Burnout and secondary traumatic stress (STS) are salient professional issues in mental health fields, including the counselling profession (Bardoshi et al., 2019; Fye et al., 2021; Quinn et al., 2019). Counsellor burnout is most commonly described as affecting personal and professional issues that result from ongoing exposure to demanding and stressful work conditions (Lee et al., 2007), while STS is an outcome from repeated and ongoing exposure to details of clients’ traumatic experiences (Bride et al., 2004; Quinn et al., 2019). There is an inherent risk of counsellors experiencing either, or both, because of the job demands associated with being a counsellor (Green et al., 2014). That is, counsellors often provide emotionally demanding counselling services to many clients with extensive and complex needs who are often called to regularly engaging in trauma-focused work (Cook, Fye, Jones et al., 2021; Green et al., 2014; Thompson et al., 2014). While burnout and STS may be a potential professional hazard for all counsellors, this risk must be contextualized by unique individual and organizational factors that heighten (or lessen) these professional issues (Lee et al., 2011). Scholars (e.g., Green et al., 2014; Hensel et al., 2015; Thompson et al., 2014) have shown that one’s risk of burnout and STS is affected based on individual (e.g., age, race/ethnicity, gender) and organizational factors (e.g., work setting, number of direct client hours, caseload, and productivity requirements).

The demands faced by frontline workers have only heightened with the emergence of COVID-19 (Litam & Balkin, 2021). Specifically, the ongoing pandemic has increased personal and professional stressors for counsellors and their clients, including increased financial hardships, social isolation, and increased mental health needs (Center for Disease Control [CDC], 2020; Litam & Balkin, 2021). As the collective mental health needs increased (CDC, 2020), professional counsellors were there to answer the call (Zhou et al., 2020). For professionals that are already overextended (Cook, Fye, Wind, 2021), the COVID-19 pandemic placed counsellors at heighted risk of burnout and STS, like many other front-line professionals (e.g., doctors, nurses, police officers; Maraqa et al., 2020; Litam & Balkin, 2021).

For those counsellors engaged in supervision, like prelicensed counselors (i.e., counsellors engaged in supervision for professional licensure), supervisors can play a critical role in helping their supervisees to develop needed skills to identify and address signs of burnout or STS (Hiebler-Ragger et al., 2021; Quinn et al., 2019). However, the degree to which existing supervision practices are effective in reducing burnout and STS are unclear (Jones & Branco, 2020; Hensel et al., 2015; Knight, 2018) noted the limitations of existing supervision practices in preparing counsellors to engage in trauma-focused work and theorized that the integration of trauma-informed supervision (TIS) into existing supervision practices may better mitigate against professional issues like counsellor burnout and STS (Jones & Branco, 2020).

In the current study, we aimed to examine the extent to which counsellor burnout and STS could be predicted by variables including individual factors (i.e., age, race/ethnicity, and gender), organizational factors (setting, caseload, number of clients, productivity requirement), COVID-19 anxiety, and TIS in a sample of counsellors engaged in supervision for licensure (i.e., prelicensed counsellors). Attending to and addressing counsellor burnout and STS are ethical obligations of counsellors to ensure that they are providing adequate care to clients (American Counseling Association [ACA], 2014). Studies that examine the influence of this constellation of variables on counsellor burnout and STS are needed to identify the unique and collective risk of these issues for counsellors. Lee et al., (2011) suggested specifically studying subsamples of counsellors in order to offer the most targeted interventions. Further, given the increased demands placed on counsellors during the COVID-19 pandemic (Litam & Balkin, 2021), it is important to examine how burnout and STS manifested for prelicensed counsellors during the height of the pandemic in order to inform future counselling and supervision practices in an ongoing or post-pandemic environment.

Counsellor Burnout and STS

Most counsellors derive satisfaction from their work and experience a strong sense of professional and personal wellness (Lawson & Myers, 2011); however, professional issues, like burnout, can manifest early in one’s professional career (Fye et al., 2021; Thompson et al., 2014). For example, scholars (e.g., Cook, Fye, Wind, 2021; Cook, Fye, Jones et al., 2021; Fye et al., 2021) found that counsellors who are engaged in clinical supervision to be independently licensed, described as prelicensed counsellors, were reporting symptoms of counsellor burnout and STS. According to Lee et al., (2007), counsellor burnout consists of exhaustion, incompetence, negative work environment, devaluing clients, and deterioration in personal life. The harmful effects of burnout for counsellors have been observed by scholars (e.g., Fye et al., 2021; Thompson et al., 2014). For example, Fye et al., (2021) found that burnout was strongly related to affective distress (i.e., depression, anxiety, and stress) in a sample of prelicensed counsellors. Further, counsellors who experience burnout may also experience low job satisfaction and are risk of leaving the field (Mullen et al., 2018).

Counsellor burnout and STS are empirically- and theoretically-related (Cieslak et al., 2014; Hensel et al., 2015); however, burnout is a professional issue, while STS is distinguished by symptoms of intrusive thoughts, emotional arousal, and avoidance (Bride et al., 2004) that aligns with the Diagnostic and Statistical Manual of Mental Disorders (DSM; American Psychiatric Association [APA], 2013) diagnostic criteria of post-traumatic stress disorder (PTSD; Hensel et al., 2015) and stems from indirect exposure to trauma (Bride & Kintzle, 2011). It is the connection between indirect trauma exposure and PTSD that distinguishes STS from other theoretically similar constructs, such as burnout, which is not specific to indirect trauma (Lee at al. 2007), or vicarious trauma, which is reflected by a worsening change in providers’ empathy towards clients (Branson, 2019; Hensel et al., 2015). STS is most conceptually related to compassion fatigue (see Hensel et al., 2015), and Fye et al., (2021) found that prelicensed counselors experienced compassion fatigue, which was related to their experience of affective distress (i.e., depression, anxiety, and stress). Prelicensed counsellors may often be providing many hours of direct services to clients who have experienced traumas (Thompson et al., 2014).

Given that most US citizens have been exposed to trauma at some point in their lives (Kilpatrick et al., 2013), it stands to reason that many clients who seek treatment will do so for issues either directly or indirectly related to their traumatic experience (Champine et al., 2019). As a result, STS may have harmful outcomes on counsellor well-being. Since the symptoms of STS are like those of PTSD (Hensel et al., 2015), emotional distress and professional impairment are possible outcomes (Bride & Kintzle, 2011). Counsellors who report higher levels of STS are also more likely to be dissatisfied at work and may decide to leave the field (Bride et al., 2004).

The incidents of STS are evident in mental health professionals, although it has only been limitedly studied in counselling professionals (e.g., Bride & Kintzle 2011). Rates of STS have been found among 22% of social workers (Quinn et al., 2019) and 19% of substance abuse counsellors (Bride & Kintzle, 2011). Given the preventiveness of these professional issues, unidentified and unaddressed burnout and STS poses an ethical risk to counsellors and clients alike (Bride et al., 2004; Fye et al., 2021).

Individual and Organizational Factors Associated with Burnout and STS

Scholars have studied the relationship between individual factors, burnout, and STS (e.g., Hensel et al., 2015; Thompson et al., 2014). For example, there is evidence that age is inversely related to burnout (Brewer & Shapard, 2004; Green et al., 2014) and STS (Hensel et al., 2015), with younger mental health professionals reporting higher levels of both professional issues. Differing levels of counsellor burnout and STS might also exist based on counsellors’ gender and race. Thompson et al., (2014) found no significant differences in levels of burnout based on counsellors’ gender; however, Lent & Schwartz (2012) found significant three-way interactions between sex, gender, and years of experience. Hensel et al., (2015) also found that female and White social workers reported higher levels of STS, although the effect sizes were small.

Researchers have also examined the effects of organizational factors, or workplace factors, on burnout and STS (e.g., Lawson 2007; Hensel et al., 2015; Quinn et al., 2019). Numerous studies (e.g., Green et al., 2014; Lawson & Myers, 2011; Lent et al. 2012; Quinn et al., 2019) have found counsellors burnout and STS differ based on work setting. For example, counsellors working in community-based settings tend to report higher levels of burnout and STS than those working in private practice, school settings, or non-community-based settings (Lawson, 2007; Lawson & Myers, 2011; Quinn et al., 2019). Relatedly, counsellors’ caseload characteristics and their productivity expectations were also investigated variable as it relates to burnout and STS (e.g., Hensel et al., 2015; Lawson, 2007; Lawson & Myers, 2011). In a qualitative study of 246 prelicensed counsellors self-reported symptoms of burnout, Cook, Fye, Jones et al., (2021) found that some counsellors reported feeling overburden by documentation, experienced issues with long shifts and no breaks, and felt pressured to see more clients. Caseload characteristics has also been conceptualized volume, direct client hours or frequency, and caseload ratio (i.e., percentage of caseload with trauma clients). Studies have shown that caseload volume and the number of direct client hours is related to counsellor burnout (Cook, Fye, Jones et al., 2021; Lawson & Myers, 2011) and STS (Hensel et al., 2015; Quinn et al., 2019). In the current study, we examined counsellors’ caseload volume, number of direct client hours, and whether they had productivity requirements.

Prelicensed Counsellors, Clinical Supervision, and TIS

Prelicensed counsellors are persons who are pursuing independent licensure in the US. These professionals have already satisfied the educational requirements but still need to complete a post-degree supervised clinical experience (Henriksen et al., 2019). The supervised clinical experience is intended to facilitate supervisees’ professional development and to ensure the client welfare, while these prelicensed counsellors gain needed clinical experiences (Bernard & Goodyear, 2019). There are several issues that may place prelicensed counsellors at increased risk of burnout and STS. For example, these counsellors often experience self-doubt about their own abilities as counsellors (Rønnestad & Skovholt, 2003). Further, they tend to work in high-stress and under resourced environments (Cook & Sackett, 2018), with issues of low wages, high productivity requirements (i.e., billable hours, documentation demands), funding issues for clients, and high turnover rates (Freadling & Foss-Kelly, 2014; Cook, Fye, Jones et al., 2021).

There is evidence; however, that supervision may be a meaningful protective factor against counsellor burnout and STS (e.g., Hiebler-Ragger et al., 2021; Quinn et al., 2019). The strength of this relationship is less clear (Johnson et al., 2020). For example, Hiebler‐Ragger et al. (2021) found that supervisees’ perceptions of a strong supervisory relationship were related to decreased burnout. In contrast, Johnson et al., (2020) found the supervisory relationship was associated with increased engagement, but unrelated to exhaustion. Regarding STS, Quinn et al., (2019) found that social workers’ favorable perceptions of the supervisory relationship were related to reduced STS. However, Hensel et al., (2015) conducted a meta-analysis of predictors of STS and found that the effectiveness of quality supervision and frequency of supervision in preventing STS was minimal (i.e., small to no effect size). Nonetheless, there is an opportunity for supervisors to facilitate the skill development of their supervisees to address these issues, and there appears to be at least some empirical evidence that supervision can positively reduce supervisees’ burnout and STS.

TIS

One possible reason for the mixed results on the effectiveness of supervision as a protective factor against counsellor burnout and STS is that existing supervision practices fail to prepare counsellors to engage in trauma-focused work (Knight, 2018) or emphasize the importance of addressing burnout and STS (Jones & Branco, 2020). To that end, scholars (e.g., Knight 2018; Jones & Branco, 2020; Jordan, 2018) have theorized the need for trauma-focused supervision practices to better address these professional issues. Knight (2018) defined TIS as “knowledge of trauma and its effects on clients, indirect trauma, core skills of clinical supervision, and core precepts of trauma-informed practice and care,” (p.18). To date, TIS is largely a theoretical construct, having only been limitedly empirically tested (e.g., Berger & Quiros 2016; Cook et al., 2022).

TIS is an extension of trauma-informed practices (TIPS), which is best practices for providing direct care to trauma survivors (Substance Abuse and Mental Health Services Administration [SAMHSA], 2014). Neither trauma-informed practices nor TIS are treatment approaches, rather they are intentional considerations by mental health professionals and their supervisors to recognize how trauma negatively impacts clients’ physical, psychological, and emotional wellbeing, while also identifying the many ways that harmful effects of trauma can manifest in clients’ lives (Goodman et al., 2016). There is consensus among scholars (e.g., Goodman et al., 2016; Jones & Branco, 2020; SAMHSA, 2014) that TIS is comprised of core principles: safety, trustworthiness, choice, collaboration, empowerment, and cultural, historical, and gender issues. Despite being the fact that TIS has limitedly been researched (e.g., Berger & Quiros 2016, Cook et al., 2022), it is possible that TIS may be a novel approach to better address counsellor burnout and TIS in counsellors engaged in supervision (Jones & Branco, 2020) and, ultimately, lead to better client from the clients whom these counsellors serve (Knight, 2018).

The Current Study

There is a need for innovative solutions to prepare counsellors to engage in trauma-informed work, while also protecting them from the detrimental effects of counsellor burnout and STS (Knight, 2018). Addressing counsellor burnout and STS is best contextualized by considering the unique individual and organizational factors that effect a given sample of counsellors (Lee et al., 2011). For those engaged in clinical supervision, the quality of their supervision may be a protective factor against counsellor burnout and STS (Knight, 2018), which is critically important given the characteristics of prelicensed counsellors and the environments in which they work (Cook & Sackett, 2018). However, the relationship between supervision and burnout and STS may be minimal (Knight, 2018; Hensel et al., 2015), and this may be explained by the fact that existing models of supervision fail to adequately emphasis trauma-informed practices (Berger & Quiros, 2016; Knight, 2018). Finally, it is important for the profession to understand the relationship that the COVID-19 pandemic had on counsellors’ experience of burnout and STS. Therefore, in the current study, we explored these relationships and conducted two hierarchical regression models to test two research questions: (1) What is the predictive relationship of individual factors (i.e., age, race/ethnicity, and gender), organizational factors (i.e., setting, caseload, number of clients, productivity requirement), COVID-19 anxiety, and trauma informed supervision on counsellor burnout? and (2) What is the predictive relationship of individual factors (i.e., age, race/ethnicity, and gender), organizational factors (i.e., setting, caseload, number of clients, productivity requirement), COVID-19 anxiety, and trauma informed supervision on counsellor STS?

Method

Participants

The participants in the current study were 282 counsellors engaged in supervision for licensure in their respective states. The age of participants ranged from 23 to 71 (M = 35.18, SD = 10.02), with an average of 4.08 (Mdn = 3.00, SD = 3.92) years of experience. Most participants identified as female (78.7%, n = 222), while 42 identified as male (14.9%), ten identified as nonbinary (3.5%), two identified as transgender (0.7%), three identified as other (1.1%), and three participants did not respond to the item (1.1%). Participants described their racial/ethnic identities as White (non-Hispanic) (75.9%; n = 214), African American/Black (8.2%; n = 23), Multiracial (5.7%; n = 16), Hispanic/Latino(a) (3.2%; n = 9), Asian/Pacific Islander (2.8%; n = 8), Native American/Alaska Native (1.4%; n = 4), none of the listed categories (1.4%, n = 4), and did not respond to the item (1.4%, n = 4). There were 220 (78.0%) participants who reported their field of study as counselling, while psychology (12.1%, n = 34), marriage and family therapy (6.4%, n = 18), social work (2.1%, n = 6) were also endorsed by participants, and four participants (1.4%) did not respond this item. Participants identified their work setting as follows: Community outpatient setting (e.g., college counselling center, mental health agency; 47.1%, n = 133), private practice (27.3%, n = 77), inpatient setting (e.g., public, private, or veteran affairs hospital; 20.6%, n = 58), school setting (3.2%, n = 9), and five participants did not respond to the item (1.8%). Participants reported an average client caseload of 25.24 (Mdn = 20.0, SD = 19.39) and providing an average of 20.44 (Mdn = 20.0, SD = 8.60) hours of direct client services per week. Participants had worked with their current supervisors for an average of 1.50 years (Mdn = 1.10, SD = 1.38) across an average of 42.87 (Mdn = 32.0, SD = 37.55) total weekly supervision sessions.

Instruments

Trauma-Informed Practice Scales – Supervision Version (TIPS-SV; Cook et al., 2022)

The TIPS-SV is a 22-item unidimensional measure of supervisees perceptions of their supervisors as adhering to the principles of TIS (Cook et al., 2022). The instrument was adapted from the Trauma-Informed Practice Scales and the Cultural Responsiveness and Inclusivity Scale, created by Goodman et al., (2016), to measure domestic violence survivors’ impressions of the staff as adhering to trauma-informed practices. The TIPS was adapted by changing the word “staff” to “my supervisor.” Participants responded to the 14 items that corresponded to the Trauma-Informed Practice Scale using a four-point Likert-type scale (0 = not at all true to 3 = very true). A sample item was, “My supervisor gives me opportunities to learn how abuse, and other difficulties affect peoples’ ability to think clearly and remember things.” The eight items that corresponded the Cultural Responsiveness and Inclusivity Scale included a fifth-point option of I don’t know, which were treated as missing data (see preliminary analysis section; Goodman et al., 2016; Cook et al., 2022). A sample item included, “My supervisor recognizes that some people or cultures have endured generations of violence, abuse, and other hardships.” Cook et al., (2022) used IRT principles to examine the psychometric properties of the TIPS-SV. Using the partial credit model, the authors found that the 22-items explained 60.35% of the variance in item responses in a unidimensional model. Inter-item correlations were all positive and high (0.44 to 0.80) and the Omega H was 0.8, with all item factor loadings ranging from 0.7 to 0.9. The authors also found evidence that the TIPS-SV can be used to detect differences participants perceptions of TIS across participants, items, and work settings (Cook et al., 2022). In the current study, the internal reliability of the TIPS-SV was 0.94, which indicted excellent consistency amongst the set of scale items. Higher TIPS-SV scores are interpreted to mean that supervisees perceive higher levels of TIS.

Secondary Traumatic Stress Scale (STSS; Bride et al., 2004)

The STSS is a 17-item measure of STS that includes three subscales – Intrusion, Avoidance, and Arousal. Bride et al., (2004) developed the items to align with DSM-IV (APA, 2000) diagnostic criteria of PTSD. We chose to use the STSS because it is the most commonly utilized measure of STS and has excellent evidence of validity and reliability (Molnar et al., 2017). Participants indicate their responses with a five-point Likert scale (1 = never to 5 = very often). Scores for the STSS are created by totaling the participants responses for each subscale or across all items (Bride et al., 2004). Bride (2007, p. 67–68) suggested the following benchmark scores for the STSS: less than 28 signals “little or no STS,” scores between 28 and 37 signals “mild STS,” scores between 38 and 43 signals “moderate STS,” scores between 44 and 48 signals “high STS,” and scores 49 and above signal “severe STS.” Bride (2007) also suggested a cutoff value of 38 and above may be evidence of PTSD, as 93% of persons who scored at least a 38 also met the core criteria of PTSD. In studies with similar samples (e.g., social workers), the STSS has evidence of both convergent and discriminant validity (Bride et al., 2004). The Cronbach alpha for the total score in the current study was 0.93, which is consistent with prior research (e.g., Quinn et al., 2019).

Counselor Burnout Inventory (CBI; Lee et al., 2007)

The CBI is a 20-item instrument of burnout and one that is widely used in the field of counselling (Bardoshi et al., 2019). The CBI includes five subscales: Exhaustion, Incompetence, Negative Work Environment, Devaluing Client, and Deterioration in Personal Life. Participants use a five-point Likert scale (1 = never true to 5 = always true), and possible total scores range from 20 to 100, with the higher the score, the higher endorsement of burnout levels. Bardhoshi et al., (2019) created percentile ranks for the CBI total score and subscales. The percentile benchmarks for the raw total score of the CBI are as follows: 80+ (99.79 + percentile), 54 (75th percentile), 46 (50th percentile), and 38 (25th percentile). The CBI has good convergent evidence with the other measures of burnout (Lee et al., 2007) and adequate to good ability to detect difference in participant responses across items (Cook, Fye, Wind., 2021). The Cronbach alphas for the total score in this study was 0.93, which was consistent with prior research (e.g., Lee et al., 2007).

Coronavirus Anxiety Scale (CAS, Lee 2020)

The CAS is a five item self-report measure of dysfunctional anxiety related to the COVID-19 pandemic. This scale was developed to measure the anxiety experienced by person as it relates to the coronavirus and the pandemic. The scale has been used in studies with participants across the US, with evidence of a single-factor and reliability (Lee, 2020). Higher scores have been shown to be associated with negative outcomes, such as impairment, alcohol use, hopelessness, and suicidality (Lee, 2020). Participants respond using a Likert-type scale (0 = not at all to 4 = nearly every day over the last 2 weeks). A sample item is “I felt paralyzed or frozen when I thought about or was exposed to information about the coronavirus.” The reliability of the scale in the current study was acceptable (α = 0.84), which is comparable to prior research (i.e., α > 0.90; Lee 2020).

Demographic Survey

The first author developed a demographic survey for this study to capture demographic information for participants, including age, gender, race, ethnicity, experience, professional affiliation. We also requested information from participants about their professional experience, including work setting, number of direct client hours per week, case load, productivity requirement, and length of the supervisory relationship (in months and number of sessions).

Procedure

After receiving IRB approval for the study, the first author requested the names and email addresses of counsellors engaged in supervision for licensure from two states in each of the four American Counseling Association (ACA) regions. The requests were granted by licensure boards in six states: Florida, Louisiana, Nebraska, Oregon, Washington, and Wisconsin. No lists were available for any of the states in the North Atlantic Region. Using the provided list, we then invited 5,394 potential participants to complete an online Qualtrics survey; however, 239 emails were unusable. Of the 5,155 usable email addresses, there were 297 usable survey responses (see Data Analysis section), which yielded a response rate of 5.5%. Data collection occurred during (November 2020 to February 2021). We offered ten $25 gift cards that were given to randomly selected participants who completed the survey. Using G*Power 3.1 (Faul et al., 2009), we performed an a priori power analysis, and found a sample size of 160 was needed to detect a moderate effect size of 0.15, with a power of 0.95 and an alpha of 0.05.

Preliminary Analysis

All analyses were conducted in SPSS (Arbuckle, 2018). Prior to this analysis, we first addressed the missing data. There were 312 participants who started the survey, but we excluded 15 participants who did not finish the survey, which yielded 297 useable responses. There were no missing data for the CBI, and low levels of missingness (i.e., less than 5%) for the STSS and CAS, which was determined to be missing completely at random (Cook, 2021). The was also no missing data for the TIPS-SV; however, we treated participants responses of I don’t know for eight items as missing data (Goodman et al., 2016). The total amount of missing data for these eight items was 9.5%. We used expectation-maximization to impute the missing values for the STSS and the I don’t know responses of the TIPS-SV, given this approach can handle large amounts of missing data for which insights into the underlying mechanism for the missing values can be gained from other participants responses (Cook, 2021). We also imputed low levels (i.e., less than 1%) of missing demographic values using a hot deck imputation method, an approach that has also been used to correct missing US Census data (Myers, 2011). We then tested the assumptions of analysis. We identified 15 outliers, using a critical value of 27.88 for Mahalanobis Distance to eliminate multivariate outliers and removing univariate outliers with extreme values (± 3 SD) for the standardized residuals. This yielded a final sample size of 282. No other assumption violations (multicollinearity, linearity, homoscedasticity) were identified.

To answer our research questions, we conducted two hierarchical regression models, one for counsellor burnout and one for STS. For Step 1, we entered the individual factors (e.g., age, race/ethnicity, gender). In Step 2, we entered the organizational factors (e.g., setting, caseload, clinical hours per week, and productivity requirement. In Step 3, we entered the CAS scores. In Step 4, we entered TIS scores. Only a small number of participants endorsed some options for the variables of race/ethnicity and gender, as such, we recoded both variables to increase cell size. For race/ethnicity, we created the category of Black, Indigenous, or People of Color (BIPOC), by collapsing the categories of Black or African American, Latino, Hispanic, or Spanish, Asian, American Indian or Alaska Native, and Multiracial. This approach is consistent with prior research (e.g., Wind et al., 2020). Further, for the gender variable, we collapsed males, transgender, and other into one category. We also dichotomized the variable, productivity requirement, by collapsing participants’ responses of no or maybe into one category. To evaluate practical significance, we interpreted the R2 effect sizes ranges of small (0.02 − 0.12), medium (0.13 − 0.25), and large (≥ 0.26) and f effect size ranges of small (0.10 − 0.24), medium (0.25 − 0.39), and large (≥ 0.40), as suggested by Balkin & Lenz (2021).

Results

The descriptive data, including means, standard deviations, ranges, and correlations for the continuous variables are included in Table I. The full results for the hierarchal regressions for counsellor burnout and STS are shown in Tables II and III, respectively.

Table I Descriptive Information and Correlations of Continuous Variables
Full size table
Table II Hierarchal Regressions of Counsellor Burnout
Full size table
Table III Hierarchal Regressions of STS
Full size table

Predictors of Counsellor Burnout

Step 1, which included the individual factors, yielded a statistically significant equation, F(3,278) = 4.75, p = .003, R2 = 0.05. Step 2 also produced a statistically significant equation F(9,272) = 7.10, p < .001, R2 = 0.19, and the inclusion of these variables explained an additional 14% of the variance in counsellor burnout, ΔF(6,272) = 4.93, p < .001. Step 3 was also statistically significant, F(10,271) = 10.99, p < .001, R2 = 0.29. Including COVID-19 anxiety in Step 3, explained an additional 10% of the variance in counsellor burnout, ΔF(1,271) = 37.37, p < .001. Finally, Step 4, TIS, was significant F(11,270) = 14.82, p < .001, R2 = 0.38. The addition of TIS in Step 4 explained an additional 9% of the variance in counsellor burnout and was statistically significant, ΔF(1,270) = 38.12, p < .001.

We interpreted the final and most parsimonious step of the model of counsellor burnout (i.e., Step 4), which included the individual factors, organizational factors, COVID-19 anxiety, and TIS. The collective variables explained approximately 38% of the variance in counsellor burnout, which was a large effect. Analysis of the beta values revealed that significant predictors in the model included age (B = − 0.25, p < .001, sr2 = 0.03), private setting (as compared to inpatient setting; B = -3.92, p = .045, sr2 < 0.01), caseload (B = 0.18, p < .001, sr2 = 0.05), COVID-19 anxiety (B = 1.52, p < .001, sr2 = 0.09), and TIS (B = − 0.41, p < .001, sr2 = 0.09). Both COVID-19 anxiety and TIS were the greatest contributors to counsellor burnout and were medium effect sizes. The effect size for caseload was also a medium effect, while age and working in a private practice setting (as compared to an inpatient setting) had small effect on counsellor burnout.

Predictors of STS

We then conducted a hierarchal regression of STS. We first entered the individual factors into Step 1, which was statistically significant, F(3,278) = 5.22, p = .002, R2 = 0.05. We then entered the organization variables into Step 2 and this equation was also statistically significant, F(9,272) = 4.57, p < .001, R2 = 0.13. The inclusion of these variables explained an additional 8% of the variance in STS, ΔF(6,272) = 4.07, p < .001. We then entered COVID-19 anxiety into Step 3, which yielded a statistically significant equation, F(10,271) = 12.05, p < .001, R2 = 0.31. The addition of COVID-19 anxiety in Step 3, explained an additional 18% of the variance in STS, ΔF(1,271) = 69.09, p < .001. Finally, we entered TIS into Step 4 and this was significant F(11,270) = 11.61, p < .001, R2 = 0.32. Adding TIS explained an additional 1% of the variance in STS and was statistically significant, ΔF(1,270) = 5.29, p < .001.

We interpreted the final and most parsimonious step of the model, Step 4, which included the individual factors, organizational factors, COVID-19 anxiety, and TIS. The collective set of variables explained 32% of variance in STS, which is a large effect size. The statistically significant variables in the model were age (B = -0.24, p < .01, sr2 = 0.03), caseload (B = 0.11, p = .009, sr2 = 0.02), COVID-19 anxiety (B = 2.11, p < .01, sr2 = 0.17), and TIS (B = − 0.15, p < .05, sr2 = 0.01). Like counsellor burnout, COVID-19 anxiety most substantially contributed to STS (i.e., large effect size). Age, caseload, and TIS were all small effect sizes.

Discussion

We aimed to understand the relationships between individual factors (i.e., age, race/ethnicity, and gender), organizational factors (i.e., setting, caseload, number of clients, productivity requirement), COVID-19 anxiety, and TIS with counsellor burnout and STS in a sample of prelicensed counsellors. Findings of the descriptive data suggests that, on average, prelicensed counsellors in this current study experienced slightly elevated but comparable levels of counsellor burnout, while their endorsement of STS as measured by the STSS was higher than prior studies (e.g., Quinn et al., 2019) and elevated using relevant benchmarks (e.g., Bardhoshi et al., 2019; Bride, 2007). Regarding counsellor burnout, the average CBI score was 48 (SD, which would be the 58th percentile per Bardhoshi et al., (2019). No comparable studies were found that reported CBI scores of prelicensed counselors prior to the pandemic. Regarding STSS, the average score for participants in the current study was 39.50 (SD = 13.10), which is interpreted as moderate STS and is higher the cutoff score for possible PTSD per Bride (2007). Relatedly, the average score is significantly than the scores found by Quinn et al., (2019) in a sample of social workers (M = 33.07, SD = 10.80). We caution readers from making direct comparisons between STSS scores observed in this study and Quinn et al., 2019, as the sample in this study were not prelicensed counsellors. However, we do hope that these comparison studies and the included benchmarks do help to better understand the CBI and STSS scores observed in the current study.

Counsellor Burnout

We found that all steps of the hierarchal regression model were significant. Therefore, we reported the final and most parsimonious step, Step 4, which included the individual factors, organizational factors, COVID-19 anxiety, and TIS. This final model of counsellor burnout collectively explained 38% of the variance. Both COVID-19 anxiety and TIS had the largest effect on counsellor burnout (i.e., medium effect size). Like other front-line professional (e.g., nurses, police officers; Maraqa et al., 2020), prelicensed counsellors experienced worry about COVID-19 during the pandemic and was related to increased burnout. Participants’ perceptions of their supervisors using TIS was inversely related to counsellor burnout, as previously theorized by scholars (Jones & Branco, 2020; Knight et al., 2018).

Although COVID-19 anxiety and TIS were the best predictors of counsellor burnout, we also found that caseload had a statistically significant and medium effect on counsellor burnout, while age and setting type (i.e., working in a private practice as compared to an inpatient facility) had a small effect on counsellor burnout. Consistent with prior research (e.g., Brewer & Shapard 2004; Green et al., 2014), age was inversely related to counsellor burnout, with younger counsellors reporting higher levels of burnout. Our finding that larger client caseloads were related to increased counsellor burnout was also consistent with prior research (e.g., Cook, Fye, Jones et al., 2021; Lawson, 2007; Lawson & Myers, 2011). Consideration of the size of client caseload, particularly with younger counsellors, may be important to addressing counsellor burnout, even if it is only to a small degree. Finally, like prior studies (e.g., Lawson 2007; Lawson & Myers, 2011), we found a statistically significant relationship between counsellor burnout and work setting, with counsellors working in inpatient settings reported higher levels of burnout as compared to those in outpatient settings. None of the other variables examined in this study were statistically or practical significant predictors of counsellor burnout.

STS

Like counsellor burnout, each of the four steps in the model of STS were significant, but it was the final step, Step 4, which included the individual factors, organizational factors, COVID-19 anxiety, and TIS. This constellation of factors explained 32% of variance in STS. Like the model for counsellor burnout, COVID-19 anxiety was the best predictor of STS (i.e., large effect size). This finding may suggest that prelicensed counsellors experience significant anxiety related to COVID-19 alongside experiencing increased STS, which is characterized by PTSD-like symptoms that result from indirect exposure to trauma (Bride et al., 2004). At the time of data collection, participants were exposed to information about COVID-19 in their personal lives as well as their professional lives, which could help to explain this finding.

Like the findings for counsellor burnout, age and caseload were also statistically significant predictors of STS and both variables had a small effect on STS. These findings are also consistent with prior research (Hensel et al., 2015; Quinn et al., 2019). However, because of the small effect size, the practical significance of these two variables may be limited. Unlike counsellor burnout, TIS only accounted for a small percentage of the variance in STS (i.e., small effect size). This finding, coupled with the finding about TIS and counsellor burnout, is a starting point to understanding the theoretical relationship between these concepts (Jones & Branco, 2020; Knight, 2018). A distinguishing feature of STS as compared to burnout is the indirect exposure to trauma (Cieslak et al., 2014). Given that TIS is rooted in educating counsellors about trauma and attending to trauma, we might expect that supervisors who utilize TIS might better prepare their supervisees to engage in trauma-focused work (Knight, 2018). However, it is possible that TIS helps prelicensed counsellor to feel more prepared and efficacious in their professional work (i.e., reduced burnout), but does not have the same practical effect on addressing the harmful effects of indirect trauma (i.e., STS; Bride et al., 2004).

Limitations and Opportunities for Future Research

We must discuss the limitations of the current study and how the findings from this study might inform future research. We attempted to access a diverse and nationally representative sample of prelicensed counsellors. Although our sample is demographically comparable to those in the counselling profession (DataUSA, 2018), the findings from this study may not be representative of counsellors of differing demographic characteristics (e.g., gender, race/ethnicity), geographic location, or professional developmental level. Relatedly, we must also acknowledge a small response rate, although the rate found in this study is comparable to prior research with prelicensed counsellors (e.g., Fye et al., 2021). We employ future scholars to explore the use of data collection methods that may yield a higher response rate and attract a more diverse sample. Relatedly, we must also acknowledge the possibility that participant interest (or lack thereof) was influenced by perceived salience of this topic.

These findings might also be influenced by the timing of data collection (November 2020 to February 2021). It would be important to understand how or if the COVID-19 anxiety that emerged during the pandemic and were related to counsellor burnout and STS will remain in an ongoing or post pandemic environment, thereby shifting the personal and workplace stressors of prelicensed counsellors. Relatedly, since our study was cross-sectional in designed, we were unable to assume causality with the results. Future research should use a longitudinal design with at least two timepoints. A noteworthy limitation is that we did not examine additional individual and organizations variables that may be relevant to both burnout and STS. For example, Hensel et al., (2015) also found that counsellors own trauma histories, including the type and extent of the exposure, played a role in STS, specifically. These variables were not measured in the current study. Future researcher should study information related to the type and extent of exposure counsellors experience when working with clients with trauma histories.

Implications for Prelicensed Counsellors and Supervisors

Informed by the findings of this study, there are implications for prelicensed counsellors and their supervisors. Specifically, understanding factors related to prelicensed counsellors’ burnout and STS are important to consider when providing ethical services to meet clients’ needs on the job (Jones & Branco, 2020). COVID-19 anxiety was the strongest predictor of counsellor burnout and STS for participants in the current study. The pandemic shifted the job demands for counsellors, and it unclear to us how or if this change will remain as the pandemic continues or in a post-pandemic environment. Nonetheless, prelicensed counsellors should monitor their own anxieties related to COVID-19 and signs of possible burnout or STS related to these worries and seek guidance from supervisors as their professional concerns arise (Cook & Sackett, 2018). Further, supervisors should also monitor their supervisees’ personal and professional wellness and engage their supervisees in an open dialogue about their feelings of burnout or STS (Cook, Fye, Wind et al., 2021), and if these feelings are related to their COVID-19 anxiety or they have been worsened since the pandemic began. This seems particularly salient given that, on average, participants in this study reported burnout scores in the 58th percentile and STSS scores that may signal moderate STS or evidence of PTSD (Bride, 2007).

Supervisors can begin to integrate TIS into their existing supervisory practices in order to help reduce or avoid burnout and STS, and hopefully to improve client care (Knight, 2018). In the current study, TIS had a stronger effect on reducing counsellor burnout as compared to the effect on STS, which was small. However, supervisors should consider the possible impact of even a small reduction in counsellor burnout or STS to their supervisees’ wellbeing and the clients whom they serve, particularly given the high anxiety related to COVID-19 that some supervisees experienced. Thus, using TIS principles as a guide, Knight (2018) as well as Jones & Branco (2020) notes that supervisors integrate TIS into their existing supervision practices by doing the following: (a) establishing safety (e.g., supervisees feel valued and able to speak honestly; (b) trust (e.g., supervisees experience a sense of trust with their supervisors); (c) choice (e.g., supervisees have choices in their supervision, to the degree it is possible; (d) collaboration (e.g., mutual sharing of ideas and egalitarian approach from the supervision; (e) empowerment (e.g., supervisees development is supported and their clinical independence is fostered); and (f) cultural, historical, and gender issues (e.g., supervisees and supervisors openly discuss topics of cultural, historical, and gender based traumas and how that manifest in the supervisory and client relationships). It is important to remember that TIS is a framework for practice as compared to a theory or treatment technique (Knight, 2018). Therefore, supervisors can easily and immediately integrate TIS into their own supervision practices. Interested supervisors should consider focusing on the following aspects of TIS, which were the four most easily endorsed items of the TIPS-SV (Cook et al., 2022): Respecting peoples’ sexual orientation and gender expression, respecting peoples’ cultural backgrounds, treating supervisees with dignity, and treating people with physical or mental challenges with dignity. We encourage readers who are interested in learning more about supervisees perceptions of TIS to review Cook et al., (2022).

Finally, age and caseload were significant predictors of counsellor burnout and STS, although the effect sizes were small to medium. Supervisors should be aware that large caseloads may exacerbate counsellor burnout and STS, particularly for younger counsellors. Although we did not study caseload characteristics in this study, a useful strategy suggested by other scholars (e.g., Hensel et al., 2015; Lawson, 2007) is for supervisors to avoid over assigning clients with complex trauma histories to counsellors, in addition to avoiding large caseload assignments. This suggestion seems acutely relevant for young or novice counsellors (i.e., prelicensed counsellors).

Conclusion

Identifying solutions to address prelicensed counsellors’ experiences of burnout and STS are of critical importance in the field of counselling. As evidenced by findings from this exploratory study, there are numerous factors that increase or reduce prelicensed counsellors’ experiences of burnout and STS. Accordingly, both supervisors and prelicensed counsellors must collectively work ensuring that clients are receiving quality clinical care and that counsellors themselves are avoiding the experience of personal issues like burnout and STS.