RESUMO
PURPOSE: Survival prediction is useful in selecting patients for palliative care or active anticancer therapy. The palliative and prognostic (PaP) score was shown to predict 1-month survival in terminally ill patients. Its application to patients with less advanced disease is a subject of debate. We assessed the value of the PaP score and of other clinical parameters in predicting survival in patients admitted in an oncological ward due to acute conditions. We also evaluated the frequency of active anticancer treatment in the last weeks of life. METHODS: All the 208 patients, consecutively admitted in a department of medical oncology and radiotherapy in a 9-month period, were included. Patients and disease features together with the PaP score were assessed and included in a multivariable model for survival prediction. RESULTS: Overall, median survival was 19 weeks and 12-week survival was 59.6%. The PaP score accurately predicted 4-week survival. Among the 39 patients who died within 4 weeks, 36% were on active treatment. The reason of admission, disease control, treatment, and PaP score were independently related to 12-week survival in the multivariate analysis; however patients with a 12-week survival lower than 30% were a minority. CONCLUSIONS: Although the PaP score accurately predicts life expectancy, its use in the setting of acute conditions seems not straightforward, due to the overall good prognosis of these patients. Active treatment in the last period of life is common. The potential reversibility of acute conditions makes prognostic measures inadequate for the purpose of treatment choices.