Owning and Reducing the Cost of Health Care in the Emergency Department

Abstract

Market forces and legislative action are restructuring US health care, with the hope of making care better and cheaper. A key element in this reformation is changing from reimbursement based on quantity and price of care and procedures to reimbursement based on quality or outcome related to price. The American College of Emergency Physicians (ACEP) has had 2 advocacy positions relating to pricing and cost: first, that emergency department (ED) charges account for less than 2% of total health costs; and second, that the marginal cost of seeing an additional nonurgent patient in the ED is low. In this issue of Annals, Lee et al address these important issues of cost of and accounting approaches for emergency care with new data and analyses. First, the authors conclude that the ED fraction of the health care dollar is probably higher than 2% and may be as high as 5% to 10%. They also address the issue of measuring cost, including marginal cost, by proposing use of a new activity based costing (ABC) accounting system. In this editorial, we examine the differences between the ACEP advocacy positions and the findings of Lee et al. We also seek to examine how these positions fit with the current restructuring of health care. The ACEP first position used the Medical Expenditures Survey to conclude that ED care is a bargain, less than 2% of overall $2.6 trillion dollar total US health care spending. ACEP was not alone among specialties and experts in using the survey to examine costs. However, Lee et al highlight many of the limitations of drawing conclusions from this survey. It is based on the memory of randomly selected subjects. Patients who die or are homeless, severely disabled, or demented by the time of the survey are underrepresented because they cannot or do not participate. These omissions make any derived figure an underestimate of the true cost of care. However, Lee et al, in describing their own methodology, use a number of assumptions and data from other sources that themselves have assumptions and unknowns. This weakness clouds the end imputed cost in a complicated fashion that cannot be unraveled. The result is that although we now know that 2% is probably not the correct figure, we do not know the exact figure. What does seem likely is that the earlier and current approaches together define the boundaries of cost, somewhere between 2% and 10% of total health care spending.