Date of Award
There has been much work to improve IT systems for managing and maintaining health records. The U.S government is trying to integrate different types of health care data for providers and patients. Health care fraud detection research has focused on claims by providers, physicians, hospitals, and other medical service providers to detect fraudulent billing, abuse, and waste. Data-mining techniques have been used to detect patterns in health care fraud and reduce the amount of waste and abuse in the health care system. However, less attention has been paid to implementing a system to detect fraudulent applications, specifically for Medicaid. In this study, a data-driven system using layered architecture to filter fraudulent applications for Medicaid was proposed. The Medicaid Eligibility Application System utilizes a set of public and private databases that contain individual asset records. These asset records are used to determine the Medicaid eligibility of applicants using a scoring model integrated with a threshold algorithm. The findings indicated that by using the proposed data-driven approach, the state Medicaid agency could filter fraudulent Medicaid applications and save over $4 million in Medicaid expenditures.
Suleiman, Muhammad Nader, "Data-Driven Implementation To Filter Fraudulent Medicaid Applications" (2014). Theses. 209.