Triage considerations in medical intensive care

Area of Science:

• Critical Care Medicine
• Health Services Research

Background:

• Improving medical intensive care unit (MICU) admission criteria is crucial for optimal patient care and resource allocation.
• Existing scoring systems like the Acute Physiologic and Chronic Health Evaluation (APACHE) have limitations in accurately stratifying patients for MICU admission.

Purpose of the Study:

• To evaluate the effectiveness of Acute Physiologic and Chronic Health Evaluation (APACHE) scores and diagnosis-adjusted mortality rates in predicting MICU admission.
• To identify potential improvements in triage procedures for MICU admissions.

Main Methods:

• Analysis of data from 2419 medical patients over a 6-month period, including those who received MICU consultation.
• Comparison of APACHE scores and diagnosis-adjusted mortality rates between patients admitted to the MICU and those not admitted.
• Utilized receiver operating characteristic (ROC) curves to assess predictive performance.

Main Results:

• Significant overlap was observed in APACHE scores and predicted mortality rates between MICU-admitted and non-admitted patients.
• No specific score or mortality threshold could reliably differentiate patients who would benefit from MICU admission.
• Diagnosis-adjusted mortality rates demonstrated predictive performance comparable to APACHE scores for MICU admission and intervention.

Conclusions:

• Current scoring systems may lack the necessary sensitivity and specificity for uniform MICU admission criteria.
• Diagnosis-adjusted mortality rates show promise for improving MICU triage, suggesting a need for diagnosis-specific predictive models.
• Prospective modeling using clinical variables and ROC curves could enhance medical triage procedures.