Associations Between Symptom Complexity and Acute Care Utilization Among Adult Advanced Cancer Patients Followed by a Palliative Care Service

TL;DR

This study shows that a new symptom complexity algorithm can help predict which advanced cancer patients are more likely to use emergency hospital services in the short term.

Contribution

A novel symptom complexity algorithm using the ESAS-r is introduced to predict acute care utilization in advanced cancer patients.

Findings

High-complexity patients were 2.83 times more likely to use acute care within seven days compared to low-complexity patients.

Symptom complexity did not significantly predict acute care use within fourteen days after adjusting for other factors.

The algorithm may help clinicians identify patients needing more frequent follow-ups to reduce unnecessary hospital visits.

Abstract

Advanced cancer symptoms can cause patients to go to the emergency department or stay in the hospital. Clinicians can use the Edmonton Symptom Assessment System—Revised (ESAS-r) to monitor these symptoms. We investigated if a new “symptom complexity” algorithm, which uses the ESAS-r to identify patients with “low”, “medium”, or “high” symptom complexity, could predict whether adult advanced cancer patients will use these hospital services. Of 559 patients, we identified 125 (22.4%) low-complexity, 180 (32.2%) medium-complexity, and 254 (45.4%) high-complexity patients. In total, 61 (10.9%) patients used these hospital services in seven days and 108 (19.3%) used them in fourteen days. High-complexity patients were 2.83 times more likely than low-complexity patients to access these hospital services within seven days. However, these groups of patients used these services equally as often within fourteen days. Subsequently, this algorithm may help clinicians identify patients who need more frequent follow-ups to prevent unnecessary hospital visits.

Among adult advanced cancer patients already accessing palliative care, symptoms can contribute to unplanned acute care utilizations, which can disrupt care and worsen patient outcomes. We examined how a novel symptom complexity algorithm, using patients’ ratings of the nine Edmonton Symptom Assessment System—Revised (ESAS-r) symptoms to assign “low”, “medium”, or “high” complexity, predicts acute care utilizations. This retrospective observational cohort study used electronic medical record data from the Durham Regional Cancer Centre in Ontario, Canada, comprising adult advanced cancer patients who completed at least one ESAS-r report between 1 January 2022 and 31 December 2023. We applied chi-squared tests, Kruskal–Wallis H tests, and multivariable binary logistic regressions to evaluate factors associated with higher odds of acute care utilization within seven and fourteen days of patients’ first ESAS-r reports after their first palliative care interaction. Of 559 included patients, 125 (22.4%) exhibited low complexity, 180 (32.2%) exhibited medium complexity, and 254 (45.4%) exhibited high complexity on their first ESAS-r report. In total, 61 (10.9%) patients accessed acute care within seven days and 108 (19.3%) patients accessed acute care within fourteen days of their first ESAS-r report. Controlling for sociodemographic and clinical covariates, compared to low-complexity patients, high-complexity patients had higher odds of acute care utilization within seven days (aOR = 2.83, 95% CI: 1.18–6.77), but not within fourteen days (aOR = 1.78, 95% CI: 0.97–3.28). Accordingly, as a clinical decision-making tool, ESAS-r symptom complexity may help identify patients who would benefit from more intensive follow-up and potentially reduce unnecessary acute care utilizations.

Figures