What Prior Authorization Outcomes Don't Show About the Work
Payers subject to CMS-0057-F published their first required prior authorization metrics this year. The disclosures cover the percentage of requests approved, the percentage denied, the percentage approved after appeal, the percentage approved after the review timeframe was extended, and the average and median time from submission to decision, reported separately for standard and expedited requests. Every one of those measures describes a decision or the time it took to reach one. None of them describes the work that happened before the decision was made. A request approved on first submission and a request approved after three documentation exchanges, a resubmission, and a handoff between departments appear in those disclosures as the same result.
Same Outcome, Different Work
Two requests can reach the same determination through very different paths. One arrives complete, moves through review without intervention, and is approved. The other requires outreach to the ordering office, a corrected submission, and clarification of which plan requirements apply before it is approved. Both count once toward the approval rate.
Turnaround time does not separate them either. It measures elapsed time, not the activity inside it. A request can sit pending for five days without anyone touching it. Another can be resolved in two days after six follow-up contacts. The longer turnaround belongs to the request that required less work.
The Prior Auth's Operating View was built around that distinction. It adds three measures to the outcome data: Authorization Workload, Failure Origin, and Resolution Burden. The framework was presented at the HFMA Great Lakes Fall Series in October 2026 as part of *What Your Denial Rate Isn't Telling You*, a session examining the operational activity that conventional authorization measures leave out.
Authorization Workload Starts Before a Request Is Submitted
Authorization Workload measures the volume and composition of authorization activity entering the operation, not only the requests that reach a determination.
Much of that activity happens before anything is submitted. Staff determine whether authorization is required, identify the plan and its requirements, collect clinical documentation, and work out which submission pathway applies. Some requests move straight from there to a decision. Others pick up exceptions, repeat contacts, and corrections along the way, and some problems are resolved before a payer ever sees the request.
A count of determinations captures the end of that activity. Authorization Workload captures what the operation was actually asked to process and where that work came in.
Where a Problem Is Found Is Not Always Where It Started
Failure Origin identifies where an authorization problem first began.
Problems usually become visible downstream, when a payer requests additional documentation, a submission is rejected, or a determination stalls. Those events show where the problem was found. A documentation request during payer review may trace back to intake, where the required information was never collected. A rejected submission may reflect an earlier misreading of plan requirements. A delay discovered during follow-up may have started at a handoff between two teams.
The point of detection and the point of origin often involve different staff, different processes, and different safeguards. Failure Origin looks past the event that exposed the problem to the first identifiable breakdown, and asks whether the available records or workflow accounts support placing it there. An operation that only tracks where problems are discovered can keep measuring the same failures without ever locating them.
Resolution Burden Is the Work Added After Something Goes Wrong
Resolution Burden measures the effort required to address an authorization problem once it occurs.
That effort can include additional documentation, repeat payer contact, internal escalation, resubmission, and coordination between clinical and administrative staff. Some problems are fixed with a single intervention. Others take several attempts or are still open when the observation period ends. Depending on what an operation records, Resolution Burden can be measured through follow-up activity, rework, handoffs, and time spent in unresolved status.
An approved request can carry a heavy resolution burden. A denied request can carry almost none. The measure does not assume every exception is a failure. It records what additional work occurred, what prompted it, and how the operation responded.
What Improving Outcome Numbers Can Hide
None of the three measures replaces approval rates, denial rates, or turnaround times. Those remain useful measures for understanding determinations and timeliness, and they are what payers are now required to publish. The Operating View describes the activity underneath them.
That matters most when outcome numbers improve. A falling denial rate does not establish that staff are doing less rework. Faster determinations do not establish that fewer interventions were needed. A shift in request volume or mix can move reported rates without the underlying workflow becoming any easier to manage.
The outcome data records that both requests were approved. It takes a different set of measures to show what each one took to get there.
Additional information about the framework is available through The Prior Auth's Operating View resources. The framework also informs The Prior Auth's research and consulting services, which examine authorization workflows, historical workload, and operational visibility.
The Prior Auth Report provides monthly analysis of published payer data, prior authorization policy, and the administrative patterns shaping utilization management.
If you'd like structured analysis delivered directly to your inbox, you can subscribe to the newsletter below.