Forecasting and Foretelling: Association Between Expected Survival Time and Characteristics of Prognosis Communication in Palliative Care Consultations (316-C)

Abstract

1.Describe two aspects of what is known about prognostication in advanced illness: that prognosis communication happens infrequently and that it is often framed optimistically.2.Describe what this study adds to the existing literature.3.Describe one strength and one limitation of the methods used to measure prognosis communication. Prognostication is an important, yet often neglected, aspect of decision-making in advanced illness involving both forecasting (forming an estimate about the future) and foretelling (communicating a forecast). To describe the relation between forecasting and foretelling in palliative care consultations with hospitalized patients. We audio-recorded 75 initial conversations occurring during palliative care consultations with hospitalized adult patients. We coded each segment (a length of uninterrupted speech) for the occurrence and linguistic framing (language cues indicating a pessimistic or optimistic perspective) of prognosis content. We linked conversations to the medical record (including standardized palliative care consultation forms) and a palliative care clinician assessment of survival prognosis. The mean frequency of physician segments containing prognosis information increased incrementally with clinical markers of shorter survival time (lower Palliative Performance Scores, PPS) [PPS > 60 = 2.4 segments; PPS 40-50 = 4.4 segments; PPS 10-30 = 6.2 segments, p < 0.05] and with physician expectations for shorter survival time [ > 2 weeks = 4.5 segments; 1 day to 2 weeks = 5.7 segments; < 24 hours = 7.7 segments, p < 0.05]. These findings persisted in multivariate analyses. The majority of segments contained no linguistic framing cues, however pessimistic cues were more common when physicians expected shorter survival. Forecasts indicating a shorter survival time are associated with more frequent foretelling in inpatient palliative care consultations. Foretelling generally did not involve optimistic/pessimistic linguistic cues; however physicians used more pessimistic cues among patients with shorter expected survival times.