By Anju Sah, DGM Operations, HGS Healthcare
Today’s payers are burdened by the complex and traditionally manual and labor-intense task of finding aberrations in services billed that could be indicative of fraud, waste, and abuse. Enter payment integrity as opportunity. In the past, healthcare organizations have leaned heavily on adjudicators to identify fraud. However, this focus has now shifted due to advancements in machine learning, analytics, and EDI, which all work to facilitate claims cost avoidance before the claim adjudication process and payment. These advancements, known as machine learning, are propelling actionable insights and maximum claims cost savings.
Machine Learning, Defined
Machine learning is an application of artificial intelligence (AI), which enables computers to learn and improve their capability over time through observation, data feed, and real-world interaction without human intervention. Machine learning uses algorithms to interpret data to look for patterns and adjust actions accordingly, enhancing human ability to resolve snags and take informed decisions. There are many use cases for machine learning in healthcare—for example, cognitive systems can help case managers to efficiently screen cases, evaluate them with greater precision, and make informed decisions. Additionally, hospital claims management is another area that stands to benefit.
Machine Learning in Payment Integrity
According to IDC Health Insights survey data, a large proportion of buyers of healthcare payment integrity solutions reported that they increased spending for these solutions in 2015. This trend is growing significantly. In fact, over the next several years, the survey indicated that many payers will continue to enhance their overall fraud, waste, or abuse (FWA) defensive capabilities. Machine learning is a key tool in revolutionizing efforts to uncover fraud, waste, errors and abuse in claims processing and the healthcare industry is applying this technology to address these expensive challenges.
Think of AI as the engine that drives critical improvements--assisting in strengthening claims management with smart audit algorithms and cognitive systems that reliably filter and reject incorrect claims, while routing innocuous claims for auto-adjudication; in turn reducing the administrative staff costs while effectively using their expertise to focus on claims that need manual review. Intelligent algorithms autonomously learn and evolve with each claim processed.
AI-led natural language processing (NLP) will assist in identifying the initial anomalies, becoming smarter with time and gaining ability to learn patterns and irregularities in claims. The AI driven pre-payment audit will eliminate fraud waste and abuse scenarios. Today’s advanced AI fraud detection systems examine every claim and every line item pre-payment. The smart and self-learning algorithms of the cognitive system filter out and reject incorrect claims, reducing the costs of redundant audits.
AI helps detect and prevent fraud, waste, and abuse at each stage of claims processing, avoiding inappropriate payment with the fraud analytics engine identifying payment integrity breaches, providing a structure for detecting fraud and managing alerts. Suspicious activities are detected and reported through usage of machine learning algorithms, coupled with techniques such as exception reporting, sophisticated data mining, analytics, and business rules. These intelligent systems identify opportunities to reduce duplicate services, identify irregularities in provider claims patterns, and efficiently allocate administrative staff based on the high demand, thus significantly influencing waste reduction, workforce optimization, and fraud avoidance.
Is Machine Learning Sufficient to Tackle the Current Challenge of PI?
While today’s innovation can drive breakthrough improvements, merely investing in AI and advanced analytics cannot address the challenges experienced in PI. To reap financial benefits, improve administrative efficiencies and control FWA, technology and analytics should be supported by reengineering through initiatives such as claims workflow improvements and service capability revamp. Today’s BPO reengineering, analytics, and industry expertise can condense resources to accelerate adjudication, enhance payment integrity, and improve payment recovery. The result is significant cost-containment and improved member and provider experience.