Spirituality is a human characteristic

Spiritual care of families in the intensive care unit* Richard J. Wall, MD, MPH; Ruth A. Engelberg, PhD; Cynthia J. Gries, MD; Bradford Glavan, MD; J....
Author: Neil Dixon
0 downloads 2 Views 309KB Size
Spiritual care of families in the intensive care unit* Richard J. Wall, MD, MPH; Ruth A. Engelberg, PhD; Cynthia J. Gries, MD; Bradford Glavan, MD; J. Randall Curtis, MD, MPH

Objectives: There is growing recognition of the importance of spiritual care as a quality domain for critically ill patients and their families, but there is a paucity of research to guide quality improvement in this area. Our goals were to: 1) determine whether intensive care unit (ICU) family members who rate an item about their spiritual care are different from family members who skip the item or rate the item as “not applicable” and 2) identify potential determinants of higher family satisfaction with spiritual care in the ICU. Design: Cross-sectional study, using data from a cluster randomized trial aimed at improving end-of-life care in the ICU. Setting: ICUs in ten Seattle-area hospitals. Subjects: A total of 356 family members of patients dying during an ICU stay or within 24 hrs of ICU discharge. Intervention: None. Measurements and Main Results: Family members were surveyed about spiritual care in the ICU. Chart abstractors obtained clinical variables including end-of-life care processes and family conference data. The 259 of 356 family members (73%) who rated

S

pirituality is a human characteristic that allows a person to find meaning and purpose in his or her life, while providing a framework for constructing a coherent world view (1–3). When patients are critically ill in the intensive care unit (ICU), they are often unable to make their own decisions due to the severity of their illnesses, limited cognitive capabilities, and the use of treatments that make communication difficult (4). In this setting, family members must often act as surrogate decision makers, while simultaneously

*See also p. 1208. From the Harborview Medical Center, Division of Pulmonary and Critical Care, Department of Medicine, University of Washington, Seattle, WA. The authors have not disclosed any potential conflicts of interest. Supported, in part, by a grant from the National Institute of Nursing Research (R01NR05226), Bethesda, MD. For information regarding this article, E-mail: [email protected] Copyright © 2007 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/01.CCM.0000259382.36414.06

1084

their spiritual care were slightly younger than family members who did not rate this aspect of care (p ⴝ .001). Multiple regression revealed family members were more satisfied with spiritual care if a pastor or spiritual advisor was involved in the last 24 hrs of the patient’s life (p ⴝ .007). In addition, there was a strong association between satisfaction with spiritual care and satisfaction with the total ICU experience (p < .001). Ratings of spiritual care were not associated with any other demographic or clinical variables. Conclusions: These findings suggest that for patients dying in the ICU, clinicians should assess each family’s spiritual needs and consult a spiritual advisor if desired by the family. Further research is needed to develop a comprehensive approach to ICU care that meets not only physical and psychosocial but also spiritual needs of patients and their families. (Crit Care Med 2007; 35:1084–1090) KEY WORDS: critical care; family research; family satisfaction; palliative care; spirituality

experiencing uncertainty and considerable psychological stress (5). During these challenging times, spirituality may be especially relevant for family members who often view the crisis through a “spiritual lens” (6). Although numerous studies have examined the relationship between spirituality and medicine in chronically or terminally ill patients, little research has focused on the spiritual needs of family members with an adult loved one in the ICU. Studies in non-ICU settings have shown that many patients and families find religion to be the single most important factor enabling them to cope with a serious illness (7). In addition, spirituality is one of the key predictors of positive long-term adaptation for family members when a patient dies (8). Because providing high-quality care for critically ill patients also requires caring for their families (9), a comprehensive approach to ICU improvement must also consider the spiritual needs of family members. The impetus for the current study arose from observations made during a previous investigation of 1,038 critically

ill patients at seven medical centers (10). In the previous study, family members were asked to rate their satisfaction with various components of care while their loved one was in the ICU. Although the overall response rate for questions was 94%, only two thirds of family members provided a rating for an item asking about satisfaction with spiritual care in the ICU. In addition, among those rating this spiritual care item, fewer than half scored their satisfaction as “excellent.” Given the importance of spiritual care for both patients and families in the ICU, we thought these findings warranted further investigation. We conducted an exploratory analysis to identify potential determinants of family satisfaction with spiritual care in the ICU. Based on our previous observations, we examined two specific questions. First, we identified whether ICU family members who rate their spiritual care are different from family members who do not rate their spiritual care (i.e., either skip or rate the item as “not applicable”). Second, among ICU family members rating their spiritual care, we examined Crit Care Med 2007 Vol. 35, No. 4

whether certain factors were associated with higher satisfaction scores. We believe that understanding these factors is an important first step for future research on this emerging topic.

MATERIALS AND METHODS Populations and Settings We used data from an ongoing cluster randomized trial aimed at improving end-of-life (EOL) care for ICU patients in 10 Seattle-area hospitals. Data are based on baseline assessments at ten of the hospitals for which chart abstraction and questionnaire data were available at the time of this analysis. The ten hospitals included a university-affiliated county hospital (65 ICU beds), two community-based teaching hospitals (44 and 45 ICU beds), and seven community-based, nonteaching hospitals (ranging in size from 15 to 32 ICU beds). The number of patients enrolled at the various sites ranged from 27 to 164. The study protocol was approved by institutional review boards at the University of Washington and each participating hospital. All patients dying in the ICU or within 24 hrs of discharge from the ICU were eligible for study. Potential patients were identified using ICU admission/discharge logs, and enrollment took place between July 2004 and November 2005. Family members were identified using two approaches. At one site, next-of-kin were identified using the hospital’s electronic medical record. At the remaining sites, survey packets were addressed to the “family of [patient name].” One to 2 months after the patient’s death, we mailed a survey packet with a consent form, cover letter expressing condolences and explaining the study, and a $10 incentive. A thank-you/reminder postcard was sent 2 wks later. If the questionnaire was not returned within the next 3 wks, we mailed a final packet with cover letter, consent form, and survey. Each family member self-reported his or her demographic characteristics, including age, sex, ethnicity, education level, relationship to patient, and whether he or she lived with the patient. The ten study sites have slightly different spiritual resources available for ICU patients and families. All hospitals have an office for a chaplain or spiritual advisor. At seven sites, a spiritual advisor performs daily rounds on all ICU patients. At two sites, a spiritual advisor rounds in the ICU two to three times each week. At one site, a spiritual advisor visits ICU patients and families on request only. At seven sites, a designated chaplain or spiritual advisor is on-call 24 hrs/day. At the remaining sites, a chaplain or spiritual advisor is only on-call during normal working hours.

Crit Care Med 2007 Vol. 35, No. 4

Data Sources Family Satisfaction in the Intensive Care Unit Questionnaire. Family members were asked to rate their satisfaction with spiritual care in the ICU as part of a 34-item survey, the Family Satisfaction in the Intensive Care Unit (FS-ICU) questionnaire. The FS-ICU is a reliable and valid questionnaire designed to measure family satisfaction with ICU care, and details of its initial development and validation have been previously described (11, 12). The instrument has been used in several Canadian and U.S. studies (13–17), and recently, we developed a validated scoring method for the FS-ICU (10). The full survey (with scoring instructions) is available on-line (18). A single FS-ICU item asked family members to rate their satisfaction with “how well the ICU staff met your spiritual/religious needs.” The item used a 5-point response scale (poor, fair, good, very good, excellent) or respondents could select “N/A,” not applicable. Later in the survey, a second item asked family members “how well did the ICU chaplain assist and support you” with the same response scale. Because the response characteristics and findings for both items were similar (data not shown), and given the broader focus on the first question, we report our results using the former. Chart Abstraction. Patient medical records were reviewed by trained chart abstractors using a standardized chart abstraction protocol. All abstractors underwent ⱖ80 hrs of formal training including instruction on the study protocol, guided practice charts, and independent chart review with reconciliation by the trainer. Abstractors were required to reach 90% agreement with the trainer before being

able to code independently. Throughout the study, 5% of charts were randomly coreviewed to ensure ⬎95% agreement on the abstracted data elements. Abstractors collected data on patient demographics, clinical variables, EOL care processes, and family conference data. Because this is an ongoing study and our analyses required both chart abstraction and family questionnaire data, we only analyzed subjects for whom both data sets were available.

Statistical Analysis Our goal was to identify potential determinants of family satisfaction with spiritual care in the ICU. Based on our specific objectives, we used a two-step approach: 1) we compared family members who rated their spiritual care with those who did not rate this item (i.e., item skipped or not applicable), and 2) among those rating their spiritual care, we compared those who scored this item excellent against respondents who rated their spiritual care as something less than excellent. These comparisons were performed for patient characteristics, family characteristics, and EOL care processes. This specific analytic approach and all variables for comparison were designated a priori. The decision to compare excellent scores with all other response categories was based on a previous publication from the FSICU developers (13). Descriptive analyses were used to summarize spiritual item responses, patient/family characteristics, and EOL care processes. Univariate comparisons were performed using Student’s t-test and chi-square or Fisher exact

Figure 1. Family satisfaction with spiritual care in the intensive care unit (n ⫽ 356).

1085

test for continuous and categorical variables, respectively. Given the skewed nature of family satisfaction data, FS-ICU scores were compared using the nonparametric Mann-Whitney U test. The total score on the FS-ICU was computed using our previously validated scoring approach (10). This method combines items from the two FS-ICU domains—satisfaction with care and satisfaction with decision making. This scoring method uses 24 items but does not include the spiritual care or chaplain items. As described elsewhere (10), FS-ICU scores are calculated by averaging available items, provided the respondent answers ⱖ18 of the items (70%). Scores are linearly transformed to range from 0 to 100 and oriented so that higher values indicate increased satisfaction. To account for data skewness, we express the FS-ICU scores using median values and interquartile ranges. Multiple linear regression was performed to adjust for potential confounders and assess the relative importance of various factors on family satisfaction with spiritual care. Acknowledging the exploratory nature of this study, our model included block entry of any variable with a p value of ⱕ.1 in the univariate analyses and any variables hypothesized a priori as meaningful. Satisfaction with spiritual care, FS-ICU scores, and ages were modeled as continuous variables. All other variables were categorical. We initially included ethnicity in the model, but given the small number of nonwhite participants (⬍15%), this variable did not lend additional information and was subsequently excluded. Analyses were conducted using SPSS 13 (SPSS, Chicago, IL), and significance was defined as a p value of ⱕ.05.

RESULTS After excluding families for whom there was no contact information, survey packets were sent to 1,074 family members of eligible patients. Among the 442 family members who returned survey packets (41% response rate), chart abstraction data were available for 356 patients. Demographic characteristics for family members returning questionnaires were compared with those who did not return questionnaires. Both groups were similar with regard to age (68.1 vs. 70.1 yrs, p ⫽ .07), sex (47.3% vs. 41.3% female, p ⫽ .08), and median hospital length of stay (4 days for both, p ⫽ .10). Survey respondents were more likely white (78.1% vs. 59.5%, p ⬍ .001) and had longer median ICU stays (2.8 vs. 2.4 days, p ⫽ .02). Figure 1 displays the frequency distributions for the spiritual care item. Overall, only 73% of family members (259 of 356) rated their spiritual care, a finding 1086

Table 1. Characteristics of study population, stratified on whether the family member rated his or her spiritual care (n ⫽ 356) Rated Spiritual Care (n ⫽ 259) Patient demographics Age in years, mean (SD) Female sex Race/ethnicity White Asian/Pacific Islander Native American/Alaskan Black/African American Hispanic Other/unknown Insurance status Private/commercial Government/public None/unknown Best LOC in last 24 hrs of life Comatose/semi-comatose Partly communicative or localizes to pain Alert and oriented Unknown ICU LOS, median days (range) Hospital LOS, median days (range) Family demographics Age in years, mean (SD) Female sex Race/ethnicity White Asian/Pacific Islander Native American/Alaskan Black/African American Hispanic Other/unknown Relationship to patient Spouse or partner Child Parent Sibling Other Lives with patient Level of education High school Some college/trade school 4-yr college degree Graduate/professional Unknown Satisfaction with ICU care, median (IQR)a

No Rating (n ⫽ 97)

68.4 (16.2) 41.7 (108)

74.6 (13.9) 40.2 (39)

78 (202) 3.5 (9) 0.8 (2) 1.9 (5) 1.2 (3) 14.7 (38)

77.3 (75) 4.1 (4) 0 3.1 (3) 1 (1) 14.4 (14)

10.8 (28) 68.7 (178) 3.9 (10)

5.2 (5) 76.3 (74) 3.1 (3)

47.5 (123) 12.7 (33) 34 (88) 5.8 (15) 3 (1–51) 4 (1–72)

39.2 (38) 14.4 (14) 42.3 (41) 4.1 (4) 2 (1–29) 5 (1–113)

56.4 (14.3) 68.8 (174)

62.8 (14.2) 62.8 (59)

81.5 (211) 3.5 (9) 3.9 (10) 3.9 (10) 1.9 (5) 5.4 (14)

83.5 (81) 4.1 (4) 2.1 (2) 4.1 (4) 2.1 (2) 4.1 (4)

43.6 (113) 34.4 (89) 4.6 (12) 6.6 (17) 10.8 (28) 55.2 (143)

41.2 (40) 34 (33) 5.2 (5) 6.2 (6) 13.4 (13) 53.6 (52)

25.1 (65) 44 (114) 13.9 (36) 14.3 (37) 2.7 (7) 80.7 (65.4–93.8)

19.6 (19) 46.4 (45) 14.4 (14) 15.5 (15) 4.1 (4) 80.4 (70–91.2)

p Value

.001 .79 .94

.21

.40

.72 .27 ⬍.001 .29 .96

.97

.71 .82

.76

LOC, level of consciousness; ICU, intensive care unit; LOS, length of stay; IQR, interquartile range. Total satisfaction was measured using the Family Satisfaction in the Intensive Care Unit questionnaire (range, 0 –100). Values are expressed as % (n), unless otherwise stated. a

similar to our previous study (10). Among families who did not rate their spiritual care, 88% indicated it was not applicable. The remainder either skipped the item or the answer was uninterpretable. Among those rating their spiritual care, responses revealed a positive skew typical of satisfaction data. Table 1 summarizes the patient and family characteristics, stratified on whether the family member respondent rated his or her spiritual care. The only differences between the two groups were that patients and family members were

slightly younger in the group rating their spiritual care (p ⬍ .001 and p ⫽ .001, respectively). FS-ICU scores were nearly identical in the two groups, suggesting no difference in overall satisfaction with care. Table 2 displays the EOL care processes for the two groups, including presence of do-not-resuscitate orders at the time of death, symptom control in the last 24 hrs of life, and family conference data. Overall, a family– clinician conference occurred for 91% of families, and a pastor or spiritual advisor was involved in 60% of deaths. As expected, family memCrit Care Med 2007 Vol. 35, No. 4

Table 2. End-of-life care processes, stratified on whether the family member rated his or her spiritual care (n ⫽ 356) Rated Spiritual Care (n ⫽ 259)

No Rating (n ⫽ 97)

p Value

41.3 (107) 15.1 (39) 85.7 (222) 19.7 (51) 54.1 (140) 67.2 (174)

43.3 (42) 12.4 (12) 82.5 (80) 24.7 (24) 45.4 (44) 40.2 (39)

.65 .52 .45 .30 .14 ⬍.001

40.9 (106) 34.7 (90) 28.6 (74) 10.4 (27)

34 (33) 34 (33) 29.9 (29) 12.4 (12)

.23 .90 .81 .60

91.1 (236) 66 (171) 81.1 (210) 17 (44) 14.9 (38) 44.8 (116) 9.7 (25)

91.8 (89) 55.7 (54) 77.3 (75) 20.6 (20) 15.5 (15) 52.6 (51) 10.3 (10)

.71 .08 .47 .41 .84 .17 .84

Living will available CPR performed in last 24 hrs of life DNR at time of death On full life support at time of death Withdrawal of life support orders used Pastor or spiritual advisor involved Symptom present at time of death Pain Dyspnea Restlessness/agitation Anxiety Family conferences Occurrence of family conference Prognosis discussed Family wanted to withdraw treatment Patient wanted to withdraw treatment Physician recommended withdrawal Decision made to withdraw treatment Any family discord

CPR, cardiopulmonary resuscitation; DNR, do not resuscitate. Values are expressed as % (n).

bers were more likely to rate their spiritual care if a pastor/spiritual advisor was involved during the last 24 hrs of the patient’s life. Table 3 summarizes the patient and family characteristics, stratified on whether the family member rated his or her spiritual care as excellent or less than excellent (very good, good, fair, or poor). Family members reporting excellent satisfaction with their spiritual care also had higher satisfaction with their ICU care, as measured using the FS-ICU. However, no patient or family characteristics were significantly associated with higher satisfaction with spiritual care. Table 4 displays the EOL care and family conference data. In general, family members were more satisfied with their spiritual care if the patient had a do-notresuscitate order at the time of death or if a pastor/spiritual advisor was involved in the last 24 hrs of life. The presence of dyspnea in the patient at the time of death, as described by the family member, was also associated with lower satisfaction ratings. In the adjusted regression model (Table 5), family members were more satisfied with spiritual care if a pastor/ spiritual advisor was involved in the last 24 hrs of life (p ⫽ .007). This finding persisted after adjusting for patient and family characteristics, EOL care processes, and total satisfaction with the ICU experience. In addition, the family member’s satisfaction with spiritual care was Crit Care Med 2007 Vol. 35, No. 4

strongly associated with his or her overall satisfaction with ICU care (p ⬍ .001). Female family members reported higher satisfaction with spiritual care than male family members (p ⫽ .03), but there were no other patient or family characteristics significantly associated with higher satisfaction with spiritual care. Overall, the model explained 51% of the variance observed (R ⫽ .712).

DISCUSSION The Robert Wood Johnson Foundation Critical Care Workgroup has identified spiritual care of families and patients as a key quality domain for EOL care in the ICU (19). In addition, recent clinical practice guidelines describe spiritual support of the family as an important task for delivering comprehensive care in the ICU (20). Anticipating future research on this topic, we sought to identify potential determinants of family satisfaction with spiritual care when a loved one is dying in the ICU. Overall, we found that family members were more likely to be satisfied with their spiritual care if a pastor or spiritual care advisor was involved during the 24 hrs before the patient’s death. In addition, female family members tended to be more satisfied with spiritual care than male respondents. As expected, families who were more satisfied with their spiritual care were also more satisfied with their total ICU experience.

We also examined whether family members who rated their spiritual care were different from family members who did not rate this aspect of their ICU experience. We undertook this analysis because previous studies show that a high proportion of family members do not provide a rating of spiritual care (10, 11). Because these data are likely not missing at random, it is important to understand this phenomenon if we are to use survey ratings of spiritual care to improve the quality of care for all family members. Overall, family members rating their spiritual care were slightly younger and their respective patients were slightly younger. In addition, family members were more likely to rate their spiritual care if a pastor or spiritual advisor was involved in the 24 hrs before the patient’s death. However, no other patient/family demographics or EOL care processes distinguished family members who rated their spiritual care. Our findings are supported by previous studies. When Abbott et al. (21) interviewed family members 1 yr after a loved one died in the ICU, half of the families spontaneously mentioned the value of clergy presence during EOL discussions. This may be because many family members use spirituality as a platform for coping with the guilt of “letting go” (22). Similarly, others have shown that women and younger individuals may be more willing to openly discuss their spirituality with healthcare providers (23, 24). Of note, Shahabi et al. (24) found that self-perceptions of spirituality were associated with being younger, whereas self-perceptions of religiousness were associated with being older. Our survey does not allow us to clearly distinguish these two concepts; future research in this area will require close attention to terminology and operative definitions. Our findings also differ from previous studies. For example, we found no evidence that education level was associated with spiritual care ratings, whereas others have found that lower education levels are associated with higher ratings of spiritual care (24, 25). However, this may be because our study was conducted in a single geographic region and the study population was predominately white. A final interesting finding is that patient dyspnea was significantly associated with lower satisfaction with spiritual care in the univariate analysis and showed a similar trend in the adjusted model. One possible explanation is that patient 1087

Table 3. Characteristics of study population, stratified on the family member’s satisfaction with his or her spiritual care (n ⫽ 259) Excellent (n ⫽ 103) Patient demographics Age in years, mean (SD) Female sex Race/ethnicity White Asian/Pacific Islander Native American/Alaskan Black/African American Hispanic Other/unknown Insurance status Private/commercial Government/public None/unknown Best LOC in last 24 hrs of life Comatose/semi-comatose Partly communicative or localizes to pain Alert and oriented Unknown ICU LOS, median days (range) Hospital LOS, median days (range) Family demographics Age in years, mean (SD) Female Race/ethnicity White Asian/Pacific Islander Native American/Alaskan Black/African American Hispanic Other/unknown Relationship to patient Spouse or partner Child Parent Sibling Other Lives with patient Level of education High school Some college/trade school 4-yr college degree Graduate/professional Unknown Satisfaction with ICU care, median (IQR)a

Less Than Excellent (n ⫽ 156)

68 (16.4) 39.8 (41)

68.6 (16.2) 42.9 (67)

77.7 (80) 2.9 (3) 1 (1) 1 (1) 0 17.5 (18)

78.2 (122) 3.8 (6) 0.6 (1) 2.6 (4) 1.9 (3) 12.8 (20)

9.7 (10) 67 (69) 2.9 (3)

11.5 (18) 69.9 (109) 4.5 (7)

51.5 (53) 13.6 (14) 29.1 (30) 5.8 (6) 3 (1–26) 4 (1–32)

44.9 (70) 12.2 (19) 37.2 (58) 5.8 (9) 3 (1–51) 4 (1–72)

56.7 (15) 72.8 (75)

56.2 (13.8) 63.5 (99)

84.5 (87) 2.9 (3) 2.9 (3) 2.9 (3) 1 (1) 5.8 (6)

79.5 (124) 3.8 (6) 4.5 (7) 4.5 (7) 2.6 (4) 5.1 (8)

42.7 (44) 33 (34) 6.8 (7) 4.9 (5) 12.6 (13) 50.5 (52)

44.2 (69) 35.3 (55) 3.2 (5) 7.7 (12) 9.6 (15) 58.3 (91)

22.3 (23) 49 (47.5) 8.7 (9) 18.4 (19) 2.9 (3) 94.2 (85.2–97.9)

26.9 (42) 41.7 (65) 17.3 (27) 11.5 (18) 2.6 (4) 71.9 (55.2–82.3)

p Value

.76 .62 .56

.83

.61

.98 .80 .78 .08 .85

.54

.25 .17

⬍.001

LOC, level of consciousness; ICU, intensive care unit; LOS, length of stay; IQR, interquartile range. Total satisfaction was measured using the Family Satisfaction in the Intensive Care Unit questionnaire (range, 0 –100). Values are expressed as % (n), unless otherwise stated. a

breathlessness at the time of death caused spiritual distress in family members. Another possibility is that breathlessness led to general dissatisfaction that was not completely controlled for in our analyses. Because this is an exploratory analysis, however, we cannot determine the exact reason for this finding. We recommend that future studies further examine the effects of patient symptom management on family satisfaction ratings. Since the early work by Molter (26), many researchers have focused on iden1088

tifying and meeting the needs of families in the ICU. However, few studies have focused on the spiritual care of families with a loved one in the ICU. One possible reason that this topic has not been studied is that clinicians under-recognize the value of spirituality, just like they underrecognize the importance of religion. For example, most patients and families find religion to be the most important factor enabling them to cope with medical illness, but only a small percentage of physicians feel this way (7). Similarly, health professionals often underestimate the

role that religious faith plays in medical decision making for patients and families (27). Finally, there is a considerable gap in religiosity levels between patients and physicians, with physicians much less likely to report a strong religious identity (28). Although spirituality and religiosity are different concepts, these studies of religion provide valuable insights into the apparent research gap. Many families and patients want physicians to ask about spirituality, especially during life-threatening illnesses (23, 29, 30). In addition, experts have called for clinicians to assess the spiritual needs of critically ill patients and their families in the ICU (31). However, most authors agree that healthcare professionals should resist the urge to undertake directive or proscriptive theological discussions with families in the ICU (2, 32– 37). Although some programs offer spiritual care training to intensivists (6), we believe that complex spiritual issues are best handled by a trained expert. In general, ICU clinicians should respectfully broach the topic with family members, and if additional assistance is required, make an appropriate referral to a spiritual specialist (38 – 40). There are limitations to this study. Because this study focused on patients dying in the ICU, these findings may not apply to families of ICU survivors. In addition, because our population was predominately white and from Seattle-area hospitals, our findings are unlikely to generalize to other U.S. regions with different racial, ethnic, or religious compositions (41). Despite obtaining a response rate that is comparable with other studies of families after the death of a patient, there is still potential for nonrespondent bias. Although this bias may limit generalizability of our findings, this should not affect the validity of the observed associations. By design, we performed multiple comparisons and our findings should be interpreted accordingly. Similarly, we are unable to determine the direction of observed associations. For example, chaplains may simply be more likely to visit families who openly discuss their spirituality. However, given the preliminary nature of this area of inquiry and the need for hypothesis generation, we believed this approach was appropriate. Finally, we used a data set that did not measure religious denomination or levels of religiosity. Given the importance of these two attributes for this field of research, Crit Care Med 2007 Vol. 35, No. 4

Table 4. End-of-life care processes, stratified on the family member’s satisfaction with his or her spiritual care (n ⫽ 259) Excellent (n ⫽ 103)

Less Than Excellent (n ⫽ 156)

p Value

38.8 (40) 13.6 (14) 91.3 (94) 18.4 (19) 52.4 (54) 75.7 (78)

42.9 (67) 16 (25) 82.1 (128) 20.5 (32) 55.1 (86) 61.5 (96)

.86 .59 .04 .68 .67 .02

42.7 (44) 27.2 (28) 25.2 (26) 8.7 (9)

39.7 (62) 39.7 (62) 30.8 (48) 11.5 (18)

.63 .04 .34 .47

93.2 (96) 68.9 (71) 81.6 (84) 20.4 (21) 12.6 (13) 40.8 (42) 12.9 (13)

89.7 (140) 64.1 (100) 80.8 (126) 14.7 (23) 16 (25) 47.4 (74) 7.8 (12)

.22 .37 .78 .23 .46 .31 .18

Living will available CPR performed in last 24 hrs of life DNR at time of death On full life support at time of death Withdrawal of life support orders used Pastor or spiritual advisor involved Symptom present at time of death Pain Dyspnea Restlessness/agitation Anxiety Family conferences Occurrence of family conference Prognosis discussed Family wanted to withdraw treatment Patient wanted to withdraw treatment Physician recommended withdrawal Decision made to withdraw treatment Any family discord

CPR, cardiopulmonary resuscitation; DNR, do not resuscitate. Values are expressed as % (n). Table 5. Multivariate model showing contributions of demographics variables and end-of-life care processes toward higher family satisfaction with spiritual care

Pastor or spiritual advisor involved Total family satisfactiona Family member age Family member sex Family member level of education Patient age Patient sex DNR at time of death Dyspnea at time of death

Coefficient

SE

p Value

8.646 0.988 ⫺0.028 7.626 0.454 0.036 4.807 4.943 ⫺5.520

3.195 0.068 0.117 3.403 1.516 0.100 3.196 4.195 3.110

.007 ⬍.001 .81 .03 .77 .72 .13 .24 .08

DNR, do not resuscitate. Total satisfaction was measured using the Family Satisfaction in the Intensive Care Unit questionnaire (range, 0 –100). a

future studies should attempt to quantify these attributes using simple screening questions (42– 44). Approximately 95% of Americans report they believe in God, with 72% calling religious faith the strongest influence in their lives and 40% attending worship in a typical week (45). Although religion is not synonymous with spirituality, both concepts share common elements, including perception of self, others, and a transcendent being or force (1–3). When discussing these issues with patients and families, however, it is probably best to use the term “spirituality” because of its broad and inclusive nature (46). In addition, it is important to recognize that although our study focused on family members, critically ill patients may also have significant spiritual needs. Crit Care Med 2007 Vol. 35, No. 4

In summary, we found that family satisfaction with spiritual care is higher when a pastor or spiritual care advisor is involved during the 24 hrs before the patient’s death. These findings suggest that if a patient is dying in the ICU, clinicians should assess the family’s spiritual needs and consult a spiritual advisor if desired by the family. For families in the ICU, critical illness is a time of crisis that challenges them to find meaning in the midst of suffering. A comprehensive approach to ICU care should aim to meet not only the physical but also psychosocial and spiritual needs of patients and their families. Based on previous reports showing physicians perform poorly when it comes to talking about EOL issues like dying and spirituality (47), we believe there is room for improvement.

REFERENCES 1. Thoresen CE, Harris AH: Spirituality and health: What’s the evidence and what’s needed? Ann Behav Med 2002; 24:3–13 2. Post SG, Puchalski CM, Larson DB: Physicians and patient spirituality: Professional boundaries, competency, and ethics. Ann Intern Med 2000; 132:578 –583 3. Dyson J, Cobb M, Forman D: The meaning of spirituality: A literature review. J Adv Nurs 1997; 26:1183–1188 4. Pandharipande P, Jackson J, Ely EW: Delirium: Acute cognitive dysfunction in the critically ill. Curr Opin Crit Care 2005; 11: 360 –368 5. Azoulay E, Pochard F, Kentish-Barnes N, et al: Risk of post-traumatic stress symptoms in family members of intensive care unit patients. Am J Respir Crit Care Med 2005; 171: 987–994 6. Todres ID, Catlin EA, Thiel MM: The intensivist in a spiritual care training program adapted for clinicians. Crit Care Med 2005; 33:2733–2736 7. Koenig HG, Bearon LB, Hover M, et al: Religious perspectives of doctors, nurses, patients, and families. J Pastoral Care 1991; 45:254 –267 8. Rabins PV, Fitting MD, Eastham J, et al: Emotional adaptation over time in caregivers for chronically ill elderly people. Age Ageing 1990; 19:185–190 9. Angus DC: Charting (and publishing) the boundaries of critical illness. Am J Respir Crit Care Med 2005; 171:938 –939 10. Wall RJ, Engelberg RA, Downey L, et al: Refinement, scoring, and validation of the Family Satisfaction in the Intensive Care Unit (FS-ICU) survey. Crit Care Med 2007; 35:271–279 11. Heyland DK, Rocker GM, Dodek PM, et al: Family satisfaction with care in the intensive care unit: Results of a multiple center study. Crit Care Med 2002; 30:1413–1418 12. Heyland DK, Tranmer JE: Measuring family satisfaction with care in the intensive care unit: The development of a questionnaire and preliminary results. J Crit Care 2001; 16:142–149 13. Dodek PM, Heyland DK, Rocker GM, et al: Translating family satisfaction data into quality improvement. Crit Care Med 2004; 32:1922–1927 14. Cook D, Rocker G, Heyland D: Dying in the ICU: Strategies that may improve end-of-life care. Can J Anaesth 2004; 51:266 –272 15. Dowling J, Vender J, Guilianelli S, et al: A model of family-centered care and satisfaction predictors: The Critical Care Family Assistance Program. Chest 2005; 128(3 Suppl): 81S–92S 16. Dowling J, Wang B: Impact on family satisfaction: The Critical Care Family Assistance Program. Chest 2005; 128(3 Suppl):76S– 80S 17. Lederer MA, Goode T, Dowling J: Origins and development: The Critical Care Family Assis-

1089

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

tance Program. Chest 2005; 128(3 Suppl): 65S–75S Family Satisfaction in the Intensive Care Unit (FS-ICU) Survey. Available at: http:// www.criticalcareconnections.com. Accessed November 1, 2006 Clarke EB, Curtis JR, Luce JM, et al: Quality indicators for end-of-life care in the intensive care unit. Crit Care Med 2003; 31:2255–2262 Davidson JE, Powers K, Hedayat KM, et al: Clinical practice guidelines for support of the family in the patient-centered intensive care unit: American College of Critical Care Task Force 2004 –2005. Crit Care Med 2007; 35: 605– 622 Abbott KH, Sago JG, Breen CM, et al: Families looking back: One year after discussion of withdrawal or withholding of life-sustaining support. Crit Care Med 2001; 29:197–201 Koenig HG: A commentary: The role of religion and spirituality at the end of life. Gerontologist 2002; 42(Spec No 3):20 –23 McCord G, Gilchrist VJ, Grossman SD, et al: Discussing spirituality with patients: A rational and ethical approach. Ann Fam Med 2004; 2:356 –361 Shahabi L, Powell LH, Musick MA, et al: Correlates of self-perceptions of spirituality in American adults. Ann Behav Med 2002; 24:59 – 68 Koenig HG: Religious attitudes and practices of hospitalized medically ill older adults. Int J Geriatr Psychiatry 1998; 13:213–224 Molter NC: Needs of relatives of critically ill patients: A descriptive study. Heart Lung 1979; 8:332–339 Silvestri GA, Knittig S, Zoller JS, et al: Importance of faith on medical decisions re-

1090

28.

29. 30.

31.

32.

33. 34.

35.

36.

37.

38.

garding cancer care. J Clin Oncol 2003; 21: 1379 –1382 Frank E, Dell ML, Chopp R: Religious characteristics of US women physicians. Soc Sci Med 1999; 49:1717–1722 Maugans TA: The SPIRITual history. Arch Fam Med 1996; 5:11–16 Ehman JW, Ott BB, Short TH, et al: Do patients want physicians to inquire about their spiritual or religious beliefs if they become gravely ill? Arch Intern Med 1999; 159: 1803–1806 Chambers N, Curtis JR: The interface of technology and spirituality in the ICU. In: Managing Death in the ICU: The Transition from Cure to Comfort. Curtis JR, Rubenfeld GD (Eds). New York, Oxford University Press, 2001, pp 193–206 Sloan RP, Bagiella E, VandeCreek L, et al: Should physicians prescribe religious activities? N Engl J Med 2000; 342:1913–1916 Savulescu J: Two worlds apart: Religion and ethics. J Med Ethics 1998; 24:382–384 Cohen CB, Wheeler SE, Scott DA: Walking a fine line: Physician inquiries into patients’ religious and spiritual beliefs. Hastings Cent Rep 2001; 31:29 –39 Astrow AB, Puchalski CM, Sulmasy DP: Religion, spirituality, and health care: Social, ethical, and practical considerations. Am J Med 2001; 110:283–287 Sloan RP, Bagiella E, Powell T: Religion, spirituality, and medicine. Lancet 1999; 353: 664 – 667 Koenig HG: MSJAMA: Religion, spirituality, and medicine. Application to clinical practice. JAMA 2000; 284:1708 Puchalski CM, Larson DB: Developing cur-

39.

40.

41.

42.

43.

44.

45.

46.

47.

ricula in spirituality and medicine. Acad Med 1998; 73:970 –974 Thiel MM, Robinson MR: Physicians’ collaboration with chaplains: Difficulties and benefits. J Clin Ethics 1997; 8:94 –103 Lo B, Ruston D, Kates LW, et al: Discussing religious and spiritual issues at the end of life: A practical guide for physicians. JAMA 2002; 287:749 –754 Putnam R: 2001 Social Capital Community Benchmark Survey. Available at: http:// www.cfsv.org/communitysurvey. Accessed November 1, 2006 Kub JE, Nolan MT, Hughes MT, et al: Religious importance and practices of patients with a life-threatening illness: Implications for screening protocols. Appl Nurs Res 2003; 16:196 –200 Lo B, Quill T, Tulsky J: Discussing palliative care with patients: ACP-ASIM End-of-Life Care Consensus Panel. American College of Physicians-American Society of Internal Medicine. Ann Intern Med 1999; 130: 744 –749 Puchalski C, Romer AL: Taking a spiritual history allows clinicians to understand patients more fully. J Palliat Med 2000; 3:129 –137 Religion in America 1993–1994. Princeton, NJ, Princeton Religious Research Center, 1994 Koenig HG: Religion, spirituality, and medicine: Research findings and implications for clinical practice. South Med J 2004; 97: 1194 –1200 Curtis JR, Engelberg RA, Nielsen EL, et al: Patient-physician communication about end-of-life care for patients with severe COPD. Eur Respir J 2004; 24:200 –205

Crit Care Med 2007 Vol. 35, No. 4