Compassion fatigue of funeral directors during and after the first wave of COVID-19

Abstract

Background

Compassion fatigue has not been studied among funeral directors. Yet, funeral directors have been exposed to the same risks for compassion fatigue as other caregivers during the coronavirus disease 2019 (COVID-19) pandemic.

Methods

An online survey was spread two times to 287 employees of funeral home DELA, in Belgium. Once during the height of the first wave of COVID-19 in Belgium, and a second time at the end of the first wave. The professional quality of life-scale 5 (PROQOL-5) was used to measure compassion fatigue, which includes burnout, compassion satisfaction and secondary trauma. Non-parametric tests were performed.

Results

In total, 104 participants answered the first survey, and 107 the second. Burnout increases from survey 1 to survey 2 (P < 0.001), while compassion satisfaction (P = 0.011) and secondary trauma decrease (P < 0.001). In survey 1, only age (P = 0.007) and gender (P = 0.040) were found to be significantly associated with secondary trauma. In survey 2, having more work experience is associated with having a higher burnout (P = 0.008) and secondary trauma (P = 0.001) score. Neither for burnout (P < 0.001), nor for secondary trauma (P < 0.001) are there any respondents in the highest category.

Conclusions

Although overall funeral directors do not have acute problems with compassion fatigue, burnout scores increase significantly after the first wave.

Introduction

The current coronavirus disease 2019 (COVID-19) pandemic is a public mental health issue.¹ Scholars have warned that the crisis might have severe mental health consequences for healthcare personnel, as they face a large number of deaths, have to work more and longer and are exposed to incisive and stressful events.^2–4 Possible consequences for healthcare personnel are depression, post-traumatic stress disorder, insomnia and a generally lower mental wellbeing.^5–9

Studies have shown that health care workers have anxiety and stress in caring for their patients and fear acquiring COVID-19 themselves.¹⁰ Due to being overexposed to patients’ suffering and dying, healthcare personnel might develop compassion fatigue. Compassion fatigue is the declining ability to be empathic for patients due to an overexposure to their suffering.^11,12 It commonly arises after a period of time in which the caregiver puts a lot of energy into caring for the patients he/she cares for.¹³

Compassion fatigue consists of two main elements: burnout and secondary trauma.^11,14,15 Burnout can come through frustration with the work environment and other environmental factors. Compassion fatigue has also been called a burnout specific to caregivers.¹⁶ Secondary trauma refers to the repeated exposure to painful details of the lives of patients, which can be exhausting for caregivers.^15,17,18 Together with this compassion fatigue, there can be compassion satisfaction, which refers to the positive feelings one can get from helping those in need. In fact, studies have shown that despite high compassion fatigue, people might keep working due to the compassion satisfaction.¹⁶

Compassion fatigue can lead to other mental health problems, such as depression or post-traumatic stress disorder.¹⁹ As this was already a reoccurring phenomenon among caregivers in stressful situations in pre-COVID-19 times,^11,12 there is a risk that it is also present during the COVID-19 crisis. A group that is on the frontline of the COVID-19 crisis and faces many of the same problems as other caregiving personnel, but has gained to our knowledge much less attention, are funeral directors.

There are only relatively few (recent) studies on the mental health of funeral directors.²⁰ The studies that have been performed, showed that funeral directors run a risk of developing mental health problems, such as depression.^21,22 The possible cause of such problems is varied. The cumulative exposure to stressful incidents,²³ the high number of funerals,²⁴ the long working hours, and the many difficulties to maintain the quality of work as before all may have an impact on the mental health, as well as an increased exposure to dead people as such. Some studies indicate there is no relationship between the latter and mental health symptoms,²¹ and others found a negative effect of such increased exposure.²⁵ Although some of these studies hint at compassion fatigue problems, to our knowledge, there have been no studies on this topic among funeral directors.

This study investigates the compassion fatigue among funeral directors in Belgium during the COVID-19 crisis.

Method

Design, population and data collection

This study is a longitudinal quantitative online survey study. The populations we have studied, are funeral directors employed by DELA, the largest funeral home organization which controls around 11.5% of the funeral industry market in Belgium.

Only funeral directors, employees in a funeral home, who were working full-time during the crisis (N = 287) were included in the study. These funeral directors were contacted through mail twice: a first time between 21 and 28 April 2020 when there were 1.290 COVID deaths in Belgium and strict COVID-measures were in place and a second time between 2 and 10 June 2020 when there were 120 deaths and the COVID-measures were again lightened.²⁶ Contacting the population twice allowed a comparison to be made between a period of possible high exposure to deceased people, and a period of low exposure.

The mail included an explanation of the study as well as a link to the online survey. Employees were able to complete the survey in French, Dutch or English. At the end of the first survey, respondents were asked if they were interested in completing a second survey. If they were, they could fill in a code based on the first two letters of their mother’s name, their house number and their date of birth, without the year of birth. In the second survey, they could fill in this code again, so that the researchers could couple it with the results of the first survey, without knowing the identity of the respondents. Internet cookies were used to prevent people from completing the survey twice.

Measures

Compassion fatigue was measured using the professional quality of life-scale 5 (PROQOL-5). It is used to measure the professional life-quality of people in stressful situations and is specifically aimed at people in a caring-profession.²⁷ As such, it has been used in many healthcare studies and is a reliable scale.²⁸ This scale has three subscales: a burnout scale, a compassion satisfaction scale and a scale measuring secondary trauma. Burnout and secondary trauma are components of compassion fatigue, whereas compassion satisfaction is a concept introduced by Stamm,²⁷ indicating positive feelings one can get from helping those in need, counterweighting the risk for compassion fatigue.¹⁵

PROQOL-5 includes 3 × 10 Likert-based statements, one per subscale, each time ranging from 1 (never) to 5 (very often), and each time with as reference point the last 30 days. Scores for each subscale range between 10 and 50, with 22 or less corresponding to a low score, 23–41 a medium score and 42 or higher a high score. For burnout and secondary trauma, the higher the score, the more indication for burnout and secondary trauma, whereas for compassion satisfaction, the higher the score, the more satisfied someone is with their work.²⁷

The PROQOL-5 has validated versions in English, Dutch and French

Demographic variables age, gender and years of working experience were also asked. For analysis purposes, age was recoded to 24–44 years and 45–64 years, while working experience was recoded to > 5 years, or 5 years or less.

Analysis

Mann–Whitney U tests were used for associations for burnout, compassion satisfaction and secondary trauma, to compare for gender, age and work experience. This was done for both surveys individually. To compare results of surveys between the two periods, several tests were used. First, Wilcoxon signed rank tests were used to compare the respondents that could be coupled over the two surveys, which was done for all main outcome variables (secondary trauma, burnout or compassion satisfaction). As mean ranks (on which the Mann–Whitney U test and signed rank Wilcoxon tests calculate the significance) might not be intuitive to interpret for the reader, the mean is also added. To compare the coupled responses on basis of the categories of PROQOL-5 subscales, a McNemar–Bowker test of symmetry was performed. If there were only two categories (e.g. due 0 responses for a category), a McNemar’s test was performed. Finally, to compare the categories for the subscales between the two samples without coupling the responses, a chi-square test for goodness of fit was used, where the results of the second sample were compared with the proportions found in the first sample.

As not all respondents have inserted the code to couple their responses, the results will be compared first without coupling of respondents. This will be referred to as “globally” comparing the results. Second, results will then be compared using the coupled respondents, using sampled t-tests, in which case P-values will also be mentioned. This to show that the general trend that might be seen in the two surveys, is supported both by the global and the coupled responses.

All analyses were performed using SPSS 26.0

Ethics

As the survey was anonymous, it was impossible to link answers to a respondent’s identity or their workplace. Furthermore, as the funeral employees were already under stress due to large number of COVID-19 deaths, the survey was made as brief as possible. The introduction screen of the survey gave more information about the study, and listed the rights of the respondents (e.g. no obligation to complete the survey) and stated that if respondents started completing the survey, they agreed to have been sufficiently informed and to take part voluntarily in the study.

The study was approved by the ethics commission of the VUB/UZ Brussels (B1432020000151).

Results

Demographics of sample

In total, 106 employees replied to the first survey (response rate of 36.9%) and 121 employees replied to the second survey (response rate of 42.2%). However, of these respondents, not everyone completed the questions on compassion fatigue, and were left out of analyses, leaving for the first survey 104 respondents and for the second survey 107. In the first survey, there are 45 male and 59 female respondents, and an almost even division for age: 51 in the age group of 24–44 years and 53 in the 45–64 age group. For work experience there are 32 respondents with 5 years or less of experience. For the second survey, the divisions are similar: 57 female and 50 male respondents, 55 between 45 and 64 years old, and 57 between 24 and 44 years old and 30 with 5 years or less work experience.

Of these respondents, 66 people could be coupled using the code they inserted. For this group, there were 37 female and 29 male respondents, 21 had <5 years of experience, while 45 had >5. Finally, 35 were younger than 45, whereas 31 were between 45 and 64 years old (Table 1).

Table 1.

Characteristics sample

	Survey 1 n = 104 (%)	Survey 2 n = 107 (%)	Coupled  n = 66 (%)
Gender
Male	45 (43.3)	50 (46.7)	29 (43.9)
Female	59 (56.7)	57 (53.3)	37 (56.1)
Age
24–34 years	22 (21.2)	25 (23.4)	14 (21.2)
35–44 years	29 (27.9)	27 (25.2)	21 (31.8)
45–54 years	36 (34.6)	34 (31.8)	22 (33.3)
55–64 years	17 (16.3)	21 (19.6)	9 (13.6)
Work experience
< 1 year	11 (10.6)	10 (9.3)	9 (13.6)
1–5 years	21 (20.2)	20 (18.7)	12 (18.2)
−10 years	30 (28.8)	30 (28)	18 (27.3)
11–15 years	12 (11.5)	15 (14)	9 (13.6)
16–20 years	9 (8.7)	10 (9.3)	5 (7.6)
> 20 years	21 (20.2)	22 (20.6)	13 (19.7)

Professional quality of life

Burnout

This burnout subscale had a Cronbach alpha of 0.727 in the first measurement, and 0.733 in the second. Globally, there is an increase of 19.4423–21.68 on the scale. This is also reflected in the increase for the coupled respondents, with a significant increase of the mean rank of 22.69–33.60 (mean 18.89–20.91; P < 0.001). Furthermore, the respondents with middle burnout scores increased from 19.2% in the first survey to 44.9% in the second survey (P < 0.001). For coupled responses, this increase was also significant (P < 0.001).

In survey 1, there are no significant associations for gender (P = 0.982), work experience (P = 0.209) and age (P = 0.557). In survey 2, this is also the case for age (P = 0.695) and gender (P = 0.525). However, work experience (P = 0.008) is significant, with a mean rank of 42.29 on the scale for those with 0–5 years of experience, and 60.05 for those with > 5 years of experience (mean 19.68–22.47).

Compassion satisfaction

The compassion satisfaction subscale has a Cronbach alpha score of 0.836 for the first measurement and 0.855 for the second. Globally, there is a decrease of the mean of 42.56–41.55 from survey 1–2. For the coupled responses, there is a significant decrease of the mean rank of 32.11–27.73 (mean 43.30–42.29) (P = 0.011). For the first survey, 69.2% of the sample has a reported high compassion satisfaction, whereas in the second survey this has decreased to 54.2% (P = 0.002). This decrease was not significant for the coupled responses (P = 0.210).

In neither survey there were any significant associations. In survey one, age (P = 0.109), experience (P = 0.474) and gender (P = 0.198) all are higher than our P-value threshold of 0.05, which is also the case for survey two: experience (P = 0.954), gender (P = 0.791) and age (P = 0.325).

Secondary trauma

This subscale had a Cronbach alpha of 0.790 for the first survey, and 0.829 for the second. Globally, there is a decrease of the mean from 22.67 to 20.88 on the secondary trauma subscale, which is also seen with the coupled respondents: a decrease from of the mean rank of 33.29–22.50 (mean 21.81–19.88; P < 0.001). For the first survey, 49% of the sample reported middle risk for secondary trauma. In the second survey, this decreased to 31.8% (P < 0.001). For coupled responses, this decrease was significant (P = 0.041).

For survey one, a significant increase was found for secondary trauma for age (P = 0.007), with an increase of the mean rank of 44.33 for those of ages 24–44, to 60.36 for those of ages 45–64 (mean 21.3333–23.9623). Furthermore, secondary trauma was also significantly different for men (45.54) and women (57.81) (mean 21.58–23.51; P = 0.040). Such significance was not found for working experience (P = 0.306). For survey 2, a significant difference was found between those who have worked 5 years or less as funeral manager, namely a mean rank of 39.61, and those with > 5 years of experience, 61.12 (mean 17.94–22.05; P = 0.001), while there was no significant associations for gender (P = 0.357) or age (P = 0.216) (Table 2).

Table 2.

PROQOL-5 subscales divided by survey and coupled responses

	Survey 1 n = 104 (%)	Survey 2 n = 107 (%)	Coupled survey 1 n = 66 (%)	Coupled survey 2 n = 66 (%)
Burnout
Low	84 (80.8)	59 (55.1)	55 (83.3)	37 (56.1)
Middle	20 (19.2)	48 (44.9)	11 (16.7)	29 (43.9)
High	0 (0)	0 (0)	0 (0)	0 (0)
Mean value	19.44	21.68	18.89	20.91
P-value chi-square test for goodness of fit*		<0.001
P-value Wilcoxon signed rank test**			<0.001
P-value McNemar–Bowker test of symmetry***			<0.001
Compassion satisfaction
Low	1 (1)	1 (0.9)	0 (0)	0 (0)
Middle	32 (30.8)	48 (44.9)	23 (34.8)	29 (43.9)
High	72 (69.2)	58 (54.2)	43 (65.2)	37 (56.1)
Mean value	42.56	41.55	43.30	42.29
P-value chi-square test for goodness of fit		0.002
P-value Wilcoxon signed rank test			0.011
P-value McNemar–Bowker test of symmetry			0.210
Secondary trauma
Low	53 (51)	73 (68.2)	38 (57.6)	48 (72.7)
Middle	51 (49)	34 (31.8)	28 (42.4)	18 (27.3)
High	0 (0)	0 (0)	0 (0)	0 (0)
Mean value	22.67	20.88	21.81	19.88
P-value chi-square test for goodness of fit		<0.001
P-value Wilcoxon signed rank test			<0.001
P-value McNemar–Bowker test of symmetry			0.041

^*To compare global results of categories in survey 2 with divisions in survey 1.

^**To detect change of mean-rank of coupled responses of survey 1 to 2.

^***To detect change in categories of survey 1 to survey 2.

Discussion

Main findings of this study

This study investigated compassion fatigue among funeral directors in Belgium, 1 week during the first peek of the corona crisis (21–28 April 2020) in our country, and more than a month later (2–10 June 2020), when the first wave was over. Our results showed an increase for burnout scores and a decrease for secondary trauma and compassion satisfaction. People with > 5 years’ work experience had a significantly higher burnout and secondary trauma scores during the period of 2–10 June 2020, while secondary trauma was significantly associated with being female and a higher age.

What is already known about this topic

Not much is known on the topic of funeral directors and mental health, and to our knowledge, nothing on the relationship between the profession and compassion fatigue. Previous studies have shown that depression, and other mental health problems might arise due to the occupation of funeral directors.^21–24

What this study adds

To our knowledge, this study is the first investigating funeral employees during the COVID-19 crisis, providing important insights in how resilient funeral employees are during such stressful times, and one of the few studies in recent years to investigate this group of professionals. Furthermore, the longitudinal design allowed to compare a point around the peak of the crisis, and a point when the crisis was already diminishing. Finally, though the sample is small in this study, other studies on caregiving personnel using the PROQOL-5 often have smaller samples, and are cross-sectional.²⁸

Following the guidelines of Stamm²⁷, the results in this study are encouraging. As there is a general high compassion satisfaction and middle/low burnout and secondary trauma across the sample, this means that the respondents are quite content in their environment, without strong feelings of compassion fatigue. However, the most concerning result in this study, was the increase of burnout in the second measurement.

Although other research has shown that burnout and secondary trauma are linked,²⁹ there was a decrease in the second survey of secondary trauma and an increase of burnout. The middle category of burnout even doubled in percentages (19.2–44.9%). Although the decrease of secondary trauma seems logical, as there is less exposure to the suffering of others (as there were less funerals), the reason for the increase of burnout is not immediately clear. As burnout is often determined by environmental factors, it might be that there was exhaustion due to all measures that stayed in place against COVID-19 during the second period, when the need for these measures might not have seemed as urgent as in the first period, as in the second period the number of deaths was relatively low. For example, despite the feeling that the crisis had passed, the measures against COVID-19 still not allowed larger funerals and closer contact with family members, hindering funeral directors to do their work with the usual respect and dignity, perhaps causing guilt among the funeral directors, which can be linked with burnout.³⁰ Although for healthcare professions, compassion fatigue and burnout is mostly found to be related to having little experience, or found to have no link at all,^17,28 having more work experience was related to higher burnout scores for funeral directors. Perhaps the more experienced funeral directors hoped for a return to the normal procedures for funerals, as they were more attached to these pre-COVID-19 procedures. Yet, as the governmental measures that limited the size of funerals continued, there might have been more frustration among these more experienced funeral directors, and higher indications of burnout, while the less experienced funeral directors might not have been so attached to the normal procedures. It might also be that there were more administrative tasks in the second period, which might feel boring compared with the intense period in the first period. This would indicate why compassion satisfaction, though still high, decreased in the second measurement, as compassion satisfaction is generally correlated with organizational support in other studies.³¹ Perhaps funeral directors felt that this support was missing in the second period, which would also indicate higher burnout scores.

As there are no other studies on funeral directors that use the PROQOL-5, we can only compare with studies on healthcare personnel. Such studies generally found much higher rates than those found here. In non-COVID-19 times, there is generally a high risk for burnout and secondary traumatic stress when studied among healthcare professionals,^13,15 whereas in the present study there was no one in the high category for burnout or secondary traumatic stress. This indicates that the risk for burnout for funeral directors is probably lower than for healthcare personnel, even in times of crisis, the numbers are still not dramatic. Although the generally low numbers in our study are rare, they are not unique, as a study by Bellolio et al.³² also found similarly low numbers for physicians in the Emergency Care in the USA. Here too no high category scores for burnout or secondary trauma were found, though high category scores for compassion satisfaction. Bellolio et al.³² did not find any demographic variables related to the high compassion satisfaction, nor an association with hours worked, and believed it showed that the respondents received positive reinforcement during their work.

As the results in this study are encouraging, the policy recommendations should reflect this. The increase of burnout can point to the importance of environmental factors contributing to these problems, which can be addressed in various ways. Managers of funeral homes can for example focus on how to lighten the administrative work for their funeral directors, how to support them logistically, etc.^31,33 Funeral directors can also support each other in difficult times. Social support is an important, and easily achieved, way of creating resilience against mental health problems in general.^2,3 As such, colleagues should keep an eye out for peers who feel frustrated with the situation or feel generally upset with the management of the funeral homes. Additionally, a way for funeral directors to protect themselves, can be achieved through self-applicable easy anti-stress techniques, such as relaxation exercises or mindfulness training. This has been shown to decrease the development of stress overall and thus improves the overall resilience, which is especially useful during times of work pressure.^34,35 Finally, if funeral directors feel the need to access more professional mental health aid, managers of funeral homes should increase access to such aid.

Limitations of this study

This study is limited due to its low response rate, probably because, during the corona crisis funeral employees were often too occupied to complete such a survey. Yet, we must be careful in generalizing the results.

Roel Van Overmeire, Phd-Researcher

Rose-Lima Van Keer, Post-doc

Marie Cocquyt, Phd-Researcher

Johan Bilsen, Professor and Head

Conflict of interest

The authors declare to have no conflict of interest.