Objective: The Cumulative Illness Rating Scale (CIRS) has been the scale most frequently used in geriatric psychiatry to measure medical burden, yet not all raters follow the guidelines to exclude from ratings acute medical conditions not expected to persist chronically. In the current pilot study we determine whether inclusion of acute medical condition undermines the usefulness of the CIRS. We examine the effects of including and excluding acute conditions on the relationship between CIRS summary scores and survival (the measure of health outcome).
Materials and Methods: Our convenience sample consists of 95 male veterans (mean age 68.4+/-6.4, range 60-88 years) admitted to acute medical/surgical inpatient units and enrolled in the UPBEAT (Unified Psychogeriatric Biopsychosocial Evaluation and Treatment) demonstration program at the Greater Los Angeles Veterans Affairs Healthcare System in California, USA. All UPBEAT patients screened positive for symptoms of anxiety, depression (Mental Health Inventory) and/or alcohol abuse (Alcohol Use Disorder Identification Test). A single clinician (IG) did a retrospective chart review and provided two sets of CIRS ratings for each patient: an inpatient rating (IP), including both acute and chronic conditions, and a post-hospitalization rating (PH), including only chronic conditions. Seven CIRS summary scores described in the literature were calculated for both IP and PH ratings. Survival time was measured in the number of days patients lived between the date of UPBEAT enrollment and end of the 24-month follow-up period. Seventy-five of the 95 patients survived throughout the 24-month follow-up period. We used Cox proportional hazards regression to estimate the association of CIRS summary scores (standardized using z-score transformation) with survival days; paired t-tests to identify statistically significant differences in means between each CIRS-IP and corresponding CIRS-PH summary score, and Spearman’s rank-order method to calculate correlation between CIRS-IP and CIRS-PH scores.
Results: For five out of the seven CIRS summary scores both IP and PH were significantly associated with survival days despite the fact that means for CIRS-IP were higher than those for CIRS-PH on most of the summary scores. Moreover, CIRS-IP scores were highly correlated with corresponding CIRS-PH scores, and even the summary score with the lowest correlation was significantly associated with survival.
Conclusion: The results suggest that in acutely hospitalized medical/surgical inpatients with symptoms of depression, anxiety and/or alcohol abuse, the CIRS may be a useful indicator for medical burden as measured by survival days over a 24-month period even when ratings include acute medical conditions.
Back to PC Wednesday Poster Sessions
Back to The Eleventh International Congress