Purpose: For cancer patients, prognostication by physicians constitutes a fundamental basis of decision-making to manage the cancer journey, especially with advanced stages of cancer. Various tools such as the palliative prognostic index (PPI) have been established to estimate patient survival. Such reliable and widely applied tools unfortunately provide unexpected results of prognosis predictions in some cases. To support disease-related decision-making, this study explores factors associated with prognosis prediction outliers in patients with advanced cancer in palliative care units (PCUs).
Methods: For this retrospective observational study conducted at a PCU, we surveyed medical records of patients with advanced cancer admitted to our facility during August 2018-June 2022. Specifically, we collected data for the PPI, laboratory data, physical status, and psychological symptoms on admission.
Results: Our survey of data of 675 patients revealed that the presence and absence of severely reduced oral intake were significantly associated, respectively, with shorter (p < 0.001) and longer (p < 0.001) than estimated survival. Moreover, symptoms other than pain and female sex were associated, respectively, with shorter-than-expected and longer-than-expected survival.
Conclusion: The oral intake status of patients with advanced cancer in a PCU plays an important role in prognosis estimation. During prognosis, effects of symptoms other than pain and sex difference must also be considered. Although PPI-based prediction can provide a foundation for care management, cancer patients in more advanced stages of disease particularly require more detailed consideration.
Keywords: Cancer; Oral intake; PPI: Palliative Prognostic Index; Palliative care unit; Prognosis; Sex difference.
© 2025. The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.