Effects of Active Patient Participation in an Inflammatory Bowel Disease Quality Improvement Program on Healthcare Utilization Outcomes

Abstract

BACKGROUND Although previous studies have shown that participating in a structured national quality improvement program improves healthcare utilization for patients with inflammatory bowel disease (IBD), it is not known whether the extent of active participation influences health outcomes. Patients in the IBD Qorus Learning Health System self-report symptoms and treatment goals prior to clinic visits via electronic surveys, which facilitates disease surveillance and patient-provider co-production of care. This study assesses the relationship between active patient participation and IBD-related healthcare utilization outcomes. METHODS We reviewed electronic health records and surveys of patients with IBD from a private community gastrointestinal practice participating in Qorus for two years after they submitted the first pre-clinic survey. The study period was from 2016 to 2021. Primary outcomes were times from the first survey to an IBD-related ED visit/hospitalization, urgent message*, and CT-scan use. Participation score was calculated as a ratio of eligible surveys (submitted within two weeks prior to clinic visits) to the number of clinic visits, with a ratio <0.5 representing inactive participation and ≥0.5 representing active participation. Cox proportional hazards regression was performed to determine the relationship between active participation and healthcare utilization, controlling for potential confounders such as IBD subtype, baseline clinical disease activity, steroid use, and six-months-prior IBD-related ED visit/hospitalization and CT scan use. The “survival time” was defined as the time from the first survey to IBD-related healthcare utilization events, loss of follow-up in the electronic health records, or two years, whichever occurred earliest. RESULTS Out of 244 patients, 122 were inactive participants and 122 were active participants. Among inactive participants, 23.8% experienced an IBD-related ED visit/hospitalization, 27.9% sent an urgent message, and 12.3% received a CT scan. Among active participants, 18.9 % experienced an IBD-related ED visit/hospitalization, 21.3% sent an urgent message, and 12.3% received a CT scan (Table 1). Active and inactive (reference) Qorus participants were not statistically different in their healthcare utilization: ED visit/hospitalization (HR 0.61[95% CI 0.34-1.09]), sending an urgent message (HR 0.68 [95% CI 0.40-1.15]), and CT scan use (HR 1.00 [95% CI 0.47-2.13]), after adjusting for confounders (Table 2). CONCLUSION Although active participants show numerically less frequent ED visit/hospitalization and urgent message than inactive participants, there was not a statistically significant association between participation and healthcare utilization. A greater sample size is needed to draw further conclusions to inform future strategies to engage patients and optimize health outcomes.