Personally Meaningful Rituals: A Way to Increase Compassion and Decrease Burnout among Hospice Staff and Volunteers

Abstract

Background: Rituals can increase a sense of connectedness, meaning, and support, especially after the death of those for whom we care. Hospice staff may benefit from the use of personal rituals as they cope with the frequent deaths of their patients, ultimately aiming to provide compassionate care while minimizing burnout.

Objective: This study investigated the role of personally meaningful rituals in increasing compassion and decreasing burnout among hospice staff and volunteers.

Design and Measurements: An online survey was completed by members of the National Hospice and Palliative Care Organization (NHPCO) which inquired about personal ritual practices, and included the Professional Quality of Life (ProQOL) scale to measure current levels of Compassion Satisfaction, Burnout, and Secondary Traumatic Stress.

Setting/Subjects: Three hundred ninety hospice staff and volunteers from across 38 states completed the online survey. The majority of participants were Caucasian and female, with an average of nine years of experience in hospice and palliative care.

Results: The majority of hospice staff and volunteers used personally meaningful rituals after the death of their patients to help them cope (71%). Those who used rituals demonstrated significantly higher Compassion Satisfaction and significantly lower Burnout as measured by the ProQOL, with professional support, social support, and age playing significant roles as well.

Conclusions: Rituals may be an important way to increase compassion and decrease burnout among hospice staff and volunteers. Organizations may benefit from providing training and support for personalized rituals among team members, especially new staff who may be at greater risk for burnout.

Background

Rituals involve traditional or symbolic activities that can vary from those of celebration (e.g., birthdays, graduations) to those of mourning (e.g., funerals).¹ Rituals can relieve anxiety and provide comfort, meaning, and support,^1,2 particularly when facing uncertainties such as those found at the end of life.^3–8 In fact, when end-of-life experiences are shared with others and reorganized in a broader context of universal meaning, they can increase comfort for patients and families, thereby becoming an important part of the overall healing process.^5,6

Hospice staff and volunteers may also benefit from ritual practices, particularly since they repeatedly experience the death of patients, and doing so may lead to a range of risks and/or benefits.^9–18 For example, some staff have demonstrated an increased risk for developing burnout as defined by feelings of reduced personal accomplishment at work, diminished career satisfaction, and poorer quality of service delivery. This type of burnout is often related to heavy caseloads or nonsupportive workplace settings.^14,19 Other staff may be at risk for secondary traumatic stress as a result of experiencing the frequent suffering of others.^20,21 Secondary traumatic stress can be similar to post-traumatic stress as demonstrated by symptoms of hypervigilance, impairment in daily functioning, intrusive imagery, and numbing.²²

Conversely, hospice staff can experience feelings of gratitude and satisfaction as a result of their work with the dying. Indeed, some hospice staff commonly describe their profession as a unique opportunity for existential growth that fosters a sense of meaning, spirituality, and wholeness.²³ Such positive growth experiences are closely related to compassion satisfaction as defined by the experience of pleasure and gratitude from helping others and feeling positive about one's work and work environment.^10,24,25

But how can we help hospice staff enhance their compassion, meaning, and satisfaction, especially as they care for the dying? Rituals may be one method for increasing the positive benefits of work, while decreasing negative aspects such as burnout or secondary traumatic stress.^14,26–30 To this end, many hospices in the United States incorporate group rituals into their organizational practices such as “Light up a Life” memorials, commemorative wreath events, or other honor ceremonies (e.g., a release of butterflies once a year). These public rituals may enhance a sense of connectedness and provide meaning for hospice staff, the bereaved families, and even the larger community. But how might personally meaningful ritual practices directly relate to increased compassion or decreased burnout for hospice providers? Since large public rituals may only occur a few times a year, might personally meaningful ritual practices—ones that can be used privately, quickly, and as often as needed, be helpful to staff as well?

This study investigated these questions by sampling hospice staff and volunteers from across the United States. It was hypothesized that the hospice staff and volunteers who engaged in personally meaningful rituals would have increased compassion satisfaction and decreased burnout. The role of additional demographic variables such as age, professional support, and years of experience were also examined given their previously demonstrated connection to general staff well-being and satisfaction.^28,31–34 Overall, the goal of this study was to shed light on how personalized ritual practices may shape both the positive and negative aspects of caring among hospice staff and volunteers in the United States.

Methods and Design

Participants

The participants were recruited online through a membership listserve of the National Hospice and Palliative Care Organization (NHPCO). All participants were hospice staff or volunteers who were emailed a description of the study and a Survey Monkey^© link to use if they wished to participate.³⁵ The recruitment email was sent in May 2014 to the primary contacts of 2478 hospice locations across the country that were members of NHPCO. The Survey Monkey link remained live for two weeks, with one reminder to complete the survey sent to all possible participants after the first week. The research design, participant recruitment plan, as well as all other facets of the study were reviewed and approved by the Human Research Protections Program and Institutional Review Board at the University of California, San Diego.

Measurements

The online survey included a total of 55 items and took ∼20 minutes to complete. The first 25 items asked for demographics, including age, gender, ethnicity, level of education, marital status, religious affiliation, years of work experience, and the characteristics of each participant's respective hospice location. Hospice staff and volunteers also rated their current level of personal, professional, and social support on a scale of 0–10 (0 = no support, 10 = strong support) and rated “how much religiosity or spirituality was a part of their lives” on a scale of 0–10 (0 = not at all, 10 = very much).

The survey then asked 10 questions about the hospice staff and volunteers’ use of personally meaningful rituals. Because there are currently no empirically validated measures of rituals in the literature, the researchers created a set of qualitative questions specifically for this study to gather the necessary information. Given the varying definitions of rituals, the following description was provided in the survey to clarify for participants:

After the death of someone close to us, we often feel the need to have a sacred ritual to honor their lives. These rituals can be done as part of a community, while others can be done by yourself. Working in hospice care means we experience the death of patients—many we may feel close to. This connection may create a desire to honor patients in some way. For example, sometimes we light a candle when a patient passes away to honor them. We may also go for a walk on the beach, write, say a prayer, journal, or create a personal keepsake that reminds us of that patient. A “ritual” here can mean anything you do to personally honor your patients.

Hospice staff and volunteers were asked to state “yes” or “no” regarding whether they used any personally meaningful rituals. If the staff and volunteers stated that yes, they used rituals, they were asked 10 additional qualitative questions to help describe their personal rituals (e.g., “Describe in detail one ritual you have used”; “How did you create this ritual?”; “Has your ritual changed over time?”).

The final portion of the survey, given to all participants, whether they engaged in personal rituals or not, involved the administration of the Professional Quality of Life Scale—version 5 (ProQOL).³⁶ This 30-item self-report scale assesses the positive and negative aspects of caring, and includes three scales that measure separate constructs: Compassion Satisfaction, Burnout, and Secondary Traumatic Stress. The ProQOL demonstrates good reliability and validity (Compassion Satisfaction α = 0.88 (n = 1130); Burnout α = 0.75 (n = 976); Secondary Traumatic Stress α = 0.81 (n = 1135). The ProQOL was further chosen for its use over the past 20 years and utilization in over 200 published articles³⁶ as a way to measure the occupational well-being of clinicians,³⁷ mental health providers,³⁸ and nurses—all of which related to the population in this sample.^39,40

Analysis

To explore the amount of variance that could be explained on the three ProQOL subscales (Compassion Satisfaction, Burnout, and Secondary Traumatic Stress) by the use of personally meaningful rituals, a single linear regression analysis was performed. After this initial analysis, Pearson Product Moment correlations were calculated among the three ProQOL subscales and the demographic variables to explore what other variables could significantly contribute to each of the subscale scores. An exploratory, stepwise (block) regression analysis was then performed with those variables found to be significantly correlated with the three ProQOL subscales to better understand the relative role of each variable.

Results

Three hundred ninety hospice staff and volunteers responded from across 38 states in the US (a response rate of 16%). As demonstrated in Table 1, most participants were working in nonprofit hospices (81%), with an average daily census of 200 or fewer patients (76%). These staff and volunteers had an average of 20 years of experience working in healthcare, with a range of 1–53 years (standard deviation [SD] = 12.66). They averaged nine years of experience working specifically in hospice and palliative care, with a range of 1–35 years (SD = 7.58). The majority of participants were female (87%), Caucasian, (88%), and married (72%). The sample was evenly distributed across the four geographic regions as defined by the United States Census Bureau (Northeast, Midwest, South, and West).⁴¹ The average age was 50 years old (SD = 12.10, range = 19–78 years). Nearly all of the participants had attended either some college (55%) or graduate school (39%).

Table 1.

Demographic Characteristics of the Hospice Staff and Volunteer NHPCO Sample

Demographic characteristics	No.	Percentage
Gender (n = 390)
Male	52	13.3
Female	338	86.7
Race/ethnicity (n = 390)
Caucasian	344	88.2
African American	15	3.8
Non-Hispanic/Latino	3	0.8
Hispanic/Latino	6	1.5
American Indian/Alaska Native	1	0.3
Asian	4	1.0
Native Hawaiian/Other Pacific Islander	1	0.3
Other	16	4.1
Geographic region (n = 386)
Northeast	84	21.8
Midwest	115	29.8
South	119	30.8
West	68	17.6
Marital status (n = 390)
Married or domestic partner	280	71.8
Separated or divorced	56	14.4
Single	54	13.8
Religious affiliation (n = 389)
Catholic or Christian	293	63.2
Jewish	3	0.8
Buddhist	6	1.5
Muslim	1	0.3
Other	37	9.5
None	49	12.6
Level of education (n = 390)
High school or less	23	5.9
Some college/completed college	215	55.1
Graduate school	152	39.0
Years worked in healthcare (n = 389)
0–10 years	122	31.4
11–25 years	138	35.4
26+ years	129	32.9
Years worked in hospice (n = 390)
0–10 years	271	69.6
11–25 years	104	26.6
26+ years	15	3.8
Type of hospice (n = 380)
For profit	61	15.3
Not-for-profit	316	81.0
Government based	3	0.8
Daily patient census at your location (on average) (n = 382)
<50	127	33.2
51–100	64	16.4
101–200	101	25.9
201–500	63	16.2
501–1000	23	6.0
1000+	4	1.0

NHPCO, National Hospice and Palliative Care Organization.

Over half of the participants reported having a Christian-based faith (63%), and when asked to rate how important religiosity or spirituality was in their life on a scale from 0 to 10, the participants reported an average score of 7.8 (n = 390, SD = 2.58, range = 0–10). When asked to rate their current level of social support on a scale from 0 to 10, participants reported a mean score of 7.5 (n = 390, SD = 2.2, range = 0–10). Finally, when asked to rate their current level of professional support, participants reported a mean of 7.0 (n = 390, SD = 2.3, range = 0–10).

Descriptive statistics were obtained for all variables, including distributions, means, medians, variances, SDs, ranges, and quartiles. When evaluated, skewness and kurtosis values were found to be within normal limits for all demographic variables as well as each of the three dependent variables in the ProQOL (Compassion Satisfaction, Burnout, and Secondary Traumatic Stress subscales).⁴² Missing data were handled by using the “exclude cases listwise” option in SPSS.⁴³ In all, 68 cases with missing data (17%) were excluded from the statistical analyses.

Use of personally meaningful rituals

Of those hospice staff and volunteers who responded to the question, the majority reported using some form of personally meaningful ritual after the death of their patients (71%, n = 248). When asked to describe their rituals, hospice staff and volunteers cited acts such as attending the funeral of patients, calling the bereaved to offer condolences, writing a poem or journaling, lighting a candle or saying a prayer, walking in a forest or near a beach, or simply picturing the deceased and wishing them well on their “next journey.” Staff and volunteers also spoke of using multiple rituals, with the following excerpts showing how various rituals were embraced depending on the situation:

I think about the person and reflect on moments spent with them. I allow myself a time of silence to be alone in a quiet space … I write a note/affirmation to the person and put it in a jar that I keep in my office. I also draw or paint … Sometimes I talk to the person and tell them how I am feeling and/or what it is that I hope for them now that they have moved on in their journey.

Immediately after a patient's death I think of that patient and family and say a prayer for their peace and strength. I take a moment to appreciate the opportunity I had to care for them and I often find myself tearful during this time. I allow myself to experience the grief and cry if needed… In addition to this, when I have a patient that I am really close to, I go for a hike after their death to a stream. I release a stick into the water in their honor and think of the gratitude I have in caring for them and the grief I have in their loss.

Note: The full qualitative analyses will be reported in a separate publication due to the large amount of data gathered in this regard.

ProQOL subscales: compassion satisfaction, burnout, and secondary traumatic stress

When measuring responses to each of the three ProQOL subscales, the majority of hospice staff and volunteers scored in the “average” to “high” range in Compassion Satisfaction with a mean score of 43 on a scale from 0 to 50. The majority of hospice staff and volunteers scored in the “low” range on Burnout, with a mean score of 20 on a scale from 0 to 50, as well “low” on Secondary Traumatic Stress as evidenced by a mean score of 19 on a scale from 0 to 50.

When comparing hospice staff and volunteers who did report using personally meaningful rituals after the death of their patients (71%, n = 248) to those who did not use rituals (29%, n = 100), significant differences were found. First, hospice staff and volunteers who engaged in rituals had significantly higher scores on Compassion Satisfaction (n = 236, X = 43.8, SD = 4.7) than those who did not use rituals (n = 92, X = 40.1, SD = 6.0); F(1,326) = 33.48, p < 0.01). Additionally, hospice staff and volunteers who engaged in rituals were found to have significantly lower scores on Burnout (n = 238, X = 19.2, SD = 5.5) than those who did not (n = 91, X = 21.4, SD = 6.4); F(1,327) = 10.01, p = 0.002. Hospice staff and volunteers who engaged in rituals were not found to have significant differences on Secondary Traumatic Stress. These findings were consistent even when the following variables were controlled for in the analyses: age, gender, race, marital status, level of education, level of social and professional support, religion/spirituality, number of years worked in hospice/palliative care, number of years worked in healthcare, and the type of hospice.

Role of ritual

Results of the single linear regression showed that 9.3% of the variance in Compassion Satisfaction was accounted for by the use of personal rituals (R² = 0.093; p < 0.01); F(1,326) = 33.481, p < 0 .01), and 2.7% of the variance in Burnout (R² = 0.027, p < 0.05); F(1,327) = 10.011, p < 0.002). The use of rituals did not significantly explain levels of Secondary Traumatic Stress (F(1,329) = 0.375, p = 0.541).

As shown in Table 2, higher levels of Compassion Satisfaction were found to significantly relate to higher levels of professional support and social support, as well as increased religiosity/spirituality, older age, and greater number of years working in healthcare. Higher levels of Burnout were significantly related to lower levels of professional support and social support, decreased religiosity/spirituality, and younger age. Higher levels of Secondary Traumatic Stress were significantly related to lower levels of professional support and social support as well as younger age.

Table 2.

Correlation Matrix: Compassion Satisfaction, Burnout, and Secondary Traumatic Stress as Correlated with Demographic and Psychosocial Variables

Variable	Professional support	Compassion satisfaction	Burnout
Religiosity/spirituality	0.180**	0.136*	−0.237**
Social support	0.571**	0.264**	−0.376**
Professional support	1	0.341**	−0.402**
Years worked in healthcare	0.090	0.169**	−0.056
Years worked in hospice	0.172**	0.096	0.072
Age	0.210**	0.032	0.052

^**p < 0.01 (two-tailed); ^*p < 0.05 (two-tailed).

Tests for the assumption of nonmulticollinearity were performed, and all variables were correlated at less than 0.70, VIF values averaged 1, tolerance values were higher than 0.2, and the Durbin–Watson statistic was 2.105. As a result, all variables were deemed to be within normal ranges of multicollinearity.⁴⁴ Missing values were handled using the “exclude cases listwise” option in SPSS.⁴³

Prediction of compassion satisfaction

As demonstrated in Table 3, the stepwise regression analyses showed that personal rituals accounted for 8.4% of the variance in Compassion Satisfaction (β = 0.290), with levels of professional support adding 9%, and age offering an additional 4%. In total, the model explained 22.6% of the variance in Compassion Satisfaction (R² = 0.226, p < 0.01).

Table 3.

Multiple Regression Models Predicting Compassion Satisfaction, Burnout, and Secondary Traumatic Stress

	R	R2	Adjusted R2	Standard error of the estimate	R2change	F change	df1	df2	Significance	β-Values
Compassion satisfactiona
Ritual	0.290	0.084	0.081	5.106	0.084	29.213	1	319	0.000**	0.290 (Ritual)
Professional support	0.427	0.182	0.177	4.831	0.098	38.291	1	318	0.000**	0.250 (Ritual), 0.316 (Professional Support)
Age	0.475	0.226	0.218	4.709	0.043	17.685	1	317	0.000**	0.229 (Ritual), 0.314 (Professional Support), 0.209 (Age)
Social support	0.482	0.232	0.223	4.696	0.007	2.774	1	316	0.097	0.227 (Ritual), 0.261 (Professional Support), 0.209 (Age), 0.098 (Social Support)
Worked in healthcare	0.482	0.232	0.220	4.703	0.000	0.048	1	315	0.826	0.227 (Ritual), 0.260 (Professional Support), 0.202 (Age), 0.098 (Social Support), 0.013 (Worked in Healthcare)
Worked in hospice care	0.488	0.238	0.224	4.693	0.006	2.351	1	314	0.126	0.228 (Ritual), 0.256 (Professional Support), 0.225 (Age), 0.106 (Social Support), 0.026 (Worked in Healthcare), 0.042 (Worked in Hospice)
Religion/spirituality	0.488	0.238	0.221	4.700	0.000	0.043	1	313	0.836	0.230 (Ritual), 0.256 (Professional Support), 0.227 (Age), 0.109 (Social Support), 0.026 (Worked in Healthcare), 0.042 (Worked in Hospice), 0.108 (Religion)
Burnoutb
Ritual	0.160	0.026	0.022	5.807	0.026	8.337	1	318	0.004*	−0.160 (Ritual)
Professional support	0.420	0.177	0.172	5.346	0.151	58.234	1	317	0.000**	−0.112 (Ritual), 0.392 (Professional Support)
Social support	0.456	0.208	0.200	5.253	0.031	12.360	1	316	0.001**	−0.108 (Ritual), 0.277 (Professional Support), −0.210 (Social Support)
Age	0.481	0.231	0.222	5.182	0.024	9.715	1	315	0.002**	−0.092 (Ritual), 0.276 (Professional Support), 0.212 (Social Support), 0.155 (Age)
Religion/spirituality	0.491	0.241	0.229	5.159	0.009	3.827	1	314	0.051	−0.077 (Ritual), 0.278 (Professional Support), 0.185 (Social Support), 0.134 (Age), 0.103 (Religion/Spirituality)
Worked in healthcare	0.495	0.245	0.231	5.151	0.005	1.951	1	313	0.163	−0.081 (Ritual), 0.282 (Professional Support), 0.184 (Social Support), 0.180 (Age), 0.103 (Religion/Spirituality), 0.083 (Worked in Healthcare)
Secondary traumatic stressc
Ritual	0.033	0.001	−0.002	5.151	0.001	0.349	1	322	0.555	0.035 (Ritual)
Social support	0.252	0.064	0.058	4.995	0.063	21.440	1	321	0.000**	0.057 (Ritual), −0.251 (Social Support)
Age	0.277	0.077	0.068	0.496	0.013	4.545	1	320	0.034*	0.070 (Ritual), 0.250 (Social Support), 0.115 (Age)
Professional support	0.284	0.081	0.069	4.965	0.004	1.363	1	319	0.244	0.074 (Ritual), 0.209 (Social Support), 0.115 (Age), −0.076 (Professional Support)

^*p ≤ 0.05; ^**p ≤ 0.01.

^aExcluded variables due to nonsignificance in the compassion satisfaction model: social support, number of years worked in healthcare, number of years worked in hospice, religiosity/spirituality.

^bExcluded variables due to nonsignificance in the Burnout model: religiosity/spirituality and number of years worked in healthcare.

^cExcluded variable due to nonsignificance in the Secondary Traumatic Stress model: Professional Support.

Prediction of burnout

Rituals accounted for 2.6% of the variance in Burnout (β = −0.160), whereas levels of professional support added 15%, social support accounted for 3%, and age accounted for 2%. In total, the model explained 23% of the variance in Burnout (R² = 0.231, p < 0.05).

Prediction of secondary traumatic stress

Rituals did not account for a significant amount of the variance in Secondary Traumatic Stress (β = 0.035). Instead, levels of social support accounted for 6.4%, and age explained 1.3% of the variance. In total, the model explained 7.7% of the variance in Secondary Traumatic Stress (R² = 0.077, p < 0.05).

Summary

The majority of the hospice staff and volunteers in this national study used a variety of personally meaningful rituals after the death of their patients. Those who used rituals showed significantly higher Compassion Satisfaction and significantly lower Burnout than those who did not use rituals. Increased Compassion was related to greater support, older age, and more years of experience, as well as increased religiosity/spirituality.

These results are similar to other studies which have shown that creating meaning through spiritual experiences and activities cultivates growth experiences among end-of-life care providers²³ and decreases negative experiences at work.^45–48 Indeed, providing meaning and spiritual practices have been found to enhance overall job satisfaction and lower labor turnover.^49–51 This study suggests that rituals may be one type of meaningful practice that reduces negative consequences and increases positive experiences at work.

With regard to Secondary Traumatic Stress, the results of this study were inconsistent with previous literature that demonstrated higher scores on measures of Secondary Traumatic Stress among healthcare professionals.^52–54 Townsend and Campbell⁵⁵ highlighted several protective factors for Secondary Traumatic Stress, which included older age, higher education levels, and peer support. The majority of the participants in this study were of older age and held over 10 years of experience in healthcare. It is possible that our sample did not capture the perspectives of those hospice staff and volunteers who were younger, with less experience, and who may currently be experiencing any type of Secondary Traumatic Stress.

Limitations and future directions

This study sheds light on the use of ritual practices among hospice staff and volunteers, yet the results need to be considered in the context of several study limitations. First, although the participants were evenly distributed across the United States, a low general response rate to the online survey was received. It is noted that online surveys tend to have lower response rates than traditional mail surveys despite reminder e-mails,⁵⁶ nonetheless, it is likely that a self-selection bias for those interested in the topic of rituals occurred among those who did respond.

Additionally, the participants’ average scores on Compassion Satisfaction were high, and their Burnout and Secondary Traumatic Stress were low, leading to a restriction of range and possible decreases in the overall generalizability of results (e.g., it is unclear why staff with greater burnout or stress did not respond to the survey, or how they may have differently answered the study questions).⁴⁴ Additionally, the specific demographics regarding individual hospice care providers are not monitored by the NHPCO at this time (e.g., average age of staff, years of experience), making it difficult to know the true representativeness of this sample. Finally, given the correlational nature of these data, it is not possible to establish causality with regard to the exact role of rituals in these participants’ work, and the precise role of rituals in adjustment, job satisfaction, personal growth, and the professional well-being of hospice staff remains unclear.

To address these limitations and to further expand our understanding of how personally meaningful rituals may shape the work of hospice staff and volunteers, several future studies are possible. First, future studies could include experimental designs that would allow for the creation of ritual interventions, with pre- and post-testing of how such practices impact various hospice staff and volunteers. Such studies could more directly discern the mechanisms of action in rituals. Do rituals provide an increased sense of connectedness, meaning, and/or spirituality, and if so, are some rituals more effective than others? Finally, qualitative interviews could be completed among those hospice staff and volunteers who demonstrate the highest levels of compassion. Such interviews would help us learn more about whether the most compassionate hospice staff or volunteers use rituals, and if so, what recommendations would they have for others?

Conclusions

This study adds to the growing body of literature that connects work satisfaction with spiritual practices, of which rituals may be an important component. Due to the significant decrease in burnout and increased compassion shown among hospice staff and volunteers who use personally meaningful rituals, hospices may want to provide trainings in this area, especially for new staff who may be at greater risk for burnout. Doing so may help provide staff and volunteers with a personalized, cost-effective way of increasing their overall work satisfaction, ultimately improving their quality of service delivery for patients and families. In sum, this study points to the positive role of rituals as a way to enhance end-of-life care practices, and suggests the need for future research into what aspects of rituals are most powerful for improving both the lives and work of hospice professionals.

Acknowledgments

This study was funded by the MAPI Research Trust—Catherine Pouget International Research Award (Montross-Thomas). Additional research resources were provided by an American Cancer Society Mentored Research Scholar Grant (Montross-Thomas MRSG-13-233-01-PCSM). The study was also made possible by the important efforts of the National Hospice and Palliative Care Organization (NHPCO) who sponsored the distribution of the online survey used in data collection among their national membership.

Author Disclosure Statement

No competing financial interests exist.