• December 7, 2022

Zero-Day Hackers Breach Samsung Galaxy S22 Twice In 24 Hours

Last year, during the Pwn2Own hacking event in Austin, Texas, the Samsung Galaxy S21 was hacked, not once but twice, across a period of just 48 hours. This year, at the …

How To Get The Most Out Of Your Branding

By Tommy Mello, owner A1 Garage Doors, a $100M+ home service business. Sharing what I’ve learned to help other entrepreneurs scale. You can buy a purse for $30. Or you can …

Cities Face Long-Term Neglect, Not Just A Real Estate “Doom Loop”

There’s been a sudden spike in worrying about city problems created by declining commercial real estate (CRE) values, especially urban office buildings where increased working from home (WFH) has reduced in-office …

Is the “digital workforce” now a reality? Witness the acceptance of an intelligent virtual assistant is a full-fledged member of the workforce at Humana, one of North America’s largest health insurers and providers. This assistant isn’t looked upon as a tool or machine, and it’s not replacing anybody. Rather, it’s a super-helpful “coworker” who grows and learns with the medical and insurance professionals working alongside of it.

The intelligent assistant, called “Allie,” was designed to alleviate the documentation burden faced by the company’s clinicians, according to Joe Bechtel, principal of enterprise automation strategy advancement for Humana. I had the opportunity to meet with Bechtel at the recent Automation Anywhere event in New York. where he discussed the process of designing and introducing Allie to his company’s workforce.

Thanks to Allie, clinicians have reduced their administrative burdens by up to 15%, enabling them to focus more directly on patient care, Bechtel says. “We first started a task force to help scale out our machine learning, desktop office automation. across the enterprise. We looked at automation as an enabler for our people — now we can have more personal interactions with them.”

As Bechtel’s team broke down and looked at their company’s quality processes, they made a startling discovery: 80% of clinician time was spent “documenting their interactions, through care plans and medical charting, rather than interacting with their patients,” he says.

Advertisement

But Allie and related AI or analytics-driven insights weren’t simply built and dropped on top of employees, Bechtel explains. Clinicians were intimately involved in the process from day one, which was key to Allie’s success. “Anytime I would go to a business unit within the company, clinicians would say, ‘you don’t understand my job. You just don’t get it,’” he says. “We overcome that by embedding them into that process — to be a part of that journey from the beginning, from ideation through design and then implementation.”

Bechtel knew the virtual assistant concept was catching on as his conversations with employees shifted. “At first, every time I stepped into the room, they would be like, ‘Joe, you’re crazy, you cannot automate those processes,’” he relates. “But then they started asking, ‘what can I name my bot?’”

While it may seem trivial, naming the bot was an important part of the adoption process, Bechtel says. “We needed to have a deliberate branding around our automation — so people won’t be thinking that this is just a fad, but this is a coworker that will evolve with them.” Out of those discussions came the name “Allie.”

As an intelligent coworker, Allie will not sit still or remain static. “Allie will grow and mature with the associates as they as they learn new skills,” Bechtel says. “So as their jobs transform, and evolve to more complicated work — not a swivel chair or copying and pasting. Allie will also evolve and upskill as we layer in document processing, and more [natural language processing] work, as our program matures. we were very deliberate on how we did that approach.”

As a bona fide coworker, Allie takes processing well beyond simple number-crunching or spitting out reports. “We start off with design thinking, like how we train both users and the system,” Bechtel says. “We would have art-of-the-possible sessions, and teach them the best flows, best use cases for automation.”

For example, Allie now assists clinicians in the claims area. “Before you would have off-label use of drugs and therapy,” says Bechtel. “The pharmacist in this case would have to look up the information, and make the right judgement call to approve or deny that claim. They didn’t have access to the stats on how often that claim would be approved in the past based on that demographics of the patient. Now, Allie can go through and collect that information, including the approval or denial rate for drug coverage.”

Allie’s purpose, he adds, is “to help clinicians, not to remove them from making a decision, but to augment them — so they can have all relevant information at their disposal.”

Advertisement

Leave a Reply

Your email address will not be published.