Journal of Gerontological Nursing

PATIENT DEATH IN A LONG-TERM CARE HOSPITAL: A STUDY OF THE EFFECT ON NURSING STAFF

Dan W Harper, PHD; Lyne D Chartrand, RN, BSCN; Stephen F Johnston, BA

Abstract

Nurses in many settings are charged with caring for the dying patient, and may be exposed to death on a regular basis. Although counted among the stressors associated with the profession (Chiriboga, Jenkins, & Bailey, 1983; Munley, 1985; Thomas, 1983), the effect of repeated exposure to death has received surprisingly little attention in the nursing research literature.

In examining 20 years of nursing literature, only four studies systematically investigating attitudes of groups of nurses were located. The most extensive of these studies is presented in a chapter in the volume entitled Dealing with Death and Dying: Nursing Skillbook Series (Chaney, 1976). Over 15,000 responses to a 70-item questionnaire designed to determine nurses' attitudes toward death and the determinants of those attitudes are presented. Unfortunately, little information is provided concerning the data gathering method; it appears that data were gathered from a number of sources over a prolonged period. The results are presented in a descriptive univariate fashion so that no attempt was made to interrelate answers to various questions. When asked about the types of patients whose deaths would elicit feelings that would make coping most difficult, nurses identified children and adolescents. The authors reported that the older the patient, the easier nurses find it to deal with the emotional stress of that patient's death. It should be noted that the population surveyed probably had limited experience with geriatric patients. The authors conclude that, among their respondents, most feel discouraged, depressed, and angry at least occasionally when caring for an incurable, terminally ill patient. They also repon that most nurses have, at least occasionally, found care of the dying to be rewarding.

In a multi-site study of hospice nurses, Chiriboga, Jenkins and Bailey (1983) reported that the more exposure to death that nurses had prior to working in this setting, the more stress they tended to experience. They also suggested that nurses characterized by certain coping strategies (e.g., rational rather than emotional) fared better than others. Surprisingly, however, degree of social support did not contribute significantly to predicting who was more or less stressed by their job.

Prevalence of grief in nursing staff due to patient loss in skilled nursing facilities and in a general hospital, was investigated by Lerea and LiMauro (1982). While virtually aU general hospital staff reported remembering grief due to patient death, only two-thirds of skilled nursing facility staff did. The authors speculate that death is more unexpected and that the patients are generally younger in the general hospital setting, accounting for the greater prevalence of grief among staff.

One hundred and fifty-seven hospice caregivers were surveyed by Adams, Hershatter and Moritz (1991) concerning the effects of repeated exposure to death, using a 92-item scale of their own design-The Impact of Patient Death Questionnaire (IPTXJ). Although the authors, using multivariate as well as univariate techniques, claimed to have demonstrated an "accumulated loss phenomenon," they did not relate any predictor variables (including "number of deaths") to measures of the effect of patient death. They did, however, report that the professional status of the caregiver was associated with degree of investment in the job. Registered nurses tend to "expect too much of themselves and are overwhelmed with patients' needs" while licensed practical nurses are "more positive about their work, and feel more confident about nursing and medical leadership" (Adams, Hershatter, & Moritz, 1991, p. 30). They also documented coping strategies that staff claimed were helpful in dealing with patient death. These strategies included socializing, involvement in religion, and the use of self-awareness and introspection.

PURPOSE

The studies mentioned above describe nursing staff in hospice, nursing home, and general…

Nurses in many settings are charged with caring for the dying patient, and may be exposed to death on a regular basis. Although counted among the stressors associated with the profession (Chiriboga, Jenkins, & Bailey, 1983; Munley, 1985; Thomas, 1983), the effect of repeated exposure to death has received surprisingly little attention in the nursing research literature.

In examining 20 years of nursing literature, only four studies systematically investigating attitudes of groups of nurses were located. The most extensive of these studies is presented in a chapter in the volume entitled Dealing with Death and Dying: Nursing Skillbook Series (Chaney, 1976). Over 15,000 responses to a 70-item questionnaire designed to determine nurses' attitudes toward death and the determinants of those attitudes are presented. Unfortunately, little information is provided concerning the data gathering method; it appears that data were gathered from a number of sources over a prolonged period. The results are presented in a descriptive univariate fashion so that no attempt was made to interrelate answers to various questions. When asked about the types of patients whose deaths would elicit feelings that would make coping most difficult, nurses identified children and adolescents. The authors reported that the older the patient, the easier nurses find it to deal with the emotional stress of that patient's death. It should be noted that the population surveyed probably had limited experience with geriatric patients. The authors conclude that, among their respondents, most feel discouraged, depressed, and angry at least occasionally when caring for an incurable, terminally ill patient. They also repon that most nurses have, at least occasionally, found care of the dying to be rewarding.

In a multi-site study of hospice nurses, Chiriboga, Jenkins and Bailey (1983) reported that the more exposure to death that nurses had prior to working in this setting, the more stress they tended to experience. They also suggested that nurses characterized by certain coping strategies (e.g., rational rather than emotional) fared better than others. Surprisingly, however, degree of social support did not contribute significantly to predicting who was more or less stressed by their job.

Prevalence of grief in nursing staff due to patient loss in skilled nursing facilities and in a general hospital, was investigated by Lerea and LiMauro (1982). While virtually aU general hospital staff reported remembering grief due to patient death, only two-thirds of skilled nursing facility staff did. The authors speculate that death is more unexpected and that the patients are generally younger in the general hospital setting, accounting for the greater prevalence of grief among staff.

One hundred and fifty-seven hospice caregivers were surveyed by Adams, Hershatter and Moritz (1991) concerning the effects of repeated exposure to death, using a 92-item scale of their own design-The Impact of Patient Death Questionnaire (IPTXJ). Although the authors, using multivariate as well as univariate techniques, claimed to have demonstrated an "accumulated loss phenomenon," they did not relate any predictor variables (including "number of deaths") to measures of the effect of patient death. They did, however, report that the professional status of the caregiver was associated with degree of investment in the job. Registered nurses tend to "expect too much of themselves and are overwhelmed with patients' needs" while licensed practical nurses are "more positive about their work, and feel more confident about nursing and medical leadership" (Adams, Hershatter, & Moritz, 1991, p. 30). They also documented coping strategies that staff claimed were helpful in dealing with patient death. These strategies included socializing, involvement in religion, and the use of self-awareness and introspection.

PURPOSE

The studies mentioned above describe nursing staff in hospice, nursing home, and general hospital settings. The primary purpose of the present study was to identify variables associated with the nurses who are most negatively affected by patient deaths in a continuing care and rehabilitation hospital. Secondary purposes included the identification of the ways in which staff are negatively affected, and the provision of a description of the ways in which staff cope with patient death in this setting.

METHOD

Subjects

Nursing staff (registered nurses, registered practical nurses, and nursing aides /orderlies) at a 516bed Canadian continuing care and rehabilitation hospital were invited to participate in this study. One hundred and forty staff (representing one-third of all the hospital's regular nursing staff) volunteered to anonymously fill out a nine-page questionnaire on their own time. Fourteen respondents had not worked on their unit for at least one year and so were unable to answer several important questions. The convenience sample used for analysis, then, consisted of the 126 respondents (described in Table 1).

Subjects represented the 13 units in the hospital where continuing care patients made up the majority of the population. Units experienced an average of 13 deaths each during the previous year.

Measures

Subjects were asked to provide background information about their nursing unit, age, gender, marital status, position, length of time worked at this facility, number of hours worked per week, and whether or not they had experienced the death of a close family member or dear friend in the past 12 months. The Multidimensional Scale of Perceived Social Support (Dahlem, Zimet, & Walker, 1991; Zimet, Dahlem, Zimet, & Farley, 1988) was administered to subjects as a way to gauge sources of potential social support.

The Impact of Patient Death Questionnaire (IPDQ) (Adams, Hershatter, & Moritz, 1991) was administered to all participating nursing staff. The 92-item EPDQ was specifically designed to yield information concerning the effect of patient death on staff, and to identify predictors of such effects. Items assess, in Likert-scale format, nursing staff attitudes and experiences with death, the ways in which they are affected, and the working and home environments of respondents. (The scale is reproduced in the original publication: Adams, Hershatter, & Moritz, 1991). For the present study, four items were added to the scale in order to reflect topics thought to be important from clinical experience, but not represented in the original IPDQ. Specifically, respondents were asked to agree or disagree with the statements: (1) "I have a greater sense of loss when a patient dies unexpectedly," (2) "It is important to me to offer patients a peaceful death when the time comes," (3) "I am upset when a patient dies," and (4) "I find my work more difficult when a number of patients die in succession." The expanded IPDQ is subsequently referred to as "IPDQ+."

Table

TABLE 1Demographic Characteristics of Subjects

TABLE 1

Demographic Characteristics of Subjects

Operational Definition of "Affected by Patient Death"

A grouping variable was created in order to classify subjects as "Affected" or "Not Affected" by patients deaths in the last 12 months. If subjects indicated that they were "Moderately" to "Extremely" negatively affected by patients deaths (Table 3, Question 1), or if they checked any of the five suggested ways of being affected by patient deaths, for example, lost time at work (Table 2, Question 2), they were considered "Affected" (Table 2). Using this criterion, approximately 42% of respondents in our sample are classified "Affected" by patient deaths in the last 12 months.

Table

TABLE 2Number (%) of Subjects Affected by Patient Deaths in the Post 12 Months

TABLE 2

Number (%) of Subjects Affected by Patient Deaths in the Post 12 Months

RESULTS

"Affected by Patient Death" versus "Not Affected by Patient Death"

For the most part there were no differences between "Affected" subjects (n=53) and "Not Affected" subjects (n=73) on any of the demographic or experiential variables assessed, including the actual number of patient deaths that staff were exposed to, and their perceived social support (Table 1). Two exceptions were: (a) length of time worked at this facility - subjects categorized as "Not Affected" by patient deaths in the last 12 months had worked, on average, approximately three years less at this hospital than "Affected" subjects, and (b) death of a friend or family member in the past 12 months - the proportion of subjects in the "Not Affected" group who had lost someone in the last year was approximately 20% lower than the "Affected" group (28.6% versus 47.2%).

Univariate analyses showed that the "Affected" and "Unaffected" groups were statistically significant in their answers to 22 IPDQ+ items. Seven of these items are related to the effect of losing patients ("Loss" variables) and 15 items concern attitudes toward work and the consequences of work-related stress ("Predictor" variables) (Table 3).

Logistic regression was used to identify variables that might predict being "Affected" by patient deaths during the last 12-month period. A forward stepwise procedure with likelihood-ratio test was utilized so that at each step, the variable with the lowest significance level for the score statistic was entered into the model. This procedure allows the predetermination of outcome groups, that is, the categorization of subjects as "Affected" or "Not Affected" by patient deaths.

The 15 IPDQ+ "predictor variables" and the two background variables which significantly differentiated the "Affected" and "Not Affected" groups were entered into the logistic regression analysis.

Four variables emerged from the logistic regression analysis as predictors of being affected by patient deaths during the last year, and they are all items from the EPDQ: (63) I feel work-related stress affects my family relationships, (17) I feel rewarded and acknowledged for my work by my co-workers, (29) The younger the patient, the harder I find it to care for him/her, and (36) The one-to-one relationship between a patient and a nurse is the most important aspect of nursing. Subjects "Affected" by patient death rated themselves as more in agreement with items 63, 29 and 36 than subjects who were considered "Not Affected," and less in agreement with item 17. Overall, approximately 77% of subjects could be correctly classified as being "Affected" or "Not Affected" by patient deaths, with this regression equation (Table 4).

Table

TABLE 3IPDQ+ Items which Significantly Differentiate Nursing Stoff who are Negatively Affected by Patient Death from those who are not (p<.05 by 2-tailed t test)

TABLE 3

IPDQ+ Items which Significantly Differentiate Nursing Stoff who are Negatively Affected by Patient Death from those who are not (p<.05 by 2-tailed t test)

Table

TABLE 3IPDQ+ Items which Significantly Differentiate Nursing Staff who are Negatively Affected by Patient Death tram those who are not (p<.05 by 2-tailed t-test)

TABLE 3

IPDQ+ Items which Significantly Differentiate Nursing Staff who are Negatively Affected by Patient Death tram those who are not (p<.05 by 2-tailed t-test)

Coping Strategies

Ninety-five subjects responded to the open-ended question, "What helps you most in dealing with patient deaths?" by spontaneously providing coping strategies. Three raters independently sorted the actual strategies given by respondents into seven categories designed to capture their basic elements. This Q-sort procedure resulted in levels of agreement that Fleiss (1981) terms "good" (69% to 78%, Kappa=.65 to .76, p<.05). The major coping strategies are presented in Table 5.

DISCUSSION

The present study found that many nursing staff who work in a continuing care facility are negatively affected by the deaths of their patients. This study was not an attempt to determine prevalence; however, it did reveal that being affected by patient death is common in this setting. Our finding that a large proportion of nursing staff (approximately 42%) were negatively affected by patient death, has potentially serious implications for the physical and emotional health of the caretakers who tend to people in the last stages of life.

Identifying specific ways in which staff are affected by patient death has proven to be problematic. "Low morale" was the only option chosen by a substantial proportion (22%) of our sample, when asked directly about the consequences of losing patients. Informal feedback concerning this questionnaire suggests that some nurses are hesitant to describe the ways in which patient death has an impact on their Uves. It could be speculated that because death is not an uncommon event in a continuing care facility, there is a higher expectation for staff to deal with the loss of patients in a more "routine" fashion.

Table

TABLE 4logistic Regression Model Predicting Group* Using IT Variables** as Predictors

TABLE 4

logistic Regression Model Predicting Group* Using IT Variables** as Predictors

The results of the univariate analyses give limited support to the "accumulated loss phenomenon" postulated by Adams, Hershatter and Moritz (1991). The present study found that nursing staff were more likely to be "affected" if they had worked in their setting longer and if they had suffered a personal loss in the last 12 months. The actual number of deaths and the perceived "rate" of deaths, however, were not predictive of being "affected." Note also that the two "accumulated loss" variables found to differentiate between groups on univariate analysis did not survive as successful predictors when entered into the logistic regression analysis.

The results of the logistic regression analysis (Table 4) paint an interesting picture of "affected" nurses. They are more likely to take work stress home, have a harder time caring for "younger" patients, and are more likely to base their nursing care on a personal relationship with the patient, than nurses who are "not affected." They also feel less "acknowledged" by their colleagues. Taken together, this may indicate the nursing style of staff who are more highly sensitive and involved with patients, and who are less able to distance themselves from work.

The most often identified strategy for coping with the loss of a patient was to concentrate on the positive impact of good nursing during that person's death. In this way the nursing staff could take comfort in the knowledge that they attempted to facilitate a peaceful and /or comfortable death. Coping strategies did not differentiate between subjects in the "affected" and "not affected" by patient death groups; however, the difference in proportions of subjects suggesting "sharing feelings" as a means of dealing with patient deaths, approached significance (p=.063). A greater percentage of subjects in the "affected" group conveyed the notion that sharing feelings and talking to others is helpful in dealing with the loss of patients. This is consistent with the IPDQ+ items that suggested that "affected" subjects were more likely to discuss work-related issues with friends and family, even though work-related stress was seen as having a detrimental effect on family relationships. Unfortunately for "affected" subjects, their need to share the negative experiences of work may not be satisfied by talking to co-workers, who they tend to see as less supportive than do "not affected" subjects.

Table

TABLE 5Major Coping Strategies in Dealing with Patient Deaths

TABLE 5

Major Coping Strategies in Dealing with Patient Deaths

We did not attempt to assess the actual effectiveness of the coping strategies employed by the respondents, and no dear patterns emerged concerning the support mechanisms that would help them cope more effectively with the loss of their patients. Spencer (1994), however, reported that more than half of the intensive care nurses she surveyed indicated that they would like more support at work, and mis support could take the form of self-help groups, counselors, or one-to-one interactions with another nurse.

GERONTOLOGICAL NURSING IMPLICAIIONS

After surveying nursing staff in a continuing care hospital setting, and taking into account the previously reported literature, we conclude mat: a) the effect of patient loss on staff is a significant issue, b) nurses are at risk for being negatively affected by patient death the longer they have worked, or if they have recently experienced a personal loss, c) a highly involved personal orientation toward nursing care puts staff at increased risk for being negatively affected by patient death, and d) although social support was not shown to reduce susceptibility to the negative effects of patient death in the present study, a supportive work environment may aid in ameliorating the death-related stress of the job. Attempts should be made to acknowledge the personnel who are most at risk, and identify individual solutions for those significant numbers of nursing staff who are affected by patient deaths.

REFERENCES

  • Adams, J.P., Hershatter, M.J., & Moritz, D.A. (1991, May /June). Accumulated loss phenomenon among hospice caregivers. American Journal of Hospital and Palliative Care, 29-37.
  • Chaney, P.S. (1976). How do others view death. In P.S Chaney (Ed.), Dealing vrith death and dying- Nursing skillbook series. Jenkintown, PA: Eugene Jackson (Intermed Communications), pp. 149-183.
  • Chiriboga, D.A., Jenkins, G., & Bailey, J. (1983). Stress and coping among hospice nurses: Test of an analytic model. Nursing Research, 32(5), 294-299.
  • Dahlem, N. W., Zimet, G.D., & Walker, R.R. (1991). The Multidimensional Scale of Perceived Social Support: A confirmation study. Journal of Clinical Psychiatry, 47(6), 756-761.
  • Fleiss, J.L. (1981). Statistical methods for rates and proportions (2nd ed.). New York, NY: John Wiley, p. 321.
  • Lerea L.E., & LiMauro, B.F. (1982). Grief among healthcare workers: A comparative study. Journal of Gerontology, 37(5), 604-608.
  • Munley, A. (1985). Sources of hospice stress and how to cope with it. Nursing Clinics of North America, 20(2), 343-355.
  • Spencer, L. (1994). How do nurses deal with their own grief when a patient dies on an intensive care unit, and what help can be given to enable them to overcome their grief effectively? Journal of Advanced Nursing, 19, 1141-1150.
  • Thomas, V.M. (1983, January-February). Hospice nursing - Reaping the rewards, dealing with stress. Geriatric Nursing, 22-27.
  • Zimet, G.D., Dahlem, N.W., Zimet, S.G, & Farley, G.K. (1988). The Multidimensional Scale of Perceived Social Support. Journal of Personality Assessment, 52(1), 30-41.

TABLE 1

Demographic Characteristics of Subjects

TABLE 2

Number (%) of Subjects Affected by Patient Deaths in the Post 12 Months

TABLE 3

IPDQ+ Items which Significantly Differentiate Nursing Stoff who are Negatively Affected by Patient Death from those who are not (p<.05 by 2-tailed t test)

TABLE 3

IPDQ+ Items which Significantly Differentiate Nursing Staff who are Negatively Affected by Patient Death tram those who are not (p<.05 by 2-tailed t-test)

TABLE 4

logistic Regression Model Predicting Group* Using IT Variables** as Predictors

TABLE 5

Major Coping Strategies in Dealing with Patient Deaths

10.3928/0098-9134-19960801-10

Sign up to receive

Journal E-contents