For health plans, developing a data-smart approach to cost reduction typically means analyzing data to identify high-cost members and more effectively engage them in their care. With healthcare costs projected to rise 6 percent this year, however, it’s clear that new approaches are needed. One emerging trend is using predictive analytics and artificial intelligence (AI) to bend the cost curve.
How can health plans most effectively achieve cost transformation with AI and predictive analytics, and where do they start? Here are two approaches to consider.
Use predictive analytics to manage risk more effectively
Predictive analytics could drive healthcare cost savings of 15 percent or more over the next five years, according to a 2018 report. Predictive analytics position health plans to draw upon one of their greatest assets — a large inventory of member data — to forecast health events before they happen. This technology also empowers health plans to pinpoint members who are emerging as high-risk and could become high-cost in the absence of care management interventions.
One health plan decreased short-term disability rates by 15 percent using predictive analytics to identify employees at high risk of short-term disability and initiate nurse-led interventions to lower their risk. Another uses medical and pharmacy claims data to identify members at high risk of developing diabetes, as well as those who likely have diabetes but are undiagnosed, and to target the right interventions to help prevent life-threatening complications.
To fully leverage predictive analytics, health plans should break down internal data silos to support the ability to aggregate, analyze, visualize and operationalize data from every area of the organization. When a plan’s Medicare Advantage, Medicaid and commercial lines do not share data insights, opportunities to improve clinical outcomes are limited — as is the potential to increase operational efficiency and lower costs.
Health plans also should invest in the right talent to support a predictive analytics approach. Such specialists include not only data scientists but also healthcare economists, who can articulate the potential impact of emerging trends and better equip health plans to design an effective response.
Integrate AI into administrative functions
Health plans could save more than $7 billion by incorporating AI into administrative functions, including claims review and processing, a 2018 report found. For example, AI can accelerate prior-authorization approval through robotic process automation that evaluates the request against the plan’s medical policy. Such technologies also help ensure the right care is provided for the right member at the right time through pathways such as:
Aiding clinical review. To ease claims-review demands on clinicians, machine learning can provide recommendations for action on pending claims. Reviewers can then approve the recommendation or take a different approach. Each time a recommendation is modified, machine-learning technologies gain intelligence to refine their approach and provide more-accurate responses.
Streamlining claims intake. Use of intelligent automation and virtual agents to receive claim information enables staff to focus on the most complex cases. This approach is widely used by property and casualty insurance companies to increase efficiency and decrease costs.
Preventing fraud. An AI-supported claims review process can help identify instances of fraudulent or erroneous billing before a claim is paid, resulting in substantial cost savings. When instances of fraud or errors — such as false billing, duplicative billing and billing for services that have not been performed — are not detected before a claim is paid, the health plan’s chances of recovering those dollars are much lower.
Creating the right infrastructure to support AI is critical. So is finding talent with skills in AI, and that’s a challenge for 38 percent of healthcare organizations, according to a survey.
Health plans should devise a strategy for attracting and cultivating IT specialists with the skills needed to launch and manage AI innovations. These positions include data scientists, data engineers and experience designers. A survey found 63 percent of companies are now providing in-house analytics training to meet the demand for AI talent.
The future is now
In 2019, focusing on decreasing utilization is no longer sufficient to significantly affect healthcare costs, given that many of the potential gains in this area already have been achieved. Health plans instead should look for new ways to support more cost-effective care by applying predictive analytics and AI to their vast trove of clinical, financial and demographic data. Such a strategy could be a game changer for healthcare cost transformation.