targeted due to their gender identity and to identify correlates of this workplace violence.

Methods This descriptive cross-sectional study used data from 4597 transgender or non-binary respondents from the 2008–2009 National Transgender Discrimination Survey. Respondents reported if they had ever experienced harassment, physical violence or sexual assault at work specifically because of their gender identity. We estimated the prevalence of each type of violence stratified by gender identity, race/ethnicity, age, educational attainment, history of working in the street economy (eg, sex industry, drug sales) and if people at work knew their gender identity.

Results Workplace violence was prevalent, with 50% of transgender and non-binary workers having ever experienced harassment, 7% physical violence and 6% sexual assault at work because of their gender identity. Harassment was common among all of these workers, but physical violence and sexual assault were more than twice as common among transfeminine and non-binary workers assigned male at birth, workers of colour, workers with low educational attainment and those who had ever worked in the street economy.

Conclusions Transgender and non-binary workers commonly face violence at work because of their gender identity. Workplace violence prevention programmes should incorporate ways to prevent gender identity-based violence and facilitate channels for workers to report the occurrence of discrimination and violence.

  • epidemiology
  • violence
  • transgender
  • non-binary

Data availability statement

Data may be obtained from a third party. The data used in this analysis are available at https://doi.org/10.3886/ICPSR37888.v1.

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

WHAT IS ALREADY KNOWN ON THIS TOPIC

  • Transgender and non-binary people experience discrimination and violence in everyday life, including in the workplace.

WHAT THIS STUDY ADDS

  • Harassment, physical violence and sexual assault due to gender identity is common among transgender and non-binary workers.

  • Physical violence and sexual assault are most prevalent among transfeminine workers, non-binary workers assigned male at birth, workers of colour, workers with low educational attainment and those who have worked in the street economy.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

  • Workplace violence prevention programmes should address the prevention of gender identity-based workplace violence, including specific elements to support workers most at risk.

Introduction

The US Occupational Safety and Health Administration (OSHA) defines workplace violence as including acts or threats of physical violence, harassment and threatening behaviour at work.1 Although OSHA has no specific standard addressing workplace violence, its General Duty Clause requires that employers provide a work environment safe from hazards that might cause serious harm, including a workplace free from violence.2 3

In addition to its acute physical or mental effects, workplace violence is associated with an increased risk of long-term consequences including depression, post-traumatic stress disorder, poor sleep, cardiovascular disease and extended leave from work.4–7 For employers, workplace violence can lead to reduced productivity, increased turnover and legal consequences.8 In the USA, an estimated 8 violent workplace crimes were reported per 1000 workers during the years 2007–2015, although this rate varied considerably by occupation; for example, workers in protective services occupations experienced workplace violence at rates over an order of magnitude higher (101 per 1000).9

Transgender and non-binary (TNB) people—those whose gender identity differs from their sex assigned at birth—experience high rates of violence in everyday life, including in the workplace.10–12 Each year, an estimated 15% of TNB workers are victims of workplace violence perpetrated specifically because of their gender identity.10 As a result of workplace violence and other forms of workplace discrimination, over three-quarters of TNB workers make efforts to avoid discrimination and violence at work by means such as hiding their gender identity, requesting a job transfer or quitting their job.10

An estimated 1.6% of US adults identify as TNB, with younger adults more likely to report being TNB (5.1% of those aged 18–29 years) than older adults (0.3% of those aged 50 years or older).13 Among the estimated 150 million adults in the US civilian workforce, this could translate to approximately 2.4 million workers at risk for workplace violence perpetrated specifically against those with TNB gender identities.14 However, TNB workers are poorly represented in the occupational health literature, and we know little about these workers’ experiences.

To better understand the occurrence of workplace violence against TNB workers, our objective was to estimate the lifetime prevalence of workplace harassment, physical violence or sexual assault perpetrated because of the worker’s gender identity, and to identify correlates of such workplace violence.

Methods

We used secondary data from the 2008–2009 National Transgender Discrimination Survey (NTDS), a study conducted by the National Center for Transgender Equality and the National Gay and Lesbian Task Force.15 People eligible to participate in NTDS identified as transgender or gender non-conforming (the terms used in the recruitment materials), were aged 18 years or older, and were residents of the 50 US states, District of Columbia, Puerto Rico, Guam, or the US Virgin Islands. To recruit participants, a link to the online survey was distributed to >800 transgender-serving community-based organisations and 150 community listservs. An additional 2000 paper surveys were distributed to organisations that serve rural, homeless and low-income transgender people to better reach populations without internet access. The survey—available in English and Spanish—focused on discrimination experienced in daily life, such as in employment, housing and healthcare.

Gender identity category was determined using questions on sex assigned at birth, gender identity (man, woman, part time one gender and part time another, gender not listed here) and how strongly respondents identified with a list of 15 terms related to gender identity (not at all, somewhat, strongly). More details are provided in the online supplemental materials. We used that information to classify respondents into one of four gender identity categories: transmasculine, transfeminine, non-binary assigned female at birth (AFAB) and non-binary assigned male at birth (AMAB) (online supplemental figure S1). We excluded people who fell outside these categories or who had incomplete data on gender identity. We selected these four gender identity categories to correspond to categories used in another large US survey of TNB people (transgender men, transgender women, non-binary AFAB, non-binary AMAB).10 We used the terms transmasculine and transfeminine because we could not reliably separate people with a binary transgender identity from those who held multiple identities. For example, transmasculine respondents would include those who identified solely as transgender men and also those who were assigned female at birth and identified both as transgender and as non-binary.

Supplemental material

Information on workplace violence was collected with the question, “Because of being transgender/gender non-conforming, which of the following experiences have you had at work?” followed by a series of statements including “I was harassed by someone at work”, “I was the victim of physical violence by someone at work” and “I was the victim of sexual assault by someone at work”. Respondents could answer ‘yes’, ‘no’ or ‘not applicable’ to each statement. To exclude participants who likely never held a job, we excluded people who answered ‘not applicable’ to every question about workplace violence or who did not answer any of those questions.

We examined six variables as potential correlates of workplace violence against TNB people: gender identity (transmasculine, transfeminine, non-binary AFAB, non-binary AMAB); race/ethnicity (Hispanic or Latinx, non-Hispanic black, non-Hispanic white and all other non-Hispanic groups); age (18–24, 25–34, 35–44, 45–54 and 55 years and older); educational attainment (less than high school, high school, some college, college degree, graduate degree); ever worked for pay in the street economy (eg, sex work, drug sales) (yes, no) and how many people at their current job knew the respondent was TNB (none, some, most, not applicable). We included the ‘not applicable’ category for the last variable to capture respondents who were currently not working but worked in the past. Further details on how these variables were collected and defined for analysis are available in the online supplemental materials.

We estimated the prevalence of each type of workplace violence separately, and how the prevalence varied by each variable of interest. We used robust Poisson regression to estimate prevalence ratios (PR) and 95% CIs for associations between the variables of interest and each type of workplace violence. No covariates were included in the model because this was a descriptive study whose purpose was to illustrate differences between subgroups, not to assess causality.16 To examine the intersection between gender identity and race/ethnicity, we reported the prevalence of each type of workplace violence stratified by combinations of both variables. Results were not presented when estimates were based on fewer than five participants or if the sample size was too small for the model to run. Analyses were conducted in Stata V.17 (College Station, Texas, USA) and SAS V.9.4 (Cary, North Carolina, USA).

Results

Of the 6456 people completing the NTDS survey, we included the 5861 (91%) classified into one of the four included gender identity groups. From that eligible population, we excluded 1264 (22%) people who marked ‘not applicable’ to all questions on workplace violence or who did not answer all of those questions. Our final analysis included 4597 respondents: 2153 transfeminine, 1258 transmasculine, 342 non-binary AMAB and 844 non-binary AFAB workers. Because some respondents did not answer all workplace violence questions, the number of people included in the analyses of each of the specific workplace violence outcomes differed.

Most participants were non-Hispanic white, in the 25–34 years age group and had at least some college education (table 1). Most respondents had worked only in the conventional economy; 17% of respondents had worked in the street economy at some point.

Table 1

Lifetime prevalence of harassment at work because of gender identity, transgender and non-binary workers, NTDS, 2008–2009

Half of the participants (50%) reported ever being harassed at work because of their gender identity (table 1). The prevalence of harassment was highest among those with less than high school education, those who had ever worked in the street economy and those who were likely currently not working (ie, answered ‘not applicable’ to the question asking if people in their current job knew that the respondent was transgender). Results for transmasculine and transfeminine workers (online supplemental table S1) and non-binary workers (online supplemental table S2) followed the same patterns.

In the NTDS population, 7% reported ever experiencing physical violence and 6% sexual assault at work because of their gender identity (table 2). The prevalence of both types of violence had similar patterns: both were more common among transfeminine and non-binary AMAB workers, workers of colour, younger workers, workers with less than a high school education, those who ever worked in the street economy and those who answered ‘not applicable’ to the question about people knowing the respondent’s gender identity at their current job (ie, they were likely currently not working). The prevalence of physical violence exceeded 15% among respondents with less than a high school education and those who ever worked in the street economy. The same patterns were evident among transmasculine and transfeminine workers (online supplemental table S1) and non-binary workers (online supplemental table S2), although the analysis for non-binary workers was limited by small sample size.

Table 2

Lifetime prevalence of physical violence or sexual assault at work because of gender identity, transgender and non-binary workers, National Transgender Discrimination Survey, 2008–2009

We then examined how the intersection of gender identity and race/ethnicity was associated with prevalence of the three types of workplace violence (table 3). Because of small sample sizes, we collapsed all workers aside from non-Hispanic white into a single category of ‘workers of colour’. Workers of colour were more likely to report all three types of workplace violence than non-Hispanic white workers. Differences were most evident for physical violence and sexual assault, for which transfeminine and non-binary AMAB workers of colour were more likely to experience these types of violence than transmasculine or non-binary AFAB workers.

Table 3

Lifetime prevalence of workplace violence by gender identity and race/ethnicity, National Transgender Discrimination Survey, 2008–2009

Discussion

A substantial proportion of TNB workers reported experiencing harassment, physical violence or sexual assault perpetrated because of their gender identity during their working lives. Our estimates did not include workplace violence from causes other than gender identity-based discrimination, meaning that the overall prevalence of workplace violence experienced by these workers was likely substantially higher.

A 2021 study also asked TNB respondents about their lifetime prevalence of workplace violence perpetrated due to their gender identity and found that 44% of respondents had been verbally harassed, 28% physically harassed and 26% sexually harassed at work at some point during their working lives.17 The types of workplace violence measured between our study and this more recent study differed, which meant we could not directly compare results. Past-year prevalence of workplace violence was collected in the 2015 US Transgender Survey. Among those who held a job within the past year, 14% were verbally harassed, 1% were physically assaulted and 1% were sexually assaulted at work within the past year because of their gender identity.10 Correlates of each type of workplace violence were not presented in that study, and we therefore could not determine whether correlates were similar for lifetime and past-year workplace violence. However, the present study and these prior studies demonstrate that workplace violence and harassment perpetrated because of a worker’s gender identity is a common occurrence among TNB workers. No study collected information on the frequency or duration of workplace violence, information that could provide a better picture of the extent that TNB workers experience workplace violence.

Transfeminine and non-binary AMAB workers of colour were the most likely to experience physical violence and sexual assault—this is consistent with findings that in the USA, black transgender women are at particularly high risk for violence due to the intersection of racism, sexism and transphobia.18 This result is similar to studies of all types of workplace violence (not only gender-identity based), where women are more likely to experience sexual assault at work than men; however, there are mixed results about if women or men are more likely to experience physical assault at work.9 19 20 Research on race and workplace violence against TNB workers has also generated mixed results. A 2021 study of lesbian, gay, bisexual and transgender workers found that workers of colour were more likely to experience workplace discrimination and harassment.17 In the 2015 US Transgender Survey, in contrast, the past-year prevalence of any type of workplace violence was similar between white, black and Latinx respondents (15% prevalence for all three groups), but substantially higher for American Indian and Alaska Native (28%) and Middle Eastern or North African (26%) respondents.10 Our study’s sample size was too small to present reliable estimates from these last two racial groups; our aggregation of multiple racial/ethnic groups might be concealing additional racial/ethnic disparities. In the US workforce overall, American Indian and Alaska Native workers are more likely to report a hostile working environment than workers of other racial/ethnic groups.21 A limitation of our analysis was that if multiple forms of discrimination were occurring in the workplace, respondents might not have been able to accurately report if acts of violence were perpetrated due to their gender identity, racial identity or some other reason.

In our study population, having less than a high school education was strongly associated with a higher lifetime prevalence of physical violence and sexual assault at work. Although educational attainment is correlated with broad occupational groups, we had no information about respondents’ jobs or workplaces and therefore could not determine if this result indicates that certain jobs or workplace characteristics pose greater risks than others with respect to workplace violence against TNB workers. In studies of all types of workplace violence (not only gender identity-related), rates differ substantially by occupational group.9 19 Future research that includes information on industry and occupation (using standard data collection that would allow linkage to other occupational data sources), workplace characteristics and the relationship between victim and perpetrator would help to better understand the causes and prevention of gender identity-based workplace violence. For example, previous research using data from the 2007–2015 National Crime Victimisation Survey (NCVS) showed that workplace crimes against men were more likely to be perpetrated by strangers, whereas crimes against women were more likely to be perpetrated by people they knew, such as clients, coworkers or supervisors.9 At that time, the NCVS did not collect information on transgender identity; now that this information is being collected, the NCVS could be an important data source for better understanding workplace violence against transgender workers.12

The NTDS was not a nationally representative survey of TNB workers, and therefore our estimates might not be generalisable to the US TNB worker population. Because this survey was advertised as a study specifically interested in learning about experiences of discrimination, people who experienced discrimination were probably more likely to participate than others; as a result, our estimates of workplace violence might be higher than in the US TNB worker population as a whole. Using survey data to study workplace violence was both a limitation and strength. Respondents might have interpreted questions about workplace violence differently, leading to misclassification; for example, ‘harassment’ was not defined in the questionnaire. Self-reported workplace violence, however, could be more completely reported in this anonymous survey than would be found in official statistics because workplace violence is often not reported to law enforcement, leading to higher sensitivity of outcome assessment—this could be even more true for TNB people, who might avoid interactions with law enforcement because of the risk for mistreatment.9 10 22 23

In this sample, 17% had worked in the street economy at some point, but the majority of respondents (83%) only ever worked in the conventional economy. The higher prevalence of workplace violence we observed among those who ever worked in the street economy is not surprising; workers in the street economy experience high rates of violence.24 25 We did not have sufficient data to determine if respondents who ever worked in the street economy also worked in the conventional economy at some point, and if violence was experienced in only one or in both of these work settings.

A limitation of this dataset was that information on workplace violence was collected over the working lifetime. Without information about length of time in the labour market, we could not take into account work history differences between respondents. As a result, we were restricted to estimating prevalence over an unspecified working lifetime. Because TNB populations are known to experience high rates of unemployment and job change, variables such as age might not be a good proxy for length of time in the labour market. Longitudinal research or studies with more information on time in the workforce will be needed to estimate how frequently workplace violence occurs. Because the outcome (workplace violence) was collected over a working lifetime and most predictor variables were collected with respect to the respondent’s current situation, of particular importance is the possibility that respondents changed their working environment in response to workplace violence. For example, a TNB worker who experiences workplace violence might move to a new job and not reveal to new coworkers that they are TNB; a previous report found that 71% of respondents had ever hid their gender identity or gender transition at work to avoid discrimination or violence, with transgender men and women more likely to report doing so than non-binary respondents.10 In our study population, people whose transgender identity was currently unknown at work still reported a high lifetime prevalence of gender identity-based workplace violence. This result suggests that these respondents could be currently concealing their transgender identity at work, perhaps in part due to previous experiences with discrimination or workplace violence. We also found that people who answered ‘not applicable’ to the question about people’s current awareness of their gender identity at work were substantially more likely than other groups to report ever experiencing workplace violence. This result suggests that previous experiences of violence might have driven these respondents out of the workforce or into jobs where they had no coworkers (eg, some forms of self-employment). Every year, an estimated 15% of TNB workers quit their job to avoid workplace discrimination.10 Pervasive workplace discrimination and violence is thought to be the reason why TNB people have an unemployment rate estimated at three times that of cisgender people.10

This analysis was descriptive and meant to highlight the scope of the problem of gender identity-based workplace violence, allowing occupational health practitioners to better protect their workforce, particularly workers at greatest risk. Further research will be needed to explain why certain groups of workers have higher risk than others. For example, a previous report found that black and Latinx NTDS respondents and those with lower educational attainment were more likely to work in the street economy than other respondents, which could partly account for the higher rates of physical violence and sexual assault observed in these worker groups.11

Violence prevention programmes are integral tools for preventing violence in the workplace, but existing programmes might not adequately address gender identity-based violence. Given the high prevalence of workplace violence experienced by TNB workers, employers in the conventional workforce can incorporate elements specific to gender identity-based workplace violence into their violence prevention programmes, including training, written antidiscrimination policies, policies that encourage reporting, identifying and training allies to take reports of workplace violence and thorough investigation of incidents of gender identity-based violence. More broadly, employer toolkits are available to help foster the inclusion of TNB workers in the workplace, with recommendations for onboarding, training, support, communication, healthcare coverage and other aspects of the work culture and operations.26 27 A study in Austria found that lesbian, gay, bisexual, transgender and intersex (LGBTI) workers were more likely to be out in the workplace if their employer fostered an LGBTI-friendly work environment and if there were antidiscrimination provisions in their labour contracts.28 In the USA, TNB workers are further protected by Title VII of the Civil Rights Act of 1964, which prohibits employment-related discrimination or harassment based on sex; ‘sex’ includes both gender identity and sexual orientation following the 2020 Bostock v Clayton County Supreme Court decision.29 For workers in the street economy, broader legal and societal interventions have been recommended to prevent gender identity-based workplace violence, including decriminalisation of sex work, educational interventions to reduce transphobia and social policies to improve economic opportunities for TNB people.30 31

Workplace violence against TNB workers due to their gender identity is prevalent. Workplace violence prevention programmes should incorporate ways to prevent gender identity-based violence, particularly for groups at greatest risk.

Data availability statement

Data may be obtained from a third party. The data used in this analysis are available at https://doi.org/10.3886/ICPSR37888.v1.

Ethics statements

Patient consent for publication

Ethics approval

This study was reviewed by the Duke University Health System Institutional Review Board (protocol no. 00108637) and was determined to be exempt from continuing review. Participants gave informed consent to participate in the study before taking part.

References

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

Footnotes

  • Contributors IE and CJ drafted the manuscript; IE, AB-G and CJ conducted the analyses; IE, AB-G, LH, SLeG, KW, TO and CJ interpreted results and edited the manuscript. CJ is the guarantor of the study.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.