Abstract
A personal history of trauma can be a risk factor for job-based secondary traumatic stress and burnout. Experts have pointed to the need to understand the prevalence of adverse childhood experiences (ACEs) among people on the frontlines of health and human services to ensure adequate supports. A small body of literature has examined the prevalence of ACEs among direct service providers, including child welfare workers and social workers. This study adds to that research by assessing the prevalence of ACEs in a sample of frontline homeless service workers. Findings suggest an ACE prevalence higher than the general population but similar to that found in other studies of direct service professionals. Implications for organizational support and trauma-informed supervision are discussed.
The Centers for Disease Control and Prevention (2019) define an adverse childhood experience (ACE) as any potentially traumatic event that occurs before age 18. Research typically considers ten categories of ACEs: lack of familial support, emotional abuse, physical abuse, sexual abuse, neglect, substance abuse in the home, mental illness in the home, domestic violence in the home, loss of a parent/caregiver through divorce or abandonment, and incarceration of a household member (Dong et al., 2004; Felitti et al., 1998). ACEs are common among the general population (Dong et al., 2004; Felitti et al., 1998; Merrick et al., 2018), and exposure to one ACE predicts exposure to others (Dong et al., 2004). In addition, studies have found particularly high ACE scores among sexual minority groups, women, racial minority groups, and individuals with lower educational attainment and lower income (Anderson & Blosnich, 2013; Cavanaugh et al., 2015; Felitti et al., 1998; Merrick et al., 2018). Concerningly, ACEs are associated with a range of physical, psychological, and behavioral health problems later in life (Bellis et al., 2019; Kalmakis & Chandler, 2015) and effects tend to worsen as the number of ACEs increases (Chartier et al., 2010; Felitti et al., 1998).
Research has also shown a relationship between ACEs and job-related problems and stressors, especially among people in helping professions (Anda et al., 2004; Ghahramanlou & Brodbeck, 2000; McKee-Lopez et al., 2019; Metzler et al., 2017; Nelson-Gardell & Harris, 2003; Siebert, 2005; Yellowlees et al., 2021). On one hand, those with a history of childhood adversity have shown a high level of empathy and concern because of their ability to connect and identify with the traumatic situations their clients have encountered (Hiles Howard et al., 2015). On the other hand, routine exposure to traumatic experiences through helping clients can have deleterious effects. Understanding the prevalence of ACEs among those working in human services organizations can inform organizational decisions about supervision, training, and other benefits to prevent burnout and ensure staff feel supported. While a small body of research has studied ACEs among medical professionals, social workers, and other direct service providers, studies have not looked at the prevalence of ACEs among homeless service workers. As such, the current study seeks to address this gap in knowledge.
ACEs Among Direct Service Providers
Research has shown that the prevalence of ACEs is often high among populations accessing social and human services, including people experiencing homelessness (Larkin & Park, 2012; Roos et al., 2013). More recently, studies have started to look at the prevalence of ACEs among those who work in human service organizations, finding that rates are typically higher than the general population. For example, studies of child welfare workers have found ACE prevalence rates of 70% and above (Esaki & Larkin, 2013; Hiles Howard et al., 2015; K. Lee et al., 2017). Studies of other types of direct and human service providers have found similarly high rates (Keesler, 2018; Steen et al., 2020). While the higher prevalence of ACEs is a concern on its own, the cumulative adversity (i.e., the total number of ACEs) across these samples of direct service professionals is also higher than community samples—with 16% (Esaki & Larkin, 2013) to 30% (Keesler, 2018) of respondents reporting four or more ACEs.
Findings related to the relationship between ACE score and demographic characteristics of direct service professionals have been inconsistent. Esaki and Larkin (2013) found no differences in ACE prevalence across gender, race, or type of position among child services workers, while Hiles Howard et al. (2015) found that ACE score was lower for younger child services workers and those with a graduate education. In a study of direct support professionals working with people with developmental disabilities, Keesler (2018) found that ACE scores were higher for women and those who had been in their current position for more than a year. While Steen et al. (2020) did not find gender differences, they did find lower rates for heterosexual participants, as well as for white participants and those younger than 40 and older than 59 (compared to those between the ages of 40 and 59).
Early Life Adversity and Choosing a Career as a Helping Professional
Early life adversity, including childhood trauma, has received extensive attention in research because of its potential to produce deleterious effects across the life course in domains of physical health, mental and behavioral health, and social functioning (Bellis et al., 2019; Chartier et al., 2010; Gilbert et al., 2009; Kalmakis & Chandler, 2015). Early life experiences can also inform career choices, which can be understood through a lens of career construction theory (Bryce et al., 2021). Career construction theory posits that careers are developed in part by the meaning we assign to types of work, which is informed by the meaning we assign to past experiences, present circumstances, and future goals. From this perspective, careers can be understood as ways of making one’s life matter to themselves and others. Careers are constructed in response to life changes and challenges, including trauma. Savickas (2013) writes that “career construction revolves around turning a personal problem into a public strength and then even a social contribution (p. 59).” Thus, a choice to pursue a career as a helping professional may be driven by the need to make meaning of early life adversity or trauma and the need to make that adversity matter by being of service to others who have experienced the same or similar adversities (Bryce et al., 2021). This developmental view of career construction might explain why the rates of ACEs among those who work in human service organizations are often higher than the general population.
In addition to understanding how ACEs might impact the decision to pursue a career in the helping professions, it is also important to recognize that ACEs can increase the risk for experiencing secondary traumatic stress and burnout at work (Ghahramanlou & Brodbeck, 2000; McKee-Lopez et al., 2019; Nelson-Gardell & Harris, 2003; Siebert, 2005; Yellowlees et al., 2021). Secondary traumatic stress is the emotional impact of hearing about the trauma experiences of other people, which can contribute to compassion fatigue—physical and emotional exhaustion resulting from long-term work with clients who are suffering (Newell & MacNeil, 2010). Burnout is a type of job stress common in human services (Newell & MacNeil, 2010) and has been described as “a state of physical or emotional exhaustion that also involves a sense of reduced accomplishment and loss of personal identity (Mayo Clinic, 2021, para 1).” ACEs have been associated with secondary traumatic stress and burnout among physicians, social workers, child welfare workers, direct support professionals, and sexual assault counselors (Ghahramanlou & Brodbeck, 2000; Keesler, 2018; Nelson-Gardell & Harris, 2003; Siebert, 2005; Yellowlees et al., 2021). For this reason, experts have pointed to the need to understand the prevalence of ACEs among people on the frontlines of health and human services to ensure adequate supports (Ghahramanlou & Brodbeck, 2000; McKee-Lopez et al., 2019; Nelson-Gardell & Harris, 2003; Siebert, 2005; Yellowlees et al., 2021).
Homeless Service Workers
Frontline service providers are integral to community efforts to end homelessness. However, burnout and job stress are significant concerns (Lemieux-Cumberlege & Taylor, 2019; Mullen & Leginski, 2010; Olivet et al., 2010; Waegemakers Schiff & Lane, 2019). Because experiences of trauma are common among people experiencing homelessness (Roos et al., 2013), experts have pushed for the integration of trauma-informed care (Hopper et al., 2010). At the same time, it is necessary to acknowledge that the providers who work directly with homeless individuals and families often experience vicarious trauma exposure from listening to their client’s life stories or supporting clients as they live through traumatic life events (Mullen & Leginski, 2010; Olivet et al., 2010). This kind of exposure can pose a significant threat to the wellbeing of workers and their ability to manage work-related stress (Baird & Kracen, 2006; Olivet et al., 2010).
The context of the COVID-19 pandemic has added additional strains to homeless service systems and workers. Organizations had to adapt quickly to evolving public health guidance, create solutions to mitigate the spread of the virus, face increasing service demands, and cope with staffing and resource limitations (Jang et al., 2021; Parkes et al., 2021; Pixley et al., 2021; Rice et al., 2020). The ongoing stress of working through the pandemic has contributed to increased stress, decreased morale, and overall burnout among providers (Rice et al., 2020). In a survey of individuals working in homeless service, supportive housing, and harm reduction programs in Canada, nearly 80% of respondents reported a decline in mental health during the pandemic. In addition, those who spent more time providing direct services were at increased risk for anxiety, stress, post-traumatic stress, and compassion fatigue (Kerman et al., 2021).
Using a cross-sectional design, the current study surveyed frontline homeless service workers across the state of Texas. The analysis presented here aimed to answer two research questions: (a) How prevalent are ACEs among those working on the front lines of homeless services? (b) Does the prevalence of ACEs differ across demographic and background characteristics?
Methods
Data were collected as part of a statewide cross-sectional survey of frontline homeless services providers in Texas. The survey was administered online using QuestionPro and was available from April 12 to June 24 of 2021.
Sample
To be eligible to complete the survey, providers had to be at least 18 years old and working in a position where they provide direct services to people experiencing homelessness or people recently housed after a period of homelessness. The survey was sent to the lead agencies of 11 homeless service networks (i.e., Continuums of Care), 407 individual agencies that provide homeless services, and 167 individual providers in agencies that provide homeless services. Three invitations were sent to each email on the recruitment list, and recipients were encouraged to forward the invite to other colleagues who may be eligible. At the conclusion of the survey, participants were given the option to enter a lottery to win a US$100 Amazon gift card. The recruitment and data collection procedures were approved by the IRB at a university in Texas.
Data Collection
The ten-item ACEs questionnaire (Anda et al., 2006; Felitti et al., 1998) was used to capture the prevalence of potentially traumatic experiences prior to age 18. The questions ask whether the respondent has experienced a range of childhood maltreatment, abuse, and family dysfunction (see Table 2 for categories). The following is an example of an ACE question: “Did a parent or other adult in the household often or very often swear at you, insult you, put you down, or humiliate you? Or act in a way that made you afraid that you might be physically hurt?” Each question is answered with a yes or no, and the total score is the number of items answered with a “yes.” Thus, a higher score reflects a higher number of adverse childhood experiences. In addition to the ACEs questionnaire, demographic information collected included age in years, gender (male, female, transgender/non-binary), race (white, Black/African American, American Indian or Alaska Native, Asian, Pacific Islander), ethnicity (Hispanic or non-Hispanic), type of position (primarily direct service vs. primarily management/supervision with at least some direct service), and years working in the homeless services sector.
The sample used here includes only survey respondents who completed the ACEs questionnaire (N = 136). The data were analyzed using univariate statistics to describe the sample and prevalence of each ACE. Bivariate approaches (e.g., correlations, anovas, t-tests) were used to examine relationships between ACE score and demographic characteristics.
Findings
Of the 136 frontline homeless services providers in this sample, most were female (69.85%) and white (75.74%). The average age was 41.82 years (SD = 13.47). More than half (57.35%) were working in primarily direct service positions while 42.65% were in positions that were primarily management or supervision but included at least some direct service. On average, participants had been working in the homeless services sector for 6.72 years (SD = 6.82). See Table 1 for complete sample characteristics.
Sample Characteristics.
Table 2 presents the distribution of ACE scores in this study. The average ACE score in this sample was 2.82 (SD = 2.36), with a range of 0 to 10. ACEs were relatively common, with 80.88% of participants reporting at least one ACE, 62.50% reporting two or more, and 38.24% reporting four or more. The frequency and percentage for each of the ten ACE categories are presented in Table 3. The most common ACE in this sample was caregiver separation or divorce (52.21%). The least common ACE was the incarceration of a household member (11.76%). Analyses used to examine the relationship between ACEs and demographic characteristics revealed that total ACE score did not differ based on age, gender, race, ethnicity, type of position, or years in homelessness services.
Total ACE Scores.
Note. ACE = adverse childhood experiences.
Adverse Childhood Experiences.
Discussion
ACEs were prevalent among this sample of frontline homelessness services workers in Texas. More than 80% had experienced at least one ACE, and 38% had experienced four or more. This prevalence rate is higher than the original ACEs studies (Dong et al., 2004; Dube et al., 2004; Felitti et al., 1998). It is also higher than prevalence rates in the Behavioral Risk Factor Surveillance System (BRFSS) survey, which uses an eight-point ACE score. Across the 23 states collecting ACE data in the BRFSS, 62% had experienced at least one ACE and 16% had experienced four or more (Merrick et al., 2018). Using BRFSS data from 2015, the Texas Department of State Health Services reported that 55% of adults had experienced at least one ACE, and just over 13% had experience four or more (Fox et al., 2019). Using the eight-point inventory in the current study (i.e., removing neglect and lack of family support), for comparison purposes, 79% had experienced at least one ACE, and 29% had experienced four or more. Altogether, this suggests that the prevalence of childhood adversity in this sample of homeless services workers in Texas is higher than the general population.
The overall prevalence of ACEs in this sample is comparable to other studies of social and human services providers, which generally report a prevalence rate between 70% and 80% (Esaki & Larkin, 2013; Hiles Howard et al., 2015; Keesler, 2018; K. Lee et al., 2017; Steen et al., 2020; Thomas, 2016). For example, studies of ACEs among child welfare workers have found overall prevalence rates from 70% to 78% (Esaki & Larkin, 2013; Hiles Howard et al., 2015; K. Lee et al., 2017). Similarly, a study of licensed social workers in the U.S. reported a prevalence rate of just over 70% (Steen et al., 2020). Notably, the percent of respondents that had experienced four or more ACEs does vary across these studies, with some reporting as few as 16% (Esaki & Larkin, 2013) and others reporting upward of 40% (Thomas, 2016). Thus, while the overall prevalence rate of ACEs in the current study is comparable to other studies of direct service providers, the rate of compound adversity is higher than that found in some samples.
The two most common ACE categories in this study were the divorce or separation of one’s primary caregivers (52.21%) and mental illness within the home (40.44%). The least common category was the incarceration of a household member. Studies of child welfare workers (Esaki & Larkin, 2013; Hiles Howard et al., 2015), direct support professionals (Keesler, 2018), and social work students (Thomas, 2016) have reported similar rankings. Like these other studies, the ranking of experiences in this study was somewhat different from that found in general population samples. For example, Dong et al. (2004) found that substance use and physical abuse were most common, and that household member incarceration was least common. For the 23 states reporting ACE data for the BRFSS, emotional abuse was most common and household member incarceration was the least common (Merrick et al., 2018).
In this sample of homeless service professionals, there were no bivariate relationships found between ACE score and age, gender, race, ethnicity, type of position, or length of time in homeless services. Findings from previous research on ACE prevalence among social and human services providers has been inconsistent in terms of demographic characteristics associated with ACE score. In a large study of licensed social workers (N = 5540), ACE scores were highest for participants ages 40 to 59 (compared to those who were 21 – 39 and 60 and older), scores for Black and Latino participants were higher than white participants, and scores were higher for gay and bisexual participants than those who were heterosexual (Steen et al., 2020). Studies using smaller samples of human services providers have reported varying results, including no differences (Esaki & Larkin, 2013) and higher scores for females (Keesler, 2018), older individuals (Hiles Howard et al., 2015; Thomas, 2016), those with young children (Thomas, 2016), and those without a graduate degree (Hiles Howard et al., 2015). Inconsistency across studies is likely due to population targeted and sampling strategies. That ACE score is not clearly dependent on any one or set of demographic characteristics supports the notion that the strategies to mitigate the effects of ACEs on worker wellbeing may be best targeted across all employees.
Limitations
The current study is limited by its small sample, convenience sampling approach, and by inherent limitations to measuring ACEs. Convenience sampling approaches are common in studies of service providers because comprehensive sampling frames are often non-existent (e.g., Kerman et al., 2021; Larkin & Park, 2012; K. Lee et al., 2017). Since there is no state-wide directory of homelessness services providers, the convenience sampling strategy used here is vulnerable to selection bias and likely under sampled some groups and over sampled others. For example, the recruitment list was developed through a process that relied on an internet-based environmental scan to identify agencies that provide homeless services. Therefore, agencies that lack a website were not included. Further, online surveys can present concerns related to accessibility, self-selection bias, and being unable to determine a response rate. The resulting sample is also relatively small. Small samples present added concerns about generalizability and confidence in findings. Thus, it is impossible to say whether the results generalize to the broader homeless services sector in Texas. It is also false to assume that ACE prevalence rates among homeless service providers in Texas would generalize to other states, since ACE prevalence varies by state (Bethell et al., 2017). Replicating this study in other states may help corroborate the findings.
In addition, the ACE questionnaire has some measurement limitations. First, as a retrospective survey it is subject to recall errors. Second, individuals may intentionally underreport ACEs, and participants with high ACE scores may be more likely to skip items or skip the questionnaire altogether (Felitti et al., 1998). Third, the ACE score is simply a tally of 10 categories of objective experiences and thus is not a comprehensive assessment of compound trauma or subjective responses to those experiences (Lacey & Minnis, 2019). In addition, there is no clear consensus on whether those ten categories are sufficient or whether some should be added or removed to strengthen the validity of the tool. For example, some important adverse experiences are not included in the ACE questionnaire, such as the death of a parent, community violence, racism, and poverty, and some have suggested expanding the ACE categories (Cronholm et al., 2015; Kim et al., 2020; E. Lee et al., 2018). In sum, the ACE score may misrepresent the actual experience of childhood adversity and trauma. Future research could focus on comparing ACE scores, lifetime trauma inventories, and post-traumatic stress symptoms in this population.
Implications for Practice
As posited by career construction theory, one explanation for why the rate of ACEs in a sample of homeless service providers is higher than that of the general population is that early life adversity can inform career choices. For example, studies have found higher rates of adversity among social work students compared to students pursuing other professions (Branson et al., 2019; Rompf & Royse, 1994). In fact, a recent systematic review found that familial dysfunction—including relationship disruptions, child abuse, and neglect—were common among those pursuing a career in the helping professions. The review highlights how choosing a helping profession might help make meaning of one’s own adversities through helping others who have experienced similar losses or trying to prevent others from experiencing the same traumas (Bryce et al., 2021). Thus, choosing a career in the social and human services may be motivated by heightened empathy for trauma survivors and the desire to help others with similar life histories, and it may serve as a pathway for personal healing.
Given that a person’s history of adversity and trauma can influence their decision to pursue a career in which they work directly with people who have similar histories, it is crucial to consider and mitigate the potential risks to worker wellbeing. Of particular importance is that a history of childhood trauma is a risk factor for secondary traumatic stress and burnout at work (Ghahramanlou & Brodbeck, 2000; McKee-Lopez et al., 2019; Nelson-Gardell & Harris, 2003; Siebert, 2005; Yellowlees et al., 2021). For example, in a sample of social workers, burnout was associated with having a troubled parent and experiencing emotional abuse as a child (Siebert, 2005). Similar findings have been reported for child welfare workers (Nelson-Gardell & Harris, 2003) and sexual assault counselors (Ghahramanlou & Brodbeck, 2000).
Direct service providers with lived experience of trauma and other adversities, especially experiences similar to the client populations they serve, have the potential to enhance service provision through deeper empathy (Bryce et al., 2021). However, it is important for individual providers and their organizations to be mindful of the heightened risk for burnout and stress. Burnout is already a significant concern for those working in homeless service settings (Lemieux-Cumberlege & Taylor, 2019; Mullen & Leginski, 2010; Olivet et al., 2010; Waegemakers Schiff & Lane, 2019), which makes it particularly important to institute strategies that can offset the negative effects of work stress.
A growing body of literature points to several strategies for mitigating such risks. Perhaps most importantly, organizations can take a trauma-informed approach to supporting employees. Studies suggest that organizational factors can reduce or increase occupational stress and burnout. For example, perceiving a workplace to be stressful and carrying a stressful client caseload can increase the risk for burnout while receiving supportive supervision can decrease the risk (Siebert, 2005). A cross-national study of homeless service providers found that inadequate supervision was associated with increased burnout while receiving useful trainings was associated with decreased burnout (Lenzi et al., 2021). In fact, inadequate training has been identified as a barrier to providing high-quality services to people experiencing homelessness (Mullen & Leginski, 2010; Olivet et al., 2010; Paat et al., 2021). In addition, a recent study of homeless service and housing providers working in Canada during the COVID-19 pandemic found that compassion fatigue and post-traumatic stress increased with the more time a person spent working in direct contact with clients (Kerman et al., 2021). These findings suggest that organizations would benefit from considering ways to strengthen the supervision and training offered to staff, especially staff who spend much of their time in the field.
While not an area that has received extensive study, there is some evidence that supervision approaches and strategies to mitigate burnout vary across programs (Olivet et al., 2010). One way to strengthen supervision to prevent burnout is through trauma-informed supervision. Trauma-informed supervision emerged from the recognition of the effects of indirect trauma on the wellbeing of providers working with trauma survivors, and it is considered an essential component of a trauma-informed organization. Like trauma-informed practice, trauma-informed supervision requires a safe working environment, an organizational climate of trust and respect, and an environment that supports collaboration, choice, and empowerment (Knight, 2018). Of course, providing effective trauma-informed supervision is predicated on having supervisors who are knowledgeable about indirect trauma effects, who are trained in how to provide good quality supervision, and who have adequate time to spend with supervisees (Berger & Quiros, 2016; Courtois, 2018; Knight, 2018). Knight (2018) argues that trauma-informed elements can be integrated into existing supervision models by including a therapeutic orientation to help supervisees recognize, normalize, and manage responses to indirect trauma experiences (Berger & Quiros, 2016; Knight, 2018). Importantly, providing practitioners with trauma-informed supervision may ultimately help improve their ability to provide trauma-informed services to their clients (Knight, 2018).
In addition to supervision, organizations might consider offering trainings based on staff needs. Many positions within the broad range of services for people experiencing homelessness are filled by individuals who do not have extensive educational or training backgrounds (Mullen & Leginski, 2010). Trainings could target specific practice skills needed for the work to help providers feel more competent in their role and in their ability to support clients with complex trauma histories. Trainings could address trauma-informed care, topics specific to homeless populations (e.g., chronic homelessness, health risks), or specific skills or practice models, such as motivational interviewing or crisis management (Olivet et al., 2010). In addition, trainings could focus on strategies for identifying and managing personal responses to the work. For example, mindfulness training has been used effectively to decrease stress and burnout for health care professionals, teachers (Luken & Sammons, 2016), and human service professionals (Crowder & Sears, 2017; McGarrigle, 2011). These strategies offer significant research potential for testing the impact of professional development, enhanced supervision models, and mindfulness interventions on burnout and post-traumatic stress.
Footnotes
Disposition editor: Sondra J Fogel
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
