RESUMEN
BACKGROUND AND OBJECTIVES: Few data exist to guide optimal communication practices for surgical oncologists. VitalTalk, an evidence-based communication skills training model for clinicians, offers the five-step ADAPT tool for discussing prognosis. This study aimed to characterize surgeon communication of pancreatic cancer prognosis using VitalTalk's ADAPT framework. METHODS: Contemporaneous audio recordings from 12 initial surgeon-patient encounters for borderline resectable pancreatic cancer were transcribed. Directed qualitative content analysis based on ADAPT (Ask, Discover, Anticipate, Provide, and Track) was used to deductively code transcripts. RESULTS: All encounters contained at least one ADAPT step while only one (8%) incorporated four or five steps. Surgeons provided prognostic information (Provide) in all but one encounter (92%); most was qualitative and clustered into themes: serious illness, surgical candidacy, prognostic ambiguity, and cancer recurrence. Surgeons elicited understanding (Ask), requested information preferences (Discover), anticipated ambivalence (Anticipate), and responded to emotion (Track) in a minority of encounters (25%-42%); of 15 patient emotional cues, six were not addressed by surgeons. CONCLUSIONS: During an initial encounter for pancreatic cancer, surgeons focus heavily on providing information but omit critical prognostic communication steps. Future studies are needed to investigate if surgeon training in palliative care-based communication is feasible and impacts patient-perceived quality of communication.