Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
J Pain Symptom Manage. Author manuscript; available in PMC 2016 July 1.
Published in final edited form as:
PMCID: PMC4492858

Preparedness for Resident Death in Long-Term Care: The Experience of Front-Line Staff



Although resident death is a common occurrence in long-term care, little attention has focused on how prepared certified nursing assistants (CNAs), who provide most of residents' daily care, are for this experience.


To identify characteristics of the resident, CNA, and care context associated with CNAs' preparedness for resident death, and to determine differential patterns for emotional versus informational preparedness.


One hundred forty CNAs completed semi-structured in-person interviews concerning their experiences around resident death. The associations of CNA characteristics (e.g., personal end-of-life [EOL] care preferences), CNAs' perceptions of resident status (e.g., knowledge of resident's condition), and the caregiving context (e.g., coworker support and hospice involvement) with emotional and informational preparedness were examined using bivariate and multivariate analyses.


CNAs who reported that their resident was “aware of dying” or “in pain” expressed higher levels of both emotional and informational preparedness. CNAs who endorsed an EOL care preference of wanting all possible treatments regardless of chances for recovery were likely to report lower emotional preparedness. More senior CNAs, both in regard to age and tenure, reported higher preparedness levels. Greater support from coworkers and hospice involvement also were associated with higher levels of both facets of preparedness, the latter in particular when hospice care was viewed positively by the CNA.


Having more information about resident status and more exchange opportunities within the care team around EOL-related challenges may help CNAs feel more prepared for resident death and strengthen their ability to provide good EOL care.

Keywords: Preparedness, resident death, caregiving, nursing assistant, direct care staff, grief, bereavement, long-term care, nursing home


Certified nursing assistants (CNAs) provide the majority of personal daily care and interact the most with nursing homes residents as compared to other staff. As a result, CNAs often develop close ties with residents they care for and sometimes see themselves as family surrogates.1,2 These relationships are found to be one of the primary reasons why nursing home staff remain on the job,3 an important factor in a work environment where retaining nursing staff is challenging.4 Having close relationships with residents means, for nursing home staff and especially for direct care workers like CNAs, that they might have reactions very similar to those of family members when a resident dies. This notion is supported by studies of grief symptoms in a range of nursing home staff members,5 as well as specifically in CNAs.6,7 Additionally, one study indicated that CNAs often feel completely unprepared for the death of their resident, and that low preparedness is associated with more intense grief after resident death.7 To our knowledge, no previous study has specifically addressed the variables related to preparedness for death among direct care staff in long-term care settings.

Overall, the topic of preparedness for death has received little research attention, even with respect to family caregivers. But the few existing studies show that low preparedness for death in family caregivers is associated with higher separation distress, traumatic distress and more symptoms of complicated grief and depression,8,9 as well as poor quality of life years after the loss.10 In a similar vein, a study on husbands' preparedness for spouses' cancer found an association between low preparedness and higher anxiety, sleep problems, emotional numbness, poor quality of life and less grief resolution in young widowers.11 Based on this sample, several important patient and caregiver factors were identified as predictors of preparedness including patient age and disease progression, health care-related information, and awareness of and willingness to talk about impending death, or efforts to avoid thinking and talk about the topic.12 Contextual factors, however, were not considered.

As the available empirical evidence suggests that being told a prognosis or accepting the inevitability of death does not necessarily translate into an overall sense of preparedness, it has been noted that preparedness might be better understood if knowledge or information about an imminent death and the emotional or mental aspect of preparedness are differentiated and explored separately.13 Subsequently, Hebert et al.14 argued that preparedness for death can be conceptualized as a multidimensional construct, and suggested that caregiver preparedness could be enhanced by targeting both cognitive/informational as well as emotional preparation. Informational preparedness refers to a clear understanding of the patient's condition, care needs and goals of care. The finding of a significant positive correlation but not complete overlap between emotional and informational preparedness for death among CNAs7 supports the notion of an emotion- versus information-based facet of preparedness.

Building on previous research related to grief and preparedness for death among family and staff caregivers,5, 7, 9, 14 we had two aims for this study: The first aim was to identify characteristics of the CNA, the resident (as per CNA's knowledge and perceptions), and the care context that may be associated with the CNA's preparedness for the resident's death. The second aim was to determine which of the associated factors relative to one another accounted for the greatest amount of variance in preparedness, and if associative patterns differ for emotional versus informational preparedness.


Recruitment and Eligibility

We recruited CNAs from three large nursing home campuses that were all part of the same health care system in Greater New York. CNAs had to have experienced the death of a resident for whom they were a primary CNA within approximately two months of a resident's death to be eligible. Resident deaths were tracked via electronic medical records and the resident's primary CNAs were identified by unit staff. CNAs were then approached on their units, given information about the study and asked whether they were interested in participating. Of the 824 eligible CNAs, 219 were approached about participating in the study. We could not reach the remaining 605 CNAs within the two-month time frame following resident death primarily because of limitations in research staffing. It is noted that this did not introduce a systematic selection based on the CNA characteristics. Of the 219 CNAs approached, 143 agreed to participate, 76 refused, and three did not complete the interview, yielding an overall response rate of 64%. With regard to age, gender, race/ethnicity and tenure, the participating CNAs were representative of the organization's CNA population.

Data Collection and Measures

Trained interviewers with a Bachelor's or Master's degree conducted the in-person interviews at a place and time of the participant's convenience. Interviews lasted on average 80 minutes. Prior to all interviews, written informed consent was obtained and participants received $30 for their time. Interviews did not occur during work hours. Data analyses were based on a selection of measures from the research interview pertaining to the focus of this paper.

Preparedness for death was assessed with two questions based on prior work examining family caregivers' preparedness for death:9,13,14 To what extent were you prepared for the patient's death mentally or emotionally? and To what extent were you prepared for the patient's death in terms of the information you had about his/her state/your understanding of the situation? Participants scored each of the two items on a 4-point Likert scale ranging from 1= not at all to 4=very.

Sociodemographic characteristics examined included age, gender, education, marital status, and race/ethnicity of the CNAs. Length of time of care provided to the deceased resident was addressed with the question: For how many months/years did you provide care to this resident?

Recent Other Deaths

CNAs were asked whether (1) or not (0) they experienced any additional deaths (at work or in their personal lives) within the past two months.

Personal End-of-Life (EOL) Care Preference

Participants were asked to indicate their agreement with a selection of EOL care preferences commonly reported and sensitive to ethnic/cultural differences.15,16 Items were assessed on a 5-point Likert scale ranging from 1= disagree a lot to 5= agree a lot. The two items most relevant as indicators of the CNAs' palliative care orientation (I want to use all possible treatment options no matter what the chance of recovery, Using pain medication is very important to me) were selected for the present analysis.

Perception of Resident at EOL

CNAs were asked to rate their perception of the resident's pain as well as the resident's awareness of dying during the last weeks of his/her life on a Likert scale ranging from 1= not at all to 4= very much.

Knowledge of Resident's and/or Family's EOL Care Preferences

CNAs' knowledge of the resident's and/or family's wishes concerning EOL care and treatment was addressed with a series of open-ended questions: Were you aware of the preferences for end-of-life care that (resident) might have had? Were you aware of the preferences for end-of-life care that (resident's) family might have had? How did you feel about the care-related decisions that (primary contact) made? Because the majority of CNAs reported not knowing about resident/family EOL care preferences or care-related decisions, responses were coded based on whether (1) or not (0) a CNA indicated having any knowledge of either aspect.

Support from Supervisor and Coworkers

CNAs were asked whether (1) or not (0) they turned to their supervisor and/or coworkers for support during the last weeks before the resident's death.

Hospice Care

CNAs were asked whether (1) or not (0) the resident had been on hospice. If so, open-ended questions were asked to further describe their experience: How were you affected by working alongside the hospice team? What were some of the positive or negative aspects for you in this situation? Responses were coded based on whether (1) or not (0) the CNA's experience with the hospice team was positive. Negative experiences with hospice were not reported; thus no such code was assigned.

Coding of Qualitative Data

Responses to the open-ended questions were written down verbatim, typed into a Word document, and imported into the computer software Atlas/ti.17 A coding system for these narrative data was developed using a qualitative analytical approach that consisted of examining data for recurrent themes, and defining codes that reflect these patterns (“clustering,”18 or “open coding”19). Interrater agreement in using this set of codes with the narrative data was adequate, with kappa coefficients consistently ranging from 0.75 to 1 (average kappa = 0.92).

To assess the data in terms of frequencies and percentages, the coded data were imported into SPSS (SPSS Inc., Chicago, IL) as categorical variables and as such included in bivariate and multivariate analyses.

Statistical Analysis

Associations of preparedness outcomes and continuous variables were evaluated with bivariate correlations, and associations with categorical variables were examined with analyses of variance (ANOVAs). Multiple regression analysis was used to assess the relative role of associated characteristics for emotional and informational preparedness. Because of the limited sample size, we were parsimonious in our selection of correlates, focusing on those characteristics that had shown marginal or significant bivariate associations with at least one of the two preparedness outcomes.


Descriptives of sample characteristics and major study variables are displayed in Table 1. Reflective of the larger population of CNAs in Greater New York, CNAs were mostly minority women. Most were high school graduates or had at least some college. A little over half the sample was married or living as married. On average, CNAs had been in the profession for 15 years and cared for the resident who died for over three years. More than half of the CNAs had additionally experienced other resident or personal deaths in the past two months. The sample consisted mostly of day and evening shift CNAs in nearly equal proportions, and a little over one-tenth of night shift CNAs. Only about a third reported having some knowledge of the resident's or resident's family's EOL care preferences or care-related decisions. Over half of the CNAs turned to coworkers for support during the last weeks of the resident's life; however, only a small number turned to their supervisor. Supervisor support, therefore, was not included in subsequent analyses. More than one-third of the CNAs indicated that the resident they cared for received hospice care, and of those, the majority reported positive perceptions of hospice.

Table 1
Descriptive Information on Sample Characteristics and Study Variables (N=140)

To address our first aim, we examined associations between CNA, resident, and care context characteristics with emotional as well as informational preparedness (Table 2). There were consistently no differences in preparedness based on shift or facility; these results, therefore, are not shown.

Table 2
Bivariate Associations of CNA Characteristics, Resident Status, and Care Context With Emotional and Informational Preparedness (N=140)

Greater emotional preparedness was reported by CNAs who were older, had a longer tenure in the job, perceived the resident as in pain, and perceived that the resident was aware he/she was dying. Greater emotional preparedness also was found for CNAs who turned to their coworkers for support in the last weeks of the resident's life, who had a resident die on hospice and viewed the hospice involvement as positive. Endorsing the personal EOL care preference to want all possible treatment regardless of the chances for recovery was associated with less emotional preparedness. In contrast, CNAs' race/ethnicity, assigned nursing home shift, length of time having cared for the resident, additional recent deaths, endorsing the personal EOL care preference that pain medication is very important, and knowledge of the resident's/family's EOL care preferences and care-related decisions were not significantly associated with emotional preparedness.

Similar characteristics showed analogous significant associations with informational preparedness, in particular longer job tenure, perceiving the resident as in pain, perceiving the resident is aware of dying, turning to coworkers for support, and resident being on hospice. There were a few exceptions to these similarities in associative patterns. Length of time caring for the resident showed a positive significant association, suggesting that length of time assigned to the resident enhanced the likelihood of gaining greater informational but not emotional preparedness. The personal EOL care preferences of wanting all possible treatments only showed a marginal negative association, suggesting a lesser role of this variable in informational preparedness. On the other hand, knowledge of resident/family's EOL care preferences and care decisions, which had no association with emotional preparedness, showed marginal positive associations with informational preparedness.

Our second aim, to determine the relative influence of associated characteristics on explaining variance in emotional and informational preparedness, was examined with multiple regression analysis. Because resident on hospice and hospice viewed positively were too highly intercorrelated to coexist as predictors in one regression, we decided to only include the evaluations of hospice rather than whether the resident was on hospice or not, as the former yielded more information about the experience of the CNA.

Findings are displayed in Table 3. The regression model for emotional preparedness explained 37% of the total variance, with significant effects for CNA age, perception of the resident being in pain, CNA endorsing an EOL care preference of wanting all possible treatment, and positive perceptions of hospice. Perceiving the resident as being in pain and viewing hospice as positive were associated with greater emotional preparedness, whereas preferring all possible treatment for oneself at the EOL was associated with less emotional preparedness.

Table 3
Regression Analysis for Emotional and Informational Preparedness (N = 140)

The regression model for informational preparedness explained 29% of the total variance, with significant effects for perception of the resident as being in pain and viewing hospice as positive, and a marginal effect for endorsing an EOL care preference of wanting all possible treatment. Knowledge about resident/family's EOL care preferences and care-related decisions did not yield significant effects in the multivariate context, suggesting a relatively lesser role of these compared to the other characteristics in the equation.


This study constitutes a first step towards developing an understanding of what is related to preparedness for resident death among front-line staff in nursing homes. Study findings are consistent with prior research on factors related to preparedness among family caregivers12 in demonstrating significant associations for personal characteristics of the CNA as well as attributes related to the resident (e.g., resident observed as being in pain). Findings go beyond previous work by also pointing to important aspects of the care context, such as coworker support and hospice involvement, and by providing insight into both similar and differential associative patterns for emotional versus informational preparedness for resident death.

CNAs who experienced their resident as in pain or aware of dying were likely to report both greater emotional and informational preparedness. Specifically, observing the resident as being in pain seemed to be most strongly related to higher preparedness levels. As optimal pain management is a core goal of palliative care efforts, and insufficient pain management in nursing homes is a widely recognized problem, pain recognition is an important topic in the field of long-term care.20,21 CNAs not only have the most daily contact with residents and often function as their advocates, alerting other staff members to potentially unmet needs and status changes,22 but pain is also often triggered during provision of daily care.23 Thus, CNAs who provide this care are best positioned to observe signs of distress, and should play an important role in the monitoring of pain. There is also evidence that CNA-generated pain assessment for cognitively impaired nursing home residents can be a more sensitive measure than the Minimum Data Set (MDS) assessments typically completed by nurses.24 Whether those who are generally more prepared for resident death are better equipped to also detect symptoms of pain, or whether seeing a resident suffering is what contributes to a sense of preparedness, is an open question that cannot be discerned with data from the present study. However, because CNAs' increased alertness to evidence of pain in residents as well as greater preparedness for dealing with resident death and dying appear to be desirable goals in an EOL care context, potentially benefitting the CNA and the resident, educational efforts targeting both seem particularly worthwhile.

Knowledge of resident/family EOL care preferences and care decisions were related only to greater informational preparedness, and evidenced a more minor role compared to, for example, pain. The lesser role of these variables might be a result of limited variability, because most CNAs indicated that they had no or little knowledge in either regard. Responses, therefore, were assigned dichotomized yes/no codes. Considering this aspect and the marginal association with informational preparedness, we would argue tentatively that being better informed about care plans might enhance CNAs preparedness and ability to advocate for a resident's needs.

The two characteristics of the care context, greater support from coworkers, and positive perceptions of hospice, were associated with higher levels of both facets of preparedness. When considered together, positive perceptions of hospice seemed to be the more influential variable relative to coworker support. As hospice is increasingly used in the nursing home context, an important question is what challenges may arise when care responsibilities are shared by hospice and nursing home teams.25 It was reassuring to find that CNAs in the present study did not seem to have negative encounters with hospice and that most of those who had a resident on hospice reported positive experiences with hospice staff. Yet, there was also a substantial minority of CNAs with a resident on hospice who seemed entirely unaffected by the presence of hospice, likely because they had limited or no interactions with them. Because an integral part of hospice goals is to also address the needs of the care team, one could argue that this should not be the case. Evidence that a positive experience with hospice is associated with greater emotional and informational preparedness for death should serve as an additional motivator to devote more attention to interactions between front-line nursing home staff and hospice teams. Here again, there is no way to discern the directionality of the association. It is possible that those who feel generally more prepared for death are also more open to the notion of hospice care and thus are more likely to have positive experiences with them. Previous work has found evidence that staff members' beliefs about hospice services can affect hospice delivery in nursing homes.26

Whereas the discussed aspects of the care context and of CNAs knowledge of resident status and care preference/plan can be targeted through educational efforts or implemented practice changes, some of the personal CNA characteristics associated with preparedness are more difficult to address, such as age, tenure in the CAN profession and length of time caring for the resident. Creating more informational exchange opportunities among nursing staff might give more senior CNAs the chance to share their knowledge and expertise with those who have not had the benefit of greater experience. Moreover, the finding that a personal EOL care preference of wanting all possible treatments regardless of chance for recovery was associated with less emotional preparedness might indicate an underlying view of EOL that blocks a more accepting view of death as a natural part of life. Such a view could make a state of greater emotional preparedness less accessible. Exploring and discussing personal EOL care views of staff and how they might influence staff perceptions in the work context might be beneficial not only with respect to preparedness for resident death and dying but also to ensure that such perceptions do not lead to responses that are less than optimal from a palliative care standpoint (e.g., underreporting of pain). Finally, more targeted education for CNAs about the benefits or lack of benefits that pursuing certain treatment options in the frail elderly (e.g., feeding tube placement or do-not-resuscitate order) may help CNAs to better understand the range of EOL treatment choices as well as recognize when resident death may be imminent.

Several potential limitations of our research deserve mention. Although we assessed CNAs' subjective perception of the resident at the EOL, we did not collect further medical information about the resident's status. Future research might draw on objective medical data to get a more detailed understanding of a resident's condition at the EOL. Another limitation is that we asked CNAs retrospectively to report their preparedness. Such retrospective assessments of preparedness for death can be biased by the person's adjustment to the loss or other current events. Additionally, we assessed each type of preparedness with only a single item. Future studies should develop a reliable multi-dimensional assessment of preparedness, both to increase our understanding of the relative importance of different facets of preparation and to yield more detailed information on preparedness that can guide the design of supportive interventions for EOL caregivers. Finally, the cross-sectional nature of the data did not allow us to discern the directionality of the identified association. Thus, we cannot determine if the characteristics associated with preparedness were aspects that influenced preparedness or if CNAs' levels of preparedness affected how they reported these characteristics, or if both was the case. However, regardless of which causes which, both enhancing CNAs' capacity for pain recognition and increasing their preparedness for resident death are efforts that point in the same direction –to strengthen the CNA's position and ability to provide the best possible EOL care. Similarly, having more information about resident status and more exchange opportunities within the care team around EOL-related challenges may help CNAs feel more prepared for resident death. Greater preparedness may in turn facilitate communication exchanges between nursing staff, and effective communication between CNAs and both coworkers and supervisors has been associated with better EOL assessments and care delivery.27 Thus, these interventions would have the potential to improve not only the experience of the CNA but also the quality of life of the resident.


This study was supported by a grant from the National Institute on Aging (1 R03 AG034076; Kathrin Boerner, Principal Investigator).


Disclosures: None of the authors has any conflicts of interest.

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.


1. Chichin ER, Burack OR, Olson E, Likourezos A. End-of-life ethics and the nursing assistance. New York, NY: Springer Publishing Company; 2000.
2. Moss MS, Moss SZ, Rubinstein RL, Black HK. Metaphor of “family” in staff communication about dying and death. J Gerontol B Psychol Sci Soc Sci. 2003;58:290–296. [PubMed]
3. Monahan DJ, McCarthy S. Nursing home employment: the nurse's aide perspective. J Geronto. 1992;18:13–16. [PubMed]
4. Oulton JA. The global nursing shortage: an overview of issues and actions. Policy Polit Nurs Pract. 2006;7:34–39. [PubMed]
5. 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. [PubMed]
6. Anderson KA, Ewen HH. Death in the nursing home: an examination of grief and well-being in nursing assistants. Res Gerontol Nurs. 2011;4:87–94. [PubMed]
7. Boerner K, Burack OR, Jopp DS, Mock SE. Grief after patient death: direct care staff in nursing homes and homecare. J Pain Symptom Manage. 2014;49:214–222. [PMC free article] [PubMed]
8. McCarthy MC, Clarke NE, Ting CL, et al. Prevalence and predictors of parental grief and depression after the death of a child from cancer. J Palliat Med. 2010;13:1321–1326. [PubMed]
9. Hebert RS, Dang Q, Schulz R. Preparedness for the death of a loved one and mental health in bereaved caregivers of patients with dementia: findings from the REACH study. J Palliat Care Med. 2006;9:683–693. [PubMed]
10. Valdimarsdόttir U, Kreicbergs U, Hauksdόttir A, et al. Parents' intellectual and emotional awareness of their child's impending death to cancer: a population based long-term follow-up study. Lancet. 2007;8:706–714. [PubMed]
11. Hauksdόttir A, Steineck G, Furst CJ, Valdimarsdόttir U. Long-term harm of low preparedness for a wife's death from cancer – a population-based study of widowers 4-5 years after the loss. Am J Epidemiol. 2010;172:389–396. [PubMed]
12. Hauksdόttir A, Valdimarsdόttir U, Furst CJ, Onelov E, Steineck G. Health care-related predictors of husbands' preparedness for the death of a wife to cancer – a population-based follow-up. Ann Oncol. 2010;21:354–361. [PubMed]
13. Hebert RS, Prigerson H, Schulz R, Arnold RM. Preparing caregivers for the death of a loved one: a theoretical framework and suggestions for future research. J Palliat Care Med. 2006;9:1164–1171. [PubMed]
14. Hebert RS, Schulz R, Copeland VC, Arnold RM. Preparing family caregivers for death and bereavement. Insights from caregivers from terminally ill patients. J Pain Symptom Manage. 2009;37:3–12. [PubMed]
15. Duffy SA, Jackson FC, Schim SM, Ronis DL, Fowler KE. Racial/ethnic preferences, sex preferences, and perceived discrimination related to end-of-life care. J Am Geriatr Soc. 2006;54:150–157. [PubMed]
16. Steinhauser KE, Chistakis NA, Clipp EC, et al. Preparing for the end of life: preferences of patients, families, physicians, and other care providers. J Pain Symptom Manage. 2001;22:727–737. [PubMed]
17. Muhr T. Atlas/ti for windows: Visual Qualitative data analysis, management & model building. Berlin: Scientific Software Development; 1997. computer software.
18. Miles MB, Huberman AB. An expanded sourcebook: Qualitative data analysis. 2nd. London: Sage Publications; 1994.
19. Glaser B, Strauss A. The discovery of grounded theory. Chicago: Aldine; 1967.
20. Ferrell BA. The management of pain in long-term care. Clin J Pain. 2004;20:240–243. [PubMed]
21. Miller SC, Vincent M, Joan T. Hospice enrollment and pain assessment and management in nursing homes. J Pain Symptom Manage. 2003;26:791–799. [PubMed]
22. Liu JYW. Exploring nursing assistants' roles in the process of pain management for cognitively impaired nursing home residents: a qualitative study. J Adv Nurs. 2013;70:1065–1077. [PubMed]
23. Lin PC, Lin LC, Shyu YIL, Hua MS. Predictors of pain in nursing home residents with dementia: a cross-sectional study. J Clin Nurs. 2010;20:1849–1857. [PubMed]
24. Fisher SE, Burgio LD, Thorn BE, et al. Pain assessment and management in cognitively impaired nursing home residents: association of certified nursing assistant pain report, minimum data set pain report and analgesic medication use. J Am Geriatr Soc. 2002;50:152–156. [PubMed]
25. Stevenson DG, Bramson JS. Hospice care in the nursing home setting: a review of the literature. J Pain Symptom Manage. 2009;38:440–451. [PubMed]
26. Welch LC, Miller SC, Martin EW, Nanda A. Referral and timing of referral to hospice care in nursing homes: the significant role of staff members. Gerontologist. 2008;48:477–484. [PubMed]
27. Zheng NT, Temkin-Greener H. End-of-life care in nursing homes: the importance of CNA staff communication. J Am Med Dir Assoc. 2010;11:494–499. [PMC free article] [PubMed]