OBJECTIVE

We assessed the relationship between emergency department (ED) crowding and delays in care in patients presenting with abdominal pain who receive abdominal computed tomography (CT).

METHODS

Prospective cohort study of adults who presented over a 1-year period to 2 urban academic EDs with abdominal pain and received CT. Each subject had 3 validated crowding measures assigned at enrollment (ED census, waiting room number, number of admitted patients). These were normalized to quartiles to signify least to most crowded. The Cuzick test was used for trend and log-linear regression and tested the association between ED crowding and time from triage to CT read. The time interval was further decomposed into triage to room, room to CT order, and order to CT read times. The adjusted analysis controlled for age, sex, race, pain score, time of day, triage level, and site.

RESULTS

767 patients were enrolled (mean age, 44 +/- 17 years; 61% female; 60% black). Median time from triage to CT read was 375 minutes (interquartile range [IQR], 276-497). Individual time intervals included triage to room (46 minutes [IQR, 16-111]), room to CT order (83 minutes [IQR, 38-151]), and order to CT read (203 minutes [IQR, 138-375]). Across waiting room quartiles, triage to CT read was associated with progressively longer times (318 vs 364 vs 414 vs 445 minutes; P < 0.001 for trend). Similar trends were noted for waiting room number and admitted patients (P < 0.001). In multivariable analysis, the association between ED crowding and time from triage to CT read remained significant and consistent across all crowding measures (P < 0.001). When decomposed into time intervals, triage to room time showed the greatest difference (22 vs 38 vs 72 vs 92 minutes; P < 0.001).

CONCLUSION

ED crowding is associated with an approximately 2-hour delay to CT interpretation availability. Attempts to reduce delays in abdominal CTs may include earlier provider evaluation and placement in the queue for scanning.