Association of area poverty level and performance of an electronic medical record tool in predicting 6-month mortality in hospitalized patients with cancer.

Abstract

Disparities in risk and access to care can lead to a greater incidence of cancer, higher stage at diagnosis, and more frequent cancer-related deaths in communities with high poverty levels. Risk prediction tools that do not take into consideration social determinants of health may underestimate mortality in these populations. We sought to determine whether an association exists between poverty and the ability of the end-of-life index (EOLI) to predict 6-month mortality in hospitalized patients with cancer. Using a sample of 802 patients admitted to oncology in an urban hospital over a 6-month period, we categorized patients according to the percentage of families living at or below the federal poverty line in their zone improvement plan codes. We applied an EOLI score of ≥40 as an indicator of higher risk of 6-month mortality. We found that the EOLI score remained equally predictive of mortality across area poverty level groups. This finding suggests that the score can be used broadly as a screening tool to predict mortality risk in hospitalized patients with cancer.