Utilization of Transport Data to Decrease Unnecessarily Repeated Laboratory Tests.

Abstract

In pediatric patients being transported for management of diabetic ketoacidosis, laboratory tests will frequently be trended throughout transportation and subsequently immediately repeated upon arrival to a particular institution. These laboratory tests may not add value to a patient's care trajectory and therefore may be unnecessary. This study examines differences between pH, sodium, potassium, chloride, bicarbonate, and glucose levels drawn during transportation and those drawn at our home institution immediately upon arrival to determine if repeating those laboratory tests upon arrival to the emergency department serves any purpose in adding to patient care. This study compares pH, sodium, potassium, chloride, bicarbonate, and glucose levels drawn during transport and at our home institution. Box and whisker plots between transport and institution laboratory values were constructed. A Wilcoxon signed rank test was performed to determine differences between pH, sodium, potassium, bicarbonate, and glucose levels, as these value sets were not normally distributed. A paired t test was performed to determine differences between transport and institution chloride values given that these value sets were normally distributed. Savings were then calculated based on charges to the patient to determine overall cost savings by not immediately repeating these laboratory tests upon presentation. Box and whisker plots showed marked similarity between laboratory tests drawn in transport and those immediately upon arrival to our ED. Paired t test did not demonstrate a statistical difference between transport and ED chloride levels ( P = 0.5699); therefore, we failed to reject the null hypothesis. Wilcoxon signed rank test did not demonstrate a statistical difference between transport and ED pH ( P = 0.1294) and potassium ( P = 0.4523) values; therefore, we failed to reject the null hypothesis. However, Wilcoxon signed rank test did demonstrate a statistically significant difference between uncorrected sodium ( P = 0.0006), corrected sodium ( P = 0.0075), bicarbonate ( P = <0.0001), and glucose levels ( P = 0.0086). Although there were some statistically significant differences between the laboratory value sets, it is arguable whether there are any clinically significant differences between them.Based on our failure to show a clinically significant difference between laboratory values drawn during transportation and those drawn immediately upon presentation to the institution, repeating laboratory draws after transportation do not add value to a patient's care trajectory. We should therefore rely on the laboratory values that were drawn from our transportation teams as part of the continuum of patient care.