Longitudinal changes in patient reported outcome measures in patients with cancer six months prior to death: A case-control study.

Abstract

1585 Background: Provider assessments consistently fall short in determining when cancer patients enter into the terminal phase of their disease. Despite prior desires to spend their last days outside an institution, patients often seek emergency room care and most report never having a palliative consultation at the time of their terminal admission. Patient reported outcome measures (PROMs) are derived directly from the patient and may provide insight into the experience near the end of life to facilitate earlier involvement of supportive oncology. This study aims to compare PROMs in the 6 months prior to death to those not dying among patients with cancer. Methods: This study uses a routine PROMs program implemented since September 2020 for all patients with a cancer diagnosis at a tertiary care hospital. This study focused on PROMs using 3 domains of the Patient Reported Outcomes Measurement Information System (PROMIS): physical function, pain interference, and fatigue; PROMs frequency mirrored patients’ oncologic visits. Using a retrospective case-control study, patients who had died within 6 months of a PROMs response (cases) were compared to controls who were alive at the time of the case’s death. Cases were matched 1:2 to controls by age at PROMs completion, gender, cancer disease site, and stage. Generalized estimating equation (GEE) models adjusted for age at PROMs completion, gender, cancer disease site, stage, and correlations between individual patient encounters were used to compare mean PROMs scores between cases and controls in each domain. Results: In total, 274 cases were compared to 270 unique controls. Univariate comparisons between cases and controls demonstrated significant differences only in Charlson Comorbidity Index (CCI), with cases demonstrating worse CCI (3.3 ± 2.5) compared to controls (2.3 ±2.2; p < 0.001). Time from diagnosis to PROMs completion was not different between groups (4.08 versus 4.10 years, p = 0.798). Over the 6 months prior to death, 10.5%/15.0%/32.0% cases had severe pain/fatigue/physical function scores compared to 3.3%/3.7%.8.7% in controls (p < 0.001). GEE models demonstrated that cases had a higher mean fatigue score of 5.34 points (95% CI = 3.81 to 6.88), higher mean pain interference score of 4.92 points (95% CI = 3.42 to 6.43), and lower mean physical function score of 7.39 points (95% CI = -8.84 to -5.93) indicating more severe symptom scores for cases in all three domains. Conclusions: On average, patients experiencing death within 6 months of PROMs completion demonstrated worse physical functioning, pain interference, and fatigue scores compared to their age, gender, cancer disease site, and stage matched controls. The differences in PROMs scores represent a first step to improving understanding of PROMs during the terminal stage of disease and may guide indicators for earlier need of supportive oncology support.