A Pragmatic Randomized Controlled Crossover Study to Evaluate the Impact of Messaging Interventions on Emergency Department Utilization Among Oncology Patients at Parkland Health

Overview:

This study is a pragmatic randomized controlled crossover trial designed to evaluate the feasibility and impact of structured follow-up messaging interventions on patient engagement with Oncology Acute Care (OAC) services and reduction in cancer treatment-related Emergency Department (ED) visits at Parkland Health.

Recruitment and Enrollment:

Eligible patients will be identified using an automated dashboard through Parkland's Epic electronic health record system based on the inclusion/Exclusion criteria. Patients meeting criteria will be automatically flagged for study enrollment. No active consent will be required as the intervention involves standard communication methods already available to patients (MyChart and text messaging), and the study poses minimal risk.

Randomization and Crossover Design:

Patients will be randomized into one of two arms:

Control Arm (Usual Care): Patients receive standard hospital discharge instructions without additional messaging.

Intervention Arm: Patients receive structured follow-up messages using existing education materials via MyChart and automated text messaging within 24, 48, and 72 hours post-discharge. Patients in the control arm who experience a subsequent cancer treatment related ED visit without documented OAC engagement will be crossed over to the intervention arm.

Messaging Intervention:

Patients randomized to the intervention arm will receive structured follow-up messages aimed at increasing awareness and utilization of Oncology Acute Care (OAC) services. Messaging will be delivered via both MyChart and automated text messages, tailored to the patient's preferred language (English or Spanish).

• Initial Message: Sent within 24 hours post discharge from the ED observation, or inpatient stay (10AM following discharge). This message will include:

• Contact information for the OAC clinic and triage line

• Guidance on when to seek urgent care through OAC versus emergency care, including a reminder that same-day OAC appointments may be available

• Follow-Up Messages: If the initial message is not viewed, it will be resent at 48 hours and again at 72 hours post-discharge. Up to 3 messages will be sent per discharge.

• Message Engagement Tracking: Once a message is viewed, the patient enters an outcome tracking period. The study team will record engagement metrics on whether messages were viewed to assess implementation fidelity.

Outcome Tracking:

The study team will monitor patient medical records for the following:

• Time to next cancer treatment-related ED arrivals (including observation and inpatient admissions)
• Contact with OAC clinic or triage line
• Total hospital visits

Follow-up will occur in 3-month intervals. Patients who engage with the OAC clinic will be removed from the observation list and considered to have completed the study. If a patient has another ED visit without OAC engagement, the observation period will repeat.

Data Collection:

Data will be extracted from Epic and include:

• Message engagement metrics (viewed/unviewed, response rates)
• OAC contact logs
• ED visit records
• Patient demographics and clinical characteristics

Data Analysis Plan:

Descriptive statistics will be used to summarize engagement and outcome data. Time-to event analyses (Kaplan-Meier curves with Cox proportional hazards models) will be used to compare time to ED arrival between arms. Mixed-effects logistic regression will be used to analyze secondary visit county outcomes. Subgroup analyses will explore differential effects by language, cancer type, and prior ED utilization.