Prior Authorization Burden Costs for Small Medical Practices

Prior authorization does not cost every practice the same amount. It costs small practices far more, proportionally, than it costs the hospital systems and large groups that dominate the conversation about administrative burden. This piece maps every layer of that cost, from the labor hours owners already track to the patient revenue and staff knowledge that disappear without ever showing up on a ledger.

Prior authorization costs small practices more than larger systems that spread the burden across departments

In most practices, at least three employees are typically involved in completing a single prior authorization request, and more than a third of practices spend a substantial amount of time per request on average. What differs is what happens to the labor once the request is submitted. A large system routes that labor to a dedicated department staffed for exactly this purpose. A small practice routes it to whichever front-desk employee or nurse happens to have a free hour, so the identical task consumes a far larger share of total operating capacity. There's no scale advantage to absorb the volume, no department to spread it across, and no slack in the schedule to hide it in.

MGMA's 2025 PA landscape report found the burden lands hardest on independent practices, and it lands hardest of all in rural and underserved areas where margins are already thin. Those are the same practices facing rising operating costs and staffing shortages severe enough to force closures. Prior authorization does not cause those closures on its own, but it adds fixed, recurring administrative weight to a business that has no cushion to absorb it.

The volume itself sets the terms. The average practice completes roughly 39 prior authorization requests per physician each week (AMA's 2024 survey), and that workload eats close to 13 hours of dedicated staff time per physician, per week. That's nearly two full business days lost, every week, for every physician on staff. A hospital system spreads 13 hours across a PA department built for exactly that purpose.

The direct costs practice owners do count: staff time and labor overhead

Staff labor is the cost every practice owner already sees, since it shows up on payroll and gets discussed at every budget review. Even so, most owners underestimate it, because they count the time spent submitting a request and miss the time spent on follow-up calls, peer-to-peer reviews, appeal documentation, and the rework that comes from a rejected or incomplete submission.

MGMA found that in most practices, at least three separate employees get pulled into completing a single PA request, and more than a third of practices report spending a substantial chunk of time per request on average. It's a relay across the front desk, clinical staff, and billing, and each handoff adds time and a chance for something to fall through.

Some practices now run full-time specialists whose entire job is chasing down approvals, an expense that did not exist a decade ago and one that shows no sign of shrinking. The cost per transaction only sharpens the picture: manual PA processing runs roughly double the cost of electronic processing, driven by labor, follow-up calls, documentation errors, and repeated resubmissions. A 2026 analysis estimates that a small, three-provider practice can spend six figures annually managing prior authorization in-house, counting wages, overhead, software, and training, before a single indirect cost gets added to the total.

MGMA's Annual Regulatory Burden Report found that a large majority of surveyed practices have hired or reassigned staff specifically to handle growing PA volume, pulling resources away from patient care to feed a workflow that generates no revenue on its own. Practices that staff a dedicated PA role have effectively turned a payer administrative requirement into a fixed payroll line, one that grows every time PA volume grows and comes with no offsetting revenue to match it.

Most owners stop their accounting right here, at the payroll line they can see and measure, and that is the mistake. Labor and overhead are the most visible cost of prior authorization, not the largest one.

Lost physician productivity: the cost layer owners least often count

Physician time costs more per hour than any other resource in a small practice, and prior authorization pulls physicians directly into work that has nothing to do with treating patients. Gathering documentation, sitting through peer-to-peer calls, drafting clinical justification letters: none of it generates revenue, and none of it appears on a line item labeled "prior authorization."

A cost analysis drawing on research into primary care physicians who manage their own PA tasks puts the productivity loss at $2,161 to $3,430 per physician per year, and that figure only covers direct time lost. It says nothing about the downstream effects on scheduling, referral capacity, or the number of patients a practice can actually see in a given week.

The switch between clinical judgment and administrative paperwork carries a cognitive cost as well. Switching between clinical judgment and administrative paperwork degrades performance on both sides of the switch, a real cost that resists measurement on any standard financial statement. The AMA's prior authorization survey found that a large majority of physicians believe PA delays actively harm clinical outcomes, so the productivity cost isn't confined to lost billable hours. It compounds into care quality itself.

A 2026 analysis built on AMA survey data found that ninety-four percent of physicians say PA contributes to burnout. In a small practice, burnout appears as reduced clinical hours, an accelerated retirement decision, or a physician leaving the group altogether, and any one of those outcomes costs more than a full year's PA labor bill combined.

Payer denial behavior turns a process cost into a revenue loss

Payer denial behavior is what makes prior authorization expensive, not the paperwork. It's the revenue that disappears once a denial goes unappealed, uncollected, and quietly written off as the cost of doing business.

Denial rates swing widely by payer and by market segment. A 2026 analysis of federal data found that among Medicare Advantage insurers, standard PA denial rates ranged from the low single digits at some plans to considerably higher rates at others, while certain ACA Marketplace insurers denied up to 25 percent of standard requests. Post-acute care takes the hardest hit: Medicare Advantage insurers denied the majority of requests submitted by long-term acute care hospitals, and more than half of requests from inpatient rehabilitation facilities. Those numbers say certain service lines carry denial risk far above what a practice-wide average would suggest, and a practice referring heavily into post-acute settings needs to know that before the denials start arriving.

Denial decisions are also arriving faster now, and not in a way that favors providers. UnitedHealthcare, Humana, Aetna, and BCBS plans have all deployed AI systems that check PA requests against internal clinical criteria databases and return decisions in hours instead of days. Speed sounds like relief, but AMA's 2025 survey data shows denial rates from AI-assisted review run substantially higher than denial rates from human reviewers. The process got faster. It did not get more favorable to the practice on the other end of it.

Payers are tightening the window to fight back at the same time they're speeding up denials. UnitedHealthcare cut its peer-to-peer review request window significantly, effective the first quarter of 2025. Humana's Medicare Advantage plans still follow the federally mandated 72-hour expedited response window for urgent requests, a CMS requirement that recent policy changes left untouched. Multiple BCBS state plans have shortened their standard appeal windows for non-urgent cases. MGMA reports that prior authorization remains the single largest administrative burden cited by medical group practices, and that burden concentrates most heavily in Medicare Advantage plans. Practices with the heaviest MA exposure end up facing the highest denial volume and the shortest window to do anything about it.

Why most denied claims are never worked

The single largest revenue loss tied to prior authorization is what happens after the denial: the overwhelming tendency of small practices to leave denials unappealed, which quietly converts recoverable revenue into a permanent write-off.

HFMA data found that a large majority of denied claims are never reworked at all. Most small practices are writing off money they are legally owed, and not because the money can't be recovered. They lack the staff hours to go after it. That's an operational choice with a price tag attached, even when no one frames it that way at the time.

The appeal paradox makes the silence especially expensive. A 2026 KFF analysis of federal data found that a large majority of PA denials appealed in Medicare Advantage get overturned, roughly two-thirds of them, compared to about half in Medicaid managed care and a smaller share in ACA Marketplace plans. Overturn rates at that scale are not a footnote. They indicate that a meaningful share of initial denials are preventable or lack clinical justification in the first place, and that a practice's decision not to appeal is, from the payer's side of the table, exactly the outcome the denial was designed to produce.

The Medicare Advantage numbers get sharper still. The same KFF analysis found that even among the small fraction of MA denials that practices bothered to appeal, roughly 80 percent came back fully or partially overturned. Any practice writing off MA denials without a fight should treat that number as a reason to examine its write-off rate as a business decision, not an unavoidable cost of doing business. Small practices absorb this loss because systematic appeal work takes exactly the kind of dedicated administrative capacity they don't have, and that gap in capacity is itself a direct cost of operating at small scale.

Patient abandonment and attrition: the revenue loss that never generates a claim

When a patient abandons treatment because of a PA delay or denial, the practice loses revenue that never enters the billing system, making it invisible to standard financial reporting and nearly impossible to attribute correctly to prior authorization burden.

A 2026 cost analysis found that a large majority of physicians report patients abandoning recommended treatment specifically because of PA barriers. Every one of those abandoned treatments is lost revenue that no denial report will ever flag, because no denial was ever entered into the system to begin with.

Patient trust erodes right alongside the revenue. Delays and denials tend to make patients blame the practice rather than the insurer actually responsible for the holdup, and that misplaced frustration depresses review scores, slows referrals, and wears down the reputation a small practice depends on to bring in new patients. It's a diffuse, slow-moving cost, and it rarely gets attributed to prior authorization on any expense report, even when PA is the root cause. Medical Economics reports that most physicians say PA barriers lead patients to abandon recommended therapies altogether, and nine in ten physicians told the AMA that authorization delays actively harm clinical outcomes. The financial loss and the clinical harm are the same event, just viewed from two different vantage points.

Small practices carry more exposure here than large systems do. The patient panel is smaller, referral relationships tend to be personal rather than institutional, and a reputation for administrative friction spreads through word of mouth faster in a tight community than it does across a large system where any single bad experience gets diluted among thousands of others.

Staff turnover as a compounding cost: how PA burden accelerates the cycle

Prior authorization work drives staff burnout, and burnout drives turnover. Each departing biller or authorization specialist takes payer-specific knowledge out the door with them, the kind of knowledge that made PA work faster and more accurate, and losing it resets the learning curve and raises the cost of every denial cycle that follows.

A 2026 analysis identifies burnout and turnover as a cost that almost never appears on a spreadsheet labeled "prior authorization," even though recruitment fees, retraining time, and the productivity dip during onboarding all trace directly back to a workflow that drives staff away. Ninety-four percent of physicians report that PA contributes to burnout in AMA's 2026 survey data, and in a small practice, one departing authorization specialist can erase years of institutional knowledge: which payers behave which way, which denial reason codes are worth fighting immediately, which documentation format each plan actually accepts before rejecting a claim on a technicality.

That knowledge loss costs more than the recruitment fee spent to replace the person. The denial rate increases and the appeal success rate declines while a new hire relearns, from scratch, patterns the previous employee had already mastered. That's the mechanism by which PA costs compound instead of holding steady year over year: each turnover event resets the learning curve and raises the error rate, the higher error rate raises denials, the higher denial volume raises rework, and the added rework drives more burnout and more turnover behind it. Left alone, that cycle does not plateau. It accelerates, and each turn of it costs more than the one before.

CMS-0057-F and the 2025 payer pledge: what changes for small practices

Reform is underway, and it deserves to be taken seriously as far as it goes. Some of those commitments take effect in 2026. Others don't land until 2027. None of them, on the timeline currently in place, will close the cost gap facing small practices in the near term, which makes prior authorization a present operational problem no matter how much improves further down the road.

The CMS Interoperability and Prior Authorization Final Rule, known as CMS-0057-F, requires covered payers starting in 2026 to return standard PA decisions within seven calendar days and urgent decisions within 72 hours. Insurers also face a first public reporting deadline, requiring them to post PA metrics, including approval and denial rates, so practices and regulators alike can finally see performance data that payers have kept opaque for years.

Those are structural changes that will matter for practices still standing when they take full effect. Faster decisions mean less time revenue sits in limbo. Public reporting means payers face a new kind of accountability for denial patterns that used to stay hidden inside internal systems. Still, the labor costs documented earlier, the six-figure administrative overhead, the physician productivity loss, the unworked denials, the patient attrition, and the turnover cycle, none of that pauses to wait for a 2026 or 2027 rule to take hold. A small practice operating today faces every one of those costs today, on the current timeline, with the current staff, under the current rules. Reform is coming. It is not here yet, and the practices absorbing this burden right now do not have the luxury of waiting for it to arrive. Altair Health, an AI-powered medical billing service backed by in-house expert billers, runs denial management and revenue cycle operations for independent and small-to-mid-size physician practices so they stop writing off recoverable revenue while regulators catch up.