Background While shared decision making (SDM) is advocated for ethical reasons and beneficial outcomes, SDM might also negatively affect patients with incurable cancer. The current study explored whether SDM, and an oncologist training in SDM, are associated with adverse outcomes (i.e., patient anxiety, tension, helplessness/hopelessness, decisional uncertainty, and reduced fighting spirit). Design A secondary analysis of a randomized clinical trial investigating the effects of SDM interventions in the context of advanced cancer. The relations between observed SDM (OPTION12), specific SDM elements (4SDM), oncologist SDM training, and adverse outcomes were analyzed. We modeled adverse outcomes as a multivariate phenomenon, followed by univariate regressions if significant. Results In total, 194 patients consulted by 31 oncologists were included. In a multivariate analysis, observed SDM and adverse outcomes were significantly related. More specifically, more observed SDM in the consultation was related to patients reporting more tension ( P = 0.002) and more decisional uncertainty ( P = 0.004) at 1 wk after the consultation. The SDM element “informing about the options” was especially found to be related to adverse outcomes, specifically to more helplessness/hopelessness ( P = 0.002) and more tension ( P = 0.016) at 1 wk after the consultation. Whether the patient consulted an oncologist who had received SDM training or not was not significantly related to adverse outcomes. No relations with long-term adverse outcomes were found. Conclusions It is important for oncologists to realize that for some patients, SDM may temporarily be associated with negative emotions. Further research is needed to untangle which, when, and how adverse outcomes might occur and whether and how burden may be minimized for patients. Highlights Observed shared decision making was related to more tension and uncertainty postconsultation in advanced cancer patients However, training oncologists in SDM did not affect adverse outcomes. Further research is needed to untangle which, when, and how adverse outcomes might occur and how burden may be minimized