Journal of Critical Care xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Journal of Critical Care journal homepage: www.jccjournal.org
Family experience with intensive care unit care: Association of self-reported family conferences and family satisfaction☆,☆☆ Sashikanth Kodali, MD a,⁎, Rebecca A. Stametz, DEd, MPH a, Amanda C. Bengier, BA a, Deserae N. Clarke, MPA a, Abraham J. Layon, MD b, Jonathan D. Darer, MD, MPH a a b
Division of Clinical Innovation, Geisinger Health System, Danville, PA Division of Critical Care, Geisinger Health System, Danville, PA
a r t i c l e
i n f o
Keywords: Family conference Family satisfaction End-of-life care Family communication Medical decision making
a b s t r a c t Purpose: A family conference is recommended as a best practice to improve communication in the intensive care unit (ICU), but this can be challenging given the setting. This study examined whether family members who reported that a family conference occurred had higher satisfaction than those who did not report that a conference was held. Materials and Methods: The study used a retrospective data analysis of family satisfaction based on family member's responses to a questionnaire. Families of all the patients admitted to ICUs of 2 quaternary hospitals in an integrated health system were surveyed. Results: The families of 457 patients who matched the inclusion and exclusion criteria were surveyed with a 55.6% response rate. Family satisfaction with decision making was significantly higher (83.6 vs 78.2, P = .0211) for families who reported that family conferences occurred. No significant difference in the satisfaction with care and overall satisfaction scores was found (84.2 vs 80.0, P = .10). Patients whose families reported a family conference were older and had higher mortality. Conclusion: This study confirms that families who report attending family conferences are more satisfied with decision making in the ICU. This study highlights the need to increase communication in ICUs. © 2014 Elsevier Inc. All rights reserved.
1. Introduction Yearly, more than 5.7 million Americans are admitted to an intensive care unit (ICU) [1]. Intensive care unit rates of mortality range from 10% to 15%, representing more than 540 000 US deaths each year [2]. Concerns with respect to quality of end-of-life care and communication with families in the ICU have been reported [3,4]. Most critically ill patients do not have decision-making capabilities and are often unable to communicate with the health care team because of their illness or the effect of treatments rendered, such as mechanical ventilation and sedation [5]. Family members of the critically ill patient are often required to act as surrogate decision makers, a role in which they may experience significant stress and emotional burden [6]. Improved communication with families of patients admitted to an ICU has been shown to reduce ICU length of stay, increase referrals to hospice, and, when appropriate, result in earlier withdrawal of life-sustaining interventions [7,8]. Effective
☆ All work was performed at Geisinger Health System, Danville PA, USA. ☆☆ Financial support was provided by Geisinger Clinic. There were no conflicts of interest for any of the authors. ⁎ Corresponding author. Geisinger Medical Center Division of Clinical Innovation, 100 N Academy Ave, Danville PA 17821. E-mail address: [email protected] (S. Kodali).
communication with families has also been shown to improve family satisfaction and psychological well-being [9]. Several areas have been identified as opportunities for improvement with regard to clinician-family communication in the ICU. Families often do not understand even basic information about the patient's diagnosis, prognosis, or treatment after 48 hours of admission to ICU [10], and physicians often miss opportunities to listen, acknowledge, and address emotions of families and do not explain surrogate decision making [11]. Poor physician communication and poor coordination of care are correlated with reduced family satisfaction with care [12,13]. Conversely, factors associated with improved family satisfaction include high-quality communication and collaboration between nursing staff and providers [14]. National guidelines with regard to palliative care for patients admitted to the ICU Include the performance of an interdisciplinary family conference for patients and families within 72 hours of admission to an ICU [15]. Despite the recognition of the positive contribution of interdisciplinary family conferences as a component of high-quality ICU care, the mechanism by which family satisfaction and well-being are improved has not been well characterized. As part of an initiative to improve the quality of palliative care for patients in the ICU, we performed a baseline survey of family experiences. We undertook analysis of the data derived from the survey and clinical data from the electronic health record (EHR) to explore the association of occurrence
http://dx.doi.org/10.1016/j.jcrc.2014.03.012 0883-9441/© 2014 Elsevier Inc. All rights reserved.
Please cite this article as: Kodali S, et al, Family experience with intensive care unit care: Association of self-reported family conferences and family satisfaction, J Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.03.012
2
S. Kodali et al. / Journal of Critical Care xxx (2014) xxx–xxx
of family conferences as reported by families with family-reported satisfaction with the care provided and the satisfaction with the decision-making process. 2. Materials and methods We conducted a retrospective cohort study using family satisfaction surveys and an EHR data pull to gather baseline satisfaction with care and the medical decision-making process for patients and families cared for in the ICU. The Family Satisfaction in the Intensive Care Unit Questionnaire24 (FS-ICU24) is a 24-item validated instrument for assessing family satisfaction in the ICU [16]. The instrument is available in public domain and can be accessed through the Canadian Researchers at the End of Life Network (http://www.thecarenet.ca). The instrument yields an overall satisfaction score (FS-Total) and has 2 subscales: satisfaction in the areas of decision making (FS-DM) and satisfaction with care provided (FS-Care). All 24 items, except for 1, use a 5-point Likert scale and ask the family member simple, easy to understand questions related to each subscale. In addition to these 24 items, there are 3 open-ended questions that solicit suggestions on how to make ICU care better. Family satisfaction scores were calculated using the validated scoring methodology of the FS-ICU24 questionnaire. An additional question that specifically asks if a family meeting was held during their loved one's hospital stay was added to the FS-ICU24. This question included a definition of what the family meeting was, specifically “A family conference is a meeting with the care staff to discuss how your loved one is doing and assist in care decisions.” Patient data were also pulled from the EHR to link to the survey. This included age, sex, disposition (death, home, transfer), length of stay, and presenting Modified Early Warning Score (MEWS). This research study was conducted in cooperation with the Geisinger Health System's (GHS's) Office of Research Protection. Approval to conduct the research was obtained from the institutional review board prior to beginning this study, under an expedited status.
contacted and who opted out prior to the mailing were excluded from the study, leaving a total of 453 NOK eligible for survey. A modified Dillman survey methodology was used for the 453 NOK whose addresses were obtained through 1 of the 2 methods described above. An initial notification letter was sent along with the FS-ICU24 and a $1 bill to improve response rates. The letter explained the study and provided a number that patients could call to opt out. The letter explicitly stated that the care received by their family member would not be affected by the decision to participate in the study. After 3 weeks, a reminder letter and an additional copy of the survey were sent to NOK who did not return the survey or opt out of the study. If there was no response 3 days after this mailing, a study team member called the NOK and verified if they had received the survey and were planning to respond. For those who returned a survey, consent was implied and thank you letters were sent to all NOK who completed and returned a survey. 2.3. Statistical analysis Analyses were performed with de-identified clinical and demographic data extracted from Geisinger's EHR. An univariate analysis of variance was performed to examine differences in each of the 3 satisfaction scores (FS-Total, FS-DM, and FS-Care) based on patient and family characteristics, ICU length of stay, hospital discharge disposition, MEWS, and whether families reported that a family conference occurred. For patients who had multiple episodes of ICU stay during a single hospitalization, cumulative ICU length of stay was used in analysis. Patient disposition was based on the final disposition from the hospital; the transfer category of the disposition included transfers to skilled nursing facilities, acute rehabilitation facilities, long-term acute care, and other acute care hospitals. The MEWS, which is shown to be a predictor of patient mortality and outcomes in ICU [17], was used as a surrogate marker for severity of illness at the time of admission. 3. Results
2.1. Study setting The GHS is an integrated health care delivery system based in Northeast and North Central Pennsylvania. Inpatient facilities in which the study took place include Geisinger Medical Center in Danville, Penn, and Geisinger Wyoming Valley in Wilkes Barre, Penn, both of which are tertiary/quaternary care inpatient hospitals. The 2 hospitals have 62 ICU beds and 161 full-time equivalents of nursing care. 2.2. Subject selection and identification Patients 18 years and older who were admitted to the Geisinger Medical Center and Geisinger Wyoming Valley ICUs who received ICU levels of care between January 1 and March 22, 2012, were included in the sampling pull. Patients with a research opt-out indicator were excluded, as were patients who lacked any next-of-kin (NOK) information in their record. For patients with multiple contacts, the NOK listed as the primary contact in the EHR were used for study purposes. Patients were identified by querying our clinical data warehouse, which stores all the clinical information from the EHR. A total of 634 patients were identified as meeting the original inclusion criteria. Only 554 of these patients had NOK name and telephone number listed in their record for contact purposes. Sequential methods were implemented to obtain mailing addresses for the family members. First, using NOK name and telephone number from the patient's EHR, a reverse look-up was performed using Whitepages. com. Next, if still unable to identify the NOK address, a telephone call was made by Survey Services at GHS' Center for Health Research. The NOK who were contacted by Survey Services were asked to participate in the study and provide an address. The NOK who could not be
Of the 453 surveys that were mailed, 16 NOK chose to opt out, leaving 437 records for analysis. Of the NOK surveyed, 243 responded, representing a 55.6% response rate. The subset of patients whose NOK responded to the survey had similar characteristics when compared with those who did not respond (Table 1). A total of 108 (44.4%) of
Table 1 Summary of patient characteristics Total Response No response P (N = 437) (n = 243; 55.6%) (n = 194; 44.4) Age category (patient; y) 18-39 40-59 60+ Patient sex Male Female Unknown Disposition Death Home Transfer ICU length of stay (h) 0-48 48-96 96+ Presenting MEWS score 0-2 3-5 6+ Unknown
.6195 47 (10.8) 142 (32.5) 248 (56.8)
29 (11.9) 76 (31.3) 138 (56.8)
18 (9.3) 66 (34.0) 110 (56.7)
250 (57.5) 185 (42.5) 2
149 (61.6) 93 (38.4) 1
101 (52.3) 93 (47.7) 1
66 (15.1) 200 (45.8) 171 (39.1)
42 (17.3) 113 (46.5) 88 (36.2)
24 (12.4) 87 (44.9) 83 (42.8)
178 (40.7) 90 (20.6) 169 (38.7)
100 (41.2) 52 (21.4) 91 (37.4)
78 (40.2) 38 (19.6) 78 (40.2)
216 (49.7) 168 (38.6) 51 (11.7) 2
117 (48.6) 94 (39.0) 30 (12.4) 2
99 (51.0) 74 (38.1) 21 (10.8) 0
.0528
.2255
.6099
.8205
Please cite this article as: Kodali S, et al, Family experience with intensive care unit care: Association of self-reported family conferences and family satisfaction, J Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.03.012
S. Kodali et al. / Journal of Critical Care xxx (2014) xxx–xxx
243 patients’ families reported that they had a family meeting. The average age of the patients whose families responded to the survey was 60.5 years, 38.4% were women, 17.3% died in hospital, and 37.4% of patients had ICU stays of 96 hours or longer. The family members who responded were more likely to be female and older, and the most common relationship listed with the patient was partner or spouse. Family satisfaction scores in the domain of decision making were significantly higher (83.6 vs 78.2, P = .0211) for families who reported that family conferences occurred (Table 2). The difference in decision-making score remained statistically significant, despite adjusting for confounding variables. No significant difference in the satisfaction with care (84.6 vs 82.1, P = .21) and overall satisfaction scores (84.2 vs 80.0, P = .0413) was found between families reporting a family meeting compared with families not reporting a family meeting. Spouses were more satisfied than parents with ICU care and the medical decision-making process. Using the same variables listed above in the Statistical Analysis section and treating the FS-DM score as continuous and binary using its 25th, 50th, and 75th quantiles as cutoff points, we examined their relationship with the FS-DM score. The logistic regression model was adjusted and found that the FS-DM score would be higher if patient's family reported that a family meeting occurred. The families who reported attending a family conference were compared with the families who reported that no conference occurred (Table 3). A χ 2 test showed that the family members reporting occurrence of a family conference were older. Similarly, the patients of the families who reported a family conference were older and had a higher rate of death when compared with patients whose families reported that no conference occurred.
Table 2 FS-ICU score ANOVA n
FS-Care
FS-DM
FS-Total
Mean ± SD P
Mean ± SD P
Mean ± SD P
Age category (family member; y) 18-39 15 78.8 (15.2) 40-59 100 81.6 (17.0) 60+ 123 84.6 (16.0) Unknown 5 82.5 (22.4) Age category (patient; y) 18-39 29 81.3 (14.3) 40-59 77 82.4 (16.6) 60+ 137 83.6 (17.0) Previous ICU experience Yes 142 82.8 (17.0) No 97 83.3 (15.7) Unknown 4 78.1 (23.2) Sex (family member) Male 66 83.9 (15.0) Female 171 82.8 (16.7) Unknown 6 75.0 (26.2) Relationship to patient Spouse/Partner 112 85.5 (14.7) Parent 35 78.8 (19.2) Sibling 26 84.5 (16.9) Child 38 81.8 (14.1) Other 28 78.3 (20.8) Unknown 4 78.1 (23.2) Family conference perception Yes 108 84.6 (17.7) No 120 82.1 (15.2) Unknown 15 77.2 (16.8) Outcome Death 42 82.7 (19.4) Home 113 81.9 (15.6) Transfer 88 84.3 (16.3) ICU length of stay (h) 0-48 82.5 (16.9) 48-96 83.9 (14.6) 96+ 82.8 (17.3)
.43 73.6 81.2 79.7 84.5
(18.5) (17.6) (18.4) (19.5)
.45
.76 78.0 (18.6) 76.9 (19.0) 82.2 (17.3)
(15.1) (16.7) (15.5) (21.0)
.58
.10
79.9 (15.2) 80.0 (16.1) 83.2 (16.2)
.33
.83 79.8 (17.9) 80.3 (18.4) 80.6 (20.1)
.98
81.5 (16.4) 82.3 (15.6) 79.2 (21.7)
.88
.45 81.5 (15.2) 79.4 (19.1) 80.6 (20.1)
.73
82.7 (14.5) 81.4 (16.7) 82.5 (20.2)
.85
.19 83.2 73.6 80.4 81.3 72.7 80.6
.0279 84.6 77.1 83.9 81.6 74.7 79.2
(15.5) (21.4) (15.5) (15.6) (24.9) (20.1)
76.7 81.4 82.7 83.3
(14.4) (18.4) (13.4) (13.7) (21.6) (21.7)
3
Table 3 Univariate analysis of family conference perception Family conference perception
Age category (family member; y) 18-39 40-59 60+ Unknown Age category (patient; y) 18-39 40-59 60+ Previous ICU experience Yes No Unknown Sex (family member) Male Female Unknown Relationship to patient Spouse/Partner Parent Sibling Child Other Unknown Disposition Death Home Transfer ICU length of stay (h) 0-48 48-96 96+ Presenting MEWS score 0-2 3-5 6+ Unknown
Yes
No
4 (3.8) 41 (38.7) 61 (57.6) 2
11 (9.3) 55 (46.6) 52 (44.1) 2
7 (6.5) 32 (29.6) 69 (63.9)
20 (16.7) 42 (35.0) 58 (48.3)
61 (57.0) 46 (43.0) 1
75 (63.6) 43 (36.4) 2
35 (32.7) 72 (67.3) 1
26 (22.2) 91 (77.8) 3
56 (52.3) 12 (11.2) 10 (9.4) 19 (17.8) 10 (9.4) 1
52 (44.1) 20 (17.0) 15 (12.7) 17 (14.4) 14 (11.9) 2
31 (28.7) 37 (34.3) 40 (37.0)
7 (5.8) 68 (56.7) 45 (37.5)
44 (40.7) 22 (20.4) 42 (38.9)
53 (44.2) 27 (22.5) 40 (33.3)
52 (48.6) 41 (38.3) 14 (13.1) 1
60 (50.4) 43 (36.1) 16 (13.5) 1
P
.0673
.0187
.3156
.0782
.4938
b.0001
.6823
.9439
4. Discussion
.0413
.21 83.6 (18.0) 77.2 (17.5) 76.1 (19.7)
.0211 84.2 (17.2) 80.0 (14.9) 77.7 (15.3)
.10
.6
78.8 (21.4) 79.9 (16.0) 80.8 (18.9)
.83
81.1 (19.8) 80.9 (14.5) 83.1 (16.3)
.63
.89 80.8 (16.9) 78.3 (19.4) 80.1 (18.7)
.72
81.5 (16.1) 82.4 (14.1) 81.8 (17.3)
.95
Family conferences offer a forum for bringing patients, family members, and health care providers together to share prognostic information, set goals of care, and offer families support. Several studies have evaluated the impact of communication strategies that include proactive family conferences upon clinical outcomes. These interventions have tended to be performed on select, high-acuity or specific patient populations including multiorgan failure [18], cerebral ischemia [19], poor prognosis based on predicted mortality [8], duration of mechanical ventilation [20], dementia [21], and in situations where conflict is perceived [22], although national guidelines recommend family conferences in all patients admitted to the ICU. This study, looking at the experiences of families of all patients admitted to the ICU, confirms that families who report attending family conferences are more satisfied with decision making in the ICU. It is interesting, although not surprising, that the mortality rate of patients whose families reported family conference occurrence is higher than that of patients whose families reported that family conference did not occur. The higher mortality and older age of the patient group whose families reported occurrence of a family meeting suggest that there may be increased focus on communication with family members of patients in end-of-life situations. In another study of 539 family members, families of patients who survived ICU stay have lower satisfaction compared with family members of the deceased [23]. However, our study did not examine the differences in family satisfaction scores between the patients who died in the ICU and the patients who survived, as this is not the primary scope of the
Please cite this article as: Kodali S, et al, Family experience with intensive care unit care: Association of self-reported family conferences and family satisfaction, J Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.03.012
4
S. Kodali et al. / Journal of Critical Care xxx (2014) xxx–xxx
study. Attending to the needs of families in the ICU is an important component of high-quality care, yet the increased burden of regularly getting all stakeholders together to meet with families is a substantial barrier to implementation. Even in structured interventions, designed to reliably deliver family conferences, rates of family meetings reach less than 70% of families [9]. Our results show an association between the presence of a family meeting as reported by the NOK/family members and increased satisfaction with the process of medical decision making. It is known that family meetings improve family well-being; however, this may be in fact resulting from the process of engaging families, specifically in the decision-making process. If scheduling a multidisciplinary family conference is overwhelming, other means of family engagement should be explored and their impact be studied. In addition, our analysis failed to demonstrate a difference in the overall satisfaction scores based on the occurrence of a family meeting as reported by family members/NOK. A review of the comments made to open-ended questions demonstrated that family members expressed their dissatisfaction with nonclinical aspects of the hospital stay. This included things like the quality of waiting room facilities, availability of information about food and lodging, and staff interactions outside patient rooms. Hospitals seeking to improve family satisfaction with care may need to look to multiple sources of dissatisfaction, rather than focusing only on care-related concerns. Another possible cause for the lack of difference may be that the self-reported rate of the family conference is very low. The system currently has no way of tracking the actual occurrence of the family conference in the medical record. This survey was a baseline measure for an intervention that will be put in place to improve communication in the ICU. Part of the intervention includes an objective way to record family conferences in the chart, in addition to making sure the meetings are as comprehensive as possible. In the postsurvey, we will be able to compare the self-reported rate of family conference with the actual rate. 4.1. Limitations Our study is limited in several ways. It focused on a fairly homogenous population related to ethnicity and geographic location. The study was performed in quaternary care settings only. Hence, the results may not be generalizable to other populations. Also, the response rate of 55.6% is relatively low, and although the responders and nonresponders were similar from a demographic perspective, there is still the potential for a response bias. In addition, we had a small population from which to survey NOK. Although this is not unusual for ICU studies, it limits the reliability of our results. The study design also allows us to evaluate only associations and not causal relationships. Moreover, our study used self-report of family conference as a means to identify whether a family conference occurred. This may be subject to recall bias. Perhaps, the family members who respond to whether a family meeting occurred or not based on our definition included in the survey may be more likely to report that they are more satisfied with their engagement with decision making. Although there is currently not a rigorous way to gather information on whether a multidisciplinary family conference occurred or not from the medical record due to unreliable documentation practices, this should be an important consideration in future studies. Finally, we surveyed during a limited time frame, which may impact results due to potential temporal variance in ICU conditions.
5. Conclusions A family conference occurrence, as reported by NOK of patients admitted to the ICU, is associated with increased satisfaction with decision making. This study emphasizes the need to increase communication with families in ICUs and the use of multidisciplinary family conference as an important modality of communication. Further studies are needed to understand the most reliable workflow for holding multidisciplinary family conferences. Other modalities besides a multidisciplinary family conference that may improve the communication with families should be explored.
References [1] Kersten A, Millbrandt EB, Rahim MT. How big is critical care in the U.S.? Crit Care Med 2003;31(12):A8–A8. [2] Angus DC, Barnato AE, Linde-Zwirble WT, et al. Use of intensive care at the end of life in the United States: an epidemiologic study. Crit Care Med 2004;32(3):638–43. [3] Azoulay E, Pochard F, Chevret S, et al. Meeting the needs of intensive care unit patient families: a multicenter study. Am J Respir Crit Care Med 2001;163:135–9. [4] Curtis JR, Rubenfeld GD. Improving palliative care for patients in the intensive care unit. J Palliat Med 2005;854:8840–54. [5] Prendergast TJ, Luce JM. Increasing incidence of withholding and withdrawal of life support from the critically ill. Am J Respir Crit Care Med 1997;155:15–20. [6] Pochard F, Darmon M, Fassier T, et al. Symptoms of anxiety and depression in family members of intensive care unit patients before discharge or death: a prospective multicenter study. J Crit Care 2005;20:90–6. [7] Norton SA, Hogan LA, Holloway RG, et al. Proactive palliative care in the medical intensive care unit: effects on length of stay for selected high-risk patients. Crit Care Med 2007;35:1530–5. [8] Lilly CM, De Meo DL, Sonna LA, et al. An intensive communication intervention for the critically ill. Am J Med 2000;109:469–75. [9] Lautrette A, Darmon M, Megarbane B, et al. A communication strategy and brochure for relatives of patients dying in the ICU. N Engl J Med 2007;356:469–78. [10] Azoulay E, Chevret S, Leleu G, et al. Half the families of ICU patients experience inadequate communication with physicians. Crit Care Med 2000;83044:3044–9. [11] Curtis JR, Burt RA. Point: the ethics of unilateral “do not resuscitate” orders: the role of “informed assent”. Chest 2007;132:748–51. [12] Breen CM, Abernethy AP, et al. Conflict associated with decisions to limit lifesustaining treatment in intensive care units. JGIM 2001;16:283–9. [13] Stricker KH, Kimberger O, Schmidlin K, et al. Family satisfaction in the intensive care unit: what makes the difference? Intensive Care Med 2009;35:2051–9. [14] Baggs JG, Schmitt MA, Mitchell PH, et al. Association between nurse-physician collaboration and patient outcomes in three intensive care units. Crit care Med 1999;27(9):1991–8. [15] Mularski RA, Curtis JR, Billings JA, et al. Proposed quality measures for palliative care in the critically ill: a consensus from the Robert Wood Johnson Foundation Critical Care Workgroup. Crit Care Med 2006;34(11):S404–11. [16] 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(1):271–9. [17] Reini K, Fredrikson M, Oscarsson A. The prognostic value of the Modified Early Warning Score in critically ill patients: a prospective, observational study. Eur J Anaesthesiol 2012;29(3):152–7. [18] Field BE, Devich LE, Carlson RW. Impact of a comprehensive supportive care team on management of hopelessly ill patients with multiple organ failure. Chest 1989;96:353–6. [19] Campbell ML, Guzman JA. Impact of a proactive approach to improve end-of-life care in a medical ICU. Chest 2003;123:266–71. [20] Dowdy MD, Robertson C, Bander JA. A study of proactive ethics consultation for critically and terminally ill patients with extended lengths of stay. Crit Care Med 1998;26:252–9. [21] Campbell ML, Guzman JA. A proactive approach to improve end-of-life care in a medical intensive care unit for patients with terminal dementia. Crit Care Med 2004;32:1839–43. [22] Schneiderman LJ, Gilmer T, Teetzel HD, et al. Effect of ethics consultations on nonbeneficial life-sustaining treatments in the intensive care setting: a randomized controlled trial. JAMA 2003;290:1166–72. [23] Wall RJ, Curtis JR, Cooke CR, et al. Family satisfaction in the ICU differences between families of survivors and nonsurvivors. Chest J 2007;132.5:1425–33.
Please cite this article as: Kodali S, et al, Family experience with intensive care unit care: Association of self-reported family conferences and family satisfaction, J Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.03.012
Contents lists available at ScienceDirect
Journal of Critical Care journal homepage: www.jccjournal.org
Family experience with intensive care unit care: Association of self-reported family conferences and family satisfaction☆,☆☆ Sashikanth Kodali, MD a,⁎, Rebecca A. Stametz, DEd, MPH a, Amanda C. Bengier, BA a, Deserae N. Clarke, MPA a, Abraham J. Layon, MD b, Jonathan D. Darer, MD, MPH a a b
Division of Clinical Innovation, Geisinger Health System, Danville, PA Division of Critical Care, Geisinger Health System, Danville, PA
a r t i c l e
i n f o
Keywords: Family conference Family satisfaction End-of-life care Family communication Medical decision making
a b s t r a c t Purpose: A family conference is recommended as a best practice to improve communication in the intensive care unit (ICU), but this can be challenging given the setting. This study examined whether family members who reported that a family conference occurred had higher satisfaction than those who did not report that a conference was held. Materials and Methods: The study used a retrospective data analysis of family satisfaction based on family member's responses to a questionnaire. Families of all the patients admitted to ICUs of 2 quaternary hospitals in an integrated health system were surveyed. Results: The families of 457 patients who matched the inclusion and exclusion criteria were surveyed with a 55.6% response rate. Family satisfaction with decision making was significantly higher (83.6 vs 78.2, P = .0211) for families who reported that family conferences occurred. No significant difference in the satisfaction with care and overall satisfaction scores was found (84.2 vs 80.0, P = .10). Patients whose families reported a family conference were older and had higher mortality. Conclusion: This study confirms that families who report attending family conferences are more satisfied with decision making in the ICU. This study highlights the need to increase communication in ICUs. © 2014 Elsevier Inc. All rights reserved.
1. Introduction Yearly, more than 5.7 million Americans are admitted to an intensive care unit (ICU) [1]. Intensive care unit rates of mortality range from 10% to 15%, representing more than 540 000 US deaths each year [2]. Concerns with respect to quality of end-of-life care and communication with families in the ICU have been reported [3,4]. Most critically ill patients do not have decision-making capabilities and are often unable to communicate with the health care team because of their illness or the effect of treatments rendered, such as mechanical ventilation and sedation [5]. Family members of the critically ill patient are often required to act as surrogate decision makers, a role in which they may experience significant stress and emotional burden [6]. Improved communication with families of patients admitted to an ICU has been shown to reduce ICU length of stay, increase referrals to hospice, and, when appropriate, result in earlier withdrawal of life-sustaining interventions [7,8]. Effective
☆ All work was performed at Geisinger Health System, Danville PA, USA. ☆☆ Financial support was provided by Geisinger Clinic. There were no conflicts of interest for any of the authors. ⁎ Corresponding author. Geisinger Medical Center Division of Clinical Innovation, 100 N Academy Ave, Danville PA 17821. E-mail address: [email protected] (S. Kodali).
communication with families has also been shown to improve family satisfaction and psychological well-being [9]. Several areas have been identified as opportunities for improvement with regard to clinician-family communication in the ICU. Families often do not understand even basic information about the patient's diagnosis, prognosis, or treatment after 48 hours of admission to ICU [10], and physicians often miss opportunities to listen, acknowledge, and address emotions of families and do not explain surrogate decision making [11]. Poor physician communication and poor coordination of care are correlated with reduced family satisfaction with care [12,13]. Conversely, factors associated with improved family satisfaction include high-quality communication and collaboration between nursing staff and providers [14]. National guidelines with regard to palliative care for patients admitted to the ICU Include the performance of an interdisciplinary family conference for patients and families within 72 hours of admission to an ICU [15]. Despite the recognition of the positive contribution of interdisciplinary family conferences as a component of high-quality ICU care, the mechanism by which family satisfaction and well-being are improved has not been well characterized. As part of an initiative to improve the quality of palliative care for patients in the ICU, we performed a baseline survey of family experiences. We undertook analysis of the data derived from the survey and clinical data from the electronic health record (EHR) to explore the association of occurrence
http://dx.doi.org/10.1016/j.jcrc.2014.03.012 0883-9441/© 2014 Elsevier Inc. All rights reserved.
Please cite this article as: Kodali S, et al, Family experience with intensive care unit care: Association of self-reported family conferences and family satisfaction, J Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.03.012
2
S. Kodali et al. / Journal of Critical Care xxx (2014) xxx–xxx
of family conferences as reported by families with family-reported satisfaction with the care provided and the satisfaction with the decision-making process. 2. Materials and methods We conducted a retrospective cohort study using family satisfaction surveys and an EHR data pull to gather baseline satisfaction with care and the medical decision-making process for patients and families cared for in the ICU. The Family Satisfaction in the Intensive Care Unit Questionnaire24 (FS-ICU24) is a 24-item validated instrument for assessing family satisfaction in the ICU [16]. The instrument is available in public domain and can be accessed through the Canadian Researchers at the End of Life Network (http://www.thecarenet.ca). The instrument yields an overall satisfaction score (FS-Total) and has 2 subscales: satisfaction in the areas of decision making (FS-DM) and satisfaction with care provided (FS-Care). All 24 items, except for 1, use a 5-point Likert scale and ask the family member simple, easy to understand questions related to each subscale. In addition to these 24 items, there are 3 open-ended questions that solicit suggestions on how to make ICU care better. Family satisfaction scores were calculated using the validated scoring methodology of the FS-ICU24 questionnaire. An additional question that specifically asks if a family meeting was held during their loved one's hospital stay was added to the FS-ICU24. This question included a definition of what the family meeting was, specifically “A family conference is a meeting with the care staff to discuss how your loved one is doing and assist in care decisions.” Patient data were also pulled from the EHR to link to the survey. This included age, sex, disposition (death, home, transfer), length of stay, and presenting Modified Early Warning Score (MEWS). This research study was conducted in cooperation with the Geisinger Health System's (GHS's) Office of Research Protection. Approval to conduct the research was obtained from the institutional review board prior to beginning this study, under an expedited status.
contacted and who opted out prior to the mailing were excluded from the study, leaving a total of 453 NOK eligible for survey. A modified Dillman survey methodology was used for the 453 NOK whose addresses were obtained through 1 of the 2 methods described above. An initial notification letter was sent along with the FS-ICU24 and a $1 bill to improve response rates. The letter explained the study and provided a number that patients could call to opt out. The letter explicitly stated that the care received by their family member would not be affected by the decision to participate in the study. After 3 weeks, a reminder letter and an additional copy of the survey were sent to NOK who did not return the survey or opt out of the study. If there was no response 3 days after this mailing, a study team member called the NOK and verified if they had received the survey and were planning to respond. For those who returned a survey, consent was implied and thank you letters were sent to all NOK who completed and returned a survey. 2.3. Statistical analysis Analyses were performed with de-identified clinical and demographic data extracted from Geisinger's EHR. An univariate analysis of variance was performed to examine differences in each of the 3 satisfaction scores (FS-Total, FS-DM, and FS-Care) based on patient and family characteristics, ICU length of stay, hospital discharge disposition, MEWS, and whether families reported that a family conference occurred. For patients who had multiple episodes of ICU stay during a single hospitalization, cumulative ICU length of stay was used in analysis. Patient disposition was based on the final disposition from the hospital; the transfer category of the disposition included transfers to skilled nursing facilities, acute rehabilitation facilities, long-term acute care, and other acute care hospitals. The MEWS, which is shown to be a predictor of patient mortality and outcomes in ICU [17], was used as a surrogate marker for severity of illness at the time of admission. 3. Results
2.1. Study setting The GHS is an integrated health care delivery system based in Northeast and North Central Pennsylvania. Inpatient facilities in which the study took place include Geisinger Medical Center in Danville, Penn, and Geisinger Wyoming Valley in Wilkes Barre, Penn, both of which are tertiary/quaternary care inpatient hospitals. The 2 hospitals have 62 ICU beds and 161 full-time equivalents of nursing care. 2.2. Subject selection and identification Patients 18 years and older who were admitted to the Geisinger Medical Center and Geisinger Wyoming Valley ICUs who received ICU levels of care between January 1 and March 22, 2012, were included in the sampling pull. Patients with a research opt-out indicator were excluded, as were patients who lacked any next-of-kin (NOK) information in their record. For patients with multiple contacts, the NOK listed as the primary contact in the EHR were used for study purposes. Patients were identified by querying our clinical data warehouse, which stores all the clinical information from the EHR. A total of 634 patients were identified as meeting the original inclusion criteria. Only 554 of these patients had NOK name and telephone number listed in their record for contact purposes. Sequential methods were implemented to obtain mailing addresses for the family members. First, using NOK name and telephone number from the patient's EHR, a reverse look-up was performed using Whitepages. com. Next, if still unable to identify the NOK address, a telephone call was made by Survey Services at GHS' Center for Health Research. The NOK who were contacted by Survey Services were asked to participate in the study and provide an address. The NOK who could not be
Of the 453 surveys that were mailed, 16 NOK chose to opt out, leaving 437 records for analysis. Of the NOK surveyed, 243 responded, representing a 55.6% response rate. The subset of patients whose NOK responded to the survey had similar characteristics when compared with those who did not respond (Table 1). A total of 108 (44.4%) of
Table 1 Summary of patient characteristics Total Response No response P (N = 437) (n = 243; 55.6%) (n = 194; 44.4) Age category (patient; y) 18-39 40-59 60+ Patient sex Male Female Unknown Disposition Death Home Transfer ICU length of stay (h) 0-48 48-96 96+ Presenting MEWS score 0-2 3-5 6+ Unknown
.6195 47 (10.8) 142 (32.5) 248 (56.8)
29 (11.9) 76 (31.3) 138 (56.8)
18 (9.3) 66 (34.0) 110 (56.7)
250 (57.5) 185 (42.5) 2
149 (61.6) 93 (38.4) 1
101 (52.3) 93 (47.7) 1
66 (15.1) 200 (45.8) 171 (39.1)
42 (17.3) 113 (46.5) 88 (36.2)
24 (12.4) 87 (44.9) 83 (42.8)
178 (40.7) 90 (20.6) 169 (38.7)
100 (41.2) 52 (21.4) 91 (37.4)
78 (40.2) 38 (19.6) 78 (40.2)
216 (49.7) 168 (38.6) 51 (11.7) 2
117 (48.6) 94 (39.0) 30 (12.4) 2
99 (51.0) 74 (38.1) 21 (10.8) 0
.0528
.2255
.6099
.8205
Please cite this article as: Kodali S, et al, Family experience with intensive care unit care: Association of self-reported family conferences and family satisfaction, J Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.03.012
S. Kodali et al. / Journal of Critical Care xxx (2014) xxx–xxx
243 patients’ families reported that they had a family meeting. The average age of the patients whose families responded to the survey was 60.5 years, 38.4% were women, 17.3% died in hospital, and 37.4% of patients had ICU stays of 96 hours or longer. The family members who responded were more likely to be female and older, and the most common relationship listed with the patient was partner or spouse. Family satisfaction scores in the domain of decision making were significantly higher (83.6 vs 78.2, P = .0211) for families who reported that family conferences occurred (Table 2). The difference in decision-making score remained statistically significant, despite adjusting for confounding variables. No significant difference in the satisfaction with care (84.6 vs 82.1, P = .21) and overall satisfaction scores (84.2 vs 80.0, P = .0413) was found between families reporting a family meeting compared with families not reporting a family meeting. Spouses were more satisfied than parents with ICU care and the medical decision-making process. Using the same variables listed above in the Statistical Analysis section and treating the FS-DM score as continuous and binary using its 25th, 50th, and 75th quantiles as cutoff points, we examined their relationship with the FS-DM score. The logistic regression model was adjusted and found that the FS-DM score would be higher if patient's family reported that a family meeting occurred. The families who reported attending a family conference were compared with the families who reported that no conference occurred (Table 3). A χ 2 test showed that the family members reporting occurrence of a family conference were older. Similarly, the patients of the families who reported a family conference were older and had a higher rate of death when compared with patients whose families reported that no conference occurred.
Table 2 FS-ICU score ANOVA n
FS-Care
FS-DM
FS-Total
Mean ± SD P
Mean ± SD P
Mean ± SD P
Age category (family member; y) 18-39 15 78.8 (15.2) 40-59 100 81.6 (17.0) 60+ 123 84.6 (16.0) Unknown 5 82.5 (22.4) Age category (patient; y) 18-39 29 81.3 (14.3) 40-59 77 82.4 (16.6) 60+ 137 83.6 (17.0) Previous ICU experience Yes 142 82.8 (17.0) No 97 83.3 (15.7) Unknown 4 78.1 (23.2) Sex (family member) Male 66 83.9 (15.0) Female 171 82.8 (16.7) Unknown 6 75.0 (26.2) Relationship to patient Spouse/Partner 112 85.5 (14.7) Parent 35 78.8 (19.2) Sibling 26 84.5 (16.9) Child 38 81.8 (14.1) Other 28 78.3 (20.8) Unknown 4 78.1 (23.2) Family conference perception Yes 108 84.6 (17.7) No 120 82.1 (15.2) Unknown 15 77.2 (16.8) Outcome Death 42 82.7 (19.4) Home 113 81.9 (15.6) Transfer 88 84.3 (16.3) ICU length of stay (h) 0-48 82.5 (16.9) 48-96 83.9 (14.6) 96+ 82.8 (17.3)
.43 73.6 81.2 79.7 84.5
(18.5) (17.6) (18.4) (19.5)
.45
.76 78.0 (18.6) 76.9 (19.0) 82.2 (17.3)
(15.1) (16.7) (15.5) (21.0)
.58
.10
79.9 (15.2) 80.0 (16.1) 83.2 (16.2)
.33
.83 79.8 (17.9) 80.3 (18.4) 80.6 (20.1)
.98
81.5 (16.4) 82.3 (15.6) 79.2 (21.7)
.88
.45 81.5 (15.2) 79.4 (19.1) 80.6 (20.1)
.73
82.7 (14.5) 81.4 (16.7) 82.5 (20.2)
.85
.19 83.2 73.6 80.4 81.3 72.7 80.6
.0279 84.6 77.1 83.9 81.6 74.7 79.2
(15.5) (21.4) (15.5) (15.6) (24.9) (20.1)
76.7 81.4 82.7 83.3
(14.4) (18.4) (13.4) (13.7) (21.6) (21.7)
3
Table 3 Univariate analysis of family conference perception Family conference perception
Age category (family member; y) 18-39 40-59 60+ Unknown Age category (patient; y) 18-39 40-59 60+ Previous ICU experience Yes No Unknown Sex (family member) Male Female Unknown Relationship to patient Spouse/Partner Parent Sibling Child Other Unknown Disposition Death Home Transfer ICU length of stay (h) 0-48 48-96 96+ Presenting MEWS score 0-2 3-5 6+ Unknown
Yes
No
4 (3.8) 41 (38.7) 61 (57.6) 2
11 (9.3) 55 (46.6) 52 (44.1) 2
7 (6.5) 32 (29.6) 69 (63.9)
20 (16.7) 42 (35.0) 58 (48.3)
61 (57.0) 46 (43.0) 1
75 (63.6) 43 (36.4) 2
35 (32.7) 72 (67.3) 1
26 (22.2) 91 (77.8) 3
56 (52.3) 12 (11.2) 10 (9.4) 19 (17.8) 10 (9.4) 1
52 (44.1) 20 (17.0) 15 (12.7) 17 (14.4) 14 (11.9) 2
31 (28.7) 37 (34.3) 40 (37.0)
7 (5.8) 68 (56.7) 45 (37.5)
44 (40.7) 22 (20.4) 42 (38.9)
53 (44.2) 27 (22.5) 40 (33.3)
52 (48.6) 41 (38.3) 14 (13.1) 1
60 (50.4) 43 (36.1) 16 (13.5) 1
P
.0673
.0187
.3156
.0782
.4938
b.0001
.6823
.9439
4. Discussion
.0413
.21 83.6 (18.0) 77.2 (17.5) 76.1 (19.7)
.0211 84.2 (17.2) 80.0 (14.9) 77.7 (15.3)
.10
.6
78.8 (21.4) 79.9 (16.0) 80.8 (18.9)
.83
81.1 (19.8) 80.9 (14.5) 83.1 (16.3)
.63
.89 80.8 (16.9) 78.3 (19.4) 80.1 (18.7)
.72
81.5 (16.1) 82.4 (14.1) 81.8 (17.3)
.95
Family conferences offer a forum for bringing patients, family members, and health care providers together to share prognostic information, set goals of care, and offer families support. Several studies have evaluated the impact of communication strategies that include proactive family conferences upon clinical outcomes. These interventions have tended to be performed on select, high-acuity or specific patient populations including multiorgan failure [18], cerebral ischemia [19], poor prognosis based on predicted mortality [8], duration of mechanical ventilation [20], dementia [21], and in situations where conflict is perceived [22], although national guidelines recommend family conferences in all patients admitted to the ICU. This study, looking at the experiences of families of all patients admitted to the ICU, confirms that families who report attending family conferences are more satisfied with decision making in the ICU. It is interesting, although not surprising, that the mortality rate of patients whose families reported family conference occurrence is higher than that of patients whose families reported that family conference did not occur. The higher mortality and older age of the patient group whose families reported occurrence of a family meeting suggest that there may be increased focus on communication with family members of patients in end-of-life situations. In another study of 539 family members, families of patients who survived ICU stay have lower satisfaction compared with family members of the deceased [23]. However, our study did not examine the differences in family satisfaction scores between the patients who died in the ICU and the patients who survived, as this is not the primary scope of the
Please cite this article as: Kodali S, et al, Family experience with intensive care unit care: Association of self-reported family conferences and family satisfaction, J Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.03.012
4
S. Kodali et al. / Journal of Critical Care xxx (2014) xxx–xxx
study. Attending to the needs of families in the ICU is an important component of high-quality care, yet the increased burden of regularly getting all stakeholders together to meet with families is a substantial barrier to implementation. Even in structured interventions, designed to reliably deliver family conferences, rates of family meetings reach less than 70% of families [9]. Our results show an association between the presence of a family meeting as reported by the NOK/family members and increased satisfaction with the process of medical decision making. It is known that family meetings improve family well-being; however, this may be in fact resulting from the process of engaging families, specifically in the decision-making process. If scheduling a multidisciplinary family conference is overwhelming, other means of family engagement should be explored and their impact be studied. In addition, our analysis failed to demonstrate a difference in the overall satisfaction scores based on the occurrence of a family meeting as reported by family members/NOK. A review of the comments made to open-ended questions demonstrated that family members expressed their dissatisfaction with nonclinical aspects of the hospital stay. This included things like the quality of waiting room facilities, availability of information about food and lodging, and staff interactions outside patient rooms. Hospitals seeking to improve family satisfaction with care may need to look to multiple sources of dissatisfaction, rather than focusing only on care-related concerns. Another possible cause for the lack of difference may be that the self-reported rate of the family conference is very low. The system currently has no way of tracking the actual occurrence of the family conference in the medical record. This survey was a baseline measure for an intervention that will be put in place to improve communication in the ICU. Part of the intervention includes an objective way to record family conferences in the chart, in addition to making sure the meetings are as comprehensive as possible. In the postsurvey, we will be able to compare the self-reported rate of family conference with the actual rate. 4.1. Limitations Our study is limited in several ways. It focused on a fairly homogenous population related to ethnicity and geographic location. The study was performed in quaternary care settings only. Hence, the results may not be generalizable to other populations. Also, the response rate of 55.6% is relatively low, and although the responders and nonresponders were similar from a demographic perspective, there is still the potential for a response bias. In addition, we had a small population from which to survey NOK. Although this is not unusual for ICU studies, it limits the reliability of our results. The study design also allows us to evaluate only associations and not causal relationships. Moreover, our study used self-report of family conference as a means to identify whether a family conference occurred. This may be subject to recall bias. Perhaps, the family members who respond to whether a family meeting occurred or not based on our definition included in the survey may be more likely to report that they are more satisfied with their engagement with decision making. Although there is currently not a rigorous way to gather information on whether a multidisciplinary family conference occurred or not from the medical record due to unreliable documentation practices, this should be an important consideration in future studies. Finally, we surveyed during a limited time frame, which may impact results due to potential temporal variance in ICU conditions.
5. Conclusions A family conference occurrence, as reported by NOK of patients admitted to the ICU, is associated with increased satisfaction with decision making. This study emphasizes the need to increase communication with families in ICUs and the use of multidisciplinary family conference as an important modality of communication. Further studies are needed to understand the most reliable workflow for holding multidisciplinary family conferences. Other modalities besides a multidisciplinary family conference that may improve the communication with families should be explored.
References [1] Kersten A, Millbrandt EB, Rahim MT. How big is critical care in the U.S.? Crit Care Med 2003;31(12):A8–A8. [2] Angus DC, Barnato AE, Linde-Zwirble WT, et al. Use of intensive care at the end of life in the United States: an epidemiologic study. Crit Care Med 2004;32(3):638–43. [3] Azoulay E, Pochard F, Chevret S, et al. Meeting the needs of intensive care unit patient families: a multicenter study. Am J Respir Crit Care Med 2001;163:135–9. [4] Curtis JR, Rubenfeld GD. Improving palliative care for patients in the intensive care unit. J Palliat Med 2005;854:8840–54. [5] Prendergast TJ, Luce JM. Increasing incidence of withholding and withdrawal of life support from the critically ill. Am J Respir Crit Care Med 1997;155:15–20. [6] Pochard F, Darmon M, Fassier T, et al. Symptoms of anxiety and depression in family members of intensive care unit patients before discharge or death: a prospective multicenter study. J Crit Care 2005;20:90–6. [7] Norton SA, Hogan LA, Holloway RG, et al. Proactive palliative care in the medical intensive care unit: effects on length of stay for selected high-risk patients. Crit Care Med 2007;35:1530–5. [8] Lilly CM, De Meo DL, Sonna LA, et al. An intensive communication intervention for the critically ill. Am J Med 2000;109:469–75. [9] Lautrette A, Darmon M, Megarbane B, et al. A communication strategy and brochure for relatives of patients dying in the ICU. N Engl J Med 2007;356:469–78. [10] Azoulay E, Chevret S, Leleu G, et al. Half the families of ICU patients experience inadequate communication with physicians. Crit Care Med 2000;83044:3044–9. [11] Curtis JR, Burt RA. Point: the ethics of unilateral “do not resuscitate” orders: the role of “informed assent”. Chest 2007;132:748–51. [12] Breen CM, Abernethy AP, et al. Conflict associated with decisions to limit lifesustaining treatment in intensive care units. JGIM 2001;16:283–9. [13] Stricker KH, Kimberger O, Schmidlin K, et al. Family satisfaction in the intensive care unit: what makes the difference? Intensive Care Med 2009;35:2051–9. [14] Baggs JG, Schmitt MA, Mitchell PH, et al. Association between nurse-physician collaboration and patient outcomes in three intensive care units. Crit care Med 1999;27(9):1991–8. [15] Mularski RA, Curtis JR, Billings JA, et al. Proposed quality measures for palliative care in the critically ill: a consensus from the Robert Wood Johnson Foundation Critical Care Workgroup. Crit Care Med 2006;34(11):S404–11. [16] 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(1):271–9. [17] Reini K, Fredrikson M, Oscarsson A. The prognostic value of the Modified Early Warning Score in critically ill patients: a prospective, observational study. Eur J Anaesthesiol 2012;29(3):152–7. [18] Field BE, Devich LE, Carlson RW. Impact of a comprehensive supportive care team on management of hopelessly ill patients with multiple organ failure. Chest 1989;96:353–6. [19] Campbell ML, Guzman JA. Impact of a proactive approach to improve end-of-life care in a medical ICU. Chest 2003;123:266–71. [20] Dowdy MD, Robertson C, Bander JA. A study of proactive ethics consultation for critically and terminally ill patients with extended lengths of stay. Crit Care Med 1998;26:252–9. [21] Campbell ML, Guzman JA. A proactive approach to improve end-of-life care in a medical intensive care unit for patients with terminal dementia. Crit Care Med 2004;32:1839–43. [22] Schneiderman LJ, Gilmer T, Teetzel HD, et al. Effect of ethics consultations on nonbeneficial life-sustaining treatments in the intensive care setting: a randomized controlled trial. JAMA 2003;290:1166–72. [23] Wall RJ, Curtis JR, Cooke CR, et al. Family satisfaction in the ICU differences between families of survivors and nonsurvivors. Chest J 2007;132.5:1425–33.
Please cite this article as: Kodali S, et al, Family experience with intensive care unit care: Association of self-reported family conferences and family satisfaction, J Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.03.012
