Healthcare Fraud Data Mining Methods: A Look Back and Look Ahead

Area of Science:

• Health Informatics
• Computer Science
• Criminology

Background:

• Healthcare fraud is a costly white-collar crime with significant financial and health impacts on the population.
• The complexity and heterogeneity of US healthcare data systems hinder the implementation of digital fraud detection advancements.
• Effective digital healthcare fraud detection is essential to mitigate societal threats and protect beneficiaries.

Purpose of the Study:

• To describe and summarize existing healthcare fraud detection systems and methods from peer-reviewed literature.
• To identify potential gaps in the real-world implementation of these systems.
• To propose future research topics to address identified gaps in healthcare fraud detection.

Main Methods:

• Systematic review and summarization of peer-reviewed articles on healthcare fraud detection.
• Tabulation of research articles detailing objectives, conclusions, and data characteristics.
• Analysis of implementation challenges and identification of research gaps.

Main Results:

• A comprehensive overview of current healthcare fraud detection techniques is presented.
• Key data characteristics and methodologies employed in existing studies are summarized.
• Significant gaps in the practical application of fraud detection systems to real-world healthcare data were identified.

Conclusions:

• There is a critical need for advanced digital systems to detect healthcare fraud effectively.
• Addressing implementation gaps requires further research into adapting detection systems to diverse healthcare data.
• Future research should focus on developing and validating robust fraud detection methods for real-world healthcare environments.