Machine Learning Reveals Demographic Disparities in Palliative Care Timing Among Patients With Traumatic Brain Injury Receiving Neurosurgical Consultation

Carlos A Aude; Vikas N Vattipally; Oishika Das; Kathleen R Ran; Ganiat A Giwa; Jordina Rincon-Torroella; Risheng Xu; James P Byrne; Susanne Muehlschlegel; Jose I Suarez; Debraj Mukherjee; Judy Huang; Tej D Azad; Chetan Bettegowda

doi:10.1007/s12028-024-02172-2

Machine Learning Reveals Demographic Disparities in Palliative Care Timing Among Patients With Traumatic Brain Injury Receiving Neurosurgical Consultation

Neurocrit Care. 2024 Dec 10. doi: 10.1007/s12028-024-02172-2. Online ahead of print.

Affiliations

¹ Department of Neurosurgery, Johns Hopkins University School of Medicine, 1800 Orleans Street, Baltimore, 21287, MD, USA.
² Division of Acute Care Surgery, Department of Surgery, Johns Hopkins University School of Medicine, Baltimore, MD, USA.
³ Division of Neurosciences Critical Care, Departments of Anesthesiology and Critical Care Medicine, Neurology, and Neurosurgery, Johns Hopkins University School of Medicine, Baltimore, MD, USA.
⁴ Department of Neurosurgery, Johns Hopkins University School of Medicine, 1800 Orleans Street, Baltimore, 21287, MD, USA. tazad1@jhmi.edu.

PMID: 39653977
DOI: 10.1007/s12028-024-02172-2

Abstract

Background: Timely palliative care (PC) consultations offer demonstrable benefits for patients with traumatic brain injury (TBI), yet their implementation remains inconsistent. This study employs machine learning methods to identify distinct patient phenotypes and elucidate the primary drivers of PC consultation timing variability in TBI management, aiming to uncover disparities and inform more equitable care strategies.

Methods: Data on admission, hospital course, and outcomes were collected for a cohort of 232 patients with TBI who received both PC consultations and neurosurgical consultations during the same hospitalization. Patient phenotypes were uncovered using principal component analysis and K-means clustering; time-to-PC consultation for each phenotype was subsequently compared by Kaplan-Meier analysis. An extreme gradient boosting model with Shapley Additive Explanations identified key factors influencing PC consultation timing.

Results: Three distinct patient clusters emerged: cluster A (n = 86), comprising older adult White women (median 87 years) with mild TBI, received the earliest PC consultations (median 2.5 days); cluster B (n = 108), older adult White men (median 81 years) with mild TBI, experienced delayed PC consultations (median 5.0 days); and cluster C (n = 38), middle-aged (median: 46.5 years), severely injured, non-White patients, had the latest PC consultations (median 9.0 days). The clusters did not differ by discharge disposition (p = 0.4) or inpatient mortality (p > 0.9); however, Kaplan-Meier analysis revealed a significant difference in time-to-PC consultation (p < 0.001), despite no differences in time-to-mortality (p = 0.18). Shapley Additive Explanations analysis of the extreme gradient boosting model identified age, sex, and race as the most influential drivers of PC consultation timing.

Conclusions: This study unveils crucial disparities in PC consultation timing for patients with TBI, primarily driven by demographic factors rather than clinical presentation or injury characteristics. The identification of distinct patient phenotypes and quantification of factors influencing PC consultation timing provide a foundation for developing for standardized protocols and decision support tools to ensure timely and equitable palliative care access for patients with TBI.

Keywords: Age factors; Cluster analysis; Critical care; Decision support techniques; Health care disparities; Machine learning; Neurosurgery; Palliative care; Prognosis; Quality of health care; Race factors; Sex factors; Traumatic brain injury.