Comfort care is a main component of the hospice care field; however, the concept is not typically used in the intensive care setting. Medical staff of ICUs strive to provide a quality of care for critically ill patients that supports healing and prevents mortality. However, they also encounter frequent patient deaths, which are often associated with withholding or withdrawing life-sustaining care (1–4). When the dying process becomes inevitable for a patient, palliative care becomes more important than curative care (5). As patients and their families are often afraid of preparing for death, appropriate comfort care in the period before death is an important aspect of intensive care.
Patients who are dying in the ICU are generally not able to participate in the decision-making process of their care, due to treatment that inhibits their communication (e.g., intubation and sedation). Patients’ families may also find it difficult to make informed decisions because of a lack of medical knowledge and emotional instability. This can be worsened by poor comfort care for the patient, which can have detrimental psychologic effects on both families and caregivers (6). Thus, ICU medical staff should understand the emotions experienced by patients’ families during the dying process and can play an important role in providing best comfort care.
Quality of death is subjectively determined by numerous factors. Patrick et al (7) developed Quality of Dying and Death (QODD), an instrument composed of 31 items covering several conceptual domains. As it is usually not feasible for patients in an ICU to respond to the QODD directly, a modified version of the questionnaire is available for use by ICU medical staff. However, the quality of death in ICUs, which is assessed by QODD questionnaire, has been rarely reported in Asian countries. In a previous study, we validated a translated Korean-language version of the medical staff QODD questionnaire (8). Furthermore, Korea is currently in the early stage after the implementation of the “well-dying” law, which seems to have a considerable effect on practice. This study aimed to understand the status of comfort care and quality of death in Korean ICUs as perceived by medical staff, and to elucidate factors affecting patient quality of death.
MATERIALS AND METHODS
Study Design and Participants
We conducted a prospective survey from June 2016 to May 2017 in four medical ICUs in the Republic of Korea: two tertiary teaching hospitals affiliated with the Seoul National University College of Medicine (Seoul National University Bundang Hospital, Seoul National University Hospital) and two secondary hospitals (Seoul Metropolitan Government-Seoul National University, National Medical Center).
We screened the data of deceased patients in the participating Korean medical ICUs and recruited the relevant medical staff. Attending physicians, fellows, residents, and nurses who were responsible for or caring for the patient at time of death were asked to complete QODD questionnaires within 48 hours of the patient’s death. Completed QODD questionnaires were stored in designated storage boxes in each ICU until the end of the study. All medical staff participants were also asked to complete a survey of their personal information.
The following patient information was obtained from medical records: sex, age, medical history, reason for ICU admission, hospital day, length of ICU stay, life-support equipment, receiving resuscitation within 24 hours before death, placement of do-not-resuscitate (DNR) order, and medication (e.g., sedatives, analgesics, and inotropes) for 48 hours before death.
We used a translated Korean version of the QODD questionnaire, Version 3.2 (9), which is designed to be self-administered by any healthcare professional who cared for a patient who died in a hospital or ICU setting. We obtained permission from the original questionnaire developers to use the questionnaire prior to commencement of the study. The detailed process of transcultural adaptation and validation of the Korean-language QODD Version 3.2 has been previously published (8). The QODD questionnaire version 3.2 consists of 20 items in three domains that are rated from 0 (terrible experience) to 10 (excellent experience). If the staff completing the questionnaire felt that they could not assess a patient’s experience in any of the items, they checked a box labeled “I do not know” which, along with any item that had no response, was deemed an invalid response. The QODD score for each patient was calculated by summing all valid ratings of the 20 designated items, dividing the total by the number of valid items, and multiplying by 10; thus, QODD scores ranged from 0 to 100 and a higher score indicated a better quality of death as perceived by ICU staff.
All data are presented as mean values ± sd for continuous variables and n (%) for categorical variables. SPSS 22.0 (IBM, Armonk, NY) was used for the statistical analyses. We performed the Student t test or Mann-Whitney U test to compare the mean value of each item and to compare each patient’s QODD score as assessed by doctors with that assessed by nurses. To analyze factors affecting quality of death, Fisher exact test or Pearson chi-square test was performed for categorical variables; Pearson correlation coefficient was calculated for continuous variables. A multiple linear regression analysis was used for identifying independent predictors of the QODD score among factors that were identified as correlating with a higher QODD score. A p value of less than 0.05 was considered statistically significant.
This study was approved by the institutional review board of each participating hospital. Within 48 hours of a patient’s death, written informed consent was obtained from responsible medical staff at time of patient’s death. All patient data were anonymized, and the requirement for patient consent was waived. All aspects of this study were conducted in compliance with the Declaration of Helsinki.
Patients and ICU Staff
From June 2016 to May 2017, 177 ICU staff that had cared for 255 deceased patients agreed to participate in this study. A total of 416 questionnaires were obtained. The same medical staff frequently cared for multiple deceased patients; thus, we limited the number of questionnaires filled by each healthcare worker per month: three per physician and two per nurse. Among 255 patients, the questionnaires of 102 patients were completed by only one of the doctors (54 patients) or one of the nurses (48 patients). One doctor and one nurse responded for 152 patients. Two doctors and one nurse responded for one patient.
The baseline characteristics of 255 deceased patients are shown in Table 1. Mean patient age was 66.6 ± 14.1 years, 62.4% were men, and 39.6% already had cancer. The most common reason for ICU admission was respiratory failure (58.0%). The mean Acute Physiology and Chronic Health Evaluation II score at ICU admission was 29.5 ± 15.0. At time of death, 89.4% of patients were receiving mechanical ventilation. Cardiopulmonary resuscitation (CPR) was performed for 18.8% of patients within 24 hours before death. The characteristics of participating medical staff are shown in Supplemental Table 1 (Supplemental Digital Content 1, http://links.lww.com/CCM/E652).
QODD Items and Score
Total QODD scores and individual scores for each questionnaire item are detailed in Table 2. The perceived overall quality of death was a mean score of 31.3 on a 0–100 scale (higher score corresponding to better quality of death). Quality of death perceived by nurses was better than that perceived by doctors (33.1 ± 18.4 vs 29.7 ± 15.3; p = 0.042). Regarding individual items of the QODD, “anyone present at moment of death” had the highest individual score from both doctors (5.8 ± 3.0) and nurses (5.9 ± 3.1). Similarly, the mean score of “having pain under control” was relatively higher than that of other items and showed no significant difference between doctors (5.0 ± 2.5) and nurses (5.2 ± 2.5) (p = 0.471). The lowest score was given for “being able to feed oneself” by both doctors (1.1 ± 1.8) and nurses (1.1 ± 2.0). In terms of patient’s autonomy for determining their death, scores for the item “discussion with doctors about wishes” were low when rated by both doctors (3.0 ± 2.7) and nurses (2.7 ± 2.9).
Factors Affecting QODD Score
Factors identified in this study as associated with better quality of death (i.e., higher QODD score) are shown in Table 3. All medical staff perceived a higher quality of death for patients who received analgesics within 24 hours before death, compared with patients who did not (33.4 ± 17.5 vs 28.4 ± 15.9; p = 0.003). However, quality of death was perceived to be lower in patients who received CPR compared with those who did not (24.3 ± 15.7 vs 33.0 ± 16.9; p < 0.001) and was also lower in patients who received inotropes compared with those who did not (30.7 ± 16.5 vs 38.9 ± 20.3; p = 0.008) within 24 hours before death. Quality of death positively correlated with age, although the correlation coefficient was low (R = 0.139; p = 0.005).
A multivariate linear regression analysis showed independent predictors of quality of death (Table 4). Performing CPR and using inotropes within 24 hours before death were significantly associated with lower quality of death as assessed by both doctors and nurses, while use of analgesics was associated with higher quality of death. In doctors’ assessments, use of inotropes and shorter ICU stay predicted a lower quality of death, whereas in nurses’ assessments, performing CPR within 24 hours before death predicted a lower quality of death and use of analgesics predicted a higher quality of death.
Medical Staff Comments for Improving Quality of Death
Participating medical staff was requested to suggest comments for improving patients’ quality of death. Multiple answers were allowed, and we corrected a total of 129 comments (Supplemental Table 2, Supplemental Digital Content 2, http://links.lww.com/CCM/E653). Written comments were sorted across five distinct categories: decision-making and DNR, communication with patients and their families, treatment and care, staff, and others. The majority (53.5%) of comments were given for decision-making and DNR.
This study aimed to understand the status of quality of death in Korean ICUs as perceived by medical staff, as well as the factors that affect the quality of death. To the best of our knowledge, this is the first study to use a transculturally adapted and validated version of the QODD in a multicenter setting among Asian countries. The mean overall QODD score as assessed by both doctors and nurses in the present study was 31.3 of a possible total score of 100, which is considerably lower than that reported in the U.S. and Dutch studies (10–13). Performing CPR or using inotropes within 24 hours of death were most strongly associated with poor quality of death, whereas using analgesics during the 24 hours before death was associated with better quality of death.
Interestingly, there were differences between doctors’ and nurses’ assessments of factors associated with quality of death. Doctors perceived that patients receiving inotropic treatment had poor quality of death in our study. As doctors often consider that the use of inotropes at the end of life might only unnecessarily prolong the duration of life without any clinical improvement, doctors tend to withdraw or decrease the dose of inotropic treatments for patients who are in progressive multiple organ failure and predicted death within several hours (14). In contrast to doctors, nurses were more likely to perceive poor quality of death among patients who received CPR within 24 hours of death, which is consistent with results from a previous study (10). Despite immediate survival after CPR, although low probability of survival (15), which is commonly accompanied by traumatic complications, can be expected (1617), nurses may be under considerable stress from participating in what they perceive as meaningless care. In addition, nurses gave overall a higher score than did doctors (Table 2). This could be because nurses, in comparison to doctors, are more continuously in contact with patients and their families. Nurses gave higher ratings for three items on the QODD: “being able to breath comfortably,” “feeling at peace with dying,” and “visits from religious advisor.” Thus, nurses had more time to care closely for patients despite the fact that they work shifts rather than in continuity and better understood experiences of dying patients in terms of spiritual link.
We found that patient quality of death in Korean ICUs, as assessed by medical staff, was notably lower than that reported in previous U.S. and Dutch studies (10–13) (Fig. 1). All 20 items of the QODD in this study received relatively low scores. In particular, “being able to feed oneself,” “being able to laugh and smile,” and “clearing up bad feelings” each received a score of less than two points. Several items related to “patient’s preparation for death” and “spending time with family” were also scored low; an item in particular, “discussion with doctors about wishes,” received a considerably lower score (mean score, 3.0 from doctors; 2.7 from nurses) than reported in previous studies from the United States (mean score: 8.64 from attending doctors, 5.93 from residents, and 7.16 from nurses) (11) and the Netherlands (median score: 9 from physicians and 8 from nurses) (12). These findings suggest that the critical care environment in Korea does not support patient-centered decision-making.
Why the medical staff consider patients’ QODD as exceptionally low in Korea is probably affected by the Boramae Hospital incident (1819). In 1997, a brain surgical patient died after being released from the Boramae hospital at the request of his legal guardian. A doctor ordered discharge against medical advice according to the continuous and obstinate request of his wife. The Korean Supreme Court convicted doctors for aiding and abetting homicide. Since this incident, doctors in Korea have been forced to continue a futile life-sustaining treatment for terminal patients due to the possibility of a legal problem. Thus, patients experiencing pain, who might otherwise be able to avoid this, and their families, have to endure great mental distress and bear economic burdens (19). Patient-centered decisions (i.e., withholding or withdrawing a life-sustaining treatment) have been rarely considered in practice, in Korea (20). Therefore, there have been a growing number of voices asking whether life-prolonging treatments are simply to delay death. Finally, in February 2018, Korea implemented a “well-dying” law, which allows terminally ill patients to withhold or withdraw the futile management; and allows doctors to formally accept advanced medical directives. One year after the enforcement, Ministry of Health and Welfare in Korea announced that approximately 35,000 terminally ill patients discontinued life-sustaining care in accordance with their or families’ wishes (21).
We conducted additional pilot survey of QODD among 80 medical staff of 43 patients who died between October 2018 and December 2018, 8 months after the implementation of the “well-dying” law. Generally, no difference was observed before (preperiod) and after (postperiod) the implementation of the law; however, the proportion of ICU admissions immediately following CPR in the ward significantly decreased (Supplemental Table 3, Supplemental Digital Content 3, http://links.lww.com/CCM/E654). The total score of QODD perceived by all medical staff showed tendency to increase (31.3 ± 17.0 vs 35.0 ± 20.0; p = 0.087). Mean scores of three items showed improvement during the postperiod: “Clearing up bad feelings” (1.7 ± 2.1 vs 2.3 ± 2.6; p = 0.038), “Visits from religious advisor” (2.3 ± 2.9 vs 3.2 ± 3.3; p = 0.030), and “Discussion with doctors about wishes” (2.9 ± 2.8 vs 4.2 ± 3.0; p = 0.003) (Supplemental Table 4, Supplemental Digital Content 4, http://links.lww.com/CCM/E655). The total score of QODD perceived by either doctors (29.7 ± 15.3 vs 30.7 ± 18.2; p = 0.694) or nurses (33.1 ± 18.4 vs 40.5 ± 21.0; p = 0.055) also showed tendency to increase. The mean score of “Discussion with doctors about wishes” had improved according to the doctors’ assessments (3.0 ± 2.7 vs 4.5 ± 3.2; p = 0.004). The mean scores of three items had improved according to the nurses’ assessments: “Clearing up bad feelings” (1.8 ± 2.2 vs 3.4 ± 3.0; p = 0.002), “Visits from religious advisor” (2.7 ± 3.1 vs 5.0 ± 3.4; p = 0.002), and “Spiritual service before death” (2.4 ± 3.0 vs 4.0 ± 3.2; p = 0.015) (Supplemental Table 5, Supplemental Digital Content 5, http://links.lww.com/CCM/E656; and Supplemental Table 6, Supplemental Digital Content 6, http://links.lww.com/CCM/E657).
Although Korea is currently in the early stage after the implementation of the “well-dying” law, we are aware that it may have a considerable effect on practice. Many patients have been treated, withheld, or withdrawn from life-sustaining care, according to their wishes. There have been active discussions between patients’ families and medical staff at the ICU field, of advanced medical directives. Results of our additional pilot surveys support these changes. We believe that the legislation seems to have a positive effect on the quality of death in the ICU. Future studies should evaluate the quality of death after the well-dying law has been in place for some years.
Medical staff’s unfamiliarity with comfort care in critical care settings can be another reason for poor quality of death in Korea. In the present study, although 89.4% of patients were receiving mechanical ventilation at time of death, a relatively small proportion of patients received sedatives (48.9%) or analgesics (60.3%) within 24 hours of death. Medical staff in this study suggested a need for more active pain control or sedation (7.0% of written comments) and a need for avoidance of invasive procedure or treatment (3.9% of written comments). Furthermore, medical staff did not have enough opportunities of comfort care (Supplemental Table 1, Supplemental Digital Content 1, http://links.lww.com/CCM/E652). As supported by the study conducted by Billings et al (22), extended coursework for students and trainees in providing comfort care would contribute greatly to improving quality of death among dying ICU patients.
Also, our study identified important aspects requiring further improvement. Items related to “patient’s preparation for death” and “family” (“spending time with family, friends,” “being touched or hugged by loved ones,” “saying goodbye to loved ones,” “visits from religious advisor,” and “spiritual service before death”) were scored considerably low. The hospitals that participated in the present study had visiting policies that allowed for approximately 30 minutes of patients’ visitation twice a day, except in certain circumstances. For several reasons, such as risk of infection, interference with care, and managing patient stress, many institutions tend to limit access to the ICU for patient’s family and friends (23). However, there is no scientific basis for these limits (24–26). More liberal visiting policies may be needed to improve quality of death in ICUs, especially in cases where a decision has been made to provide a patient with only palliative care. Preparing separate places for patients and their families may also improve comfort care for patients in the ICU.
This study has some limitations. First, as our study was performed in the medical ICUs of acute care hospitals, our findings may not be generalizable for all ICUs. Results of this study represent patients’ experiences in both secondary hospital ICUs and tertiary referral hospital ICUs, where medical staff typically experience high rates of patient mortality. Thus, our findings may still be valuable despite this limitation. Second, selection bias may have affected our findings, as most doctors who participated in this study were residents. A previous study reported that quality of death was assessed differently by residents and attending physicians (11). However, as residents generally have more direct contact with dying patients than do attending physicians, the assessments in this study may be useful for improving quality of death in ICUs.
We found that performing CPR and using inotropes within 24 hours of death were associated with poorer quality of death, whereas using analgesics 24 hours before death was associated with better quality of death, as assessed by medical staff. Nurses’ assessments of patient quality of death were better than doctors’ assessments, but overall quality of death assessed by healthcare professionals in Korean medical ICUs was considerably lower than that reported in other countries. Provision of appropriate comfort care, avoidance of unnecessary life-sustaining care, and permission for frequent visits from patients’ families may be the most useful interventions that should be targeted for improving quality of death. Also, it is expected that the new legislation would positively affect the quality of death in Korean ICUs.
We sincerely thank all the medical staff who participated in the survey. And we specially thank Dr. Tae Yun Park, MD. She made a large contribution in collecting additional data and writing revised.
1. Vincent JL, Marshall JC, Namendys-Silva SA, et al.; ICON investigators: Assessment of the worldwide burden of critical illness: The Intensive Care Over Nations (ICON) audit. Lancet Respir Med 2014; 2:380–386
2. Adhikari NK, Fowler RA, Bhagwanjee S, et al. Critical care and the global burden of critical illness in adults. Lancet 2010; 376:1339–1346
3. Prendergast TJ, Puntillo KA. Withdrawal of life support: Intensive caring at the end of life. JAMA 2002; 288:2732–2740
4. Ferrand E, Robert R, Ingrand P, et al.; French LATAREA Group: Withholding and withdrawal of life support in intensive-care units in France: A prospective survey. French LATAREA Group. Lancet 2001; 357:9–14
5. Cook D, Rocker G. Dying with dignity in the intensive care unit. N Engl J Med 2014; 370:2506–2514
6. Connor SR, Tecca M, LundPerson J, et al. Measuring hospice care: The National Hospice and Palliative Care Organization National Hospice Data Set. J Pain Symptom Manage 2004; 28:316–328
7. Patrick DL, Engelberg RA, Curtis JR. Evaluating the quality of dying and death. J Pain Symptom Manage 2001; 22:717–726
8. Cho JY, Lee J, Lee SM, et al. Transcultural adaptation and validation of quality of dying and death questionnaire in medical intensive care units
in South Korea. Acute Crit Care 2018; 33:95–101
10. Hodde NM, Engelberg RA, Treece PD, et al. Factors associated with nurse assessment of the quality of dying and death in the intensive care unit. Crit Care Med 2004; 32:1648–1653
11. Levy CR, Ely EW, Payne K, et al. Quality of dying and death in two medical ICUs: Perceptions of family and clinicians. Chest 2005; 127:1775–1783
12. Gerritsen RT, Hofhuis JGM, Koopmans M, et al. Perception by family members and ICU staff of the quality of dying and death in the ICU: A prospective multicenter study in The Netherlands. Chest 2013; 143:357–363
13. Gerritsen RT, Koopmans M, Hofhuis JG, et al. Comparing quality of dying and death perceived by family members and nurses for patients dying in US and Dutch ICUs. Chest 2017; 151:298–307
14. Hall RI, Rocker GM. End-of-life care in the ICU: Treatments provided when life support was or was not withdrawn. Chest 2000; 118:1424–1430
15. Lemiale V, Dumas F, Mongardon N, et al. Intensive care unit mortality after cardiac arrest: The relative contribution of shock and brain injury in a large cohort. Intensive Care Med 2013; 39:1972–1980
16. Kashiwagi Y, Sasakawa T, Tampo A, et al. Computed tomography findings of complications resulting from cardiopulmonary resuscitation. Resuscitation 2015; 88:86–91
17. Kim MJ, Park YS, Kim SW, et al. Chest injury following cardiopulmonary resuscitation: A prospective computed tomography evaluation. Resuscitation 2013; 84:361–364
18. Shin DC. Perspectives on death: Korea’s first court decision supporting death with dignity, its meaning and future prospect. Japan Med Assoc J 2009; 52:2
20. Shin K, Mok JH, Lee SH, et al. The current status of medical decision-making for dying patients in a medical intensive care unit: A single-center study. Korean J Crit Care Med 2014; 29:160–165
22. Billings ME, Engelberg R, Curtis JR, et al. Determinants of medical students’ perceived preparation to perform end-of-life care, quality of end-of-life care education, and attitudes toward end-of-life care. J Palliat Med 2010; 13:319–326
23. Liu V, Read JL, Scruth E, et al. Visitation policies and practices in US ICUs. Crit Care 2013; 17:R71
24. Malacarne P, Corini M, Petri D. Health care-associated infections and visiting policy in an intensive care unit. Am J Infect Control 2011; 39:898–900
25. Fumagalli S, Boncinelli L, Lo Nostro A, et al. Reduced cardiocirculatory complications with unrestrictive visiting policy in an intensive care unit: Results from a pilot, randomized trial. Circulation 2006; 113:946–952
26. Berwick DM, Kotagal M. Restricted visiting hours in ICUs: Time to change. JAMA 2004; 292:736–737