Do Hospitals Respond to Price Changes? A Study of Hospital Acquired Condition Payment Change in California

The Centers for Medicare & Medicaid Services (CMS) developed the Hospital-Acquired Conditions (HAC) —Present on Admission (POA) program for Medicare population in response to the Deficit Reduction Act (DRA) of 2005. According to the program, after October 2008, inpatient claims can no longer be assigned to higher-paying MS-DRGs on the basis of preventable complications that are acquired during the hospital stay. The goal is to reduce high cost or high volume conditions that result in the assignment of a case to a DRG that has a higher payment when present as a secondary diagnosis through the application of evidence-based guidelines (CMS, 2014).

In this paper, we study the impact of this Medicare policy change on various outcomes, including hospital’s nominal response such as coding practice, and their real response such as intensity and quality of health care services they provide. We focus on the sample of patients that receive total knee or hip replacement, which was the most common hospital procedure covered by Medicare in 2013, accounting for nearly 450,000 inpatient admissions and $6.6 billion in spending (Purvis et al., 2015).

We use data from the 2005 to 2011 Healthcare Cost and Utilization Project (HCUP) inpatient claims data for the state of California. These data contain patient characteristics, such as age, gender, race/Ethnicity, as well as their medical information, such as insurance status, diagnoses, procedure, DRG codes, charge, length of stay, hospital discharge status, and readmission. Using difference-in-differences specification, we compare the change of outcomes for patients over 65 relative to that for patients under 65. We also stratify the sample according to the percentage of Medicare patients that hospitals treat.

Results indicate that those over 65 are less likely to experience falls and trauma, a popular complication among this group of patients, relative to those under 65 after the adoption of program. They are also less likely to be readmitted to hospitals within 30 days or 90 days, compared to their younger counterparts. We do find evidence that hospital uses more resource to treat those patients, and these impacts are more prominent among hospitals with higher Medicare share of patients. The results of this study have important policy implications for the effective design of value-based purchasing programs under the current health care reform.