Objectives: To update previous guidelines to score the Cumulative Illness Rating Scale (CIRS) and test their usefulness in hospitalized elderly patients.
Design: The CIRS was scored retrospectively in a cohort of elderly patients followed for 18 months.
Setting: An acute internal medicine ward in an academic tertiary care hospital.
Participants: Three hundred eighty-seven patients aged 65 and older.
Measurements: The CIRS was retrospectively scored for the enrolled patients. Intrarater and interrater reliability were calculated. Two illness severity indices (total score (TSC) and severity (SV)) and one comorbidity index (CM) were obtained. Clinical features and comprehensive geriatric assessment (CGA) variables were also used. All patients underwent an 18-month follow-up for mortality and rehospitalization.
Results: Intrarater and interrater reliability of the CIRS scored following the guidelines was good (intraclass correlation coefficients of 0.83 and 0.81, respectively). The TSC, SV, and CM correlated with clinical features (laboratory values, medication usage, and length of in-hospital stay) and CGA variables (cognitive impairment, depression and disability). All three indices were able to predict 18-month mortality and rehospitalization rates.
Conclusion: This study confirmed the validity of the CIRS as an indicator of health status and demonstrated its ability to predict 18-month mortality and rehospitalization in hospitalized elderly patients. The availability of detailed guidelines for scoring the CIRS can improve its usefulness and facilitate more-widespread use for research and clinical aims.