How artificial intelligence can save health insurers $7 billion

Insurers can save up to $7 billion over 18 months using technologies driven by artificial intelligence, according to an Accenture report.

About three-quarters, or 72 percent of payer executives, said within the year, AI will be one of their top three strategic priorities for their organization.

Accenture identified six areas where AI can make a difference in an insurer's operating model, and said the top three are in anticipating and resolving customer questions, improving the benefits loading and design process and accelerating prior authorization and clinical review of claims.

Money saved by using AI in the six areas include $2.1 billion in managing customer interactions; $1.4 billion in managing membership and billing; $1.1 billion in managing and support reimbursement by automating claims processing and reviews; $1 billion in managing network and providers; $.9 billion in performing health management to engage members in improving outcomes with intelligent solutions; and $.5 billion in managing quality improvement and compliance via automated reporting and regulatory updates.

In workforce management, the result of automating core administrative functions using AI equates to unlocking $15 million in operating income for every 100 full-time employees, the report said.

The savings from automation can be reinvested for more customer support services, which is important, the report said, since payer loyalty is low.

Consumers expect more as they pay more out-of-pocket for their healthcare. Thirty-seven percent of millennials cite service as a driver for switching health plans, according to the report.

The top areas for customer satisfaction that drive net promoter scores are evaluating insurance, buying insurance and resolving administrative functions. Payers can invest in technologies to provide a richer shopping and support experience to members, the report said.

AI can be implemented in applying advanced call analytics to deploy proactive customer outreach, combined with automated communications. This can deflect the potential influx of avoidable calls, while improving overall satisfaction by anticipating the needs of customers, the report said.

AI can be used to prevent duplication downstream in the benefits capture process. A benefits-capture utility can be developed using natural language processing to validate benefits data entry.

AI can identifying and compile claims, which is now often a manual process.

Prior authorization currently requires multiple stakeholders to approve requests for treatment. To automate and standardize medical policy, robotic process automation can be deployed to auto-approve requests. Intelligence automation and virtual agents can streamline the intake of information for the initial steps of eligibility and prior authorization requests, allowing agents to focus on more complex cases.

Machine learning can streamline the process of reviewing claims by providing a recommendation for handling a pending claim that a human can approve or modify. Over time, self-learning will enable faster and more accurate responses to claims needing clinical review, the report said.

To be implemented, AI transformation needs a stable and efficient infrastructure. Payers must identify gaps such as data quality and security.

The organization must have the governance structure to manage projects through to implementation.

The right vendor must be identified and there must be a roadmap to implementation, the report said.

Accenture is a consultant and services company. The report used information from a 2017 Accenture survey of  10,000 consumers to understand how their insurance companies perform across nine healthcare consumer experience touchpoints.

Twitter: @SusanJMorse
Email the writer: susan.morse@himssmedia.com