Factors associated with unplanned healthcare encounters in pediatric cancer patients

Pediatric cancer therapies can cause complications that drive unplanned healthcare encounters (e.g., emergency visits and unplanned admissions). Characterizing their frequency and risk factors is important for resource planning, family counseling, and targeted mitigation strategies. In a secondary analysis of data from a cluster-randomized controlled trial of symptom monitoring in pediatric cancer patients, the primary objective was to describe the characteristics of unplanned healthcare encounters. The secondary objective was to identify risk factors associated with unplanned healthcare encounters.

This was a sub-analysis of a cluster-randomized trial of 10 sites randomized to symptom screening and 10 sites randomized to usual care. We included English- or Spanish-speaking pediatric patients newly diagnosed with cancer who were 8-18 years of age receiving any cancer treatment. At symptom screening sites, enrolled participants were reminded to complete symptom screening with Symptom Screening in Pediatrics Tool (SSPedi), a validated symptom screening and assessment, three times weekly for 8 weeks. The healthcare team was notified for severely bothersome symptoms. Emergency department visits, and unplanned clinic visits and admissions were determined from families and charts. Encounter documentation was reviewed to determine if the unplanned encounter was related to SSPedi symptoms.

Among the 444 participants from 20 participating sites, there were 652 unplanned encounters and the most common were emergency room visits (n = 269) followed by unplanned clinic visits (n = 143) and unplanned admissions (n = 240). Pain was the primary reason for emergency room visits (59, 13.3%), unplanned clinic visits (52, 11.7%), and unplanned admissions (19, 4.3%). Nausea/vomiting was the next most common symptom for emergency room visits (40, 9.0%), unplanned clinic visits (4, 0.9%), and unplanned admissions (14, 3.2%). In multivariable analysis, leukemia diagnosis (P = 0.0003), site randomization (P = 0.018), and increasing physician full-time equivalents per 100 new cancer diagnoses (P = 0.032) were associated with more unplanned encounters.

Unplanned healthcare utilization is common among newly diagnosed pediatric cancer patients. The association between unplanned healthcare utilization and standardized symptom screening and increasing physical full-time equivalents warrant further investigation. Key aspects to the implementation of routine symptom screening among pediatric patients with cancer will be resource allocation to prevent and manage common symptoms including those that may lead to unplanned healthcare visits.