Validation of PROFUND prognostic index over a four-year follow-up period.

Abstract

BACKGROUND/OBJECTIVES: The PROFUND index stratifies accurately the 12-month mortality risk of polypathological patients (PPs), but its fitness over a longer follow-up period remains unknown. We aimed to explore the calibration and discrimination power of PROFUND index over 4-years, in order to assess its follow-up interval generalizability. DESIGN: Multicenter prospective cohort-study. SETTING: 33 Spanish hospitals. PARTICIPANTS: PPs included after hospital discharge, outpatient clinics, or home hospitalization. MEASUREMENTS: Mortality over a 4-year follow-up period. METHODS: PROFUND index calibration was assessed by risk-quartiles predicted/observed mortality (Hosmer-Lemeshow goodness-of-fit test), and its discrimination power by ROC curves. RESULTS: A total of 768 patients were included (630 [82%] of them completed the 4-year follow-up). Global mortality rate was 63.5%. When assessing individual patient scores, mortality was 52% in the lowest risk group (0-2 points in PROFUND score); 73.5% in the low-intermediate risk group (3-6 points), 85% in the intermediate-high group (7-10 points); and 92% in the highest risk group (≥11 points). Accuracy testing of the PROFUND index showed good calibration (P=.8 in the Hosmer-Lemeshow goodness-of-fit test), and also a good discrimination power (AUC=0.71 [0.67-0.77] in ROC curve). CONCLUSIONS: The PROFUND index maintained its accuracy in predicting mortality of polypathological patients over a 4-year follow-up period. This index may be of potential usefulness in deciding the most appropriate health-care interventions in populations with multimorbidity.