Development of prognostication model of 60-day survival in ambulatory cancer patients receiving palliative care

e24091 Background: Prognostication by predicting the likelihood of surviving at certain time points is an important aspect for clinician-patient communication, inform medical decisions and facilitate advance care planning. Understanding disease trajectory has also been reported to improve patient satisfaction in medical encounters and reduce anxiety and depression of carer due to unprepared clinical deterioration. The main problem of applying currently available tools is that the included prognostic factors were designed to look at the more-ill patients in ward settings and sometimes needs specialized lab-tests, and is difficult to apply to out-patients and in real-world public care setting. Methods: 240 consecutive patients attending our oncology palliative care new-case assessment clinic from 1st January 2016 to 31st December 2018 were included. Information regarding underlying cancer, symptoms collected on baseline survey and laboratory results including blood counts, renal and liver function were collected. Survival outcomes were dichotomized at 60 days. Significant prognostic factors were identified by univariate analysis. Multivariable logistic regression were used for model building. Model fitness was assessed by Hosmer-Lemeshow test and R squared statistics, predictive properties assessed by area-under curve (AUC), sensitivity and specificity. The robustness of the prediction model was confirmed by bootstrapping. Results: 240 patients were included. The median KPS is 70. Prevalence of symptoms including tiredness (n = 135, 55.6%), anorexia (n = 106, 43.6%), shortness of breath on exertion (n = 76, 31.3%), peripheral edema (n = 58, 23.9%) and nausea (n = 35, 14.4%); prevalence of laboratory abnormalities including elevated bilirubin (n = 94, 38.7%), low albumin (n = 94, 38.7%), leucocytosis (n = 68, 28.0%), thrombocytosis (n = 61, 25.1%) and lymphopenia (n = 53, 21.8%). Hospital depressive screening scale, PHQ-9 score (cutoff = 8) was elevated in 81 patients (33.3%). At 60 days, there were 70 death events. On univariate analyses, all the above factors were significant predictors. On multivariate analysis and prognostic model development, the most significant prognostic factors were self-reported presence of poor appetite and edema, laboratory test of elevated WCC above normal, lymphocyte below normal and bilirubin above normal. A prognostic model built upon these five factors showed high sensitivity of 73% and specificity of 55% in predicting survival at 60 days. The significance and prediction property were maintained after bootstrapping operation. Compared to the parameters employed in palliative care prognostic index (PPI), the AUC of current predictive model is superior (0.782 vs 0.692). Conclusions: We hope to come up with a tailored prognostic tool for real-world ambulatory palliative care setting. This study is supported by KCC research grant.