AI and Data Analytics Improve the Employee Health Experience

photo of a doctor talking to a patient

Data is the backbone of health care. From individual diagnoses to population health measures to coding and billing, each piece of information and every data trend helps point the way. But with so much personal information included in that data, privacy and security are key, as are data integrity and HIPAA (Health Insurance Portability and Accountability Act) compliance. 

Artificial intelligence (AI) and automation present new opportunities to improve health care based on large stores of data. Today’s AI technology is making it easier to provide personalized care, understand health equity, avoid bias, and cut costs.

All data tells a story

Every health care journey — individual or public — is ultimately a story. Those stories are best understood by looking at the data, says Ian Blunt, vice president of analytics at Highmark Health. Highmark Health is the parent company of Highmark Inc. and its affiliate Blue companies, which are independent licensees of the Blue Cross Blue Shield Association. 

“When we layer multiple member journeys on top of one another, we can see trends, predict outcomes, and positively influence those outcomes. The data improves care for individuals, while also helping to manage costs for the employer,” Blunt explains in Highmark Health’s “Health Care Reinvented” podcast series.

Better employee health powered by AI

Technologies like AI, machine learning (ML), and natural language processing (NLP) power Blunt’s advanced analytics work. His team of more than 80 data scientists, researchers, and engineers has oversight of enterprise data analytics for the entire Highmark ecosystem. 

Individual member data merges into a centralized system programmed with alerts for different scenarios. For example, if an employee with coverage from Highmark Inc. or one of its affiliate health plans experiences a serious health event, analytical models predict what may happen next based on individual historical data and population health data. 

“In this instance, we can look at hospitalization and discharge rates for the specific condition and see when readmissions are most likely to occur, whether they spike at seven, 14, or 21 days after initial hospitalization,” says Blunt. 

Understanding that data allows for earlier intervention, whether that is a follow-up appointment, referral to a specialist, or recommendation to talk to a health coach. The result for the member is mitigation of negative outcomes, a healthier life, and lower costs of chronic illness and/or (re)hospitalization.

Using data to uncover bias

Blunt’s team takes precautions to mitigate any kind of bias in their data analytics work. They employ a three-part process that begins with the use case for the data. “We immediately look at the ethical approach and search for challenges against the use case,” Blunt explains. “We ask if there is unintended bias.” 

The second step involves developing predictive analytical models for the use case. The data analytics team investigates the data to make sure that it applies to different groups across socioeconomic segments in varied locations. This step helps reveal inequities and find ways to better serve marginalized communities. 

Finally, Blunt’s team monitors all analytical models once they are deployed. They look for changes that may be leading to weighted outcomes and reassess the original use case for adjustments to data sets.

The impact and promise of AI

Beyond the work of Blunt’s team, Highmark is deploying AI in other ways that support improved health care delivery and lower costs.

One of these advanced tools (from a third-party solutions firm) is in use at in-network provider Allegheny Health Network (AHN). The tool uses AI to predict infectious disease events before they occur, helping AHN hospitals plan ahead for staffing, bed space, and essential resources. 

Highmark used AI to save $245 million in 2021 by preventing health care waste, fraud, and abuse and resolving billing errors. The technology makes it easier to detect insurance fraud and errant claims, while simultaneously guarding against emerging scams and protecting members. 

“There is a lot of promise for AI and new technology developments,” Blunt concludes. “It’s almost limitless. But we need to proceed in ways that are safe and treat everyone’s data with maximum privacy and respect. The next two decades are going to be truly spectacular for health care advancement and improvement.” 

Listen to the “Health Care Reinvented: Data Analytics with Ian Blunt” podcast for more insights and explore the full podcast series for the latest Highmark news.

All references to “Highmark” in this communication are references to Highmark Inc., an independent licensee of the Blue Cross Blue Shield Association, and/or to one or more of its affiliated Blue companies. 

Find a plan that fits your business needs

Talk to your client manager today and we’ll help you find the right health plan for your company.