Measuring comprehensive outcomes in palliative care: validation of the Korean version of the Good Death Inventory.

CONTEXT: No systematic or comprehensive attempts have yet been made to assess quality of death as an indicator of palliative care outcomes in Korea, and no validated instruments exist for the assessment of a good death in Koreans.
OBJECTIVES: This study examined the validity and reliability of the Korean version of the Good Death Inventory (GDI), which was developed in Japan to evaluate the quality of death from the perspective of bereaved family members.
METHODS: Forward and backward translations and a pilot test were conducted. In a multicenter cross-sectional survey, a questionnaire packet, including the GDI, overall quality of life during the last week, and overall satisfaction with care, was mailed to bereaved family members (n=501) of patients who had died from cancer two to six months before the study. Descriptive analyses were performed, including response rate, mean, median, skewness, and kurtosis for each item. The reliability of the GDI was tested by Cronbach's alpha. The dimensional structure was assessed using confirmatory factor analyses. Concurrent validity was tested by correlation with the overall quality of life and overall satisfaction with care.
RESULTS: Participants were able to complete the GDI, and the compliance rates were satisfactory. Cronbach's alpha coefficient for internal consistency was 0.93 overall and ranged from 0.69 to 0.94 for subdomains. The hypothesized 18-factor model of a good death appeared to fit the data (goodness of fit index [GFI]=0.964; adjusted GFI index=0.960; normal fit index=0.952). The overall scores on the GDI correlated with patients' quality of life (0.56; P<0.001) and overall satisfaction with care (0.44; P<0.001).
CONCLUSION: The Korean version of the GDI is a reliable and valid measure of the comprehensive outcomes of palliative care from the perspective of bereaved Korean family members.