Association Between Patient-Reported Outcomes-Derived Symptom Complexity and Overall Survival Among Patients With...

Cancer Care Alberta (CCA) developed a patient-reported outcomes (PROs)-derived algorithm that assigns patients a low, moderate, or high symptom complexity score based on the number and severity of symptoms reported on a routine questionnaire. This study investigated the algorithm's prognostic utility by examining the association between symptom complexity and overall survival (OS).

This study included patients aged ≥18 years who had their initial oncology consultation between October 1, 2019, and April 1, 2020, and who completed a PRO questionnaire within 30 days of the consultation. Vital status was assessed on April 30, 2023. Analyses included Kaplan-Meier curves with log-rank tests to estimate OS and Cox proportional hazards models to identify factors associated with increased risk of death. Multiple demographic and clinical variables were included as covariates in the Cox models, including sex and tumor group.

The study included 5,841 patients. Most (61.3%) presented with low symptom complexity, whereas 21.6% scored moderate and 17.2% scored high. Patients with different complexity levels also differed in terms of sex, tumor group, and other factors. Patients with higher baseline complexity had significantly shorter median OS: 25.7 weeks for high complexity, 39.0 weeks for moderate, and 64.7 weeks for low complexity (P<.01). The Cox model revealed that, compared with patients with low complexity, those with moderate complexity had a 69% higher risk of death (hazard ratio [HR], 1.69; 95% CI, 1.50-1.90; P<.01) and those with high complexity had a 142% higher risk of death (HR, 2.42; 95% CI, 2.15-2.72; P<.01).

Higher symptom complexity, assessed following a patient's initial oncology consultation, is associated with shorter OS and higher risk of death. This study demonstrates the utility of CCA's symptom complexity algorithm in predicting survival based on one PRO assessment. This information could support clinicians' decision-making regarding treatment plans, symptom management, and supportive care.