The Revenue Cycle and Medicare: An Analysis of Chronic Condition Treatments and the Effect on Hospital Revenue (Preprint)

BACKGROUND Medicare is a federal social insurance program aimed to supply health insurance to older and disabled people. In 2019, over 61million people were enrolled in Medicare [22]. The Medicare program is mostly funded through payroll taxes and Social Security income deductions, with beneficiaries being responsible for a portion of coverage costs [18]. Medicare uses a variety of administered price systems to pay health care providers and faces major challenges with setting this amount in order to avoid distorting the care patients receive, and not to overpay. (Newhouse, 2005). This study analyzes Medicare data from the Centers for Medicare & Medicaid Services (CMS) database and draws conclusions about the relationship between Medicare reimbursements for chronic disease treatments, and the sociodemographic characteristics of beneficiaries receiving the treatment. OBJECTIVE The objective of this study is to determine if there is a relationship between sociodemographic characteristics (age, gender, location, race) of beneficiaries with chronic diseases and total Medicare reimbursement amounts. In the assessment of the relationship between several factors and Medicare payment amounts, this study was conducted to determine whether location significantly impacted the amount of revenue disbursed after accounting for other factors such as race, age and gender. To emphasize, Medicare data that originated from the CMS databases were utilized to assess the impact of location while controlling other factors. METHODS A primary search of the CMS data was conducted using the terms “chronic conditions” online. The data set used for this study was the Medicare Physician and Other Supplier National Provider Identifier (NPI) Aggregate Report, Calendar Year 2017 [15]. This data was then extracted to a Microsoft Excel spreadsheet to be analyzed and filtered. A sample size of the data was analyzed for this paper using SAS software, and a regression model was created to illustrate the relationship among the variables. The variables utilized were age, gender, race, socioeconomic status (rural vs. city) to evaluate what hospitals would receive the maximum reimbursement from Medicare beneficiaries. The chronic conditions that were used to research and examine Medicare reimbursement were heart failure, diabetes, and cancer. RESULTS When we break the total Medicare payment amounts down, the top five cities in Georgia that received the largest payment were Tucker ($67,991,109.45), Atlanta ($59,302,111.16), Alpharetta ($19,822,926.69), Brookhaven ($12,082,039.90), and Atlanta ($10,519,988.37). The number of Medicare beneficiaries at these hospital locations are 519,239; 16,651; 79,904; 14,516, and 52, respectively. All five of these cities also have a provider RUCA of 1, meaning they are in an urban area. The top five cities in Georgia who received the lowest payment were Augusta ($0), Decatur ($0), Perry ($44.06), Loganville ($48.27), and Greensboro ($48.27). While three of these cities were labeled as urban with a provider RUCA < 4, these three cities would be seen as outside the perimeter (OTP) of the city. The city of Greensboro has a provider RUCA of 7.1, which sets this hospital location as rural. With this data,we can assume that hospitals located in rural areas receive a significantly lower amount of Medicare revenue than hospitals located in urban areas, or areas closer to the capital city. The number of beneficiaries at hospitals receiving lower payments also tend to be significantly low. CONCLUSIONS The top three chronic illnesses having the highest economic impact on the U.S. healthcare system are heart (cardiovascular) disease, cancer, and diabetes, respectively. The purpose of this study was to analyze Medicare data from the Centers for Medicare & Medicaid Services (CMS) database and draw conclusions about the relationship between the total Medicare payments for chronic disease treatments, and the sociodemographic characteristics of beneficiaries receiving the treatment. For cardiovascular diseases, hospitals located in urban areas and age of the beneficiary were not reliable indicators of Total Medicare Payment Amount. Gender and race both were reliable predictors of Total Medicare Payment Amount. For diabetes, hospitals located in urban areas also was not a reliable predictor of Total Medicare Payment Amount. Age and race were statistically significant, meaning they are reliable predictors of Total Medicare Payment Amount. From these results, we are 95% certain that gender, age and race play a role in the amount of reimbursement a hospital receives. For Diabetes, age, gender and race were all reliable predictors of Total Medicare Payment Amount. Urban areas were not a reliable predictor of Total Medicare Payment Amount for beneficiaries with diabetes. One thing was consistent throughout all diseases, hospitals located in urban areas did not have a significant impact on Total Medicare Payment Amount. For Cancer, gender and non-Hispanic white beneficiaries were reliable predictors of Total Medicare Payment Amount. Black or African American beneficiaries, American Indian/Alaksa Native beneficiaries, Hispanic beneficiaries and urban were not reliable predictors of Total Medicare Payment Amount.