Streamlining Prior Authorization: Technical Answers from Our WEDI Webinar (Part 1)

In August 2025, we partnered with WEDI to host a two-part webinar series on CMS-0057 implementation, focusing on how payers can streamline prior authorization while managing delegated vendor relationships. The response was outstanding, with technical teams, policy leaders, and operational stakeholders asking detailed questions about the practical realities of implementation.

The questions revealed a common theme: organizations understand the regulatory requirements, but they need clarity on how the technology actually works in practice. How do you extract policies from existing documents? What happens when clinical guidelines live with third-party vendors? Can the system handle urgent cases differently than routine requests?

We’ve compiled the most substantive questions from Part 1 into this guide, organizing them by topic to help you navigate the technical considerations of electronic prior authorization implementation. Whether you’re evaluating vendors, planning your infrastructure, or working through integration challenges, these answers provide 

Policy Management & Extraction

How are the policies being extracted from the documents? What is the backend system and how do you test for accuracy of the extraction process?

At Itiliti Health, we’ve built tooling that parses plain-text policies and helps a policy author within the payer organization see a comparison of the source document and the resultant ingested version. For clinical criteria, we allow the user to see original criteria side by side with the breakdown of the component requirements (as individual computable questions), in order to create the machine-readable version of the policy. When implementing policies in our PA Checkpoint™ tool, many organizations are able to standardize and improve how they write their criteria. For prior authorization decisions we also support the ability to highlight parts of non-coded documents that may satisfy individual criteria within the policy through our Clinical Assistant feature.

Integration with Utilization Management Systems

Does payer UM have a particular admin dashboard managing assigning PA to nurses and then if escalated to MD? And do you handle both pharmacy UM and medical UM? If yes, would drugs be coming in medical UM as well or all go through pharmacy UM? How do you manage?

Itiliti Health solutions do not replace the payer UM system. Payers often have dedicated Utilization Management workflows for concurrent review when handling inpatient and urgent authorizations. Similarly, requests for medication prior authorizations are typically categorized separately for a dedicated formulary team. Within a Coverage Requirements Discovery (CRD) request, FHIR distinguishes Service Request, Medication Request, and Device Request. When Itiliti Health processes these requests, we can invoke APIs in the UM system and/or those provided by delegated vendors. Imaging services and management of drugs under medical benefit are the most commonly delegated types of review.

Is Itiliti Health developing a Utilization Management tool?

Itiliti Health adds functionality that is outside the scope of most payers’ existing UM systems. We integrate with Intermediaries, Provider EHRs, and Delegated Vendors using the standards defined in the Da Vinci Implementation Guides. As UM systems improve their conformance with new FHIR standards, new implementations will become easier, and our existing customers will continue to leverage Itiliti Health’s Policy Management, Auto Authorization, and Clinical Assistant features.

Working with External Guidelines & Delegated Vendors

Do you have your own decision tree on top of CDS like InterQual and could you apply AI on them for a better review process or do those CDS parties not allow it? How do you combine Payer and CDS guidelines?

Itiliti Health has our own decision tree for payer guidelines. For external vendor policies, Itiliti Health provides the much-needed “first stop”, in order to route a request based on the matching policy. We use a transparent user-controlled algorithm (not AI) to handle policy route determination. For third parties like InterQual, Itiliti Health performs as a proxy to retrieve the questionnaire from the third party and deliver the questionnaire to the provider. When the submission is returned to Itiliti Health, we route the completed questionnaire to the third party for use in their decision system. This is necessary since third parties have strict limits that do not allow their policies to be viewed or used in other companies’ technology. For the Itiliti Health Clinical Assistant decision tree, during the review of Prior Authorization Submissions, staff can use Itiliti Health’s Clinical Assistant (with AI) to parse large volumes of submitted data against specific criteria to highlight relevant information that may satisfy individual criteria.

Compliance & Turnaround Times

Are there any turnaround time requirements to be met either by health plans and/or by delegated vendors?

CMS-0057 specifies the following:

Prior Authorization Decision Timeframes: We are requiring impacted payers (excluding QHP issuers on the FFEs) to send prior authorization decisions within 72 hours for expedited (i.e., urgent) requests and seven calendar days for standard (i.e., non-urgent) requests.

The Da Vinci Implementation Guide also acknowledges that:

The timeframes for evaluation will also differ [from payers’ existing adjudication engines]. Many payers have an asynchronous prior authorization process where processing may take several minutes or even longer. CRD timelines are much shorter: 5 to 10 seconds, depending on circumstances.

Can your system flag urgent cases or help prioritize authorizations for time-sensitive care?

Yes! The specification allows providers to communicate urgency, expected date, and service type, all of which can be used to route the prior authorization submission into the appropriate queue inside Utilization Management.

Handling Real-World Scenarios

How would your CRD response handle a case where the servicing provider is unknown at the time the ordering provider submits the request?

The performing date and performing location are often unknown at the time of ordering. When these data elements are not present, we evaluate the procedure code and expected date to return as much information as possible to help the ordering provider know what conditions apply. For example, in the CRD response, we can answer “PA Needed” if that procedure always requires prior authorization, and “conditional” if that procedure is exempt from prior authorization when performed at certain surgical centers. Importantly, we return those conditions and policy instructions to help the provider understand what to do next.

Continue the Conversation

These questions from Part 1 focused on the foundational technical elements: policy management, UM integration, and delegated vendor coordination. In Part 2, we explore the mechanics of connecting these systems, including how CRD requests handle errors and missing data, the role of prefetch bundles, routing logic for third-party vendors, and EHR certification timelines under HTI-4.