Using admission characteristics to predict return to the community from a post-acute geriatric evaluation and...

Area of Science:

• Geriatric Medicine
• Health Services Research
• Outcome Prediction

Background:

• Post-acute care units play a crucial role in patient recovery and transition.
• Accurate prediction of discharge outcomes is essential for resource allocation and patient management.
• Existing tools like CIRS and NSI require evaluation for their predictive power in this setting.

Purpose of the Study:

• To compare the predictive capabilities of the Cumulative Illness Rating Scale (CIRS) and the Nursing Severity Index (NSI) for discharge outcomes.
• To develop a multivariate model to predict discharge outcomes from a geriatric post-acute (GEM) unit.

Main Methods:

• Retrospective chart review of 452 patients admitted to a 20-bed post-acute GEM unit.
• Data collected at admission included demographics, CIRS, NSI, functional status, and social support.
• The sample was divided into derivation (n=298) and validation (n=154) cohorts.

Main Results:

• In the derivation cohort, 75.8% of patients returned to the community.
• The final logistic regression model included NSI, severe CIRS items, age, and social support.
• The model predicted 87.7% of discharge outcomes in the validation cohort.

Conclusions:

• A logistic regression model incorporating illness severity, functional status, social support, and age effectively predicts discharge outcomes.
• Both NSI and CIRS are valuable components in a predictive model for geriatric post-acute services.
• This model can potentially refine patient selection for post-acute care.