Driving Faster Prior Authorizations in Healthcare with NLP and AI Technology
Pushback around prior authorization trends continues to make headlines, from a federal push to shorten prior authorization response times to a controversial attempt to require authorizations for gastroenterology procedures, a policy that has since been altered. But amid calls for meaningful change, there’s a way health systems can ease this process now, and it involves using Natural Language Processing (NLP) and Artificial Intelligence (AI) to ensure payers receive patient data in a structured format to speed decision-making processes…
One of the biggest challenges health systems face when it comes to prior authorizations is that the data doesn’t arrive in a form that health plans can immediately act upon. While most of these requests are sent electronically, typically, they take the form of an unstructured PDF. This makes it difficult for payer systems to ingest the data, requiring at least some manual data entry on the payer side before it can be sent for review. And even when a request is marked “urgent,” the request might not make it into the appropriate queue for urgent response.
Combine this with data errors resulting from manual entry and the struggle to keep up with health plans’ varying rules and forms for prior auth, and it’s a recipe for delayed care. More and more, prior authorization is required not just for specialty procedures like chemotherapy, but also for basic imaging and medication refills. These scenarios not only result in frustration for patients and providers, but also put lives at risk, such as when a cancer patient must wait for an approval before another session of chemotherapy can be delivered.
There is a better way—and it’s accessible to providers now.
The Power of NLP and AI to Speed Prior Authorizations
Seven out of 10 physicians say the volume of prior authorizations increased from 2021 to 2022, an MGMA survey found. Nine out of 10 blame prior authorization requirements for care delays, and 82% say breakdowns in this process led patients to abandon treatment. The impact on care outcomes is steep: One out of three physicians say delays caused by prior authorization led to a serious adverse event.
NLP and AI technology, when combined with a powerful integration engine can be a health systems’ best bet for streamlining the prior authorization process.
NLP and AI work together by transforming unstructured PDFs—even handwritten documents—into the format health plans want to receive, which varies by payer. This facilitates faster decision-making, speeding referrals to specialists and access to treatment. It also protects continuity of care and improves patient outcomes, when the patient is able to receive the right care at the right time.
At Consensus, our Clarity solution leverages the most advanced NLP and AI capabilities available to extract clinical information and patient demographic data, like the patient’s name, birthdate and member ID number, from the PDF and apply it to the right fields in a digital form. This ensures the data can be easily consumed by the health plan’s IT system. And, unlike other commercial offerings, Clarity leverages machine learning to recognize data fields and increase confidence scores, which in turn, leads to faster prior authorization decisions.
As mentioned previously, many authorization requests are sent to payers in unstructured data formats including pdf images from cloud faxes. Our solution eFax ®, the most widely used platform for transmitting ePHI by fax for more than 25 years, can be combined with Clarity and our powerful integration engine Conductor, to take data exchange capabilities one step further. This level of innovation allows health systems to send a message using one type of protocol and convert the message to a more advanced structured standard without doing any heavy lifting. This means a digital fax can be converted to an HL7, FHIR or X12 message. Intelligent data extraction can provide vital data leading to actionable insights that streamline prior authorizations, so that patients can get the treatment they need sooner, and avoid any potentially serious health events. It’s Time for a Smarter Approach
During a panel discussion at ViVE this past spring, we were discussing data-sharing protocols when I asked attendees, “How many of you, as patients, have faced issues with getting your prior authorization approved by a health plan?” Quite a few hands rose. It’s an issue that affects all of us, because as healthcare professionals and leaders, we are all ultimately patients as well. I am sure we all have stories of delayed care, if not for ourselves, for a loved one
As we advocate for change in the industry, including decreased turnaround times for prior authorizations and greater use of gold-carding to fast-track requests, we must also leverage the technologies at our disposal to solve challenges with prior authorization in the near term. NLP combined with AI, in an integrated setting, provides the basis for a more proactive approach to improve outcome and patient satisfaction.
Click here to find more information on Consensus’ approach to NLP and AI for prior authorization. I would love to hear from you if you have any comments or questions about the use of NLP and AI solutions in healthcare.
