Hepato-pancreato-biliary
Lindsay K. Dickerson, MD (she/her/hers)
Resident
University of Washington Department of Surgery
Seattle, Washington, United States
Lindsay K. Dickerson, MD (she/her/hers)
Resident
University of Washington Department of Surgery
Seattle, Washington, United States
Lindsay K. Dickerson, MD (she/her/hers)
Resident
University of Washington Department of Surgery
Seattle, Washington, United States
Griffen I. Allen, MD
Resident
University of Washington Department of Surgery, United States
S. Shahmir B. Chauhan, MD
Postdoctoral Research Fellow
University of Washington Department of Surgery, United States
Trisha A. Lipson, BA
Medical Student
University of Washington, United States
Bill Young, BS, BA
Medical Student
University of Washington, United States
Kathleen M. O'Connell, MD, MPH
Assistant Professor
University of Washington Department of Surgery, United States
Jonathan G. Sham, MD, MBEE, FACS, FSSO
Assistant Professor
University of Washington Department of Surgery, United States
Effective prognostic communication by surgeons caring for patients with pancreatic cancer is imperative given the high morbidity, mortality, and social burden of the disease, but there is little evidence of optimal methods. VitalTalk, an evidence-based communication skills training program for clinicians, provides a five-step tool called ADAPT to guide complex conversations about prognosis. This pilot study aimed to characterize surgeon communication of pancreatic cancer prognosis using VitalTalk’s ADAPT framework.
Methods:
A single-institution qualitative content analysis and retrospective chart review was performed for patients with newly diagnosed borderline resectable pancreatic cancer evaluated in a multidisciplinary clinic. Audio recordings from 12 surgeon-patient encounters were transcribed. Deductive coding using the VitalTalk ADAPT tool was carried out in iterative rounds, with review by content experts (KOC, JGS).
Results:
100% of encounters contained at least one ADAPT step (Ask, Discover, Anticipate ambivalence, Provide information, and Track emotion), five of 12 (42%) included three or more steps, and only one encounter (8%) incorporated four or five steps. Surgeons provided prognostic information in nearly all encounters (92%). The majority of prognostic information conveyed was qualitative in nature, including the likelihood of cancer recurrence, the gravity of the surgery, surgical candidacy, and prognostic ambiguity. Although surgeons incorporated Ask and Discover in a minority of encounters (42% and 25%, respectively), quantitative prognostic data were not offered unless the patient explicitly expressed this preference. Of the 13 patient emotional cues identified in six encounters, surgeons addressed all or select cues (62%) in four encounters but did not respond to cues in five instances (38%). Seven of the 12 patients (58%) received surgery, eight (67%) had specialized palliative care involvement at any time during treatment, and six (50%) are currently alive.
Conclusions:
During an initial surgeon-patient encounter for pancreatic cancer, surgeons focus heavily on conveying information and omit key prognostic communication steps. Future studies are warranted to investigate whether surgeon training in garnering patient understanding, eliciting preferences for prognostic discussions, and responding to emotional cues improves surgeon-patient communication and impacts outcomes in pancreatic cancer.