How do hospitals respond to price changes in emergency departments?

Little is known about how prospective provider payment affects the provision of services led by unpredictable demand. We investigate hospital responses to a 32% increase in price for two treatments in emergency departments in England in April 2011 using data on 11,532,304 attendances (79 hospitals) between 2009/2010 and 2013/2014. We compare changes in the volumes of these two treatments to a treatment not attracting additional reimbursement using a difference-in-differences framework. Additional reimbursement led to 76% and 152% increases in the volumes of the two incentivised treatments. Hospitals received an additional £64.4 M between April 2011 and March 2014 for providing these treatments, of which 40% (£25.9 M) was attributable to the unanticipated hospital response to the price increase. We use time in treatment to distinguish real increases in treatment from reductions in undercoding or increases in upcoding. The association between the recorded receipt of these treatments and time spent in treatment was the same before and after the price increase, and there was no association between hospital-specific increases in recorded treatment volumes and changes in treatment times. The persistence of the treatment time increment suggests the increase in recorded treatment was a real increase in provision of treatments.