Impact of end-of-life care quality in intensive care units on adjustment of bereaved family members

Project: National Health Research InstitutesNational Health Research Institutes Grants Research

Project Details


Death in an intensive care unit (ICU) is common and costly. One in five Americans dies in an ICU, accounting for 11-20% of all healthcare costs, with approximately 25% of Medicare expenditures spent in the final month of life on ICU care. Comparable data are not available for Taiwan. Such high ICU expenditures do not translate into high quality ICU care at end of life (EOL). Patients dying in ICUs suffer significant pain or other distressing symptoms and receive high intensity, often futile life-sustaining treatments, which patients do not prefer. Moreover, ICU patients’ families are left with tremendous emotional distress long after the patient’s death. Therefore, EOL care in ICUs has been recognized worldwide as a priority area for ICU quality improvement to promote patients’ and families’ values and preferences, improve the quality of death and dying, reduce bereavement grief, and counteract escalating and unsustainable ICU-care expenditures. Such a priority is relevant to Taiwanese healthcare systems and critical to Taiwan’s National Health Insurance. The proposed study addresses several substantial gaps in current knowledge about quality of EOL care in ICUs. First, few studies worldwide have explored EOL-care quality in ICUs and bereavement adjustment for ICU decedents’ family members, with none from Taiwan, despite its having the largest number of ICU beds per 100,000 population worldwide. Second, the few existing studies have substantial methodological insufficiencies. For example, almost all studies are cross-sectional, with small samples, low response rates, and varied intervals between measurements of perceived EOL-care quality in ICUs and psychological distress during bereavement. The only longitudinal study, a design needed to assess dynamic changes in psychological adjustments during bereavement, had high attrition rates. No study has prospectively identified EOL-care quality in routine ICU practice. In addition, bereavement adjustment has been inadequately conceptualized for bereaved family members of ICU decedents, i.e., little is known of their quality of life (QOL) as an outcome variable, their prior psychiatric history and emotional problems have been omitted as confounders, and their anxiety, depression, posttraumatic stress disorder (PTSD), and complicated grief have been treated as independent psychological problems despite the well-recognized comorbidity of depression and anxiety as well as depression and PTSD. Therefore, the specific aims of this 5-year prospective, longitudinal study are to (1) explore the EOL care quality received by patients who died in ICUs (decedents), (2) describe these decedents’ quality of death and dying, (3) identify ICU decedents’ family members’ distinct patterns of psychological distress (i.e., symptoms of anxiety, depression, PTSD, and complicated/prolonged grief) over the first 2 years of bereavement, and (4) examine the associations of EOL-care quality with QOL and distinct patterns of psychological distress of ICU decedents’ family members over their first 2 years of bereavement. Data for Aim 1 and 2 will be obtained by reviewing medical records and surveying 332 consecutive ICU decedents’ family members, respectively. Data on EOL-care practices in ICUs; family members’ perceptions of the quality of decedents’ death and dying; and bereaved families’ QOL and psychological outcomes, i.e., anxiety and depressive symptoms, PTSD, complicated/prolonged grief will be collected by well-trained, experienced ICU nurses using well-established instruments. Each indicator of EOL-care practices in ICUs will be identified by daily review of patients’ medical records until their death. Bereaved family members will be interviewed at 1, 3, 6, 13, 18, and 24 months after patient death, with family members’ perceived quality of dying and death being evaluated at the first month post-death only. EOL-care quality received by ICU decedents (Aim 1) and ICU decedents’ quality of death and dying (Aim 2) will be examined by descriptive statistics. Distinct patterns of psychological distress (i.e., symptoms of anxiety, depression, PTSD, and complicated/prolonged grief) (Aim 3) and associations of EOL-care quality and quality of death and dying in ICUs with identified psychological distress patterns over the first 2 years of bereavement (Aim 4) will be examined by latent growth mixture modeling using categorical latent class indicators, i.e., dichotomized scores from each instrument. Associations of EOL care quality in ICUs with family members’ QOL over the first 2 years of bereavement (Aim 4) will be examined by multivariate multiple regression modeling with generalized estimating equations. Our findings may highlight directions for clinical interventions to improve EOL-care quality care in ICUs, avoid futile life-sustaining treatments that patients and families may not prefer, counteract escalating EOL-care expenditures, and facilitate bereavement adjustment by meeting the unique needs of bereaved family members in each distinct pattern of psychological distress and by improving their QOL to benefit both bereaved family members of ICU decedents and Taiwanese society at large.

Project IDs

Project ID:PG10901-0406
External Project ID:NHRI-EX109-10704PI
Effective start/end date01/01/2031/12/20


  • End-of-life care
  • quality of care
  • intensive care
  • critical care
  • bereavement grief


Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.