Enhancing Abuse/Fraud Detection in OPD Insurance Through Rules-Based Customer Risk Scoring Approaches

Abstract

This thesis investigates the challenges of outpatient (OPD) health insurance fraud detection and proposes a solution that uses a rules-based approach to score every single customer based on their policy utilization patterns & behavior and assigns a risk score to enhance abuse / fraud detection in the field. The research is motivated by the growing prevalence of abuse / fraud in the OPD insurance industry and the need for more effective abuse / fraud detection methods to protect both insurers and policyholders. The study aims to identify the key characteristics of OPD insurance abuse/fraud and develop a comprehensive set of rules for customer risk scoring for abuse/fraud detection. The research questions focus on the effectiveness and efficiency of the proposed solution.
The study uses a large secondary dataset of OPD insurance claims and policyholder data and shows that the combination of rules leads to a more robust customer risk scoring which helps in raising alerts and signals to prevent abuse and/or fraud. The proposed solution is expected to have a significant impact on the OPD insurance industry, including the identification of high- risk customers, discovery of their syndicates and nexus, blocking their policies, recovering lost money, and reducing operational expenses.
We propose a novel approach utilizing a set of pre-defined rules to flag customer data points indicative of potential risk factors. These flags can be assigned weights based on their relative
importance in predicting claim behavior. A customer's risk score will be calculated as a normalized value between 0 and 100, with higher scores indicating a greater risk of claims.
We will be deriving the Customer Risk Score (RS) and it will be calculated using a weighted/non-weighted sum of the rules.
The masked and anonymized data on Outpatient Department (OPD) claims will be used to develop and validate the risk scoring model. Analyzing this data will allow us to evaluate the effectiveness of the proposed rule-based approach in identifying high-risk customers within the insurance population.
In summary, this thesis contributes to the development of more effective abuse / fraud detection methods in the OPD insurance industry and provides valuable insights for practitioners in the field. By using a rules-based approach, the proposed solution offers an effective and efficient solution to identify and prevent abuse / fraud in the industry. The findings of this study have important implications for the OPD insurance industry and pave the way for further research in this area.