How artificial intelligence is building bridges between practices and payers

Whether in movies, books, or the news, we continually hear about how artificial intelligence (AI) creeping into our daily lives might eventually become an existential threat. Skeptics question whether machines that mimic human intelligence should be allowed to make decisions currently left to highly skilled personnel, but within dentistry, AI is turning out to be a great coworker.

As most dentists know, dental insurers (i.e., payers) perform utilization reviews on a portion of the claims they receive. The purpose of these reviews is to ensure reimbursed services are necessary, appropriate, and meet the payer’s clinical guidelines and group contract requirements. Payers require dental offices to provide radiographs, periodontal charts, and other forms of documentation in order to review certain procedures. Specific requirements vary by payer and plan and, along with review processes, they periodically change. This inconsistency has been a source of frustration for dentists for decades, and payers have been frustrated by mounting claim review costs as additional time and communication are needed to resolve issues resulting from insufficient or noncompliant documentation. But machines don’t get frustrated; in fact, they excel at complex, repetitive, detail-oriented tasks such as screening mountains of documentation and analyzing millions of medical images. When used appropriately, AI can step in and deliver immediate benefits that reduce frustration and friction between providers and payers.

Related reading:

Ensuring complete claims

Dental office staff do their best to include all necessary documentation, but claim preparation can be confounding when requirements vary by payer. Claims may be denied, processing suspended, and reimbursement delayed until all requirements are met. Meanwhile, payers don’t have the resources to fully review the massive volume of claims streaming into their systems, so they often pay noncompliant claims. One major US payer used AI to learn that over 75% of their dental claims were paid without required radiographs or periodontal charts attached.¹ Well-designed AI can ease the administrative burden on both providers and payers by ensuring that each claim submitted includes all required data and documentation, thereby freeing administrative staff from time-consuming phone calls and follow-ups. AI can screen claims and immediately determine whether the required attachments are included (e.g., if the relevant teeth and anatomical structures are visible on the images provided). When a claim is missing a required attachment, AI can automatically search the payer’s database to see if it might have been submitted with a prior claim, or it can trigger an alert to let the dental office know what’s needed. AI can also use natural language processing to analyze narrative content to ensure it fully describes the clinical condition. When AI is incorporated into the claim preparation process, it can check for missing documentation before the claim is even submitted to the payer.

Increasing claim review efficiency

Payer systems are often highly automated, and most claims complete the adjudication process without any human intervention. Utilization reviews are performed on only a small sampling of claims, so many claim issues go undetected. Furthermore, most claims selected for review are paid,² so time and effort are often wasted. AI can instantaneously analyze every claim submitted to even the busiest payers. It can flag and annotate questionable claims and prioritize them to help payers improve efficiency and effectiveness, with their highly skilled reviewers only receiving claims that are likely to have issues. Such process improvements on the payer side benefit providers by facilitating faster payment of clean claims. They may also benefit patients by helping keep both providers’ and payers’ administrative costs under control.

Assuring claim review consistency

From a provider’s perspective, since most claims are processed and paid without human intervention, when delays do occur, they may seem arbitrary. But if AI is incorporated into the claims processing workflow, both payers and providers can be assured that when human review is invoked, it’s necessary and appropriate. AI uses statistical models to accurately and consistently assess radiographs and determine whether a procedure qualifies for payment under a payer’s clinical guidelines. Services that meet guidelines can be automatically approved, while those that do not appear to do so can be routed to the payer’s clinical staff for review. In cases such as periodontitis, where human review can be subjective,³ AI’s data-driven clinical assessment capabilities can help support clinicians’ decisions.

Prequalifying treatment plans

Today, dental office staff do their best to figure out if and how much of a patient’s recommended treatment will be covered by their insurance plan, but the answer is often unclear until a procedure has already been performed, the claim submitted, and the payer responds. That’s really frustrating for providers and patients. AI can quickly prescreen radiographs to determine whether a proposed service will meet a participating payer’s guidelines, and how much will be reimbursed by a particular patient’s plan, before service is rendered.

AI bridges the gap

Dental AI can deliver many benefits. It can be a detail-oriented administrative workhorse that outperforms an army of staff. It can provide impartial second opinions, offering clinical decision support. It can conserve cash flow and accelerate reimbursement. And, perhaps most importantly, AI can help fill the goodwill gaps between payers, providers, and patients.

Editor's note: This article appeared in the July 2022 print edition of Dental Economics magazine. Dentists in North America are eligible for a complimentary print subscription. Sign up here.

References

1. Internal data. A NovoDynamics NovoHealth dental claim quality assessment study. 2021.

2. Johnston JW. Delta Dental of Michigan takes a bite out of fraud with AI technology. NHCAA 2020 Annual Training Conference presentation. November 19, 2020.

3. Internal data. A NovoDynamics NovoHealth dental clinical assessment study. 2018