Predictors of burnout and health status in Samaritans’ listening volunteers

Abstract

Samaritan listening volunteers provide emotional support to people in distress or suicidal. Samaritans’ has high volunteer turnover, which may be due to burnout. This study evaluated the role of demographic and psychosocial factors in predicting Samaritans listening volunteers’ burnout and health status. Samaritans’ listening volunteers (n = 216) from seven branches across UK completed an online survey to assess their levels of burnout (emotional exhaustion, depersonalisation, personal accomplishment), subjective health status, coping, empathy and social support. Overall, listeners showed low levels of burnout and good health. Regression analysis revealed that higher emotional exhaustion was predicted by younger age and avoidant coping style; higher depersonalisation was predicted by lower empathy fantasy and higher avoidant coping style; lower personal accomplishment scores were predicted by higher empathy personal distress and worse health status was predicted by more hours per week spent on listening duties, lower social support and higher avoidant coping style. Overall, different factors influenced different facets of burnout. However, higher use of avoidant coping style consistently predicted higher burnout and worse health status, suggesting avoidant coping is an important target for intervention.

Keywords:

• Burnout
• health status
• volunteers
• Samaritans
• coping
• social support
• empathy

Introduction

Burnout is linked with a range of organisational and health outcomes such as staff turnover, physical illness and depression (e.g. Bianchi, Boffy, Hingray, Truchot, & Laurent, 2013; Federici & Skaalvik, 2012; Melamed, Shirom, Toker, Berliner, & Shapira, 2006; Shirom, Toker, Melamed, Berliner, & Shapira, 2013). Studies on employee burnout have focused mainly teachers, nurses, informal caregivers and general practitioners exploring rates of burnout and the effectiveness of a range of interventions including meditation and related approaches such as mindfulness (e.g. Dharmawardene, Givens, Wachholtz, Makowski, & Tjia, 2016; Federici & Skaalvik, 2012; Shirom et al., 2013). Some research has also explored burnout in mental health professionals (MHP) including psychiatrists, non medical MHPs and psychiatric trainees indicating high levels of burnout possibly due to issues such as working with demanding or violent patients, low pay, lack of positive feedback, working with little training or experience, high job responsibility and patient suicides (e.g. Jovanović et al., 2016; Kumar, 2011; Rössler, 2012; Volpe et al., 2014). Many of these factors are also of relevance to the focus on the current study, namely crisis line volunteers working for Samaritans.

One key predictor of burnout is age and experience. For example, a meta analysis by Brewer and Shapard (2004) found negative correlations between age and years of work experience and burnout (emotional exhaustion). There is also a key role for psychosocial predictors of burnout. Thornton (1992) found avoidant coping predicted higher burnout whereas Ben-Zur and Michael (2007) found problem-focussed coping predicted lower depersonalisation and greater personal accomplishment, suggesting that problem-focusing coping may buffer against burnout.

Studies also indicate a role of for social support. For example, Ben-Zur and Michael (2007) found that more personal resources including social support related to lower burnout. In addition, although empathy has not featured prominently in burnout research, it has been shown to prevent or reduce risk of burnout (Day & Chambers, 1991) possibly by increasing compassion satisfaction (Thomas, 2013; Wagaman, Geiger, Shockley, & Segal, 2015). Likewise, in a small sample of Samaritan volunteers, Lydon (2009) reported that empathy was protective against the impact of supporting others. Finally, there is also a potential role for emotional expression which has proven positive impacts on physical and mental health outcomes (Pennebaker & Chung, 2011).

The focus of the present study is crisis-line providers working for Samaritans. Samaritans’ vision is that fewer people die by suicide. With 201 branches in UK and Ireland, Samaritans provide free, 24/7, confidential, emotional support to ‘callers’ by telephone, letter, text message, email, face-to-face in their branches and by outreach including prisons and music festivals. In 2014, Samaritans had over 3 million ‘caller’ contacts (Samaritans, 2015a). Samaritans are constantly recruiting new volunteers due to a high level of staff turnover which takes huge resources to train and supervise the new recruits (Samaritans, 2015b). It is likely that this level of turnover relates to burnout and health status given the stressful nature of the role of listeners. This study investigated the role of demographic and psychosocial factors (coping, social support, empathy) in predicting burnout and health status in Samaritans listening volunteers in UK.

Method

Design

A cross-sectional design using an online survey between June and July 2015.

Participants

Two hundred and ninety-six Samaritans’ volunteers entered the study site but complete data was received from 216 participants (73%). All participants were on-duty, listening volunteers ranging in age from 18 to 80 years (18–25 (n = 35); 26–35 (n = 51); 36–45 (n = 63); 46–55 (n = 52); 56–65 (n = 13); 66–75 (n = 1); 75 + (n = 1)). Mean = 38.7 years, SD = 11.81. Of these 65 (30.4%) were male, and 150 (69.1%) were female (1 was not disclosed). The minority lived alone (n = 51; 23.5%) whilst 166 (76.5%) lived with others. Only 25 (11.5%) participants kept a written diary for recording thoughts and feelings. Mean hours per week volunteering on listening duties was 4.17 h (SD = 1.7) ranging from 1 to 15 h per week.

Measures

The following constructs were measured. Reliability was assessed using Cronbach’s alpha.

1. Profile characteristics: Age, gender, living status, and the average time spent on listening duties each week were assessed.

2. Psychosocial variables: (i)Social Support: using the Short Form Social Support Questionnaire (SSQ6; Sarason, Shearin, Pierce, & Sarason, 1987; α = .93). (ii) Empathy: using the Interpersonal Reactivity Index (IRI; Davis, 1983); Perspective Taking (α = .65), Fantasy (α = .71), Empathic Concern (α = .64) and Personal Distress (α = .65); (iii) Coping: using the Brief COPE (Carver, 1997) grouped into two subscales; approach (α = .81) and avoidant (α = .69). Participants were also asked if they kept a written diary for recording thoughts and feelings (Yes/No) to assess diary keeping.

3. Outcome variables: Outcome variables were measured for the past 4 weeks:

(i) Health status: using a single item health status measure (Ware, Kosinski, Dewey, & Gandek, 2001). Higher scores indicate worse health. (ii) Burnout: using the Maslach Burnout Inventory (MBI-HSS; Maslach, 1982) modified for Samaritans’ listening volunteers. This 22 item questionnaire has three subscales: Emotional Exhaustion (α = .83), Depersonalisation (α = .66) and Personal Accomplishment (α = .84).

Procedure

On-duty, listening volunteers, as opposed to support volunteers or those currently ‘off-rota’ were sent an invitation by their Branch Director including a link to the information sheet and survey. There are a total of 201 Samaritans’ branches across the UK. Seven of these participated in the study. At the end of 2015 there were 16,941 registered listeners and 15,011 active listeners working for Samaritans.

Results

Describing outcome variables

Key outcome variables are shown in Table 1.

Table 1. Burnout and health status over the past 4 weeks.

	Population norms*	Mean/SD (n = 216)	N (%)	Normal range
Emotional exhaustion	20.99	6.59 (5.7)	Low = 203 (94%)	0–16
			Moderate = 13 (6%)	17–26
			High = 0 (0%)	27 or over
Depersonalisation	8.73	3.47 (3.54)	Low = 184 (85.2%)	0–6
			Moderate = 27 (12.5%)	7–12
			High = 5 (2.3%)	13 or over
Personal accomplishment	35.58	32.03 (8.11)	Low = 55 (25.5%)	39 or over
			Moderate = 52 (24.1%)	32–38
			High = 109 (50.5%)	0–31
Health status	N/A	2.23 (1.0)	N/A	N/A

* Data from Maslach Burnout Inventory manual.

Compared to population norms, participants showed low levels of emotional exhaustion and depersonalisation, and moderate to high personal accomplishment. When classified into groups, the majority of participants showed low emotional exhaustion and depersonalisation and moderate to high personal accomplishment. Mean health status scores were 2.23, indicating that most participants had very good health in the past 4 weeks.

Predicting burnout and health status

Multiple regression analyses were used to predict burnout subscales and health status from demographic characteristics and psychosocial predictors (see Table 2).

Table 2. Predictors of burnout subscales and health status in the past 4 weeks.

Variable	Emotional exhaustion	Depersonalisation	Personal accomplish.	Health status
Age	−.23***	ns	ns	ns
Gender	ns	ns	ns	ns
Living arrange.	ns	ns	ns	ns
Hrs/wk listening	ns	ns	ns	.15*
Diary keeping	ns	ns	ns	ns
Social support	ns	ns	ns	−.15*
Emp persp taking	ns	ns	ns	ns
Emp fantasy	ns	−.19***	ns	ns
Emp concern	ns	ns	ns	ns
Emp personal distress	ns	ns	−.14*	ns
Avoidant coping	.15*	.24***	ns	.18**
Approach coping	ns	ns	ns	ns
Adj. R^2 final model	9.7%	12.7%	6.3%	6%

Notes: Values are standardised regression β-coefficients, ns is non- significant variables removed from the model.

* p < .05

** p < .01

*** p < .001.

Burnout

Emotional exhaustion

Younger age and higher use of an avoidant coping style predicted higher emotional exhaustion accounting for 9.7% of the variance (F = 2.9; p = .001)

Depersonalisation

Lower empathy fantasy and greater avoidant coping predicted greater depersonalisation scores accounting for 12.7% of the variance (F = 3.6; p = .0001).

Personal accomplishment

Lower empathy concern predicted higher personal accomplishment accounting for 6.3% of the variance (F = 2.2; p = .01).

Health status

Worse health status was predicted by longer hours per week spent listening, less social support and more avoidant coping accounting for 6% of the variance (F = 2.15, p = .01)

Discussion

The present study indicated that Samaritan listeners showed low levels of burnout and good health status. In terms of burnout, emotional exhaustion was predicted by younger age and using an avoidant coping style, depersonalisation was predicted by low empathy fantasy and avoidant coping and personal accomplishment was predicted by lower empathy personal distress. In term health status, poorer health status was predicted by more hours per week spent on listening duties, less social support and greater avoidant coping. Avoidant coping was thus a consistent predictor of two of the three burnout subscales and worse health status with some role for age, social support and empathy. There were no effects of gender, living arrangements or diary keeping. These results find reflection in previous research highlighting the role for age and experience (Brewer & Shapard, 2004) and social support (Wang, Wu, & Liu, 2003) and empathy (Thomas et al., 2013; Wagaman et al., 2015). They also particularly reflect previous work on the importance of coping mechanisms, particularly avoidant coping (Ben-Zur & Michael, 2007; Thornton, 1992).

To conclude, the present study indicated low levels of burnout and good health status in Samaritan listeners. The results also showed that although different factors influenced different facets of burnout, avoidant coping emerged as a consistent predictor of burnout subscales and poorer health status. Samaritan listeners show a high turnover rate which places additional stress upon the organisation. Previous research has highlighted a role for meditation based interventions as effective for those in high stress occupations (e.g. Dhamawardene et al., 2016). The results from the present study indicate that for this specific population, avoidant coping would be a useful target for intervention as a means to improve listeners health and well being and encourage sustained working within the organisation.

Disclosure statement

No potential conflict of interest was reported by the authors.