Peer-to-Peer Scheduling Is a Provider-Abrasion Metric

Ask a UM medical director what a peer-to-peer review is for and you will get a good answer: it is the moment a coverage decision meets clinical context that the paperwork didn't carry. Ask the ordering physician on the other end of that call what a peer-to-peer is and you will often get a different answer: forty minutes on hold, a callback window that landed mid-clinic, and a reviewer from an unrelated specialty reading the same criteria the denial letter already quoted.

Both descriptions are accurate, which is the problem. The clinical value of the peer-to-peer conversation is real; the logistics wrapped around it are where the value leaks out. And because the logistics are what the provider experiences, P2P scheduling is not an administrative detail — it is one of the most direct provider-abrasion metrics a plan owns.

What a P2P is, and where it sits in the lifecycle

A peer-to-peer review is a clinician-to-clinician discussion between the ordering or treating provider and a payer clinical reviewer, typically offered when a prior authorization request is heading toward — or has just received — an adverse determination. Depending on plan policy and program rules, it can function as a pre-denial consultation (a chance to supply the missing clinical context before the determination is issued) or a post-denial reconsideration conversation that may or may not change the outcome.

The regulatory floor underneath it is thinner than most people assume. For Medicare Advantage, 42 CFR 422.566(d) requires that an expected adverse medical-necessity determination be reviewed by a physician or other appropriate health care professional with expertise in the field of medicine appropriate to the service — but that reviewer need not be the same specialty as the treating physician, and federal rules do not mandate that a peer-to-peer conversation occur before an initial denial. On reconsideration, 422.590(h) requires a different physician with appropriate expertise. Some state UM statutes go further on reviewer qualifications and P2P availability for commercial lines; verify your own states rather than assuming a uniform rule. In practice, then, most of the P2P experience — who is offered one, when, with whom, inside what window — is plan policy. Which means it is designable.

One clock-related caution: under CMS-0057-F's timeframes (72 hours expedited, 7 calendar days standard, live since January 2026 for most impacted payers), a pre-denial P2P has to fit inside the determination window. An offered-but-unscheduled P2P does not pause the clock. Plans that treat "P2P offered" as a reason a request can sit are converting a goodwill gesture into a turnaround-time breach.

Why scheduling friction turns into abrasion — and appeals

Walk through the failure chain from the provider's side:

• The window problem. The plan offers a P2P window — often 24 to 72 hours, sometimes business hours only — that assumes the physician can step out of clinic when the reviewer calls. A surgeon in the OR or a primary care physician with a full panel cannot. The window closes; the denial stands by default.
• The matching problem. The reviewer who does connect is licensed and credentialed but practices nowhere near the clinical question. An oncologist arguing a regimen with a reviewer who has never prescribed it experiences the call as theater, not review. Federal MA rules explicitly do not require specialty match, so if you want matching, you have to build it.
• The phone-tag problem. Missed call, voicemail, callback to a main line, second missed call. Every failed connection consumes the window and hardens the provider's belief that the process is designed to run out the clock.
• The preparation problem. Neither side arrives with a shared packet. The provider doesn't know which criterion failed; the reviewer doesn't have the note that answers it. Ten minutes of a fifteen-minute call go to reconstructing the file. CMS-0057-F's specific-denial-reason requirement helps here — the provider should at least know why — but a reason code is not a prep packet.

Every failed or hollow P2P has a downstream cost with a paper trail: the dispute that could have resolved in a fifteen-minute conversation becomes a formal appeal, with its own regulatory clocks, staffing burden, and — since the CMS-0057-F metrics era began — a publicly reported outcome. Industry physician surveys, including the AMA's annual prior authorization survey, have consistently reported prior-auth process friction as a major source of practice burden and payer dissatisfaction; the P2P call is one of the few moments in that process where a plan is talking to a physician live, in real time. Making that moment a scheduling failure is choosing abrasion at the highest- leverage point available.

Design P2P ops like a scheduling product

Treat P2P as a product whose users are ordering clinicians, and most of the fixes become obvious — they are the same patterns as any appointment-booking system:

• Bookable slots, not callback roulette. Publish reviewer availability as discrete slots the provider's office can book — through the portal, through intake staff, at minimum through a scheduling desk that commits to a specific time. A booked slot converts an open-ended window into an appointment both sides can keep.
• Specialty-aware routing. Maintain a reviewer roster tagged by specialty and common request types, and match at booking time. Where a true specialty match isn't staffable, disclose that at booking rather than letting the provider discover it on the call.
• Prep packets, both directions. At booking, generate a shared one-pager: the request, the specific criterion at issue, the denial reason, the documentation received, and what's missing. Send it to both participants. The call should start at the disagreement, not at the case number.
• Delegate-friendly logistics. Let office staff book, reschedule, and join for the administrative preamble. Requiring the physician to personally navigate the plan's phone tree is friction with no clinical function.
• Escalation and no-show rules. Define what happens when the reviewer misses the slot (automatic rebook with priority, not silence) and when the provider does (one grace rebook inside the clock). Symmetry here is cheap and buys credibility.
• Clock awareness by design. Every P2P offer and booking should carry the underlying request's deadline, so scheduling never quietly outruns the determination window — the same discipline your SLA dashboard applies to pends and expedited requests.

What to measure

If P2P is a product, it gets product metrics. A workable starter set, all computable from scheduling and UM system data you already have:

• Offer-to-connect rate — of P2Ps offered, how many conversations actually happened. This is the single best abrasion proxy in the set.
• Time-to-connect — median hours from offer to completed call, and the share completed inside the determination window.
• Attempts per connection — average dial/callback cycles per completed P2P. Phone tag, quantified.
• Specialty-match rate — share of completed P2Ps where reviewer specialty matched the request's clinical domain.
• Outcome shift — share of completed P2Ps where the determination changed (either direction), tracked separately pre- and post-denial. A near-zero shift rate with high P2P volume says the calls are ritual; a very high shift rate says your first-pass review is missing information it could have collected upstream.
• Downstream appeal rate — appeals filed after completed P2Ps versus after missed ones. This ties the scheduling investment to the metric your CFO and your public reporting both care about.

Review these monthly with the medical directors, the way you review turnaround compliance. When outcome-shift is concentrated in one service line, that's a criteria or documentation-requirements problem to fix upstream. When offer-to-connect is the weak number, that's scheduling — and scheduling is fixable with product work, not with more clinical FTEs.

The plans that treat the P2P call as an obligation to survive will keep generating appeals and resentment from the same conversations that better logistics would turn into resolutions. The scheduling layer is the cheapest part of the whole apparatus to fix — and the only part the provider actually touches.

Verify reviewer-qualification and reconsideration requirements against 42 CFR 422.566 and 422.590 and your applicable state UM statutes; P2P availability rules vary by program and state.