Abstract
Context
Direct care workers in long-term care can develop close relationships with their patients and subsequently experience significant grief after patient death (Boerner et al., 2015). Consequences of this experience for employment outcomes have received little attention.
Objectives
To (1) investigate staff, institutional, patient, and grief factors as predictors of burnout dimensions among direct care workers who had experienced recent patient death, (2) determine which specific aspects of these factors are of particular importance, and (3) establish grief as an independent predictor of burnout dimensions.
Methods
Participants were 140 certified nursing assistants and 80 homecare workers having recently experienced patient death. Data collection involved comprehensive semi-structured in-person interviews. Standardized assessments and structured questions addressed staff, patient, and institutional characteristics, grief symptoms and grief avoidance, as well as burnout dimensions (depersonalization, emotional exhaustion, and personal accomplishment).
Results
Hierarchical regressions revealed that grief factors accounted for unique variance in depersonalization, over and above staff, patient, and institutional factors. Supervisor support and caregiving benefits were consistently associated with higher levels on burnout dimensions. In contrast, coworker support was associated with a higher likelihood of depersonalization and emotional exhaustion.
Conclusion
Findings suggest that grief over patient death plays an overlooked role in direct care worker burnout. High supervisor support and caregiving benefits may have protective effects with respect to burnout, whereas high coworker support may constitute a reflection of burnout.
Keywords: Burnout dimensions, employment outcome, grief, bereavement, patient death, caregiving, nursing assistants, homecare workers, direct care staff
Introduction
Due to the acceleration of the average life expectancy and related growth of the elder population, addressing long-term care (LTC) needs has become an increasingly vital issue. Yet, the current LTC workforce faces major barriers in meeting these needs, including insufficient training and support mechanisms for staff (1, 2), high levels of burnout (3), and high turnover rates in particular among direct care staff who provide the bulk of hands-on care (4). As the demand for direct care workers grows exponentially, both among certified nursing assistants (CNAs) in nursing homes and home care workers or home health aides (HHAs) in the community, identifying ways to ameliorate these workforce challenges is crucial.
Given the high prevalence of chronic illness and rising incidence of dementia in the aging population, direct care workers in LTC increasingly care for very ill elders in need of palliative care as they approach the end of life (5, 6). As a result, these workers are increasingly exposed to patient death and dying in the LTC work context. While the lack of attention to this issue has been lamented (7), a few studies have provided evidence that direct care workers often develop close relationships with their patients and subsequently experience significant grief after patient death (8, 9, 10). Consequences of this experience for employment outcomes have received even less attention. One study identified grief symptoms as a contributor to burnout among CNAs, but this conclusion was based on a general sample of CNAs who may or may not have had a recent patient death; thus questions about grief could not be related to any particular loss (8). Another study, using an ethnographic approach, found that home care aides attributed client death to their experience of burnout (11).
The purpose of our study was to examine burnout outcomes in the context of a recent patient death. Burnout negatively affects core prerequisites of high quality care and is therefore an important outcome for the direct care workforce. Burnout has been found to be associated with low morale and job satisfaction as well as high absenteeism and turnover (12, 13, 14). A key component of burnout, depersonalization, reflected in a sense of cynicism and detachment, is particularly concerning as it can result in personal distancing and reduce a staff member’s capacity to provide compassionate care. Depersonalization, as well as emotional exhaustion, are unfortunately common and experienced more frequently among direct care workers than among other healthcare professionals (15). One study found that staff members with high levels of burnout were more likely to condone resident abuse in the nursing home (16).
Since burnout is seen as a stress response, potential contributing factors have been conceptualized within stress-process type models. Stress process models focusing on caregiving typically consider caregiver characteristics (e.g., basic demographics), as well as care recipient/relational (caregiver to care recipient relationship) and context factors to explain why the impact of caregiving stressors on outcomes varies from person to person (e.g., 17). Similarly, work-specific models such as the job demands-resource model of burnout (18) focus on the links between person characteristics and both workplace demands and available resources to explain burnout reactions. Guided by this literature, we designed our study to include staff factors (characteristics of the person), institutional factors (characteristics of the workplace), and patient/relational factors (characteristics of/relationship with the patient). Since our focus in this study was direct care staff’s response to patient death, we had a particular interest in the role of grief for burnout dimensions. We included both the extent of grief experienced as well as efforts to avoid grief. We regarded the latter as pertinent because a key professional expectation for direct care staff is to be able to control one’s experiences and expression of feelings, and to perform daily care tasks in an uninterrupted fashion regardless of the circumstances (7).
Taken together, our primary aim was to investigate staff, institutional, relational, and grief factors as predictors of burnout dimensions among direct care workers who had experienced the recent death of a patient in their care. To be able to assess the role of the care setting (nursing home vs homecare) as one of the institutional factors, we included both CNAs and HHAs in this study. We sought to determine which specific aspects among the considered factors that had been identified as relevant in previous caregiving, bereavement, or workforce research (8, 18, 17, 3, 19, 20, 22) might play a role in the prediction of burnout dimensions. Based on prior work, we anticipated that support in the work context would be associated with lower levels on burnout dimensions (23, 15, 24, 3). We also suspected that experiencing the caregiving role as positive and meaningful might show protective effects (19, 20). Finally, we wanted to establish whether or not grief factors would explain independent variance in the burnout dimensions, after accounting for staff, patient/relational, and institutional factors.
Methods
Recruitment and Eligibility
This study is part of a larger mixed-method study that looked at bereavement in direct care workers (CNAs and HHAs; 9, 25, 26, 27, 28). For the nursing home sample, we recruited actively employed CNAs from three large nursing homes that were all part of the same care system in Greater New York. Patient deaths were tracked via electronic medical records. CNAs who had experienced within the past two months the death of a patient for whom they had primary responsibility were approached on the units, informed about the study, and asked if they were interested in participating. Of the 824 CNAs meeting eligibility criteria, we approached 219; 143 agreed to participate, 76 refused, and three did not complete the interview, resulting in a response rate of 64%. The remaining 605 CNAs could not be reached within two months of resident death, largely because of the CNAs’ schedules or limitations in research staffing. Enrolled CNAs were representative of the organization’s CNA population with regard to age, gender, race/ethnicity, and length of employment.
Procedures for HHA recruitment were modified to accommodate the homecare context. Additionally, HHAs could choose to complete the interview in Spanish, since English language proficiency is not a job requirement for HHAs and the pool of potential participants included individuals whose primary language was Spanish. We recruited from three homecare agencies. Administrative staff informed us when client deaths occurred and asked the primary HHA of the deceased client if study staff may contact them. If the HHA agreed, we followed up with a phone call to explain the study and schedule an interview. We attempted to contact a total of 122 HHAs. Of those, 38 could not be reached within two months of the client’s death. Of the 84 we were able to approach, 80 agreed to participate and four refused, resulting in a response rate of 95%. A comparison between enrolled HHAs and the larger pool of HHAs from the participating agencies indicated that the HHA sample, too, was representative of the population they were drawn from in terms of age, gender, and length of employment. A comparison with respect to race/ethnicity, however, indicated a difference in the balance of Black versus Hispanic HHAs. Whereas our study sample was 67% Black and 29% Hispanic, the larger pool of HHAs was 33% Black and 64% Hispanic.
Data Collection and Measures
Interviews were conducted by trained interviewers with a Bachelor’s or Master’s degree, lasting on average 80 minutes. Prior to all interviews written informed consent was obtained and participants received $30 for their time. Interviews were conducted in-person outside of work time, at a place and time of the participant’s convenience. Data analyses were based on a selection of measures from this interview pertaining to the focus of this paper.
Staff factors
Sociodemographic characteristics assessed included age, gender, education, marital status, and race/ethnicity. Years in profession and number of other patient deaths in past months were both assessed with single-item questions.
Institutional factors
Care setting was either nursing home (1) or homecare setting (0). Training regarding patient death was assessed with a single-item: “What, if any, training or preparation did you get from your agency to help you deal with dying and death of patients?” Since reports of training were overall low, we created dummy codes to indicate whether (1) or not (0) any training had been reported.
Supervisor and Coworker support were measured with one item each drawn from an assessment of support availability and adequacy validated and widely used in family caregiving studies (developed by 29): “To what extent do you feel you can turn to your supervisor for support? To what extent do you feel you can turn to your coworker for support?” Responses ranged from (1) not at all, to (4) very much. Higher values represented greater perceived availability of supervisor and coworker support.
Patient/relational factors
Months caring for patient was assessed with a single item asking about length of time assigned to the patient.
Relationship quality was assessed with a four-item scale successfully used in a previous large-scale caregiving study (30) to measure rewarding aspects in the relationship between caregiver and care recipient. Staff members were asked how often (a) they felt happy with their relationship with the resident, (b) the patient made them feel good about themselves, (c) they felt very emotionally close to the patient, and (d) they felt bored with the patient. Responses ranged from (1) never to (4) always. Higher scores indicated closer relationships. Cronbach alphas were 0.71 (CNAs) and 0.76 (HHAs).
Caregiving benefits were assessed with an 11-item scale that has emerged as a predictor of bereavement outcomes in previous studies (19, 20, 21). Each item began with the stem “Providing help to (name) has …,” followed with specific items such as “made me feel useful” and “enabled me to appreciate life more.” Responses ranged from (1) disagree a lot, to (5) agree a lot. Higher scores indicated greater caregiving benefits. Cronbach alphas were 0.80 (CNAs) and 0.78 (HHAs).
To assess patient suffering, participants were asked to rate the extent to which the patient suffered during the last weeks of life on a scale from (0) not suffering at all, to (10) suffered terribly.
Grief factors
Grief symptoms were assessed with the 13-item version of the Texas Revised Inventory of Grief (31), a validated scale to assess current symptoms associated with separation distress. Responses ranged from (1) completely false to (5) completely true. Cronbach alphas in the present study were 0.91 (CNAs) and 0.76 (HHAs).
Grief avoidance was assessed with two-items (avoid thinking and avoid talking about deceased) (32) rated on a 5-point scale for frequency of occurrence (almost never - almost constantly). Cronbach alphas in the present study were 0.83 (CNAs) and 0.81 (HHAs).
Burnout dimensions
The widely used Maslach Burnout Inventory (33) assesses three dimensions of the burnout syndrome: emotional exhaustion (e.g., Working with patients all day is really a strain for me), depersonalization (e.g., I don’t really care what happens to my patient), and personal accomplishment (e.g., At my work, I am confident that I am effective at getting things done). Items are answered in terms of frequency with which these feelings are experienced (ranging from 0 never to 6 everyday). Based on recent psychometric analyses and recommendations by Poghosyan, Aiken, and Sloane (34), the emotional exhaustion scale was computed with 7 items, depersonalization with 7 items, and personal accomplishments 8 items. Cronbach alphas in the present study ranged from .71 – .87 for CNAs and HHAs.
Statistical Analysis
Data analyses included frequency and mean level comparisons, bivariate correlations and hierarchical regression. First, Chi-square and t-tests were used to compare CNAs and HHAs on all major study variables. Since the two groups did not differ significantly on most of the relevant grief and employment indicators, subsequent bivariate and multivariate analyses examining associations of staff, institutional, patient/relational, and grief factors with burnout dimensions were conducted for the total sample of N = 220, accounting for care setting (nursing home vs homecare) as an institutional factor. Hierarchical regression models included staff factors (block 1), institutional factors (block 2), patient/relational factors (block 3), and grief factors (block 4). Grief factors were entered as the final block to be able to assess the unique contribution of this block over and above the prior blocks. Coefficients shown in Table 3 reflect the final models.
Table 3.
Depersonalization | Emotional Exhaustion | Personal Accomplishment | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
| ||||||||||||
B | SE | β | ΔR2 | B | SE | β | ΔR2 | B | SE | β | ΔR2 | |
Staff factors: | .05a | .10c | .08b | |||||||||
Age | −.02 | .04 | −.03 | .02 | .08 | .02 | .07 | .05 | .11 | |||
Education | −.25 | .26 | −.06 | 1.79 | .51 | .23c | .54 | .36 | .10 | |||
Marital status (married) | 1.37 | .78 | .12a | .97 | 1.50 | .04 | −.58 | 1.05 | −.04 | |||
Hispanic (yes) | 2.48 | 1.33 | .17a | 4.47 | 2.57 | .16a | −2.66 | 1.80 | −.13 | |||
Black (yes) | .23 | 1.18 | .02 | 3.45 | 2.28 | .13 | −.3.18 | 1.60 | −.18b | |||
Years in profession | −.04 | .06 | −.06 | −.09 | .11 | −.07 | −.08 | .08 | −.09 | |||
# of other patient deaths | .08 | .48 | .01 | .27 | .94 | .02 | 2.11 | .65 | .29c | |||
| ||||||||||||
Institutional factors: | .06b | .08c | .10d | |||||||||
Care setting (nursing home) | .81 | 1.09 | .07 | 3.07 | 2.10 | .14 | −3.45 | 1.48 | −.22b | |||
Training re: patient death | .65 | .80 | .06 | −1.08 | 1.54 | −.05 | −1.16 | 1.08 | −.08 | |||
Supervisor support | −.72 | .35 | −.15b | −1.54 | .68 | −.16b | 1.81 | .48 | .28d | |||
Coworker support | .71 | .36 | .16b | 1.50 | .69 | .17b | .46 | .48 | .07 | |||
| ||||||||||||
Patient/relational factors: | .03a | .05b | .05b | |||||||||
Months caring for patient | .40 | .78 | .04 | 1.84 | 1.50 | .09 | .77 | 1.05 | .06 | |||
Relationship with patient | .06 | .30 | .01 | −.71 | .57 | −.08 | −.31 | .40 | −.05 | |||
Caregiving benefits | −.55 | .18 | −.21c | −1.10 | .36 | −.20c | .74 | .25 | .20c | |||
Patient suffering | −.19 | .11 | −.12a | −.30 | .21 | −.09 | .23 | .15 | .10 | |||
| ||||||||||||
Grief factors: | .06c | .03b | .01 | |||||||||
Grief symptoms | .04 | .03 | .09 | .06 | .06 | .07 | .03 | .04 | .05 | |||
Grief avoidance | .54 | .19 | .21c | .60 | .37 | .12a | −.12 | .26 | −.03 | |||
| ||||||||||||
Total R2 | .20d | .25d | .22d |
ΔR2 = R2 Change;
P < 0.10,
P < 0.05,
P < 0.01,
P < 0.001.
Results
Reflective of the larger population of direct care workers in Greater New York, CNAs and HHAs were primarily female and of minority backgrounds (Table 1). Educational levels were also similar; most were high school graduates or had at least some college. HHAs were significantly younger than CNAs, and more likely to have never married than CNAs. HHAs were more likely to identify as Hispanic, and CNAs were more likely to be Black. Relative to HHAs, CNAs had been in the profession longer and had cared for the patient longer. Both groups showed similar levels of grief after patient death, as well as similar levels regarding two burnout dimensions, depersonalization and personal accomplishment. However, CNAs evidenced significantly higher levels of emotional exhaustion compared to HHAs.
Table 1.
CNAs (N=140) | HHAs (N=80) | ||||||
---|---|---|---|---|---|---|---|
| |||||||
Variable | Mean (SD) | Range | n (%) | Mean (SD) | Range | n (%) | Significance |
Gender (female) | 125 (89) | 77 (96) | χ2 (1, n = 220) = 3.29a | ||||
Age | 50.5 (8.9) | 24–69 | 43.2 (12.5) | 19–66 | t(124) = 4.55d | ||
Ethnicity (Hispanic) | 16 (11) | 23 (29) | χ2 (1, n = 219) = 10.31c | ||||
Race | χ2 (4, n = 216) = 13.70d | ||||||
Black | 117 (86) | 52 (67) | |||||
White | 3 (2) | 8 (10) | |||||
Pacific Islander/Native Am. | 2 (1) | 3 (4) | |||||
Other | 16 (12) | 15 (19) | |||||
Education | χ2 (7, n = 220) = 7.23 | ||||||
Grades 7–9 | 10 (7) | 8 (10) | |||||
Grades 10–11 | 9 (6) | 8 (10 | |||||
GED | 13 (9) | 4 (5) | |||||
HS graduate | 54 (39) | 25 (31) | |||||
Some college | 42 (30) | 25 (31) | |||||
College graduate | 10 (7) | 9 (11) | |||||
Marital status | χ2 (4, n = 219) = 19.52c | ||||||
Married | 67 (48) | 21 (26) | |||||
Living as married | 7 (5) | 2 (3) | |||||
Divorced/separated | 35 (25) | 24 (30) | |||||
Widowed | 8 (6) | 2 (3) | |||||
Never married | 22 (16) | 31 (39) | |||||
Years in profession | 15.22 (7.4) | 1–35 | 6.5 (6.6) | .16–29 | t(218) = 8.74d | ||
Other patient deaths past months | 1.77 (.89) | 1–6 | 1.15 (.45) | 1–3 | t(218) = 5.8d | ||
Training re: patient death (employer) | 59 (42) | 37 (46) | χ2 (1, n = 219) = 0.35 | ||||
Supervisor support | 2.8 (1.3 | 2.65 (1.2) | t(218) = 0.94 | ||||
Coworker support | 3.34 (.96) | 2.14 (1.2) | t(138) = 7.61d | ||||
Months caring for patient | 38.86 (36.9) | 1–150 | 18 (29) | .03–168 | t(216) = 6.1d | ||
Relationship with patient | 10.65 (1.8) | 4–12 | 10.45 (1.9) | 3–12 | t(218) = 0.76 | ||
Caregiving benefits | 52.35 (4.5) | 27–55 | 52.70 (3.7) | 38–55 | t(218) = 0.59 | ||
Patient suffering | 4.38 (3.3) | 0–10 | 5.28 (3.5) | 0–10 | t(217) = −1.87a | ||
Grief symptoms | 31.49 (13.2) | 13–65 | 33.16 (12.5) | 13–65 | t(217) = −0.92 | ||
Grief avoidance | 3.99 (2.1) | 2–10 | 4.36 (2.1) | 2–10 | t(219) = −1.27 | ||
Burn-out (MBI) | |||||||
Depersonalization | 4.74 (5.9) | 0–26 | 3.71 (4.6) | 0–24 | t(217) = 1.35 | ||
Emotional Exhaustion | 18.12 (10.9) | 0–26 | 12.94 (10.4) | 0–37 | t(217) = 3.42c | ||
Personal Accomplishment | 38.89 (8.0) | 11–48 | 40.54 (6.9) | 13–48 | t(217) = −1.54 |
CNA = certified nursing assistant; HHA = home health aide; GED = General Educational Development (legal equivalent to high school diploma); MBI = Maslach Burnout Inventory;
P < 0.10,
P < 0.05,
P < 0.01,
P < 0.001.
Hispanic ethnicity, being married, as well as higher levels of grief and grief avoidance showed significant positive associations with depersonalization (Table 2). More emotional exhaustion was associated with higher education, more months caring for patient, nursing home setting, and greater support availability from coworkers. Greater support availability from supervisor and caregiving benefits showed negative associations with depersonalization and emotional exhaustion, and positive associations with personal accomplishment.
Table 2.
Variable | Depersonalization | Emotional Exhaustion |
Personal Accomplishment |
---|---|---|---|
Gender (female) | .06 | .02 | −.06 |
Age | −.07 | .03 | .09 |
Ethnicity (Hispanic) | .15b | .05 | −.07 |
Education | .00 | .27d | .04 |
Marital status (married) | .14b | .13a | −.04 |
Years in profession | .03 | .12a | −.09 |
Other patient deaths past months | .03 | .10 | .11 |
Care setting (nursing home) | .09 | .23c | −.10 |
Training re: patient death (employer) | .01 | −.13a | .03 |
Supervisor support | −.17b | −.22c | .25d |
Coworker support | .11 | .16b | .06 |
Months caring for patient | .05 | .15b | .04 |
Relationship with patient | −.04 | −.11 | .06 |
Caregiving benefits | −.17b | −.18c | .25d |
Patient suffering | −.06 | −.06 | .05 |
Grief symptoms | .17b | .08 | .06 |
Grief avoidance | .26d | .13a | −.03 |
P < 0.10,
P < 0.05,
P < 0.01,
P < 0.001.
Predicting depersonalization with hierarchical regression analysis (Table 3), staff factors (Hispanic ethnicity and being married) made only a marginal contribution. Institutional factors explained 5% of the variance, primarily due to the influence of support availability: greater supervisor support was linked to lower levels of depersonalization, while greater coworker support was linked to higher levels of depersonalization. Patient/relational factors only explained 3% of additional variance, with a negative significant effect of caregiving benefits and a negative marginal effect of patient suffering. The final block, grief factors, explained an additional unique 6% of the variance, with a positive significant effect of grief avoidance: more grief avoidance was linked to higher levels of depersonalization.
Considering the prediction of emotional exhaustion, institutional and patient/relational factors played a similar role in terms of the strength and direction of the significant effects. We found similar patterns for support availability and caregiving benefits, but the proportion of variance explained was greater, with 8% explained by institutional and 5% explained by patient/relational factors. Staff factors explained 10% of the variance, primarily due to a positive significant effect of education; grief factors explained a smaller portion of the variance in emotional exhaustion, with a positive marginal effect of grief avoidance.
Regarding prediction of personal accomplishment, institutional factors explained the largest proportion of variance as in the prior models, with a negative significant effect of nursing home setting and a positive significant effect of support availability from supervisor: CNAs reported lower levels and those with greater supervisor support reported higher levels of personal accomplishment. Staff factors explained 8%, primarily due to the influence of race and number of other patient deaths: black staff members reported lower personal accomplishment and those who had experienced more other patient deaths in the past months reported higher levels of personal accomplishment. Patient/relational factors explained an additional 5% of variance, as in the other models, due to the positive effect of caregiving benefits. Grief factors yielded no unique explanation of variance in this model.
Even though there were some differential findings across the three outcomes, it is noted that the total amount of variance explained was similar across the models (.20–.25; P < 0.001), and institutional and patient/relational factors explained unique variance in all three models, with a consistent role of support availability and caregiving benefits. Grief factors evidenced a particularly influential role in the prediction of depersonalization.
Discussion
Our study highlights the role of grief, support in the workplace, and meaning derived from caregiving for burnout in the context of patient death among direct care workers in nursing homes and homecare. It is notable that grief factors indeed explained variance in burnout dimensions over and above other person, relationship and institutional factors, albeit significantly only for depersonalization, and marginally for emotional exhaustion. This is in line with a prior study (8), which also linked grief symptoms with burnout for depersonalization only. Beyond this finding is the new insight that grief avoidance relative to grief symptoms was more influential in predicting depersonalization. It suggests that those who made efforts to avoid their grief may have been more likely to report instances of depersonalization. The strong focus on controlling emotions as a way of dealing with staff reactions to patient death (7), may deserve some further reflection, as it could come with a cost in terms of the staff members’ ability to care for and about patients that is not desirable in the LTC context.
Generally, findings highlighted the importance of institutional factors for burnout dimensions. This is not surprising as support in the work context has been established as an important resource that can reduce or prevent burnout (23, 3). However, the new, at first glance unexpected, finding was that while availability of supervisor support had a positive effect throughout, coworker support showed the opposite pattern, being associated with more depersonalization and emotional exhaustion. It is possible that those who experienced burnout symptoms more frequently were simply more likely to reach out to their coworkers for support. However, an alternative or complementary explanation may be that the benefits of support depend on what the support entails and whether it is perceived as helpful (35, 36). But even the perception of helpfulness may not be a sufficient indicator for how effective available support is with regard to certain employment outcomes. For example, sharing frustration or “venting” among coworkers may give some relief in the moment, but may also reinforce general negativity in the workplace. Underscoring the adverse effect of coworker support, Willemse, et al. (24) found that in high strain situations, when CNAs experience high job demands and low decision making authority, high coworker support negatively impacted the aides’ personal accomplishment. Future research is needed to clarify which types of support and provision by whom may have the most positive impact on employment outcomes. Given the emerging evidence that support can not only have positive but also negative effects in this context, more research is urgently needed to clarify benefits and potential risks.
Besides support, the only other aspect that emerged as consistently associated with the burnout dimensions was caregiving benefits, an indicator of the extent to which the role of providing care is experienced as meaningful. Along with supervisor support, it was the only other variable that showed a consistent significant and positive effect on burnout dimensions. While relationship quality with the patient did not show an effect, it is noted that caregiving benefits are associated with better relationship quality; caregivers are more likely to experience their role as meaningful when the relationship to the care recipient is positive and close (19, 20). Thus, the potentially protective role of caregiving benefits could be reduced in a professional environment that discourages bonding with patients to avoid emotional upheaval among employees. Administrators and supervisory staff in LTC should consider this notion in light of the finding for grief avoidance, that strengthening direct care staff’s ability to manage their emotions related to patient death may be more effective than trying to prevent grief or the relationships that cause grief.
Several limitations deserve mention. Even though all predictor groups contributed to explaining variance in at least two of the three burnout dimensions, supporting the notion that these different types of factors should be considered in this type of study, the total amount of variance explained was small. This suggests that there were other important aspects that we did not include or that subsequent research should pursue the testing of more complex models. For example, prior work has shown associations between job satisfaction and burnout (14). Perhaps job satisfaction acted as a mediator between our predictors and outcomes. Moreover, we were not able to account for several job characteristics, such as number of patients the direct care worker was responsible for, case mix/patient diagnoses, or other work demand factors, for which prior evidence suggests associations to burnout, low job satisfaction, and turnover among direct care staff (37, 38). Another limitation is that findings were based on cross-sectional data, thus we cannot draw conclusions about what causes burnout in the context of patient death. But this study does identify which characteristics co-occur with higher levels of burnout symptoms, and these insights, at a minimum, provide us with cues for future research and yield “food for thought” for clinical staff and administrators in LTC.
With the growing demand for direct care workers, the negative impact of burnout can be devastating for an industry that is desperate for high quality caregivers. Our findings indicate a need for employers to recognize the potential negative effect of grief on direct care workers and to acknowledge the emotional impact of patient loss. Our study further shows the positive role of the supervisor as a resource for direct care staff, highlighting the critical nature of the supervisor-direct care worker relationship. Moreover, study findings provide cues that the perceived benefits of caregiving for this workforce may not only be positive on their own, but may also be protective against burnout. Employers can assist their direct care employees by providing opportunities for grief support, promoting supervisory support, and finding ways to increase the experienced benefits of providing care, such as encouraging a strong client-worker relationship.
Acknowledgments
This study was supported by a grant from the National Institute on Aging (1 R03 AG034076), as well as by several private donors (Kathrin Boerner, Principal Investigator).
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Disclosures
None of the authors had any conflicts of interest.
References
- 1.PHI. U.S. home care workers: Key facts. [Accessed January 19, 2017];2016 Available from: http://phinational.org/sites/phinational.org/files/phi-home-care-workers-key-facts.pdf.
- 2.Barbarotta L. Direct Care Worker Retention: Strategies for Success. [Accessed March 16, 2016];Institute for the Future of Aging Services and the American Association of Homes and Services for the Aging. 2010 Available from: http://www.leadingage.org/uploadedFiles/Content/About/Center_for_Applied_Research/Publications_and_Products/Direct%20Care%20Workers%20Report%20%20FINAL%20(2).pdf.
- 3.Woodhead EL, Northrop L, Edelstein B. Stress, social support and burnout among long-term care nursing staff. J Appl Gerontol. 2016;35:84–105. doi: 10.1177/0733464814542465. [DOI] [PubMed] [Google Scholar]
- 4.Dill JS, Cagle J. Caregiving in a patient’s place of residence: turnover of direct care workers in home care and hospice agencies. J Aging Health. 2010;22(6):713–733. doi: 10.1177/0898264310373390. [DOI] [PubMed] [Google Scholar]
- 5.Centers for Disease Control and Prevention. Chronic disease overview: The leading cause of death and disability in the United States. [Accessed January 19, 2017];2016 Available from: https://www.cdc.gov/chronicdisease/overview/index.htm.
- 6.Hebert LE, Weuve J, Scherr PA, Evans DA. Alzheimer disease in the United States (2010–2050) estimated using the 2010 census. Neurol. 2013;80:1778–1783. doi: 10.1212/WNL.0b013e31828726f5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Moss MS, Moss SZ. Nursing home staff reactions to resident deaths. In: Doka KJ, editor. Disenfranchised grief: New directions, challenges, and strategies for practice. Champaign, IL: Research Press; 2002. pp. 197–216. [Google Scholar]
- 8.Anderson KA. Grief experiences of CNAs: relationships with burnout and turnover. J Gerontol Nurse. 2008;34:42–49. doi: 10.3928/00989134-20080101-01. [DOI] [PubMed] [Google Scholar]
- 9.Boerner K, Burack OR, Jopp DS, Mock SE. Grief after patient death: Direct care staff in nursing homes and homecare. J Pain Symptom Manage. 2015;49:214–222. doi: 10.1016/j.jpainsymman.2014.05.023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Rickerson EM, Somers C, Allen CM, et al. How well are we caring for caregivers? Prevalence of grief-related symptoms and need for bereavement support among long-term care staff. J Pain Symptom Manage. 2005;30:227–233. doi: 10.1016/j.jpainsymman.2005.04.005. [DOI] [PubMed] [Google Scholar]
- 11.Stacey CL. The caring self: The work experiences of home care aides. Ithaca: ILR Press; 2011. [Google Scholar]
- 12.Braithwaite M. Nurse burnout and stress in the NICU. Adv Neonatal Care. 2008;8:3343–347. doi: 10.1097/01.ANC.0000342767.17606.d1. [DOI] [PubMed] [Google Scholar]
- 13.Laschinger HKS, Leiter M, Day A, Gilin D. Workplace empowerment, incivility, and burnout: Impact on staff nurse recruitment and retention outcomes. J Nurs Manag. 2009;17:302–311. doi: 10.1111/j.1365-2834.2009.00999.x. [DOI] [PubMed] [Google Scholar]
- 14.Ybema JF, Smulders PG, Bongers PM. Antecedents and consequences of employee absenteeism: A longitudinal perspective on the role of job satisfaction and burnout. Eur J Work Organ Psy. 2010;19:102–24. [Google Scholar]
- 15.Hare J, Pratt CC. Burnout: Differences between professional and paraprofessional nursing staff in acute care and long-term care health facilities. J Appl Gerontol. 1988;7:60–72. [Google Scholar]
- 16.Shinan-Altman S, Cohen M. Nursing aides’ attitudes to elder abuse in nursing homes: The effect of work stressors and burnout. Gerontologist. 2009;49:674–684. doi: 10.1093/geront/gnp093. [DOI] [PubMed] [Google Scholar]
- 17.Pearlin LI, Lieberman MA, Menaghan EG, Mullan JT. The stress process. J Health Social Behav. 1981;22:337–356. [PubMed] [Google Scholar]
- 18.Demerouti E, Bakker AB, Nachreiner F, Schaufeli WB. The job demands-resources model of burnout. J Appl Psychol. 2001;86:499–512. [PubMed] [Google Scholar]
- 19.Boerner K, Schulz R, Horowitz A. Positive aspects of caregiving and adaptation to bereavement. Psychol Aging. 2004;19:668–675. doi: 10.1037/0882-7974.19.4.668. [DOI] [PubMed] [Google Scholar]
- 20.Schulz R, Boerner K, Shear K, Zhang S, Gitlin LN. Predictors of complicated grief among dementia caregivers: a prospective study of bereavement. Am J Geriatr Psychiatry. 2006;14:650–658. doi: 10.1097/01.JGP.0000203178.44894.db. [DOI] [PubMed] [Google Scholar]
- 21.Tarlow BJ, Wisniewski SR, Belle SH, Rubert M, Ory MG, Gallagher-Thompson D. Positive aspects of caregiving: Contributions of the REACH project to the development of new measures for Alzheimer’s caregiving. Res Aging. 2004 Jul;26(4):429–53. [Google Scholar]
- 22.Boerner K, Schulz R. Caregiving, bereavement, and complicated grief. Bereave Care. 2009;28:10–13. doi: 10.1080/02682620903355382. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Chappell NL, Novak M. The role of support in alleviating stress among nursing assistants. Gerontologist. 1992;32:351–359. doi: 10.1093/geront/32.3.351. [DOI] [PubMed] [Google Scholar]
- 24.Willemse BM, Jonge J, Smit D, Depla MF, Pot AM. The moderating role of decision authority and coworker- and supervisor support on the impact of job demands in nursing homes: A cross-sectional study. Int J Nurs Stud. 2012;49:822–833. doi: 10.1016/j.ijnurstu.2012.02.003. [DOI] [PubMed] [Google Scholar]
- 25.Barooah A, Boerner K, van Riesenbeck I, Burack OR. Nursing home practices following resident death: The experience of Certified Nursing Assistants. Geriatr Nurs. 2015;36:120–125. doi: 10.1016/j.gerinurse.2014.11.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Riesenbeck I, Boerner K, Barooah A, Burack OR. Preparedness for resident death in long-term care: The experience of front-line staff. J Pain Symptom Manage. 2015a;50:9–16. doi: 10.1016/j.jpainsymman.2015.02.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Riesenbeck I, Boerner K, Barooah A, Burack OR. Coping with client death: How prepared are home health aides and what characterizes preparedness. Home Health Care Serv Q. 2015b;34:204–219. doi: 10.1080/01621424.2015.1108890. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Boerner K, Gleason H, Barooah A. Home health aides’ experience with client death: The role of employer policy. Home Healthc Now. 2016;34:189–196. doi: 10.1097/NHH.0000000000000360. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Cantor MH, Brennan M. Social care of the elderly: The effects of ethnicity, class, and culture. New York: Springer; 2000. [Google Scholar]
- 30.Williamson GM, Shaffer DR. Relationship quality and potentially harmful behaviors by spousal caregivers: how we were then, how we are now. The Family Relationship in Late Life Project. Pcychol Aging. 2001;16:217–226. [PubMed] [Google Scholar]
- 31.Faschingbauer TR, Zisook S, DeVaul RA. The Texas revised inventory of grief. In: Zisook S, editor. Biopsychosocial aspects of bereavement. Washington, DC: American Psychiatric Press; 1987. pp. 111–124. [Google Scholar]
- 32.Bonanno GA, Papa A, Lalande K, Nanping Z, Noll JG. Grief processing and deliberate grief avoidance: A prospective comparison of bereaved spouses and patients in the United States and China. J Consult Clin Psychol. 2005;73:86–98. doi: 10.1037/0022-006X.73.1.86. [DOI] [PubMed] [Google Scholar]
- 33.Maslach C, Jackson SE, Leiter MP. Maslach burnout inventory manual. 3. Palo Alto, CA: Consulting Psychologists Press; 1996. [Google Scholar]
- 34.Poghosyan L, Aiken LH, Sloane DM. Factor structure of the Maslach Burnout Inventory: An analysis of data from large scale cross-sectional surveys of nurses from eight countries. Int J Nurs Stud. 2009;46:894–902. doi: 10.1016/j.ijnurstu.2009.03.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Chou RJ, Robert SA. Workplace support, role overload, and job satisfaction of direct care workers in assisted living. J Health and Soc Beh. 2008;49:208–222. doi: 10.1177/002214650804900207. [DOI] [PubMed] [Google Scholar]
- 36.Gleason HP, Boerner K, Barooah A. Supporting home health aides through a client’s death: The role of supervisors and coworkers. Geriatr Nurs. 2016;37:278–283. doi: 10.1016/j.gerinurse.2016.04.001. [DOI] [PubMed] [Google Scholar]
- 37.Banaszak-Holl J, Hines MA. Factors associated with nursing home staff turnover. Gerontologist. 1996;36:512–517. doi: 10.1093/geront/36.4.512. [DOI] [PubMed] [Google Scholar]
- 38.Bishop CE, Squillace MR, Meagher J, Anderson WL, Wiener JM. Nursing home work practices and nursing assistants’ job satisfaction. Gerontologist. 2009;49:611–62. doi: 10.1093/geront/gnp040. [DOI] [PubMed] [Google Scholar]