Role of multidimensional assessment of frailty in predicting short-term outcomes in hospitalized cancer patients: results of a prospective cohort study.

AIMS AND BACKGROUND

The study analyzed the value of physical and psychosocial frailty assessment in predicting the need for supportive care and the risk of short-term failures after discharge in hospitalized cancer patients.

METHODS AND STUDY DESIGN

Frailty was assessed in 350 consecutive patients using a multidimensional tool. Patients were followed for 4 months after discharge to record the occurrence of chemotherapy interruption, urgent hospital readmission or death. The association between patient characteristics and the outcomes were analyzed with either logistic or Cox multivariable models.

RESULTS

About 40% of patients were classified as frail, with a higher prevalence of clinical frailty (alone or together with psychosocial frailty). Psychosocial frailty was positively associated with the need for supportive care at discharge (adjusted OR, 3.46; 95% CI, 1.55-7.76) but did not predict a worse prognosis when other important clinical factors were considered. However, the need for supportive care at discharge, in addition to advanced disease and reduced performance status, was a strong predictor of short-term hospital readmission or death (HR 7.50; 95% CI, 3.12-18.02).

CONCLUSIONS

A more comprehensive assessment of frailty in cancer patients can aid in the timely identification of the need for supportive care after hospital discharge and improves the prediction of the short-term risk of hospital readmission or death.