Al-driven software can help labs expertly navigate the prior authorization landscape
Uses for artificial intelligence (AI) are being explored and implemented throughout the healthcare sector, as new and exacting ways are being developed to guide treatment and support the clinical decision making process. Simultaneously, AI is also being introduced into the patient access and revenue cycle management arenas, with promised results including enhanced revenue , reduced denials, and far less burdensome workflows.
AI-driven technology represents a disruptive opportunity to revisit and transform the way business is done, bringing both apprehension and enthusiasm in equal parts. Imagining the possibilities, there is a sense of trepidation that Al could replace jobs and make human intellect obsolete. On the flip side, however, few such transformations have ever offered such beneficial reimbursement opportunities to patients, clinicians, laboratory entities, and insurance payor simultaneously.
Growing Complexities in Clinical Lab Testing
While traditional chemistry and immunodiagnostics continue to be the foundational pillars of laboratory testing, molecular and genetic diagnostics are the rising stars, with new and better tests becoming more common every day. Today, molecular and genetic testing is responsible for an estimated $8.7 billion in laboratory revenue; that tally is expected to grow to $12.9billion in 2024.1
The rapid increase in molecular testing is creating a healthcare reimbursement conundrum that is growing exponentially unmanageable. Today, there are approximately 75,000 genetic tests available on the market.2 Seeking to improve patient care and outcomes, both primary care providers and specialists are finding more and more opportunities to use molecular testing in their diagnostic processes. Such expanding use, in turn, creates growing problems for insurance payors, who are trying to contain costs while ensuring that medically sound and appropriate patient care is being delivered.
Payors’ Preferred Management Tool: Prior Authorization
With the goal of balancing the promising and increasingly numerous diagnostic testing choices against existing levels of contractual obligation, insurance payors are using prior authorization as a key tool for managing physician behaviors and patient expectations. To meet the needs of an aging population that will require more disease identification, and the desire of younger clinicians for increasingly advanced diagnostics, it seems inevitable that payors’ use of prior authorization will become more prevalent.
In 2018 and 2019, insurance payors responded to these trends by requiring prior authorization for a number of common procedural terminology (CPT) codes representing molecular and genetic testing. The list of affected tests encompasses tier 1 and tier 2 molecular pathology procedures (common and increasing complexities, respectively), genomic sequencing procedures, multianalyte assays with algorithmic analyses that include molecular pathology testing, and tests requiring the use of healthcare common procedure coding system (HCPCS) U and M modifier codes.3-5
Lab Operations Under Pressure
Over the past decade, the clinical lab sector has already been buffeted by a number of macroeconomic trends that have increased lab workloads while reducing revenues. The sector is continuing to absorb the requirements of an aging population that’s living longer and requires more care, the growing use of high deductible healthcare plans that transfer additional costs to patients, and the perpetual decline in reimbursement pay ments resulting from implementation of the Protecting Access to Medicare Act of 2014 (PAMA).6,7
Now, hospital-based clinical laboratories and reference labs alike are also straining under the administrative burdens placed on them by increasing prior authorization obligations. While prior authorization mandates issued by insurance payors started as a technique for accomplishing utilization review, they have evolved into a time consuming adnlinistrative burden directly affecting clinical labs’ bottom lines.
Affecting the entire healthcare industry, adverse clinical implications of prior authorization were detailed in a 2018 survey conducted by the American Medical Association.8 Among the results of that survey, 91% of respondents reported that obtaining prior authorization bad resulted in care delays, and 28% of respondents said that the often-delayed process had led to a patient having a serious adverse event (ie, hospitalization, a life-threate1ling event, or death).
Specific Pain Points for Clinical Labs
As an ancillary service, clinical laboratories are in the unique situation of performing vital patient testing without having any direct contact with the patient. Such a workflow makes labs dependent on the ordering hospital or physician to provide accurate patient demographics and insurance information when forwarding specimens and testing materials. Labs need this information in order to properly bill insurance payors and process prior authorization requests.
Additionally, when the ordering provider does not initiate prior authorization, it becomes the lab’s responsibility to seek approval from the patient’s insurance company. Since insurance companies typically recognize the sample collection date as the date of service for billing purposes, labs often find themselves submitting prior authorization requests after the fact.
Such ‘retroactive’ prior authorization requests are becoming less viable. United Healthcare and Anthem, two of the country’s largest payors, are no longer granting appeals for claims that have been denied because of a retroactive prior authorization request.9 This policy decision significantly undercuts a laboratory billing technique that has long been used to secure reimbursement, and necessitates more write-offs from denied claims.
Although significant campaigns to reform prior authorization policies and practices are under way through the American Medical Association and in Congress, little change has so far been accomplished.10-12 Meanwhile, 87% of participants industry wide are still manually processing prior authorization requests using labor-intensive methods such as phone calls and faxes that can require several hours to several weeks to complete.13,14
Applying Artificial Intelligence to Clinical Laboratory Billing
Analysts believe that emerging strategies using Al can play a role in improving business performance in many fields. AI techniques are especially suited for identifying ways to improve workflows and reduce time consuming, repetitive tasks that are currently being performed manually.
When using AI-driven software, a pre-defined set of algorithms uses statistical analysis to unlock data insights and then supports data-driven decisions that improve the timeliness and accuracy of targeted outcomes. AI programming can combine large chunks of data through iterative processing, enabling the software to ‘learn’ by memorizing patterns in the data (Figure 1).15
Seen through the lens of laboratories challenged to obtain prior authorizations for their services, Cloud-based Al-driven software can be integrated bidirectionally, using a facility’s laboratory information system for accessioning and client management. When a patient order arrives, tests requiring prior authorization can be identified electronically; provider/facility detail, patient demo graphics, and test/diagnosis information are then collected and added to the record; and the request is then submitted to the insurance payor electronically, in real time.
AI-driven software is especially suited to the lab testing environment, where billing and payment involve thousands of insurance groups and plans, each with unique guidelines. AI-driven software using machine learning capabilities can continuously update insurance information and clearing house parameters, automatically determine whether prior authorization is required, and route the request to the appropriate insurance payor portal. Such a system enables prior authorization approvals that used to take hours or days to be completed in seconds, with an accuracy rate better than 98% (Figure 2).16
Implementing a Secure Cloud-Based Solution
lnfinx’s prior authorization software is a seamless and scalable solution that uses Health Level 7 or application programming interface-based integration, and is compatible witl1 all leading electronic health record, electronic medical record, and laboratory information systems. The software embeds all protected health information in layers of security that are compatible with electronic data interchange requirements, and stores the data on the Cloud, using 64-bit and 256-bit encryption that guarantees compliance with the requirements of the Health Insurance Portability and Accountability Act of 1996 (HIPM; Figure 3).17,18
With such a system in place, prior authorizations can be tracked using real-time analytics and followed-up continuously, with the status reported back to the lab upon completion. When appeals are needed, the lab can define how to handle them, and the system can seamlessly execute the appeal protocol. This feature eliminates the need for staff to spend countless hours on hold or faxing information to the many different insurance payors, each with their own set of guidelines. Changes to demographic, procedure-level, or service-level information can be critical to the prior authorization process. Using an AI-driven solution, unexpected changes can be accommodated and retransmitted in real-time directly to tile affected insurance payor (Figure 4).
While AI-driven software is capable of doing amazing things in the prior authorization process, having a support team of highly trained specialists is required. AI must be supplemented with human intelligence to handle emergent or complex requests and exceptions . Building a team of trained specialists helps to ensure that a laboratory has complete coverage of its prior authorization processes.
Efficient Prior Authorization Increases Revenue
Without a doubt, clinical laboratories are under severe pressure from changes in fee schedules and reduced reimbursement models that are being implemented industrywide. With future healthcare outcomes reliant on more and better diagnostic tools, labs find themselves in a financial bind that’s only going to get worse.19
LabCorp and Quest Diagnostics, both publicly traded, have indicated that they expect to encounter significant fee schedule reductions due to PAMA. LabCorp announced that it had a fourth-quarter, year-over-year decline of 4% in 20181.9 With PAMA-mandated reductions scheduled each year through at least 2023, labs are looking for creative ways to strengthen their bottom lines.19
Using an AI-driven automated solution for prior authorizations can have a strong positive effect on a lab’s revenues. The 2019 CAQH Index, a widely respected bellwether, has estimated that the current cost for manual processing of a prior authorization is $14.24, while the cost for electronic processing is $1.93-leaving a potential industry wide savings of $12.31 for each occurrence. While the CAQH me thodology accounts for a lot of variables that can differ from region to region-including salary costs, insurance company variants, and procedure/test mix the cost savings revealed are without a doubt significant and worth consideration.
The business of diagnostic testing has 1 grown exponentially more complicated with the introduction of new and more elaborate molecular and genetic testing capabilities. Insurance payors have sought to manage the use of such advanced technologies by inter posing prior authorization requirements that have significantly increased laboratories administrative burden.
Clinical labs can address such obligations by embracing AI-driven software that offers the potential to recognize subtantial cost savings through decreasing workloads, significantly reduced denials and accounts receivable rework, and improved revenue capture.
Automating the prior authorization process not only improves the current operations of a clinical laboratory but better positions that organization as circumstances continue to evolve in the future.
AI-supported functionalities in the health care billing lifecycle will continue to evolve toward a richer ability to evaluate clinical detail, and to be certain of medical necessity for prior authorizations before they are sub mitted for approval.
Navaneeth Nair is vice president for products at Infinx Healthcare. He has been in the health care technology field for more than 16 years, working collaboratively with some of the largest healthcare payor in the industry He ha spent the past 7 years immersed in the appli cation of artificial intelligence to amplify business process improvements.