LAUSR

CONTEXT Quality improvement in end-of-life care generally acquires data from charts or caregivers. "Tracer" methodology, which assesses real-time information from multiple sources, may provide complementary information. OBJECTIVES The objective of this study was to develop a valid brief audit tool that can guide assessment and rate care when used in a clinician tracer to evaluate the quality of care for the dying patient. METHODS To identify items for a brief audit tool, 248 items were created to evaluate overall quality, quality in specific content areas (e.g., symptom management), and specific practices. Collected into three instruments, these items were used to interview professional caregivers and evaluate the charts of hospitalized patients who died. Evidence that this information could be validly captured using a small number of items was obtained through factor analyses, canonical correlations, and group comparisons. A nurse manager field tested tracer methodology using candidate items to evaluate the care provided to other patients who died. RESULTS The survey of 145 deaths provided chart data and data from 445 interviews (26 physicians, 108 nurses, 18 social workers, and nine chaplains). The analyses yielded evidence of construct validity for a small number of items, demonstrating significant correlations between these items and content areas identified as latent variables in factor analyses. Criterion validity was suggested by significant differences in the ratings on these items between the palliative care unit and other units. The field test evaluated 127 deaths, demonstrated the feasibility of tracer methodology, and informed reworking of the candidate items into the 14-item Tracer EoLC v1. CONCLUSION The Tracer EoLC v1 can be used with tracer methodology to guide the assessment and rate the quality of end-of-life care.