Diagnosis-based payment systems can create incentives to upcode patients to a higher level of severity to increase payment. In some instances, upcoding can be a form of fraud if providers code patients to a higher complexity than is appropriate, whereas in other instances, upcoding can accurately reflect patient acuity. We estimated the increase in Medicare Severity Diagnosis-Related Group (MS-DRG) upcoding during the period 2011–19, using all-payer discharge-level data from five states. During this period, the number of discharges with the highest MS-DRG coding intensity increased by 41 percent. Adjusting for changes in patient characteristics, length-of-stay, and hospital characteristics, we estimated that the increase would have been 13 percent in the absence of changes in coding behavior. We estimated that in 2019, the increase in upcoding (relative to 2011 coding practices) was associated with $14.6 billion in
hospital payments, including $5.8 billion from private health plans, $4.6 billion from Medicare, and $1.8 billion from Medicaid. These findings can contribute to the growing body of evidence supporting the design of payment models that limit distortions in payment and resource allocation.
Upcoding Linked to up to Two-Thirds of Growth in Highest-Intensity Hospital Discharges in Five States, 2011–19