Enhancing utility of interfacility triage guidelines using machine learning: Development of the Geriatric...

Area of Science:

• Trauma Surgery
• Geriatric Medicine
• Epidemiology

Background:

• Undertriage of injured older adults to tertiary trauma centers (TTCs) is a significant issue, particularly in rural areas where initial transport is often to non-tertiary trauma centers (NTCs).
• Current interfacility triage guidelines lack the ability to predict individual risk or prioritize risk factors for older adults.

Purpose of the Study:

• To develop a novel transfer risk score to aid in the secondary triage of injured older adults to TTCs.
• To simplify the process of identifying older patients who require transfer to a higher level of trauma care.

Main Methods:

• Retrospective prognostic study using data from the Oklahoma State Trauma Registry (2009-2019).
• Included injured adults aged 55 years or older initially transported to an NTC.
• Machine-learning techniques, including random forests, were employed to develop the risk stratification model.

Main Results:

• The study analyzed 5,913 patients, with 32.7% experiencing mortality or serious injury (Injury Severity Score ≥16) requiring intervention.
• The final prognostic model achieved an area under the curve of 75.4%, identifying key predictors: airway intervention, femur fracture, spinal cord injury, low Glasgow Coma Scale score, and need for hemodynamic support.
• A risk score of 7 demonstrated 78% sensitivity and 56% specificity for identifying patients needing transfer.

Conclusions:

• A validated risk score can significantly enhance secondary triage for injured older adults, improving access to appropriate trauma care.
• This study presents the first risk stratification tool specifically designed for older adults requiring transfer to a higher level of trauma care.