An intelligent authorization platform relies on AI and machine learning to extract patient-specific data from a number of sources including the EHR. This helps health plans better anticipate, manage and optimize member outcomes.
By Brian Covino, M.D., FAAOS, Chief Medical Officer & Matt Murphy, EVP of Network & Clinical Content
In 2018, six national advocacy associations—representing physicians, health plans, pharmacists, hospitals and medical groups—released a consensus statement outlining recommendations for reforming prior authorization (PA). Yet four years later, PA remains an impediment to delivering timely, necessary patient care.
In 2021, 93% of physicians surveyed by the American Medical Association (AMA) reported that PA led to care delays at least some of the time, and 34% of physicians said PA has contributed to a serious adverse event for a patient in their care. While many health plans are focused on improving PA efficiency through the digitization of submission and clinical review, this merely speeds up a broken process. It does not transform PA into a more valuable tool for care management.
Contrary to popular belief, the PA process provides a valuable opportunity to encourage high-value care choices. With greater interoperability, providers and health plans can share patient-specific clinical data to see the full context of the patient’s care path—including services offered by other providers—and ensure optimization from start to finish.
Making evidence-based choices for an entire episode of care
Currently, health plans treat each PA request as an individual transaction with a binary outcome: “approve” or “deny.” By unlocking a richer view of their members’ clinical record, however, health plans can enable proactive management of each patient’s unique care journey.
An intelligent authorization platform relies on AI and machine learning to extract patient-specific data from a number of sources including the EHR. With each member’s longitudinal care history in hand, health plans can better anticipate, manage and optimize member outcomes.
By automatically suggesting additional services that might be appropriate for a bundled authorization, an intelligent platform can reduce the time and cost of PA while promoting high-value care decisions. Instead of submitting several disconnected preauthorization requests for one patient, physicians can get multiple services approved upfront for an entire episode, speeding patient access to the most appropriate care.
Using evidence-based criteria, an intelligent platform can also guide physicians toward the highest-value decisions. A physician might be prompted to change an inpatient surgery request to a more appropriate outpatient setting, or to skip straight to a gold-standard imaging modality instead of requesting multiple, low-value tests. This not only ensures appropriate utilization for the broader population, but also promotes conservative therapy and reduces unnecessary testing.
Providing the transparency that fosters physician trust
When it comes to PA, the systemic disconnect between health plans and providers is due in part to the inherent opacity of the clinical policies that drive decisioning. A health plan’s utilization management (UM) program is intended to evaluate the efficiency, appropriateness and medical necessity of the treatments and services its members receive. Of course, the standard of care may vary by specialty, geography or other factors not accounted for in clinical policy design.
The lack of in-workflow transparency into clinical policy leaves much to be desired by providers. At the moment, providers do not always know which services even require PA, and almost a third (29%) believe that PA criteria are “rarely or never evidence-based.” While health plans do in fact use evidence-based criteria to guide their policies, clinicians typically have no insight into how these policies are determined or what they entail.
As a result, clinicians do not always know whether their PA request will meet the health plan’s standards, or if they’re including the right documentation for the request. Without knowledge of the member’s care history and anticipated care plan, health plans are also operating in the dark. This lack of transparency drives delays and denials that could easily be avoided, which in turn erodes provider-payer trust.
To make UM a more collaborative process, full transparency is essential. The 2018 Consensus Statement on Improving the Prior Authorization Process outlined five key areas of PA reform: selective application of PA; continuity of care; program review; automation; and transparency, including the clear articulation of “prior authorization requirements, criteria, rationale and program changes.”
Creating a more transparent PA process reduces friction, as it becomes clear to physicians that a plan’s approval requirements are anchored by best practices. When physicians understand the clinical rationale behind the plan’s policies and know in advance what is required for an approval, they are more apt to entertain a specific high-value care suggestion. By offering providers the choice to accept or reject clinical recommendations—and by showing them why a recommendation has been made in the first place—an intelligent authorization platform can begin to move the needle on improving patient outcomes.
Speeding up the revenue cycle without increasing costs
When care is delayed or denied, provider reimbursement is also delayed, and in some cases is never realized. While the current PA process already sits far upstream in the revenue cycle, it does not use clinical data to influence the care a patient receives, which means it does not necessarily promote better overall health outcomes.
Our current PA process also has little impact on member costs. While some health plans will call members to inform them they can receive an approved procedure more cheaply at a different in-network facility than the one requested, this intervention would be more helpful if it occurred automatically during the PA process. An intelligent authorization model can deploy health plan-specific rules to prompt the physician to select preferred ancillary providers, thus impacting the overall cost of the procedure for patients and encouraging cost efficiency among providers.
The PA process is highly fragmented, which also drives delays and lost revenue. Providers typically deal with between five to 10 health plans, each of which likely has its own portal, fax process and 278 EDI process, not to mention wide variance in coverage policies. While some technology solutions attempt to integrate these processes with a front end that functions for multiple health plans, these solutions do not support meaningful clinical intelligence. PA typically requires much denser clinical information, which cannot be accommodated by EDI alone.
By enabling a single user experience for all authorization requests, an intelligent authorization platform saves time for providers, accelerating the revenue cycle. Such a platform promotes a clinically intelligent care path that anticipates the likely course of a patient’s journey, recommends an evidence-based course of action, and aligns the patient and health plan with every provider relevant to the episode of care. The result is better outcomes at a lower cost of care, before administrative efficiencies are even figured into the equation.
Payment processes can then be accelerated significantly for providers who consistently follow these evidence-based care paths. By automatically expediting PA requests from physicians in the highest tiers of performance, physicians whose requests fall outside of accepted clinical standards can be encouraged to change their behavior. Rewarding high-performing physicians helps health plans incentivize high-value care choices, reduce variation and satisfy providers.
If the healthcare system is going to resolve these entrenched challenges, we must ensure that PA is fully functional for both providers and health plans—and that it centers on improving care. When we stop thinking of PA as a payer-imposed roadblock and begin thinking of it as an opportunity for true care management, we can begin to optimize care for each individual patient.
Share:
About the Author: Cohere Health
