SPICT as a predictive tool for outcomes of 1-year in patients with neurological disorders: a cohort study
摘要
The Supportive and Palliative Care Indicators Tool (SPICT) is designed to identify patients with deteriorating health and palliative care needs. Its predictive validity in a dedicated neurological inpatient population, particularly for a comprehensive set of clinical and functional outcomes, remains underexplored.
MethodsIn this prospective cohort study, we enrolled 337 patients with neurological disorders during hospitalization. Baseline assessments included a demographic characteristics questionnaire, disease-specific scales, and the SPICT. Physicians also subjectively assessed each patient’s palliative care needs. All patients completed one-year follow-up assessments after discharge, including length of hospital stay, number of unplanned hospitalizations, complications, Karnofsky Performance Status (KPS), Barthel Index (BI), frailty status, vital status, palliative care referral willingness, and palliative care referral.
ResultsAt baseline, 152 patients (45.1%) were SPICT-positive, whereas physicians identified only one patient (0.3%) as having palliative care needs, indicating poor agreement (κ = 0.007, p = 0.451). At one-year follow-up, 301 patients were included in this study, and SPICT-positive patients demonstrated significantly worse outcomes compared to the SPICT-negative group: longer hospital stay, greater complications and mortality, increased frailty severity, and significantly lower KPS and BI scores (all p < 0.001). No significant difference was found in number of unplanned hospitalizations, palliative care referral preference, and palliative care referral (p > 0.05).
ConclusionThese findings support the use of SPICT as a systematic, evidence-based tool for early identification of neurological disorders patients at higher risk for poor one-year outcomes. SPICT is not intended as a referral engine but rather as an upstream screening tool to facilitate goal of care discussions, advance care planning, and longitudinal monitoring for neurology patients.