The exasperation among doctors frustrated by prior authorization battles with health insurance providers is not unique. Prior authorization frustrates everyone in the healthcare industry — to say nothing of patients. But don’t forget about the health insurance company frustrations in this, too. For every doctor tired of submitting more information to justify warranted care, a payer is feeling the same frustration, having to make repeated requests for clinical information or waiting at the other end of a fax machine (a fax machine!) to obtain the information they need to ensure they are not spending their customers’ premium dollars on unnecessary care. What was once “healthy tension” in our healthcare system has become untenable. And the administrative costs associated with perpetuating these processes can no longer be sustained.
A recent piece in the New York Times cites a warning from Dr. Jeff Levin-Scherz, a health policy expert at the Harvard T.H. Chan School of Public Health: “AI tools may make it easier for ill-intentioned players to bill for medically unnecessary treatments.” Another physician, Dr. Robert Wachter, the chair of the medicine department at the University of California, San Francisco, predicted insurance companies would simply respond with more sophisticated AI tools of their own, and “their AI will deny our AI and we’ll go back and forth.”
These sentiments perpetuate a false paradigm. Trading blows isn’t the solution — it perpetuates the problem. It’s time to end the arms race, especially since we now have the digital tools available to do exactly that.
The mindset of using AI as a tool in the battle of the bots is not pervasive among the health plan leaders we speak with, nor is it a path toward efficiency. Using AI and related tools to simply get quicker at fighting the other side isn’t a solution, and it certainly doesn’t take advantage of the true opportunities AI presents for health care organizations, both in terms of realizing new administrative efficiencies and improving the manner in which healthcare organizations each conduct their necessary business functions.
Rather than embrace a strategy that relies on arcane tools of the past to torment one another, health care organizations today – health systems and health plans – are using AI in much more meaningful ways. These include avoiding unnecessary fights altogether and working more collaboratively in their shared clinical decision-making processes. As Chris Bond of America’s Health Insurance Plans notes in the Times, insurance companies aren’t sitting still.
For example, a national health plan began working with a regional health system six years ago to use predictive analytics that consider the clinical documentation of individual patients while receiving care in acute care hospitals to help determine inpatient care vs. observation (outpatient) status. The hospital and health insurer compared the final status of the patients they had agreed on historically against what status the AI-based tool would have predicted for those same patients. As a result, the two organizations were able to identify about 20% of cases where they had agreed on inpatient or observation, and thus did not warrant the significant resources required to review and reach status determination.
Applying AI in this manner reduced the time to reach inpatient or observation status from 55 minutes per case to nine minutes.
Dr. Azlan Tariq, a rehabilitation medicine practitioner in Illinois, notes that “it wouldn’t be fair if the physicians didn’t have A.I.” Perhaps this is a common concern among private-practice physicians who don’t have the full resources of a health system or hospital network, whose administrators and other staff have been incorporating AI technology into their processes and procedures for months, if not years. Making AI available to every healthcare provider — from the smallest private clinic to the largest national system — is a noble goal that doesn’t necessarily require an escalating arms race, with payers on one side and providers on the other.
In reality, AI is becoming more pervasive and accessible within all professions. Healthcare is an early example of two parties in the same professional space (payers and providers) using shared instances of AI to resolve their competing interests. While hospitals and insurance companies have greater resources to experiment and implement these tools compared to physician practices, the days of creating and maintaining processes and procedures that unnecessarily drive up the administrative cost of delivering and paying for care are nearing an end.
When there was no better way of doing things, staying the course made sense. But new tools are available today that can help providers and health plans work better – and together – and that’s a much more exciting future of health care than perpetuating the provider/payer arms race.
About Michael Drescher
Michael Drescher is a seasoned healthcare and communications professional with more than 20 years’ experience helping organizations succeed and navigate complex environments. He currently serves as Vice President of Payer Strategy at Xsolis.
