An Investigation of Secondary Traumatic Stress and Trauma-informed Care Utilization in School Personnel

Abstract

Trauma-informed practices in schools are designed to address the impacts of trauma on students and increase supports for school personnel who are delivering this care The impact of a trauma-informed school-based intervention. Journal of Adolescence, 43, 142–147, Mendelsonet al., 2015. Research has established that professionals trained to implement the approach may have secondary traumatic stress (STS) reactions that could interfere with successful implementation (Stevens al., 2020). In this study it was hypothesized that increased use of trauma-informed care strategies would be associated with decreases in total STS scores, as well as all STS subscale scores at the end of a system’s transformation initiative, controlling for sex, age, education, years worked in schools, and exposure to student trauma awareness at baseline. The Trauma Sensitive Schools Checklist (TSSC) and the Secondary Traumatic Stress Scale (STSS) were used to measure study outcomes in a sample of 205 school personnel at baseline and follow up. Statistically significant improvement in STSS scores and TSSC score were noted from Time 1 to Time 2. As hypothesized, improvements in TSSC scores were associated with decreased levels of STS over time, controlling for the covariates. However, the symptom domains of intrusion and arousal impacted this relationship in a differential manner than avoidance and alterations in cognitions and mood. This study provides evidence that increased use of trauma-informed care practices can positively impact the STS levels of school personnel, though special attention should be paid to those with high levels of intrusion or arousal.

Introduction

The 2015 National Survey of Children's Exposure to Violence documents that 24.5% of youth age 0–17 witness violence in their lifetime, 15.2% are maltreated, and 5% experience sexual offenses, with rates increasing significantly during adolescence (Finkelhor et al., 2015). Children and youth whose exposures involve interpersonal violence within the family and/or sexual assault and physical abuse by paramours, are at highest risk for PTSD (Breslau et al., 2004; Copeland et al., 2007; McLaughlin et al., 2013). Posttraumatic Stress Disorder (PTSD) prevalence rates determined by epidemiological studies note that 30% to 70% of children with such exposures will develop PTSD (Lewis et al., 2019; McLaughlin et al., 2013). Even those who do not qualify for a diagnosis of PTSD may report clinically significant symptoms and functional impairment (Carrion et al., 2002; McLaughlin et al., 2015).

One documented area of disturbance for children and youth with trauma conditions relates to school functioning. Studies involving elementary school students demonstrate that as symptoms of traumatic stress increase, reading, math and science achievement scores decrease (Goodman et al., 2012). Childhood traumatic stress has been noted to be highly correlated with impairments in working memory and expressive language skills, as well as lower scores on IQ and academic achievement tests (Perfect et al., 2016) and increased odds of having an Individualized Education Program for learning or behavior problems (Goodman et al., 2012). However, research with school personnel document uncertainty and perceived incompetence in addressing the needs of students with trauma conditions (Alisic, 2012; Alisic et al., 2012; Chafouleas et al., 2016).

In response to these concerns, efforts towards implementing trauma-informed practices in schools to address the impacts of trauma on students and increase supports for school personnel who are delivering this care are on the rise (Alisic et al., 2012; Mendelson et al., 2015). According to Substance Abuse and Mental Health Services Administration (2014) “a program, organization, or system that is trauma-informed realizes the widespread impact of trauma and understands potential paths for recovery; recognizes the signs and symptoms of trauma in clients, families, staff, and others involved with the system; and responds by fully integrating knowledge about trauma into policies, procedures, and practices, and seeks to actively resist re-traumatization" (p.9). The integration of trauma-informed care in education is further spurred by preliminary evidence demonstrating that trauma-specific interventions delivered in the school environment improve access and spur treatment completion, reduce traumatic stress reactions, and improve functioning among students (Rolfsnes & Idsoe, 2011). Even so, rigorous investigations of the impact of school-based trauma-informed care efforts are lacking (Maynard et al., 2019).

Although trauma-informed care in education may be considered a student-focused approach, one cannot expect school personnel to enact such principles if the organizational environment does not extend and apply these principles to the staff. This is especially important as research has emerged documenting secondary traumatic stress symptoms (e.g. trauma responses in those exposed to the trauma details of others) in school personnel indirectly exposed to student trauma (Stevens et al., 2020; Berger et al., 2016; Smith Hatcher et al., 2011). These exposures can occur during school-based counseling sessions when details of a traumatic encounter may be disclosed; seeing the aftermath of trauma in the emotional dysregulation and behavior of a child (e.g. decompensation, hoarding, self-harming, extreme aggression), hearing or retelling trauma details when working with other professionals and through the trauma narratives of caregivers. This process of emotional contagion can occur outside the awareness of the professional, but may have mild to severe impact on their levels of distress and ability to function (Motta, 2020; Sprang et al., 2021). The limited capacity model of attention suggests that task performance may be inhibited if an individual is preoccupied with trauma-related memories and reactions (Bailey et al., 2015). In this scenario, bottom-up, attention-grabbing stimuli (e.g. stress responses, trauma symptoms) may override top-down decisions driven by an individual’s personal or professional goals (e.g. desire to act in a trauma-informed manner) (McEwen et al., 2015). Similarly, when faced with threatening stimuli, individuals are naturally driven by two motivations- proactive survival behaviors or avoidance of harmful situations, the latter of which can activate a defensive cascade of responses that can deplete coping resources when the threat intensity is high (Cacioppo et al., 2011; Ito & Cacioppo, 2005). Studies note that these processes can influence performance outcomes such as learning, persuasion, and a professional’s sense of enjoyment in their work (Bailey et al., 2015; Fisher et al., 2018).

There is also evidence that the psychological strain caused by high job demands and inadequate resources leads to negative organizational outcomes such as poor performance and absenteeism (Bakker & Demerouti, 2007; Bakker et al., 2014), burnout (Aloe et al., 2014; ); low job satisfaction (McCarthy et al., 2017) and attrition (Harmsen et al., 2018). Applied to a trauma context, the burden of indirect exposure and the resulting traumatic stress in an environment where trauma-informed care is not adequately applied to students and staff, may result in a job demand-resource need imbalance that produces such negative outcomes. Unfortunately, there is little research available to elucidate the interplay between secondary traumatic stress, in total and by domain, and the utilization of trauma-informed care strategies in school personnel. To address this gap in the literature, this study investigates whether increases in TSSC total scores are associated with decreased levels of total STS and distress in each symptom cluster in school personnel. It was hypothesized that improvements in total TSSC scores would be associated with decreases in total STS scores and all subscale scores at the end of the initiative, controlling for sex, age, education, years worked in a school setting, and exposure to student trauma awareness at baseline.

Methods

Procedure

This study involves secondary analysis of data collected from 249 participants (72.6% of possible respondents) who were involved in two, statewide Trauma-informed Practices for Educators learning collaboratives. The purpose of this initiative was to create trauma sensitive schools, by allowing teams of school personnel to learn together, and move beyond trauma awareness to becoming trauma-informed in practice. The trauma informed care model utilized in this study was based on the framework proposed by Harris and Fallot (2001) as well trauma theory and research developed over the past decades. This approach is hallmarked by data that suggests that if trauma experiences are not processed and resolved they are stored as physiological and psychological reactions to situations (benign and otherwise) that can a manifest in ways that seem unrelated to the trauma event (e.g. oppositional behavior) (van Loon et al., 2004). Intrusive memories, thoughts and sensations can be evoked by typical interactions and relationships that may seem unrelated, unprovoked or unpredictable to school personnel (Streb et al., 2017). Understanding trauma theory, and the neurobiology of a traumatic stress response allows reactions to be interpreted with a new lens, and provides a pathway to intervention that is informed by several decades of trauma research (Zaleski et al., 2016). The learning collaborative model of training involved four, day-long learning sessions that provided information about foundational aspects of trauma, the neurobiological and developmental consequences of trauma, trauma-informed care strategies organized by multi-tiered systems of support, secondary traumatic stress and strategies to attenuate the effects of indirect exposure, how to integrate trauma-informed care into school safety legislation mandates, opportunities for case application, goal setting and the creation of school-specific action plans. Three to five topical learning calls were also included to supplement the information provided in the learning sessions. A multidisciplinary faculty of trauma experts representing the fields of psychology, social work and education delivered the content and served as coaches and consultants to the participants.

The data collected represents baseline and end of the learning collaborative responses to a cloud-based survey that was voluntary and anonymous but linked (baseline to post) by a six-item code generated by and known only to respondents. Only those completing surveys at each time interval were included in the study so that the full effect of time could be measured. The study was approved by the appropriate university Institutional Review Board.

Measures

The Trauma-Sensitive Schools Checklist (Lesley University and Trauma and Learning Policy Initiative, 2012) is a 26-item measure that assesses a school’s status and progress toward becoming a trauma-sensitive school. A trauma-sensitive school is operationally defined by this measure as a safe and respectful school culture that facilitates healthy and caring relationships between students and school staff, and an environment that promotes emotional and behavioral self-regulation, academic success, and physical and emotional well-being. While the TSCC contains five subscales: school-wide policies and practices (8 items), classroom strategies and techniques (8 items), collaborations and linkages with mental health resources (5 items), family partnerships (3 items), and community linkages (2 items), we calculated a total TSSC score to obtain a global measure of the extent to which the school culture is optimized to promote healthy and caring relationships between students and school staff. Participants received a total score by multiplying the total possible items by the number of choices they could select from: “1” indicating the element is not at all in place, “2”, the element is partially in place, “3”, the element is mostly in place, and “4”, the element is fully in place. In this case, the highest score participants could receive is 104 (26 items*4 possible choices). Internal consistencies for TSSC subscales were noted in a study by Baker et al. (2020) to be good to excellent (αs = 0.83–0.93).

The Secondary Traumatic Stress Scale (STSS) for the DSM-5 developed by Bride (2013) contains 21-items that relate to the professional’s self-reported symptoms of STS in four domains (intrusion, avoidance, negative mood and cognition, and alterations in arousal and reactivity). A 5-point Likert scale was used to determine the frequency of symptoms with 1 representing never and 5 signifying very often. Scores for each of the subscales were calculated. Possible total scores on the STSS range from 20–100, with higher scores indicating higher levels of STS. Item #5 (I felt discouraged about the future) on the DSM-5 version of the STSS does not map directly onto the most current symptom criteria for PTSD and was not included in the calculation of scores in this study. With one exception of avoidance at Time 2 (0.85 for T1 and 0.50 for T2), alpha coefficients for intrusion (0.92 for T1 and T2), negative cognition and mood (0.95 for T1 and 0.93 for T2), arousal (0.93 for T1 and 0.95 for T2), and total STSS (0.96 for T1 and 0.92 for T2) demonstrated a high level of internal consistency.

Demographic information was solicited regarding sex (0 = other; 1 = female; 2 = male), actual age; and the number of years the individual has worked in a school setting. Additionally, respondents were queried about their highest level of education; and amount of exposure to details of student trauma (1 = once a day, 2 = once a week; 3 = once a month; 4 = less than three times over the school year; 5 = between 4 and 8 times over the school year).

From the initial dataset of 249 responses, cases were deleted due to missing Time 2 TSSC data (n = 21), and missing observations for sex and age (n = 21). A disproportionate number of participants (n = 2) reported “other” for sex; and thus, were also deleted, thereby leaving a total analytic sample of 205 individuals. Secondly, descriptive statistics of the final analytic sample were conducted (see Table 1). Means and standard deviations for each of the single item responses for Time 1 and Time 2 are displayed in Table 2. Paired Samples T-tests were conducted to determine if mean scores differed significantly from Time 1 to Time 2. Finally, Generalized Linear Mixed Models (GLMM) were conducted for the total TSSC score using STATA 17 software (StataCorp, 2021). GLMM’s were conducted to account for the non-independence of repeat observations from the same subject. The total TSSC score served as the dependent variable while each of the STS subscale scores (intrusion, avoidance, negative cognition and mood, and arousal) were included in the model as independent variables. Analyses controlled for sex, age, education, years worked in a school setting, and exposure to student trauma awareness at baseline.

Table 1.

Descriptives (N = 205)

Variable	Mean (std) Min, Max
Age	40.32 (8.04) 23,61
Years in Education	14.19 (7.81) 0.75,30
N (%)
Sex
Female	171 (83.41)
Male	34 (16.59)
Other	0
Job Role
Teacher	44 (21.46)
School-based Mental Health	84 (40.98)
Administration	43 (20.98)
Other Certified	26 (12.68)
Other Classified	8 (3.9)
Education
Associate degree or Less	3 (1.46)
Bachelor's Degree	15 (7.32)
Master's Degree	60 (29.27)
Rank 1	122 (59.51)
Doctorate	5 (2.44)

Table 2.

Mean Scores for Time 1 and 2. (N = 205)

		Time 1	Time 2
Variable		Mean (std) Min, Max
Trauma Sensitive School Checklist (TSSC) SCORE		50.6 (16.08) 26,96	71.71 (16.39) 29,102
Secondary Traumatic Stress Scale (STSS)
Intrusion		11.09 (3.55) 5,20	7.39 (3.27) 2,20
Avoidance		3.93 (1.76) 2,12	2.4 (1.08) 1,7
Cognitions and Mood		13.67 (4.97) 6,29	9.69 (4.07) 2,28
Arousal		12.8 (4.27) 6,24	11.05 (4.12) 2,24
Total STSS score		41.49 (12.65) 20,73	30.49 (9.15) 12,59

Paired Samples T-test showed that each of the mean scores improved from Time 1 to Time 2 (p < .001)

Results

Table 1 provides descriptive information about the sample. On average, the sample was 40 years of age, and garnered 14 years of experience working in an educational setting at Time 1 data collection. Results further revealed that over three-fourths of the sample included females (n = 171, 83%) and school-based mental health providers and social workers (n = 84, 41%). Most of the sample (n = 122, 60%) reported “Rank 1” as their highest level of education. Rank 1 is a post-graduate certification that teachers receive.

Table 2 provides mean scores for TSSC and STSS scores. Participants received an average TSSC score of 50.6 (SD = 16.08) out of a possible total score of 104 at Time 1. This score improved significantly to 71.71 (SD = 16.39) at Time 2. Mean scores for total STSS scores as well as for each of the STSS subscales, including intrusive thoughts, avoidance, negative cognitions and mood, and arousal decreased significantly from Time 1 to Time 2.

Table 3 includes odds ratios, robust standard errors, 95% confidence intervals, and p-values from the Generalized Linear Mixed model. First, results showed a significant association between total TSSC scores and STSS intrusion. At baseline, as intrusion increased, total TSSC scores decreased (B = 0.97, p < 0.05). At Time 2, TSSC scores did not change, even as STS intrusion increased (B = 1.04, p = 0.045). Secondly, results showed a significant association between TSSC scores and STS arousal. At baseline and Time 2, as arousal increased, total TSSC scores increased (B = 1.06), p < 0.001). However, there was less of a change at Time 2 (B = 0.097, p = 0.056), suggesting that TSSC scores did not change as much over time, when STS arousal lingered.

Table 3.

Generalized Linear Mixed Model Predicting TSSC Total Score (N = 205)

	Odds	Robust	[95% conf. interval]		p
	Ratio	Std Err
Time (post)	2.09	0.28	1.61	2.73	0.00
Age	1.01	0.01	0.99	1.02	0.55
Sex—Male	0.99	0.13	0.76	1.28	0.93
Job Role
School-based Mental Health	1.31	0.17	1.01	1.69	0.04
Administration	1.35	0.21	1.00	1.83	0.05
Other Certified	1.08	0.21	0.74	1.58	0.68
Other Classified	0.98	0.29	0.55	1.74	0.95
Education
Bachelor's Degree	1.60	0.48	0.89	2.89	0.12
Master's Degree	1.21	0.34	0.70	2.10	0.49
Rank 1	0.93	0.26	0.53	1.61	0.78
Doctorate	1.38	0.54	0.64	2.97	0.41
Trauma Awareness
At least once a week	1.19	1.21	0.97	1.45	0.09
At least once a month	1.25	0.22	0.89	1.78	0.20
Less than 8 times over the school year	1.55	0.41	0.93	2.61	0.09
Years in Education	1.01	0.01	0.99	1.03	0.40
Secondary Traumatic Stress (STS)
STS Intrusion	0.97	0.01	0.94	0.99	0.02
STS Intrusion*Post Time	1.04	0.02	1.00	1.07	.045
STS Avoidance	1.01	0.03	0.96	1.06	0.84
STS Avoidance*Post Time	1.02	0.04	0.95	1.10	0.62
STS Negative Cognition & Mood	0.98	0.02	0.95	1.01	0.15
STS Negative Cognition & Mood*Post Time	1.01	0.01	0.98	1.04	0.45
STS Arousal	1.06	0.02	1.03	1.10	0.00
STS Arousal*Post Time	0.97	0.02	0.94	1.00	.056

Third, results showed no significant associations between total TSSC scores and 1) STSS avoidance, and 2) STSS negative cognition and mood. Given the varied findings for each subscale, the STSS total score variable was not included in the final GLM model. Graphs depicting significant results for the two subscales (intrusion and arousal) are included in Figs. 1 and 2. Finally, none of the identified covariates were significantly related to the specified outcomes.

Fig. 1.

Percentage of trauma-sensitive school checklist total score (four categorical options for each of the 26 items = total possible score of 104). Results depict the percentage for each single item response for “STS–Intrusion” at Time 1 (pre) and Time 2 (post). Analyses control for covariates

Fig. 2.

Percentage of trauma-sensitive school checklist total score (four categorical options for each of the 26 items = total possible score of 104). Results depict the percentage for each single item response for “STS – Arousal” at Time 1 (pre) and Time 2 (post). Analyses control for covariates

Discussion

While trauma-informed schools are designed to interrupt the pathway from child traumatic stress to poor student outcomes, the model inherently relies on school personnel to enact these strategies and become the stewards of implementation and sustainment (McInerney & McKlindon, 2014). Investigations regarding the effectiveness of trauma-informed care approaches must consider the role that secondary traumatic stress plays in the adoption and utilization of these strategies. This study is one of the first to examine the interplay between secondary traumatic stress symptoms in school personnel, and the enactment of trauma-informed care practices in an educational setting.

The study utilized a sample of school personnel who in the aggregate were experiencing mild to moderate symptoms of secondary traumatic stress at baseline. Although measurement differences prevent direct comparison of STS scores across studies, the results of this research are consistent with other reported sample means (Christian-Brandt et al., 2020; Castro Schepers & Young, 2022), but lower than those noted by Borntrager et al. (2012) and Abraham-Cook (2012). All subscale and total scores of the STSS improved in a statistically significant manner from baseline to the end of the initiative, suggesting participation in the trauma-informed program was positively associated with decreased symptom severity. Although an understudied phenomenon, emerging research converges with this finding. Castro Schepers and Young (2022) report a positive relationship between participation in trauma-informed practices seminars and decreased STS levels. Possible reasons for these type of improvements are noted in a qualitative study where teachers reported increased regulatory abilities, enhanced understanding of their own stress responses, feeling calmer, improved relational abilities (e.g. stronger workplace relationships, more able to be a role-model, higher involvement in healthy relationships) and more confidence, gratitude, and resilience following participation in a trauma-informed practices initiative (Brunzell et al., 2021). It may be that trauma-informed care training improves professional outcomes by promoting proactive skills and strategies to deal with child trauma, and in a parallel process, allows these professionals to cultivate and grow their own internal assets and resources (Kern, 2014). So in addition to some symptom relief, school personnel may be enhancing their well-being by applying the same trauma-informed care principles to themselves that they are using with their students.

It was hypothesized that improvements in total TSSC scores would be associated with decreases in STS by the end of the initiative, controlling for sex, age, education, years worked in a school setting, and exposure to student trauma awareness at baseline. Total TSSC scores improved from baseline to follow up at a statistically significant rate, but at Time 2, it appears that TSSC scores are impacted by STS symptoms of intrusion and arousal. While the TIPE learning collaborative might be providing educators the skills to remain as trauma-informed as they were in the past, even when they experience intrusion and arousal, it is clear TSSC scores do not improve significantly when professionals are experiencing high levels of these symptoms. These findings are consistent with what we know about the neurobiological stress response, that is, once activated (in this case by indirect exposure), higher-order tasks such as critical thinking and empathic responding can be stalled, as the autonomic nervous system redirects energy toward survival (McEwen et al., 2015). The differential pattern of responding by symptom cluster is not unfamiliar, as a recent study found that higher levels of intrusion and arousal were more likely to predict PTSD in ways that alterations in cognition and mood and avoidance did not (Sprang et al., 2021). In this manner, it may be that the clinical disturbance associated with intrusion and arousal is dysregulating in a way that makes those symptoms less compatible with enacting trauma-informed care behaviors, and/or applying the skills to oneself. Additional research is needed to determine why avoidance and alterations in cognition and mood did not significantly impact trauma-informed care efforts in the same way as the other symptom clusters, and if a typology of STS is emerging across studies. A partial answer may lie in the adaptive pathways activated by or associated with symptom severity in specific clusters. Recent research has documented the salience of emotional regulation strategies used in response to secondary traumatic stress, and found that expressive suppression was positivel associated with high arousal symptoms, but only marginally related to other trauma reactions (Măirean, 2016). This underscores the need to consider the primary coping response of those impacted by indirect exposure, and how this may impact their receptivity to and enactment of trauma-informed care strategies. Of particular interest would be examination of how these domain specific findings might vary based on the use of cognitive reappraisals versus expressive suppression in affected groups.

Limitations

This study utilized survey methodology (allowing for voluntary participation) that can be influenced by self-selection bias. However the overall response rate of over 70% suggests these effects were somewhat attenuated. The training approach used was grounded in trauma theory and practice and framed by well-accepted trauma informed care principles, however the results of this study cannot be generalized to scenarios where different content or implementation models are applied. Additionally, these data reflect school personnel’s self-reports of trauma-informed care actions, and therefore could be influenced by social desirability concerns. The anonymous nature of the survey was designed to minimize these effects, though the inclusion of additional measures to assess the quality and fidelity of trauma-informed care implementation would strengthen future research efforts. Due to the number of covariates included in the present study and sample size considerations, organizational level determinants were not included despite their importance in the overall success of trauma-informed school integration. Next steps with this dataset include the investigation of macro-level factors that may influence the identified outcomes and interactions between variables.

Conclusion

While trauma-informed care is universally available and recommended, the professionals trained to implement the approach may have secondary traumatic stress reactions that could interfere with successful implementation. This study provides preliminary evidence that as an individual increases their use of trauma-informed practices, their personal wellbeing improves, a testament to the fact that the skills are being integrated into professional life and practice, and beyond a student body only application. However, those with higher intrusion and arousal symptoms of traumatic stress may not benefit in the same way. Future research should investigate potential strategies to address these specific symptom clusters so that the promise of trauma-informed care can benefit students who are experiencing trauma as well as those that educate and support them.

Declarations

Conflict of Interest

On behalf of all authors, the corresponding author states that there is no conflict of interest.