Diagnostic Discordance in Pediatric Critical Care Transport: A Single-Center Experience.

Objectives: The aims of the study were to describe diagnostic discordance rates at our pediatric tertiary care center between the reason for transfer of critically ill/injured children (determined by the referring institution) and the inpatient admission diagnosis (determined by our accepting institution), to identify potential factors associated with discordance, and to determine its impact on patient outcomes.

Methods: We conducted a retrospective chart review of all critically ill/injured children transferred to the Johns Hopkins Children's Center between July 1, 2017, and June 30, 2018. All patients whose initial inpatient disposition was the pediatric intensive care unit were included.

Results: Six hundred forty-three children (median age, 51 months) from 57 institutions (median pediatric capability level: 3) met inclusion criteria: 46.8% were transported during nighttime, 86.5% by ground, and 21.2% accompanied by a physician. Nearly half (43.4%) had respiratory admission diagnoses. The rest included surgical/neurosurgical (14.2%), neurologic (11.2%), cardiovascular/shock (8.7%), endocrine (8.2%), infectious disease (6.8%), poisoning (3.1%), hematology-oncology (2.2%), gastrointestinal/metabolic (1.9%), and renal (0.3%). Forty-six (7.2%) had referral-to-admission diagnostic discordance: 25 of 46 had discordance across different diagnostic groups and 21 of 46 had clinically significant discordance within the same diagnostic group. The discordant group had higher need for respiratory support titration in transport (43.9% vs 27.9%, p = 0.02); more invasive procedures and vasopressor needs during the day of admission (26.1% vs 11.6%, P = 0.008; 19.6% vs 7%, P = 0.006); and longer intensive care unit (ICU) and hospital stays (5 vs 2 days; 11 vs 3 days, P < 0.001). When compared with respiratory admission diagnoses, patients with cardiovascular/shock and neurologic diagnoses were more likely to have discordant diagnoses (odds ratio [95% confidence interval], 13.24 [5.41-35.05]; 6.47 [2.48-17.75], P < 0.001).

Conclusions: Seven percent of our critically ill/injured pediatric cohort had clinically significant referral-to-admission diagnostic discordance. Patients with cardiovascular/shock and neurologic diagnoses were particularly at risk. Those with discordant diagnoses had more in-transit events; a higher need for ICU interventions postadmission; and significantly longer ICU stays and hospitalizations, deserving further investigation.