Secondary Trauma And Burnout On Relationship Satisfaction In Helping Professions

Participants

The sample consisted of 120 participants; 83.3 percent were female working in the helping professions: 25.8 percent as psychologists and 74.2 percent as social workers. Purposive sampling method was used where the participants were selected from the list of licensed psychologists and social workers who were be employed in psychological counselling centres, school counselling centres, and social care institutions for children, youth and adults; The mean age was 38.88 years (SD =11.16) with a range of 22-67 years. The majority of the sample had a Master's degree (61.62 percent) and the rest had a Bachelor's degree. The average years of professional experience was 10.34 (SD = 9.30) and they worked approximately 31.44 (SD = 10.31) hours per week. The majority of the sample reported that they were married (71. 67 %) during the survey, 10.83 % of participants were in a relationship, 4.16% reported a common-law partner and 13.33% were currently not in a relationship (single/separated/divorced/widow). The majority of participants (92.12 %) worked in public institutions. About a half of the sample (44.16 %) had low caseload (0 to 10 patients per week) or medium (11-27 patients per week; 41.67 %), and 14.17 % had high caseload (more than 28 of patients per week).

Measures

The Demographic Information Questionnaire (DIQ) was created specifically for the purpose of this study. Items on the DIQ comprised standard demographic information regarding respondents’ age, gender, family status, children, and occupation. In addition, the DIQ gathered information regarding aspects of respondents’ professional activities such as length of work experience, average number of work hours, a field of practice, and typical work-related activities.

The Professional Quality of Life Scale (ProQOL) constructed by Stamm (2010) assesses the positive and negative consequences of working with traumatized people. This 30-item scale contains three subscales: the first is compassion satisfaction, which is characterized as satisfaction from providing compassion and help, as well as the pleasure of the work that one performs. The second scale measures burnout and feelings of hopelessness and problems at work. The third subscale measures compassion fatigue. Cronbach's alpha values are .88 for the compassion satisfaction subscale, .81 for the secondary traumatic stress subscale, .75 for burnout subscale (Stamm, 2010). In the current research, alpha reliabilities for compassion satisfaction, secondary traumatic stress, burnout, and overall scale were .80, .81, .72, and .88, respectively.

The Brief Trauma Questionnaire (BTQ; Schnurr, Vielhauer, Weathers, & Findler, 1999) is a brief ten-item self-report measure used to determinate if respondents have ever experienced a traumatic event (experiencing combat, unwanted sexual contact or sudden death of a close friend or family member). BTQ items target three levels: exposure to trauma, life threat or serious injury of the trauma. Interrater reliability reported by Schnurr et al. (2002) as kappa coefficients were above .70 (range = .74–1.00) which indicate substantial agreement. In the current research, alpha reliability for overall scale was .68.

The Dyadic Adjustment Scale (DAS; Spanier, 1976) is a measure that assesses the quality of marriage and other similar dyads. DAS comprises 32 items of various response formats developed to measure dyadic adjustment comprising four subscales, namely Dyadic Consensus, Dyadic Satisfaction, and Dyadic Cohesion, and Affective Expression. The Affective Expression subscale was found to produce scores with poor Cronbach's alpha across studies (Graham, Liu, & Jeziorski, 2006); therefore this subscale was not used in the research reported here. In the present research, the DAS version comprised of these subscales: Dyadic Consensus (the degree to which the couple agrees on matters of importance to the relationship), Dyadic Satisfaction (the degree to which the couple is satisfied with their relationship), Dyadic Cohesion (the degree of closeness and shared activities experienced by the couple). Spanier (1976) reported the internal reliability of DAS as Cronbach’s alpha (α = .96). In the present research, the internal reliability of a total score of DAS was Cronbach’s alpha .82, and for subscales as follows: consensus dimension (α = .84), satisfaction dimension (α = .71), cohesion dimension (α = .82).

Data Collection and Analysis

The questionnaire forms were sent to the participants who had provided written approval by conventional mail with a return envelope. Of the 180 respondents addressed, 136 returned questionnaires and 5 were excluded for data incompleteness. Male (N = 11) were also excluded due to low frequency. Descriptive statistical analysis of data was performed by means of Pearson and Spearman correlations, ANOVA, and Hierarchical Multiple Regression. SPSS version 21.0 was used. A p-value level <0.05 was adopted for the statistical significance.

Primary, secondary traumatic stress, and burnout and relationship satisfaction

The occurrence of primary traumatic events was relatively low (AM = 1.33, SD = 0.73); approximately one life traumatic event was experienced by the participants. A low frequency of life trauma occurred in 46.6 percent, 14.4 percent reported a moderate life trauma, 3.3 percent a high occurrence of the primary trauma, and no traumatic events occurred for 35.6 percent. The level of secondary traumatic stress was perceived as moderate (AM = 17.18, SD = 6.85) according to the criterion set up by Stamm (2010) as 22 or less. Similarly, the level of burnout was also low in comparison to abovementioned criterion (AM = 18.71, SD = 5.50). Compassion satisfaction was perceived as moderate (AM = 39.15, SD = 4.78).

Correlation analysis yielded some interesting results. Significant correlation was found between secondary traumatic stress and age of participants (r = -0.31, p < 0.01), years of practice (r = - 0.16, p < 0.05), overall primary trauma history (r = 0.26, p < 0.01). However, no significant correlations were found among secondary traumatic stress and dyadic consensus in relationship (r = 0.09, p = 0.19) and dyadic satisfaction in relationship (r = 0.07, p = n.s.). Burnout was negatively associated with level of education (r = - 0.23, p < 0.01). No significant correlation was found among the dyadic consensus in the relationship, dyadic satisfaction in the relationship and the level of education and caseload (correlation coefficients ranged from 0.03 to 0.09).

Differences in secondary traumatic stress and burnout by gender, education level, specialization and family characteristics

The performed statistical analysis did not yield the significant differences in secondary traumatic stress according to gender (F = 2.62, p = 0.11), family status (F = 2.54, p = 0.16.). On the other hand, significant differences were revealed in secondary traumatic stress due to the variable "having children"; participants without children experienced a higher level of secondary traumatic stress (AM = 19.79, SD = 8.04) in comparison to their counterparts with children (AM = 16.16, SD = 6.09).

With regard to burnout, no differences were observed due to gender (F = 0.58, p = 0.49) or family status (F = 2.31, p = 0.18). Participants who lived with partners reported AM = 18.21 (SD = 5.39) on burnout and participants who do not live with partners reported a higher burnout level (AM = 20.91, SD = 5.55), with significant difference (F = 4.17, p = 0.03). Additionally, significant statistical difference was observed for 'with children' and 'no children' (F = 4.53, p = 0.02) where participants with children reported lower levels of burnout (AM = 17.65, SD = 4.78) than participants who had no children (AM = 21.45, SD = 6.33). Conversely, there were no significant gender differences in compassion satisfaction (F = 0.54, p = 0.46).

Statistically significant differences were revealed in the level of secondary traumatic stress and burnout in terms of occupation and education. Higher levels of secondary stress and burnout were perceived by social workers compared with psychologists (F = 4.29, p = 0.01). Furthermore, the professionals with the bachelor degree reported a higher level of burnout and secondary traumatic stress in comparison with those with a master’s degree (p ranged from 0.01 to 0.02).

Regression analyses

A three-step hierarchical multiple regression was conducted to test the proposed model and control for the effect of demographic and professional variables (Table 2 ). Since no prior hypotheses had been made to determine the order of entry of the predictor variables, a direct method was used for the multiple linear regression analyses.

Model 1 (Table 2 ) included exposure to trauma, a threat of serious injury, and overall primary trauma. Model 2 included all variables in model 1 plus the variables of the proposed model: dyadic consensus, dyadic cohesion, and dyadic satisfaction. Model 3 included the participant variables: work hours, caseload, household chores, overall primary trauma, dyadic, consensus, dyadic cohesion, and dyadic satisfaction. The dependent variable was Secondary Traumatic Stress. Results of model 3 indicated that the combined effect of work hours, caseload, household chores, overall primary trauma, dyadic consensus, dyadic cohesion, and dyadic satisfaction explained 13% of the variance in Secondary Traumatic Stress (F = 2.31, p = .034). The participant variables were predictors of STS as a whole set with the strongest predictor overall being primary trauma (β = 0.23).

Model 1 (Table 3 ) included exposure to trauma, a threat of serious injury, and overall primary trauma. Model 2 included all variables in model 1 plus the variables of the dyadic adjustment: dyadic consensus, dyadic cohesion, and dyadic satisfaction. Model 3 included the participant’s variables: work hours, caseload, household chores, overall primary trauma, dyadic, consensus, dyadic cohesion, and dyadic satisfaction. The dependent variable was Burnout.

Results of model 3 indicated that the combined effect of work hours, caseload, household chores, overall primary trauma, dyadic consensus, dyadic cohesion, and dyadic satisfaction explained 18% of the variance in Burnout (F = 3.31, p = .003). The participant variables were predictors of Burnout as a whole set with the strongest predictor being overall primary trauma (β = 0.21) which was similar to STS.

In searching for predictors of relationship satisfaction, additional regression analyses were conducted. Results demonstrated significant correlations between dimensions of relationship satisfaction and burnout. Burnout was confirmed as the only significant (negative) predictor of consensus (R2 = 0.30, F = 9.69, p = 0.00; β = 0.26) and satisfaction in relationships with a partner (R2 = 0.32, F = 2.60, p = 0.03; β = 0.22). The higher level of consensus and satisfaction with a partner was associated with lower level of burnout.

Secondary traumatic stress and relationship satisfaction

Cerney (1995, p. 140) alluded that caregivers “may traumatize their families by their chronic unavailability and emotional withdrawal, perhaps in the same way that trauma victims sometimes traumatize those around them”. The results of this study however, did not suggest a significant correlation between secondary traumatic stress and relationship satisfaction (for all dimensions). One possible explanation for this result could be that the helping professionals in the current study did not perceive that they underwent a high level of secondary traumatic stress. However, they reported moderate and high levels of compassion satisfaction, which could have a positive impact on participants' partner relationships. On the other hand, as was suggested by regression analyses, the secondary traumatic stress could be explained by the combined effect of several variables, such as of working hours, caseload, household chores, overall primary trauma, dyadic consensus, dyadic cohesion, and dyadic satisfaction. Contrary to our findings are the results of a study conducted by Adams, Figley, & Boscarino (2008) who found that lifetime trauma, and having a high percentage of clients who were victims of violence were not statistically associated with secondary trauma.

Burnout and relationship satisfaction

Correlation analysis between burnout and relationship satisfaction dimensions has yielded a significant association among burnout and such variables as consensus and satisfaction dimensions. Regression analysis suggested that burnout was predicted by the combined effect of consensus and satisfaction dimension of partner relationship, as well as the variables associated with work and family. Therefore, these results are consistent with the research assumptions. The results are similar to other studies. For example, Jayaratne et al. (1986) found that social workers who experienced more intense burnout were more likely to demonstrate a lower marital satisfaction as well as depression and anxiety. Several studies investigated both strengthening and stressors factors that may affect therapists' family life. The most stressful indicators were found to be "little time left for own marriage\family", "little energy left for own marriage\family" (Duncan & Goddard, 1993). Another study noted that a higher percentage of female participants expressed concerns about how traumatic case material might impinge on personal relationships with partners and children (Killian, 2008).

Demographic and work characteristics

Several studies have examined aspects of academic training and occupational role as factors in the development of secondary traumatic stress. Steed and Bicknell (2001) found that psychologists had lower levels of secondary traumatic stress than social workers, but no difference was found between bachelor and master’s level graduates. In contrast, levels of education and academic discipline were unrelated to secondary trauma symptoms in child welfare workers (Nelson-Gardell & Harris, 2003). In concordance with these findings, we have found that psychologists perceived a lower level of secondary traumatic stress than social workers did.

Limitations of research

The findings of the present research should be interpreted with several limitations concerning the characteristics of the sample, cross-sectional method and results. First, as mentioned above, the sample size may affect the likelihood to receive significant results and variability in the levels of burnout and secondary trauma although the proportions of psychologists and social workers and gender proportions reflected represented approximately the ratio in the population. Second, additional limitations in the present study is that the participants did not report on extreme secondary traumatic stress or burnout and did not classify under high levels of secondary traumatic stress. Third, the cross-sectional design did not allow the answer of questions of causality of results.

Intervention strategies for reducing and preventing secondary traumatic stress and burnout

Providing helping professionals with a better understanding of serious mental illness and training them in a broader range of self-care activities, could help them to be more positive in their attitudes towards the clients that they work with and experience less negative effects of stress resulting from their caring role. We recommend preventive activities aimed at building resilience, creating boundaries between work and family, work and leisure. Emphasis should be on a proactive, self-regulated approach to early detection of symptoms of secondary trauma and burnout and the timely implementation of well-being activities.