Clinical decision support for service coverage of AYA palliative care: an interrupted time series.

Abstract

Volunteer-based palliative care programs often lack the tools to ensure consistent service delivery across their full patient census. This study assessed the impact of a clinical decision support system (CDSS) on the daily percentage of eligible patients visited in a hospital-based adolescent and young adult (AYA) palliative care peer support program. We analyzed data from Streetlight at University of Florida (UF) Health from November 14, 2021 to November 14, 2025. On November 14, 2022, a CDSS feature was introduced to display each patient's cumulative percentage of days visited on the program census. Segmented generalized linear regression was used to estimate changes in the level and slope of daily visit coverage before and after implementation. Estimates were adjusted for census size, volunteer staffing, and temporal patterns. CDSS implementation was associated with an immediate 7.6 percentage-point increase in visit coverage (95% CI: 2.5 to 12.6), equivalent to approximately 1.7 additional patients visited per day at a median census of 22 patients. This corresponds to roughly 440 additional patient visits per year assuming weekday-only operation. The intervention also significantly improved the post-implementation trend, reversing a pre-existing decline in visit coverage. CDSSs can leverage real-time program data to improve service delivery in palliative care support programs by increasing coverage and complementing human decision-making.