Improving timeliness for acute asthma care for paediatric ED patients using a nurse driven intervention: an interrupted time series analysis

Abstract

Asthma is the most common chronic paediatric disease treated in the emergency department (ED). Rapid corticosteroid administration is associated with improved outcomes, but our busy ED setting has made it challenging to achieve this goal. Our primary aim was to decrease the time to corticosteroid administration in a large, academic paediatric ED. We conducted an interrupted time series analysis for moderate to severe asthma exacerbations of one to 18 year old patients. A multidisciplinary team designed the intervention of a bedside nurse initiated administration of oral dexamethasone, to replace the prior system of a physician initiated order for oral prednisone. Our baseline and intervention periods were 12 month intervals. Our primary process measure was the time to corticosteroid administration. Other process measures included ED length of stay, admission rate, and rate of emesis. The balance measures included rate of return visits to the ED or clinic within five days, as well as the proportion of discharged patients who were admitted within five days. No special cause variation occurred in the baseline period. The mean time to corticosteroid administration decreased significantly, from 98 minutes in the baseline period to 59 minutes in the intervention period (p < 0.01), and showed special cause variation improvement within two months after the intervention using statistical process control methodology. We sustained the improvement and demonstrated a stable process. The intervention period had a significantly lower admission rate (p<0.01) and emesis rate (p<0.01), with no unforeseen harm to patients found with any of our balance measures. In summary, the introduction of a nurse initiated, standardized protocol for corticosteroid therapy for asthma exacerbations in a paediatric ED was associated with decreased time to corticosteroid administration, admission rates, and post-corticosteroid emesis.

Figure 1

Statistical process control (Xbar) chart of the mean times to steroid administration in the baseline (Prednisone) and intervention (Dexamethasone) phases.