Compassion Fatigue in Portuguese Firefighters: Preliminary Results of the Compassion Fatigue Scale ^†

Abstract

The present study had the purpose of creating and studying the psychometric properties of the Compassion Fatigue Scale (CFS) in a sample of 286 firefighters, 69% of whom were males and 78% had previous exposure to a limit situation. All participants were assessed after their informed consent through a protocol composed of a sociodemographic data section, the CFS and the Professional Quality of Life scale (ProQOL). The CFS evidenced a two-dimensional structure, Compassion Fatigue and Compassion Satisfaction, with Cronbach’s alphas superior to 0.90 and correlations in the expected direction with ProQOL. Females showed higher levels of compassion fatigue than males, demonstrating that CFS can be a reliable and valid self-report measure of compassion fatigue in firefighters.

1. Introduction

Burnout and, specifically, compassion fatigue, a distinct form of burnout that can occur in persons regularly exposed to trauma and suffering (as is the case with hospital emergency and pre-emergency professionals), is an emergent health topic, due to its medium- and long-term impact on health. However, to the best of our knowledge, there is only one measure of compassion fatigue, directed to the general population.

Compassion fatigue is a specific form of burnout that can arise in individuals, professionals or even informal caregivers, resulting from systematic exposure to the suffering of others, which can lead to vicarious traumatization and cumulative burnout [1,2,3]. Thus, compassion fatigue is more likely to occur in health professionals and the pre-hospital emergency area, and in professionals in security, protection and social services, who are increasingly exposed to risks related to the social determinants of health throughout development. Literature reviews carried out in this area among health professionals have shown that the determinants of compassion fatigue include individual, organizational and ecosystem-related factors [4] and that, among the main consequences are its impact on the quality of care provided on personal and family life [4,5] and on mental health [5]. Firefighters are one of the professions constantly exposed to extreme situations that can lead to vicarious traumatization, as they are constantly called upon to intervene in situations of which they may have little or no knowledge and which require rapid and effective decision-making [6]. Studies carried out with firefighters highlight the negative impact on well-being caused by exposure to traumatic events and the greater likelihood of developing compassion fatigue [7]. On the other hand, compassion satisfaction, that is, the positive feeling of accomplishment and gratification that results from providing help to others and its impact on other people’s lives, may constitute a protective factor that can buffer compassion fatigue [8].

However, despite the relevance of the topic, there are still few studies conducted with firefighters, highlighting the need to develop measures to assess this specific population. The measure probably most widely used in this context is the ProQOL [9], a self-report measure of professional quality of life which measures burnout, secondary traumatic stress (compassion fatigue) and compassion satisfaction. In Portugal, to our knowledge, there is only one measure adapted for this purpose in firefighters, the Compassion Fatigue-Short Scale [10], which, in the original, was constructed to assess compassion fatigue in social workers [11] but does not assess compassion satisfaction.

Thus, the main goal of the present study was to develop and study the preliminary psychometric qualities of a compassion fatigue self-report in firefighters. It was expected that the obtained dimensions would present adequate levels of reliability and validity.

2. Materials and Methods

2.1. Participants

Participants were included in this cross-sectional study if they were active firefighters over 18, working in any region of the country.

2.2. Measures

The ProQOL is a self-report measure of compassion fatigue, composed of 30 items which assess the frequency of each experience in the 30 days before the assessment in a 5-point Likert scale (1, Never; 5, Very Frequently) in three dimensions: compassion satisfaction (CS), burnout and secondary traumatic stress. The ProQOL presents adequate psychometric properties, with Cronbach’s alphas superior to 0.71 [12]. The CFS was developed within the scope of the present study, and the results of its psychometric properties will be presented in the Results Section. The studied measure consisted of 28 items for which participants were asked to rate the frequency with which they usually experienced the described in each item, according to a 5-point Likert scale (1, Never; 5, Always).

2.3. Procedures

An initial set of items was developed and reviewed to reflect a comprehensive set of experiences across different categories of compassion fatigue. These items were first analyzed for their face and content validity. The research protocol, composed of an informed consent section, a sociodemographic section and both self-report measures, was then subjected to validation, and a pilot study was carried out with the main goal of clarifying potential language and response doubts and understanding the duration of the protocol administration. After small revisions and institutional board approval, the protocol was disseminated through different firefighters’ corporations all over the country, with a request addressed to each commander to disseminate the study among the firefighters of their corporation. The protocol was available online between May and October 2024 and had an average filling time of 10 min.

2.4. Statistical Procedures

Face and content validity were assessed through interrater agreement. The internal structure of CFS was studied through exploratory factor analyses. Reliability was studied through Cronbach’s alphas and intercorrelations, and concurrent validity was studied through Pearson’s correlations. Comparisons between groups were studied through independent Student’s t tests (assuming normality), and descriptive statistics were calculated for CFS norms. A confidence level of 95% was considered for each of the analyses.

3. Results

3.1. Sociodemographic Characteristics of the Participants

A convenience sample composed of 286 volunteer and professional firefighters from different corporations across the country was studied, which included 88 (31%) females and 198 (69%) males, aged between 18 and 63 years old (M = 38.48, SD = 10.19). Most of the participants reported working as firefighters for 8 to over 20 years (66.8%). Most of the participants spent at least 12 years in school (96%), and more than half of the participants were married or reported living with another person (53%). Also, 78% of the participants reported being already exposed to limit situations, and more than half of the participants reported working more than 35 h/week (53%).

3.2. Face and Content Validity of the Compassion Fatigue Scale

An initial pool of 35 items was developed and submitted to a panel of two non-experts and two experts to assess their facial and content validity, respectively. Based on the interrater agreement (IA) results, 7 of the 35 items were excluded due to ambiguity or because they did not represent the dimensions of compassion fatigue that were intended to be assessed (IA inferior to 49%), 17 of the items were included in the preliminary test version (IA superior to 75%), and 11 of the items were subjected to reformulation (IA between 50 and 74%). The final testing version was composed of 28 items.

3.3. Internal Structure of the Compassion Fatigue Scale

The internal structure of the 28 items of the CFS was studied through exploratory factor analyses with principal axis factorization and Varimax rotation. Based on principal components, the factorial solution was determined according to the following criteria: KMO and Bartlet’s test (KMO = 0.91, χ² (35) = 4574.49, p < 0.001), eigenvalues superior to 1, explained variance superior to 50%, loadings superior to 0.40, ambiguity (difference superior to 0.30 when loading on more than one factor) and scree plot.

Only one item was excluded, and the results of a scree test pointed to a structure composed of two factors. A subsequent exploratory factor analysis was conducted with the 27 retained items. The first factor, Compassion Fatigue (CF), was composed of 17 items related to emotional, cognitive, physiological and social distress, and the second factor, Compassion Satisfaction (CS), was composed of 11 items related to positive feelings about the impact of work on other persons. Both factors explained about 48% of the total variance.

3.4. Reliability

Internal consistency values, Cronbach’s alphas, were equal to 0.91 (CF) and 0.93 (CS). Inter-item correlations varied from 0.12 to 0.77, and item-total correlations were between 0.45 and 0.87. The intercorrelation between the two dimensions was equal to 0.18 (p = 0.002).

3.5. Concurrent Validity

Positive moderate correlations were found between CFS-CF with ProQOL Burnout (r = 0.36, p < 0.001) and Secondary Traumatic Stress (r = 0.63, p < 0.001). Positive moderate correlations were also found between CFS-CS with ProQOL-CS (r = 0.61, p < 0.001). A negative moderate correlation was obtained between CFS-CS and ProQOL Burnout (r = −0.49, p = 0.001). CFS-CF was independent of ProQOL-CS (r = 0.007, p > 0.05), and CFS-CS was independent of ProQOL Secondary Traumatic Stress (r = −0.04, p > 0.05).

3.6. Comparisons Between Groups

Comparisons between groups were carried out through t tests for independent samples. Significant sex differences were found for compassion fatigue, t (284) = 3.49, p = 0.0005. Female firefighters reported higher levels of compassion fatigue (M = 46.9, SD = 10.3) compared to male firefighters (M = 42.9, SD = 5.3). No significant sex differences were found for compassion satisfaction t (284) = 1.44, p = 0.151. Also, no significant group differences were found in compassion fatigue, t (284) = −0.29, p = 0.77, and compassion satisfaction, t (284) = −1.17, p = 0.24, when comparing firefighters who reported already being exposed to a limit situation with firefighters not exposed to a limit situation.

3.7. Norms

The resulting norms in the present sample are presented in

Table 1. CFS norms.

	N	M	SD	Range	P25	P50	P75	Sk	K
Compassion fatigue	286	43.69	10.57	17–82	36	43	50	0.50	0.74
Compassion satisfaction	286	41.99	10.00	10–50	39	44	47	−1.70	3.74

Note. Sk = Skewness; K = Kurtosis.

.

4. Discussion

The present study had the main goal of developing and studying the preliminary psychometric qualities of the CFS in firefighters. As expected, the obtained dimensions revealed adequate reliability and validity. The internal structure was adequate, and although there were some differences (compassion fatigue and burnout seem to be understood as different components on the same dimension), it seems to reflect the theoretical model [2,3] on which it was based. Correlations with the ProQOL and sex differences were also in the expected direction and according to the literature [10,13]. Despite the importance of the results, the present study has limitations, related to the sample (not random) and study nature (cross-sectional). Future studies are needed to replicate the invariance of the structure and assess its temporal stability. Although exploratory, the results are of most relevance for theory and professional practice. Interventions aimed at preventing compassion fatigue should promote self-care, the identification and recognition of its early manifestations and communication (horizontal and vertical) and support. Also, there is a need for systematic monitoring of these interventions to identify the mental health impact of CF and to provide targeted, timely and appropriate resources for its prevention.

5. Conclusions

The obtained results demonstrated that CFS can be a reliable and valid self-report measure of compassion fatigue in firefighters, and results highlight the relevance of the inclusion of sociodemographic variables in the interventions designed to prevent compassion fatigue in vulnerable populations.