Abstract
389
Background: Integrating electronic patient-reported outcomes (ePROs) into electronic health records (EHRs) provides real-time symptom monitoring to enable early detection and timely intervention, which can reduce health care resource use (HCRU). Patients with cancer have high rates of HCRU that include emergency department (ED) and urgent care (UC) visits and planned and unplanned hospitalizations. As part of the Northwestern University Improving the Management of Symptoms During and Following Cancer Treatment (NU IMPACT) study, we examined the longitudinal associations between ePRO completion and HCRU. Methods: This observational study included adult cancer patients who received an EHR-initiated ePRO invitation linked to an NU oncology visit (maximum once per month). Sociodemographic, cancer diagnosis, and ePRO completion (any and proportion completed/ offered) up to 1-year prior to the index date (i.e., date first eligible for an ePRO during the study period), and frequency/types of HCRU (i.e., ED, UC, hospital admissions) in the year after the index date were extracted from the EHR. Multivariable regression models adjusted for NU Clinic, sociodemographic information, and cancer type were used to examine associations between ePRO completion and counts of ER/UC and planned hospital admissions in the year after the index visit. Negative binomial regression was used for ER/UC count outcome and zero inflated negative binomial regressions for all and unplanned hospitalization count outcomes. Results: Patients (N = 20,160), on average, were 64 years old, female (69.7%), White (79.5%), non-Hispanic or Latino (89.7%), English-speaking (94.9%), married (63%) and had Medicare (57.4%). Breast cancer (34.7%) was the most common site. Patients completing any ePRO prior to the index date (36%) compared to no ePRO (64%) had lower rates of HCRU for ER/UC visits (IRR = 0.94, 95% CI: 0.89 - 0.98, p = 0.008), planned (IRR = 0.83, 95% CI: 0.74 – 0.92, p < 0.001), and unplanned (IRR = 0.81, 95% CI: 0.72 – 0.92, p < 0.001) hospital admissions. Higher ePRO completion rates were associated with lower HCRU for ER/UC (IRR = 0.88, 95% CI: 0.83 – 0.94, p < 0.001), planned (IRR = 0.70, 95% CI: 0.60 – 0.82, p < 0.001), and unplanned hospital admissions (IRR = 0.70, 95% CI: 0.59 – 0.84, p < 0.001). Participants who either did not complete ePROs or had lower completion rates, and had public insurance or self-pay, experienced higher HRCU compared to those with private insurance (p < 0.05). Conclusions: Patients who do not complete ePROs may be at a higher risk of increased HCRU compared to those who do complete ePROs. The proportion of ePRO completion is a predictor of HCRU. Understanding factors that can optimize ePRO uptake may lead to improved HCRU. Clinical trial information: NCT03988543 .