In his rehabilitation medicine practice in Illinois, Dr. Azlan Tariq typically spent seven hours a week battling insurance companies reluctant to pay for his patients’ treatments.
He often lost.
There was the 45-year-old man who spent five months in a wheelchair while his insurer rejected appeal after appeal for a prosthetic leg. Or the stroke survivor who was readmitted after a fall after his insurer determined his rehabilitation “could be done at home.”
Over the course of Dr. Tariq’s 12-year career, these stories became more common: The list of treatments requiring pre-approval from insurers seemed to grow longer and longer, and the denials seemed to increase.
So, in an effort to spare his patients what he considered subpar care and himself mountains of paperwork, Dr. Tariq recently turned to an unlikely tool: generative AI.
For a growing number of doctors, AI chatbots — which can draft letters to insurers in seconds — are opening a new front in the battle to approve costly claims, accomplishing in minutes what years of advocacy and attempts at health care reform have failed to do.
“We didn’t have any legislative or policy tools to defend ourselves,” Dr. Tariq said. “This is finally a tool I can use to defend myself.”
Doctors are turning to the technology even as some of the nation’s largest insurance companies face class-action lawsuits alleging they used their own technology to quickly reject large batches of claims and exclude seriously ill patients from rehabilitation treatment.
Some experts fear that the prior authorization process could soon turn into an AI “arms race,” in which robots compete for insurance coverage. Among doctors, there are few things that generate as much hate.
“If you want to see a doctor go crazy at a cocktail party, mention prior authorizations,” said Dr. Robert Wachter, chairman of the department of medicine at the University of California, San Francisco.
This process was designed by insurance companies to reduce health care costs by limiting doctors’ use of unnecessary and expensive treatments.
But Dr. Jonathan Tward, a Utah-based radiation oncologist, said insurers often require such extensive documentation and paperwork — even for standard cancer treatments — that he feels like he’s on the losing side of a “war of attrition.”
Physicians and their staff spend an average of 12 hours a week submitting prior authorization requests, a process widely viewed as burdensome and detrimental to patients’ health among physicians surveyed by the American Medical Association.
With the help of ChatGPT, Dr. Tward now types a few sentences describing the purpose of the letter and the types of scientific studies he wants to reference, and a draft is produced in seconds.
He can then ask the chatbot to quadruple the processing time. “If you put all kinds of barriers in front of my patients, when I get back to them, it’s going to take me a long time,” he said.
Dr. Tariq said Doximity GPT, a HIPAA-compliant version of the chatbot, has helped him cut the time he spends on prior authorizations in half. More importantly, he said, the tool, which draws on his patients’ medical records and insurer coverage requirements, has made his letters more efficient.
Since using AI to write prior authorization requests, he said about 90% of his coverage applications have been approved by insurers, up from about 10% previously.
Generative AI has proven particularly useful for doctors in small practices, who typically don’t have time to appeal an insurer’s decision, even if they believe their patient’s treatment will suffer.
Nearly half of the doctors surveyed by the AMA said that when they did not appeal a claim denial, it was at least partly because they did not have the time or resources to go through the insurance company’s lengthy appeals process.
Dr. Michael Albert, an obesity medicine specialist in Oklahoma, said AI has helped his small, resource-limited telehealth practice go from having almost no insurance denials to fielding 10 to 20 calls a week.
Now, Dr Albert said, it could “operate at the same level as companies that have essentially infinite resources”.
Although AI is still primarily used by tech-savvy individual physicians, a growing number of companies are trying to bring the technology to the mainstream.
Epic, one of the nation’s largest electronic medical records companies, has rolled out a prior authorization tool that uses AI to a small group of doctors, said Derek De Young, a developer working on the product.
Several Major health systems are testing Doximity GPT, which was created to help with a number of administrative tasks, including prior authorizations, a company spokeswoman said.
But insurance companies aren’t sitting idly by. Chris Bond, a spokesman for America’s Health Insurance Plans, said insurers welcome attempts to streamline the process, including those involving “the appropriate use of AI.”
Dr. Jeff Levin-Scherz, a health policy expert at the Harvard T.H. Chan School of Public Health, said he believes most health insurance plans are at least evaluating how to use artificial intelligence in their claims review process, if they aren’t already using it.
At the same time, these AI tools could make it easier for malicious players to bill for medically unnecessary treatments, Dr. Levin-Scherz noted.
As doctors use AI to write prior authorization letters faster, Dr. Wachter said he has “tremendous confidence” that insurance companies will use AI to better deny them.
“There’s automatically a conflict,” said Dr. Wachter, author of a book on digital technology in medicine. “Their AI is going to negate ours, and we’re going to go back and forth.”
Dr. Wachter said he hopes that one day, with more advanced AI technology, insurers and providers could create a new system that relies less on sending letters back and forth and more on automation.
Insurance coverage could be automatically approved based on an algorithm’s analysis of the most recent scientific publications. An AI tool could also inform a doctor that the $100,000 drug he or she is about to prescribe could be replaced by a $250 pill that is just as effective.
Until then, many doctors have a new, albeit temporary, advantage in the battle for insurance claims.
“It’s clearly a race for AI,” Dr. Tariq said, “but I think it wouldn’t be fair if doctors didn’t have AI.”