A Combined Scoring System of Palliative Prognostic Index and Laboratory Prognostic Score

Abstract

Purpose

We hypothesized that combining the Palliative Prognostic Index (PPI; maximum score: 15) with the Laboratory Prognostic Score (LPS; maximum score: 10), which is based on blood test parameters, would yield superior prognostic accuracy compared with each score alone. This study aimed to evaluate the prognostic performance of the combined PPI-LPS score and compare it with that of PPI, LPS, and the Palliative Prognostic Score (PaP).

Methods

The PPI-LPS score was calculated as the arithmetic mean of scaled PPI and LPS values: Round[((PPI/1.5) + LPS)/2]. A total of 1061 terminally ill patients with cancer admitted to an acute care hospital with a palliative care unit were included. Prognostic performance was compared among PPI, LPS, PaP, and PPI-LPS in terms of (1) discrimination of 30-, 60-, and 90-day mortality; (2) accuracy of median survival prediction; and (3) categorical survival classification as “days” (0-13 days), “weeks” (14-55 days), or “months” (>55 days).

Results

The PPI-LPS score demonstrated the highest discriminative ability for 30-, 60-, and 90-day mortality (accuracy: 74%, 80%, and 80%, respectively). It also achieved superior accuracy in predicting median survival (34%) and categorical survival classification (agreement: 61%) compared with PPI, LPS, and PaP.

Conclusion

The combined PPI-LPS score provides better prognostic performance than PPI, LPS, or PaP in terminally ill patients with cancer, offering a simple and clinically useful tool for survival estimation in palliative care settings.