Variation of the Supply and Cost of Treat-and-Release Visits to Hospital Emergency Departments

OBJECTIVE: To analyze variation across hospitals in the number of hospital emergency (ED) visits and to test an economic planning model that constrains the supply of visits to keep financial losses under a tolerable limit. An appendix with rigorous development of the model and implications will be provided. DATA SOURCE: Discharge records from the Healthcare Cost and Utilization Project for patients visiting EDs in 8 states in 2003. More than 11 million visits were included; 550 hospitals had complete discharge information with linked accounting reports. STUDY DESIGN: A hospital could incur an unacceptable level of losses due primarily to revenue for the uninsured being less than the cost. A two-stage model is used to first fit average cost to the area wage index, diagnostic casemix, comorbidity and hospital characteristics. Then the fitted average cost is used in a regression of the annual number of visits at a hospital on a number of measures about the hospital and the area, e.g., alternative sources of care in the area, demographics, household income and related measures. PRINCIPAL FINDINGS: A lower cost of production and higher inpatient bed capacity are strongly associated with a greater supply of treat and release (T&R) ED visits, holding constant the number of patients admitted to inpatient units from the ED, area population, age mix, and other hospitals with EDs. Income levels and poverty rates are not significantly associated with the number of T&R visits supplied. CONCLUSION: A model of limited planned supply is consistent with the data and supports the widespread view and evidence that hospital-based ED capacity is not sufficient for relatively frequent peak-load situations and rare disasters. Policies to increase supply might find the databases and our methods for cost estimation useful. Because hospitals are affected by payer mix in the ED and potential losses, strategies to increase capacity for emergent or non-urgent care in a community should anticipate the rational reactions of hospital managers. Even an attractive policy to increase supply of service outside hospitals for non-urgent care could affect the payer mix in the hospital ED and therefore the hospital's planned supply of ED capacity.