Study objective: To study the prevalence, risk factors, and consequences of ambulance diversion.
Design: Observational cohort analysis from January 1, 1986, to December 31, 1989.
Setting: Population-based study of a large urban region located in Northern California.
Patients: Individuals transported by ambulance to any of 13 hospitals in the region (n = 153,167).
Measurements: Diversion defined as the patient not being transported to their initially intended hospital because the hospital was unable to accept patients because of temporary emergency department closure. Ambulance run time recorded by radio contact was documented in ambulance registry. "Transport-associated deaths" were measured as any deaths occurring in the field, while en route, or soon after arriving at the ED.
Results: During the four-year interval, total diversions increased by 453% (n = 718 in 1986 versus 3,973 in 1989; P < .005), thereby affecting one in nine transports during the last quarter of 1989. Diversion was more common in elderly patients (odds ratio, 1.17; 95% confidence interval, 1.11, 1.23), during the winter (odds ratio, 1.36; 95% confidence interval, 1.31, 1.44), and at night (odds ratio, 1.30; 95% confidence interval, 12.4, 1.37). Compared with their nondiverted counterparts, diverted transports had longer times at the scene (13.5 versus 12.4 minutes; P < .005) and greater transport times (13.3 versus 11.6 minutes; P < .005). We did not find a significant increase in the rate of transport-associated deaths (0.460 deaths per 1,000 population in 1986 versus 0.464 deaths per 1,000 population in 1989; P = NS).
Conclusion: Ambulance diversion is a common and increasing event that delays emergency medical care.