Introduction

Burnout is a common condition among those in the serving professions. The concept of burnout grew out of research by Maslach et al. in the 1970s who noted a constellation of symptoms—low sense of personal accomplishment, high emotional exhaustion, and a high degree of depersonalization (the quality of treating another person as an object) (Maslach and Leiter 1996). While burnout is tightly correlated with depression (rho = .85, P < .01) (Maslach and Leiter 1996), an important difference is that burnout is job-related; leaving one’s job usually results in resolution of symptoms. Burnout has been studied among many in the helping professions, including primary care physicians (Freeborn 2001; Winefield and Anstey 1991), nurses (Catalan et al. 1996), teachers (Friedman 2000; Burke and Greenglass 1993), and social workers (Kadushin and Kulys 1995; Oktay 1992; Evans et al. 2006). However, the clergy represent a unique vocation with a unique set of required emotional demands and job skills. Administrative demands are often significant in the setting of a significant burden of individual counseling. Also, clergy may feel pressured to model normative Christian behaviors, scrutinized by their congregations.

At the same time, clergy often have near-total autonomy with great latitude in their schedule. This freedom, balanced with an encompassing set of emotional and administrative demands, allow clergy to employ various behaviors to cope with job stress. Scholarly reading, mentorship, independent bible study, retreats, as well as time out with family, can be integrated into their regimen. Among clergy, there are few studies that correlate job burnout with specific behaviors.

Many studies among clergy examine overall job satisfaction amidst stressors in an effort to understand the burnout syndrome. A study examining clergy for the United Church of Canada revealed a high level of job satisfaction and personal commitment—80% were “satisfied with their pastoral charge.” Paradoxically, the clergy scored higher on depression (78th percentile) and stressful symptoms (88th percentile) than the general population. Also, 75% felt that they “have too much work for one minister to do” and 60% “feel that they do not have adequate time to think and contemplate their role as a minister”(United Church of Canada—Warren Shepell Research Group 2003).

This paradox of experiencing high job stress but relatively high job satisfaction is found in a study that evaluated job stress and burnout among 237 mental health social workers. About 47% met criteria for a potential psychological disorder based on General Health Questionnaire-12 score—three times the prevalence of the general population. Yet, 47% were “mostly satisfied” with their job, 35% were ambivalent, and 19% were dissatisfied. Feeling undervalued in the workplace, perceived high job demand, lack of support, and limited decision-making ability were correlated with high prevalence of stress and emotional exhaustion (Evans et al. 2006).

There are only a few studies that correlate coping strategies, burnout, and job satisfaction among the clergy. A study by Rodgerson and Piedmont (1998) among 252 American Baptist ministers found that clergy who took an approach of “collaboration” with God (i.e., the problem solving process is held jointly with God) or “deferring” to God (i.e., the problem is deferred completely to God) had a strong correlation against emotional exhaustion and depersonalization, and positive impact with personal accomplishment. This is one of the few studies that evaluates religious coping among ministers.

Similarly, a study among Presbyterian ministers, elders, and members showed that “positive” religious coping—such as “looking to God for strength”—correlated with improved well-being. Conversely, “negative” religious coping—such as “feeling God was punishing me”—correlated with worse depressive symptoms. Interestingly, “positive” religious coping strategies had a greater effect on well-being among clergy than among members, while “negative” coping had a greater impact on depressive symptoms among clergy than church members. Although statistically significant, these correlations were weak suggesting other coping strategies might have a stronger impact on emotional stress (Pargament et al. 2001).

A study of several hundred clergy by Lee and Iverson-Gilbert (2003) across five protestant denominations showed that greater perceived intrusive demands of a congregation correlated with a lower sense of well-being and life satisfaction, and greater burnout. Interestingly, the greater impact on well-being was not the frequency of congregational demands, but rather the minister’s perception of those demands. A negative interpretation from personal criticism and family criticism significantly lowered a minister’s sense of well-being. The study also revealed that the greater number of supportive relationships of a minister—both within and outside of the congregation—correlated with greater well-being, life satisfaction, and optimism, and lower sense of burnout.

Employing emotional copings strategies, and specific behaviors, that protect against burnout is important not only for the long-term viability of the minister’s career, but also the long-term health of the church. Another important study by Randall (2004) showed in a cohort of 340 Anglican priests that higher levels of emotional exhaustion and depersonalization and lower levels of personal accomplishment correlated closely with a consideration to leave the ministry.

A study evaluating spirituality, emotional coping strategies, and burnout among Methodist ministers found that a greater spirituality score – as measured by a validated instrument—correlated with greater emotional exhaustion and also greater personal accomplishment. This suggests that, while exhausted, clergy felt a sense of satisfaction to their work. Specific emotional coping strategies were correlated with protecting against burnout: strategic planning, active coping, acceptance, and positive reframing (Doolittle 2007).

While emotional coping strategies and demographic variables have been well studied among the clergy, specific behaviors have not. Whereas personality types and demographics are often stable over time (one cannot change one’s gender, for example, or readily effect marital status), how clergy spend their time is an open, ready opportunity. Identifying specific behaviors, in addition to emotional coping strategies, is important for the management of burnout. This paper demonstrates that several specific behaviors are tightly correlated with—and may be protective against—the burnout syndrome among parish-based clergy of the New York Annual Conference of the United Methodist Church.

Design, Participants, and Setting

The association of behaviors, burnout, and spiritual attitudes was investigated by inviting the 358 parish-based clergy of the New York Annual Conference of the United Methodist Church, the administrative body of a mainline protestant denomination to complete a survey.

Participation was voluntary and anonymous. All identifying information was stored separately from the returned responses. Only subjects who fully completed the survey were included in the final analysis.

The study design and materials were approved by the Human Investigation Committee of the Yale University School of Medicine.

Survey Instrument

The survey collected demographic data and included two validated instruments: the Maslach Burnout Inventory (Maslach and Leiter 1996) and the Hatch Spiritual Involvement and Beliefs Scale (SIBS) (Hatch et al. 1998). The survey inquired about a broad range of behaviors that may impact the burnout syndrome. The survey sought responses regarding relational activities, such as spending time with family; professional activities, such as continuing education and having a mentor; and even physical activities, such as how often a minister exercised. These behaviors were identified based on a review of the literature as well as collaboration with members of the Bishop’s Cabinet of the New York Annual Conference.

The Maslach Burnout Inventory (MBI) is a leading instrument in the evaluation of the burnout syndrome among several human service and helping professions (Maslach and Leiter 1996). A total of 22 items address three domains—emotional exhaustion, depersonalization, and personal accomplishment. Each item is scored on a 1–5 Likert scale with standardized cutoffs—high, moderate, and low—for each burnout domain (Maslach and Leiter 1996).

The Spiritual Involvement and Beliefs Scale (SIBS), developed by Robert Hatch in 1998, investigates broad aspects of spirituality resonant with, but not specific to, the Christian tradition. Many prior studies used attendance at religious services or belief in God as the primary marker of spirituality (Comstock 1972; Paloutzian 1982; Zuckerman et al. 1984).

The SIBS consists of 26 questions on a 1–5 ordinal scale (strongly agree, agree, neutral, disagree, or strongly disagree) that address four distinct aspects of spirituality. The domain of internal beliefs evaluates one’s concept of God and spirituality. External practices evaluate how often one participates in worship. The domain of humility and forgiveness explores the relational qualities of faith, values, and reconciliation. Existential beliefs explore one’s concept of the “spiritual forces” that influence one’s life. The SIBS is tightly correlated with other instruments in the literature, but incorporates a broader definition of spirituality (Hatch et al. 1998). The composite SIBS score as well as the distinct domains are widely used in the literature to evaluate the complicated, multivariate nature of a person’s spiritual life.

Analyses

Data from completed questionnaires were entered into a secure Microsoft Excel Database (Microsoft Corporation 2000). Statistical analysis was performed using the JMP 4 statistical software package (JMP Institute 2001).

First, composite scores of the Maslach Burnout Inventory (emotional exhaustion, depersonalization, personal accomplishment) and the Hatch Spirituality Inventory were calculated. In the MBI, standardized cutoffs allow participants to be characterized as high, medium, or low for each burnout domain. For the purposes of generating bivariate likelihood ratios, those in the “low” and “medium” burnout category were grouped together and compared with those in the “high” burnout category (high emotional exhaustion, high depersonalization, and low personal accomplishment). Second, the Student’s t-test and the chi-square tests were used to compare the means of Maslach Burnout Inventory categories with demographic variables and specific coping behaviors. Third, multivariate regression was employed to control for demographic variables.

Results

Of the 358 surveys mailed, 222 (62%) completed all parts.

Table 1 shows demographic information for the respondent population. The average age of the ministers was 52 (±10) with 18 (±13) years in the ministry; 65% were male and 65% were married.

Table 1 Demographics of clergy participants
Full size table

While 13% considered themselves “burned out”, 23 % considered themselves “depressed” when asked.

Among the Maslach Burnout domains, 19% were highly emotionally exhausted, 26% were moderate, and 55% were low; 10% met the cutoff for high depersonalization, 30% were moderate, and 60% were low; and 43% showed a high sense of personal accomplishment, 46% were medium, and 11% were low.

Table 2 details which behaviors correlate with burnout variables. In multivariate modeling, demographic variables (age, gender, marital status, children, income, solo pastorate, income, and ethnicity) had a non-significant impact on the burnout domains.

Table 2 Correlation between burnout axes and self-assessed attitudes and behaviors
Full size table

Discussion

Pastors had good insight in identifying themselves as being burned out. Those pastors who identified themselves as being “burned out” were 30.6 times (P < .0001) as likely to meet criteria for high emotional exhaustion, and 16 times (P < .0001) as likely to meet criteria for high depersonalization. Pastors who self-identified themselves as being depressed were 41 times (P < .0001) as likely to have high emotional exhaustion and 21 times as likely to have high depersonalization. Those ministers presently serving a “traumatic church”—one characterized by high stress and conflict—were 11.9 times (P < .0001) as likely to have high emotional exhaustion. Interestingly, the effect appears to carry forward. Ministers who served a traumatic church in the past were 10.5 times (P < .0001) as likely to have high emotional exhaustion, even though they are no longer serving that congregation. The residual toxic effect of a past traumatic church negatively impacts their present vocation. Identifying a “traumatic church” experience early and intervening with a mentor or possibly professional counselor may have an impact in the on-going emotional health of clergy. Indeed, participating in regular counseling correlated with a high degree of emotional exhaustion (LR 5.2, P < .05) and great depersonalization (LR 4.1, P < .05). This finding suggests that ministers have self-identified their emotional exhaustion and have appropriately sought help.

Yet, interestingly, the domain of personal accomplishment had no correlation with self-identification as being depressed, burned out, or having a traumatic church. Self-identified satisfaction with one’s spiritual life was a very powerful predictor for not suffering emotional exhaustion, depersonalization, or low personal accomplishment. Clergy satisfied with their spiritual life were 1/20th (P < .0001) as likely to suffer emotional exhaustion. Efforts to enhance a clergy’s spiritual life may have important implications for their overall job satisfaction.

In general, several behaviors had a markedly greater impact on emotional exhaustion than on depersonalization and personal accomplishment. There are a few important features of the landscape of behaviors and burnout among clergy. First, ministers able to exercise three times a week were one-fourth (LR 0.25, P < .05) as likely not to have high emotional exhaustion. Also, having a mentor (LR 0.25, P < .05), engaging in scholarly reading (LR 0.14, P < .01), and going on a retreat twice a year (LR 0.20, P < .05) were associated with lower emotional exhaustion scores. Those behaviors associated with lower depersonalization were engaging in a regular bible study (LR 0.28, P < .05), going on a retreat at least twice a year (LR 0.09, P < .001).

Those behaviors associated with protecting against a low sense of personal accomplishment were being a mentor to another person (LR 0.20, P < .05), and taking regular time out for oneself (LR 0.22, P < 0.05) Having a rich network of relationships, family, retreats, mentors—outside of their role of minister—may protect against emotional exhaustion.

Taken in sum, a pattern emerges among those behaviors which may protect against burnout: balance and variety. Indeed, in multivariate modeling, the specific behaviors, along with the SIBS score, account for 62% of the variance for emotional exhaustion (P < .0001), 42% of the variance for depersonalization (P < .0001), and 42% (P < .0001) for personal accomplishment. In contrast, demographic variables had a non-significant impact on burnout.

Burnout may be influenced by one’s ability to establish healthy boundaries and time-management skills. Clergy who create the space to engage in outside activities may also have the ability and the emotional capacity to disengage from their ministerial demands, thereby re-energizing themselves. They may pour themselves out for their congregations, but then re-energize by engaging with the outside activities. These outside activities likely support and sustain their sense of call. Fostering healthy relationships at home, for example, models healthy relationships at church. Engaging in scholarly reading deepens one’s intellectual framework for ministry. Fostering a mentoring relationship provides support for the on-going emotional demands of ministry. In sum, these outside behaviors do not detract from a clergy’s call, but rather supports the call to ministry.

While many behaviors had a strong correlation with the burnout domain, it is equally important that many behaviors did not. Continuing education activities, outside reading, and writing for scholarly journals did not seem to impact burnout. This is likely due to the homogeneity of responses for those items. Most people pursue outside activities and reading, and very few write for scholarly journals.

There are two main weaknesses to this study. First, this study correlates data, but does not imply causation. While one may infer, for example, that semi-annual retreats may prevent emotional exhaustion, this may not necessarily be true. The correlation between counseling and emotional exhaustion, for example, suggests a selection bias: burned out pastors would be more likely to seek counseling in an attempt to cope. Counseling, unlikely, causes emotional exhaustion.

Second, this study examines specific behaviors but leaves many other aspects unexamined. A prior study evaluated emotional coping strategies that revealed active coping, positive reframing, planning, and acceptance to be protective against burnout variables (Doolittle 2007). Yet, there are many factors that may impact emotional health: friendships, chronic physical illness, satisfaction with marriage, etc. This study evaluated behaviors that may be readily implemented, but these behaviors themselves may only be markers of stronger moral value systems and emotional strength.

Nonetheless, the implications of this study are important. First, it may be more important for a church governing body to enhance “relational” activities, such as convening a retreat, rather than explicitly “educational” ones. Second, if a church is identified as a “traumatic church” marked by strife and conflict, assertive intervention by denominational leaders to mitigate the ongoing conflict is merited. Third, affirming a “balance and variety” model of personal growth may protect against burnout, strengthening the health of the clergy and, by extension, the church.