Burnout and Moral Distress Among Social Workers Working with Children and Families Versus Those Who Do Not

Objectives: Burnout is of international concern among social workers, and recently moral distress (MD) has been identified among this professional group. Little is known about how burnout and MD experiences differ between social workers serving children and families (CF) and social workers in other domains. Less is known about the potential relationship between burnout and MD across these subgroups of social workers. Methods: This brief report examines if the levels of, and associations, between MD and burnout differ between a sample of Finnish CF social workers (n = 199) compared social workers in other domains (n = 168). Results: Based on multivariate analyses of covariance and hierarchical regression analyses, we found that working with children and families did not moderate the associations between MD and burnout. However, working with children and families was associated with higher levels of exhaustion, MD frequency, and distress. MD frequency and MD distress were also both significant predictors of burnout among the sample of social workers. CF social workers had higher levels of exhaustion compared to the other social workers. Implications:


Introduction
Burnout is a significant concern among social workers internationally (Brown et al., 2019;Fong et al., 2022;Sochos & Aljasas, 2021); and child serving social workers have long been considered at particularly high risk (Anderson, 2000).Maslach and Leiter (2016) describe burnout as a three-dimensional syndrome resulting from prolonged workrelated exposure to interpersonal stressors.It is characterized by overwhelming exhaustion, feelings of cynicism, detachment from the job, a sense of ineffectiveness, and lack of accomplishment.Burnout is a threat to service quality, employee well-being and retention among social workers and helping professionals serving children and youth (Brown et al., 2019;Kim & Kao, 2014;Mor Barak et al., 2001).Higher levels of burnout have been shown among child welfare social workers as opposed to social workers in other domains (Sprang et al., 2011).While several studies have shown only moderate rates of burnout (Hamama, 2012), it remains an important concern given the consequences for individual social workers, those they serve, and the high costs of turnover on service systems.Thus, gaining a better understanding of the factors connected to burnout among social workers is necessary.
Recently, moral distress (MD) has emerged as another potential risk faced by social workers.MD refers to a situation at work in which three factors are present: (1) A professional believes a certain course of action is the right one to take; (2) They are constrained from taking that course of action by factors related to their work (i.e., policy, culture, funding, rules, etc.); and (3) They experience distress as a result of the constraint that disallows them from practicing ethically (Morley et al., 2019).Social workers have been shown to suffer from MD, largely due to the resources available to them being insufficient to practice in the way they believe they should (Mänttäri-van der Kuip, 2016).MD has also been identified within child welfare practice among professionals serving children and youth (Brend, 2020).However, scant research investigating MD or the relationship between burnout and MD exist among social workers (Dalmolin et al., 2014;Dzeng & Wachter, 2020;Fumis et al., 2017).This brief report aims to answer the question: Do the levels of, and associations between, MD and burnout differ between social workers who work with children and families compared to those who do not?Preliminary data gathered using a novel instrument (Moral Distress Instrument; MDI; Mänttäri-van der Kuip & Brend, 2021) was analyzed to address this research question and determine if MD and burnout differed between the two groups.Burnout and MD were further analyzed in relation to these sub-samples to increase understanding about social workers' experiences of these phenomena when working with children and families.

Method Sample
Social workers registered with the Finnish union of social workers (n=367) in 2020-2021 were given a written summary of the project and practices used to protect confidentiality (see Table 1 for the descriptive statistics).If they gave informed consent, they were then invited to answer the online questionnaire.Their union (Talentia) reviewed the study and the guidelines of the Finnish Advisory Board on Research Integrity were followed.Individual questionnaire items were also voluntary, as any question could be skipped.The sample size represents 14% of Talentia members in clinical practice and based on estimates of the sociodemographic characteristics of Talentia members at that time, this sample was representative of union members.The average number of years of social work experience was 14.4 years (SD = 10.2) with a range of 0-43 years and the average length of time in the current position for those who were employed was 7.7 years (SD = 8.4) with a range of 0-42 years

Moral Distress
MD was measured with the MDI, developed by Mänttäri-van der Kuip and Brend (2021).The scale consists of seven "A-items" capturing different forms of constrained moral agency: (A1) Have you ever been unable to do your job in the way you believe it should have been done?; (A2) Have you been pressured, obligated, or forced to do something at work that did not seem like the right course of action?; (A3) Have you been in a situation at work that required you to act despite being unsure about what the right course of action was?; (A4) Have you witnessed things happening at work that you believed to be wrong but felt powerless to change?; (A5) Have you encountered situations at work that have caused you to compromise your professional values or ethical principles?; (A6) Have you encountered situations at work that have caused you to compromise your personal values or ethical principles?; (A7) Have you encountered situations at work in which you knew the right thing to do, but felt you were unable to do it?(A-items scale: 0= Never, 1= A few times a year or less, 2= Once a month or less, 3= A few times a month, 4= Once a week, 5= A few times a week, 6= Every day).If the response to these A-items was 1-6, the respondents were also asked to estimate if these events caused them any discomfort and if so, how manageable they found it (B-items).Ratings of discomfort are conforming to the American Psychological Association (2022) definition of distress "a type of stress that results from being overwhelmed by demands, losses, or perceived threats".We used the word "discomfort" to enable a range of potential experiences to be reported, and distress was represented by two response options on a 0-4 scale: Option 3, "Yes, and my discomfort was difficult to manage"; and Option 4, "Yes, and I was unable to manage my discomfort" (Mänttäri-van der Kuip & Brend, 2021).This scale has been validated among Finnish social workers and its validation study is currently under review.The MDI was shown to be a reliable and valid measure among the Finnish sample: the Crohnbach's alphas were .87 for the A-items (measuring the frequency of compromised moral agency) and .88 for the discomfort items; and confirmatory factor analysis supported the hypothesized two-factor model with factor loadings varying between .56 and .86.To calculate item indexes for each individual item, the scores from the answers to all A and B items were multiplied (A x B).Thus, each item index had the range from 0 to 24.Finally, all seven item indexes were added together (AB1 index + AB2 index + AB3 Index + AB4 Index + AB5 Index + AB6 Index+ AB7 Index).This MDI total score had a range from 0-168, and the average length of time in the current position for those who were employed was 7.7 years (SD = 8.4) with a range of 0-42 years.

Burnout
Burnout was measured with the Finnish 9-item version of the Bergen Burnout Inventory (BBI-9; Feldt et al., 2014), for the full-scale version, see Matthiesen and Dyregrov (1992) and Salmela-Aro et al. (2011).The three core dimensions of burnout that are covered by this short version include: exhaustion at work (e.g., "I am snowed under with work"); cynicism toward the meaning of work (e.g., "I feel dispirited at work and I think of leaving my job."); and a sense of inadequacy at work (e.g., "I frequently question the value of my work.").Higher mean scores on a 6-point frequency-based scale, from 1 (completely disagree) to 6 (completely agree), indicate higher levels of burnout.The Cronbach's alphas for the subdimensions of this scale were .73(exhaustion), .83(cynicism), and .77(inadequacy).To isolate the sub-samples needed for this analysis a new variable was coded.All different job positions involving work with children and families were coded as 1 and all others as 0.

Results
Correlations and ANOVAs were run to test which background variables might associate with the focal variables (MD and burnout and their subdimensions) and should be controlled for in the subsequent analyses.No significant associations were found for: age, gender, supervisory tasks, education, work sector (public/private/voluntary sector), type of working hours/shifts, and full or part-time work.The type of work contract had a significant association with MD.Those with a fixed-term contract experienced higher MD frequency and distress (i.e., the whole MDI score was higher) compared to a permanent work contract.Surprisingly, those respondents who were currently without a job contract experienced the highest levels of MD (mean of the full MDI score = 51.75)compared to those in a permanent contract (MDI mean = 27.96) or in a fixed-term contract (MDI mean = 36.68).Finally, the amount of work experience correlated with MDI: correlation between previous work experience in social work and MDI score was -.20 (p < .001),with MDI (frequency subscale) -.20 (p < .001),and with MDI (distress subscale) -.19 (p < .001).Work experience with the current employer correlated with the MDI score (-.12) and with the MDI frequency (-.13).The more previous work experience, the less MD.
To test the differences of MD and burnout between the two groups (working/not working with children and families), MANCOVAs were run with SPSS.The following variables were tested as separate outcomes: the MDI frequency subscale; the MDI distress subscale; and the burnout dimensions of exhaustion, cynicism, and inadequacy.Working with children and families (1 = yes, 0 = no) was used as the independent variable, and the following covariates were entered to the models: work contract (recoded and left out those without a current work contract, 1 = fixed, 0 = permanent); work experience (in years); and work experience with the current employer (in years).
The only significant difference was that social workers working with children and families had higher levels of exhaustion compared to the other social workers.No other differences in MDI or burnout dimensions reached statistical significance.Work contract type or work experience with the current employer did not have significant associations with any of the outcomes in the MANCOVA, so these variables were excluded from the following analyses (see Table 2).Finally, we ran interaction analyses in SPSS to test whether the associations between MDI and burnout dimensions differed between the two groups (working/not working with children and families).We used hierarchical regression analysis, where the first step included the significant background variables (work experience, working with children and families).In the second step, we entered the subdimensions of MDI (frequency and distress subscales), and in the third step we entered the interaction terms (MDI frequency subscale × working with children and families; MDI distress subscale × working with children and families).For exhaustion, working with children (β = .15,p < .001,S.E.= .11),MDI frequency subscale (β = .48,p < .001,S.E.= .09)and MDI distress subscale (β = .46,p < .001,S.E.= .09)were significant predictors with the explained variance of R 2 = .31(for the frequency subscale) and R 2 = .25(for the distress subscale).For cynicism, only MDI frequency subscale (β = .39,p < .001,S.E.= .10)and MDI distress subscale (β = .30,p < .001,S.E.= .09)showed significant associations.The explained variance was R 2 = .18(for the frequency subscale) and R 2 = .13(for the distress subscale).Finally, for inadequacy, similarly only MDI frequency subscale (β = .48,p < .001,S.E.= .10)and MDI distress subscale (β = .36,p < .001,S.E.= .10)showed significant associations with the explained variance of R 2 = .18(for the frequency subscale) and R 2 = .17(for the distress subscale).However, no interactions reached statistical significance.

Discussion and Implications
Our analysis supports burnout and MD as separate phenomena that are associated with each other.As a positive finding, the overall level of MD was relatively low among the studied social workers.Thus, most of them were able to do their daily work without having to bypass or violate their moral values.What was surprising was the finding that social workers with permanent positions had the lowest levels, contract workers had higher levels, and social workers who were currently without a job contract had the highest levels of MD.This trend raises several questions.First, does workplace security or stability protect against MD by providing better support and guidance, thereby enabling permanent workers to act morally?Second, are individuals with temporary contracts more at risk of compromising their moral agency, perhaps due to concerns about job continuity or future prospects?In Finland, those entering the field for the first time typically work in temporary positions.This situation prompts the question: Is it more challenging for them to act or speak out in morally charged situations, such as witnessing unethical practices at work? Surprisingly, the highest levels of MD were observed among social workers who were currently without a work contract, leading to the question: Does MD cause social workers to leave their jobs?These questions all warrant further attention in future research on MD.
When evaluating the generalizability of our findings it is necessary to keep in mind the restricted sample size and the cross-sectional research setting, which are clear limitations of this study.In future research it would be beneficial to study the association between burnout and MD among social workers working with children and families and among those who do not with longitudinal data.This would enable a better examination of the potential causal relationship between these two phenomena.Also, social workers registered with Talentia may not regularly update their status or contact information and their rate of turnover appears to be quite high, therefore, estimates of the true number of social workers in the sub-samples is inexact.In conclusion, MDI frequency and MDI distress were both significant predictors of burnout among the sample of social workers.In this sample, social workers working with children and families reported higher levels of exhaustion compared to the other social workers.However, no other differences in MDI or burnout dimensions reached statistical significance.Working with children and families did not moderate the associations between MDI frequency/MDI distress and burnout dimensions (exhaustion, cynicism, inadequacy).Working with children was associated with higher levels of exhaustion in addition to MDI frequency and distress.

Table 2 .
Means of MDI and the burnout dimensions among the sub-samples To calculate the total index to represent the MDI total score, all seven item indexes (A x B = item index) were added together.The MDI total score has a range of 0 to 168.