Validation of toolkit after-death bereaved family member interview

The purpose of this study was to examine the reliability and validity of th e Toolkit After-Death Bereaved Family Member Interview to measure quality o f care at the end of life from the unique Perspective of family members. Th e survey included proposed problem scores (a count of the opportunity to im prove the quality of care) and scales. Data were collected through a retros pective telephone survey with a family member who was interviewed between 3 and 6 months after the death of the patient. The setting was an outpatient hospice service, a consortium of nursing homes, and a hospital in New Engl and. One hundred fifty-six family members from across these settings partic ipated. The 8 proposed domains of care, as represented by problem scores or scales, were based on a conceptual model of patient-focused, family-center ed medical care. The survey design emphasized face validity in order to pro vide actionable information to healthy care providers. A correlational and factor analysis was undertaken of the 8 proposed problem scores or scales. Cronbach's alpha scores varied from 0.58 to 0.87, with two problem scores ( each of which had only 3 survey items) having a low alpha of 0.58. The mean item-to-total correlations for the other problem scores varied from 0.36 t o 0.69, and the mean item-to-item correlations were between 0.32 and 0.70. The proposed problem scares or scales, with the exception of closure and ad vance care planning, demonstrated a moderate correlation (i.e., from 0.44 t o 0.52) with the overall rating of satisfaction (as measured by a five-poin t, "excellent" to "poor" scale). Family members of persons who died with ho spice service reported fewer problems in each of the six domains of medical care, gave a higher rating of the quality of care, and reported higher sel f-efficacy in caring for their loved ones. These results indicate that 7 of the 8 proposed problem scores or scales demonstrated psychometric properti es that warrant further testing. The domain of closure demonstrated a poor correlation with overall satisfaction and requires further work. This surve y could provide information to help guide quality improvement efforts to en hance the care of the dying. (C) U. S. Cancer Pain Relief Committee, 2001.