Clinical value of second opinions in oncology: A retrospective review of changes in diagnosis and treatment recommendations

Abstract

AbstractBackgroundData on the clinical value of second opinions in oncology are limited. We examined diagnostic and treatment changes resulting from second opinions and the expected impact on morbidity and prognosis.MethodsThis retrospective cohort study included patients presenting in 2018 to a high‐volume cancer center for second opinions about newly diagnosed colorectal, head and neck, lung, and myeloma cancers or abnormal results. Two sub‐specialty physicians from each cancer type reviewed 30 medical records (120 total) using a process and detailed data collection guide meant to mitigate institutional bias. The primary outcome measure was the rate of treatment changes that were “clinically meaningful”, i.e., expected to impact morbidity and/or prognosis. Among those with treatment changes, another outcome measure was the rate of clinically meaningful diagnostic changes that led to treatment change.ResultsOf 120 cases, forty‐two had clinically meaningful changes in treatment with positive expected outcomes (7 colorectal, 17 head and neck, 11 lung, 7 myeloma; 23–57%). Two patients had negative expected outcomes from having sought a second opinion, with worse short‐term morbidity and unchanged long‐term morbidity and prognosis. All those with positive expected outcomes had improved expected morbidity (short‐ and/or long‐term); 11 (0–23%) also had improved expected prognosis. Nine involved a shift from treatment to observation; 21 involved eliminating or reducing the extent of surgery, compared to 6 adding surgery or increasing its extent. Of the 42 with treatment changes, 13 were due to clinically meaningful diagnostic changes (1 colorectal, 5 head and neck, 3 lung, 4 myeloma; 3%–17%) .ConclusionsSecond‐opinion consultations sometimes add clinical value by improving expected prognoses; more often, they offer treatment de‐escalations, with corresponding reductions in expected short‐ and/or long‐term morbidity. Future research could identify subgroups of patients most likely to benefit from second opinions.