Patterns in Symptoms Preceding Acute Care in Patients With Cancer

Area of Science:

• Oncology
• Health Services Research
• Data Science

Background:

• Patients with cancer frequently require unplanned acute care (emergency department visits, hospitalizations) due to disease or treatment complications.
• Understanding the patterns of symptoms preceding these acute care events is crucial for improving patient outcomes and quality of life.
• A knowledge gap exists in characterizing these preceding symptoms and their documentation across diverse patient populations.

Purpose of the Study:

• To characterize symptoms documented in the 30 days before acute care encounters in patients with cancer.
• To quantify differences in symptom documentation based on sociodemographic and cancer-related characteristics.

Main Methods:

• A retrospective cohort study was conducted at a single tertiary-care institution.
• Natural language processing (NLP) analyzed clinical documentation to identify symptoms preceding acute care.
• Logistic regression examined associations between patient/cancer characteristics and symptom burden.

Main Results:

• Over 70,000 acute care visits were analyzed, revealing over 850,000 documented symptoms.
• Pain, nausea, and vomiting were the most frequent symptoms.
• Female patients, racial/ethnic minorities, and Medicaid-insured patients had a higher documented symptom burden preceding acute care.

Conclusions:

• The study identified common symptoms preceding acute care in cancer patients and highlighted disparities in symptom documentation.
• Further research is needed to develop interventions that reduce patient burden and acute care utilization.
• Improving symptom recognition and documentation is key to enhancing care quality and patient outcomes.