Data mining is not a new concept and it has been traditionally used in the financial, marketing, retail and manufacturing sectors. It is useful with regard to credit scoring and detecting fraud in financial organizations; direct marketing, cross-selling and up-selling for marketers; market segmentation and store lay out for retailers and in the areas of maintenance scheduling and quality control for manufacturers. In healthcare also, data/knowledge discovery is becoming increasingly popular. For example, medical billing data is mined to identify high risk areas, understand new trends and practices, and evaluate innovative policies and procedures.
Major Applications of Data Mining in Health Systems
In Identifying Treatment Effectiveness: Mining of data can help trace the effectiveness of medical treatments. By comparing and contrasting causes and symptoms, or courses of treatments, the process can identify which course of action is more effective. For example, when patient groups are treated with different drugs for the same disease, it will help determine the treatment that works best and is more cost-effective.
In Health Care Management: Data mining applications can be used to identify and track chronic disease states, identify high risk patients, provide adequate information to physicians and help provide better healthcare at lower cost. Moreover, it can identify areas for quality improvements, such as treatment guidelines, disease management, and cost management. Better disease management helps to avoid clinical complications and reduce the number of hospital admissions and claims.
In Customer Relationship Management: Customer interactions in the healthcare sector may occur through call centers, offices of physicians, healthcare departments, billing departments, ambulatory care settings, and inpatient settings. To establish and maintain good relations with customers, analysis of customer and patient related data becomes important. Commercial organizations in the healthcare industry are using data mining techniques to identify healthcare usage patterns, determine preferences, and identify the present and future needs of individuals. They can identify the products that a patient is likely to purchase, determine whether a patient is expected to comply with the prescribed treatment or not, and so on.
In Fraud Detection: Fraud detection mainly uncovers the issues of duplicate billing, up-coding, over payments, and many more. Data mining and data processing helps to establish certain rules and standards, and any fraudulent activity is identified through sophisticated data mining techniques that find any abnormal or unusual patterns in medical claims submitted by physicians, clinics, laboratories, etc. Estimated values from medical billing experts reveals that the U.S. Government is losing around 30 cents of every dollar to medical billing scams. It is a punishable offense to create false medical billing notes and obtain payment.
Just as in other business sectors, data discovery applications have great significance in the healthcare sector. To ensure error-free data, providers have to find effective ways for better capturing, storing, preparing and mining of data.