The AMA prior authorization survey released by the American Medical Association on May 13, 2026 was fielded in December 2025 among 1,000 practicing physicians across specialties and practice settings, and it was the first AMA survey taken after the June 2025 pledge in which roughly 60 health insurers promised to reduce and streamline prior authorization requirements over 2025 through 2027. The AMA's framing was blunt: the pledge, so far, has not changed what physicians experience.
The numbers will be familiar to anyone who runs a practice. Physicians and their staff spend about 13 hours a week on prior authorization work per physician. Forty percent of practices employ staff whose only job is prior authorization. Ninety-five percent of respondents said the process delays access to necessary care, and 92 percent said it negatively affects patient outcomes. Twenty-six percent reported that a prior authorization led to a serious adverse event for a patient. Thirty-two percent said requests are often or always denied. Eighty-eight percent said the requirements lead to higher overall resource use. And only 33 percent believe the insurer pledge will make a meaningful difference.
We think the AMA prior authorization survey is useful for two reasons that have nothing to do with lobbying. First, it gives an independent practice a benchmark to hold its own numbers against. Second, it is a reminder that almost no practice actually knows its own numbers. This article covers what the survey found, what it does and does not tell you, and how to count your own prior authorization hours in a way that leads somewhere.
Key takeaways
- The AMA survey released May 13, 2026 reports about 13 hours a week of physician and staff time per physician on prior authorization, with 40 percent of practices employing dedicated staff.
- Only one in three physicians believe the June 2025 insurer pledge will make a meaningful difference; the survey was fielded six months after the pledge.
- The survey is a national average across specialties; your practice's burden depends on payer mix and service mix, and can be measured in two weeks.
- Counting hours by payer and by service type turns a complaint into a negotiating point and a staffing decision.
- The regulatory deadlines that will actually change the workflow (CMS decision timeframes already in force for 2026 and the January 2027 API requirements) apply only to certain payer types.
What the AMA prior authorization survey found
| Finding | Figure | What it means operationally |
|---|---|---|
| Physician and staff time on prior authorization per physician | About 13 hours a week | Roughly a third of one full-time employee per physician |
| Practices with staff working exclusively on prior authorization | 40 percent | A dedicated role is now normal, not exceptional |
| Physicians reporting delays in access to necessary care | 95 percent | Scheduling and referral workflows carry the delay |
| Physicians reporting negative impact on outcomes | 92 percent | Clinical, not just administrative, consequences |
| Physicians reporting a serious adverse event caused by prior authorization | 26 percent | One in four; a risk management concern for every practice |
| Requests often or always denied | 32 percent | Appeals are a permanent workload, not an exception |
| Physicians who think the insurer pledge will make a meaningful difference | 33 percent | Practices should plan on current burden, not promised relief |
The survey sample is 1,000 physicians drawn to cover primary care and specialty, employed and independent, across regions. It is a self-report survey, and the AMA is an advocacy organization, so the numbers are best read as a physician-perspective benchmark rather than a payer audit. That said, the findings have been stable across the AMA's surveys for years, and they match what we see in the practices we work with.
What it does not tell you
The survey cannot tell you which payers, which services or which staff roles account for your practice's hours. A national average of 13 hours a week blends a dermatology practice with heavy biologic authorizations, a family medicine group where most of the work is imaging and referrals, and an orthopedic practice negotiating surgical approvals. Your number could be six hours or twenty-five.
It also cannot tell you how much of the burden is payer-imposed and how much is self-inflicted. In our experience, a meaningful share of prior authorization time in a typical practice goes to requests submitted with incomplete clinical information, requests for services the payer does not actually require authorization for, and status checks made by phone when the payer portal would have answered in a minute. None of that is the payer's fault, and all of it is fixable.
And the survey says nothing about the regulatory side. The CMS Interoperability and Prior Authorization final rule (CMS-0057-F, published January 2024) already requires Medicare Advantage plans, Medicaid and CHIP managed care plans and Marketplace plans to decide standard requests within seven calendar days and urgent requests within 72 hours, to give a specific denial reason, and to publish their prior authorization metrics. Those requirements took effect for 2026. The electronic prior authorization API requirements arrive January 1, 2027. Commercial employer-sponsored plans, which are most of the volume for many practices, are outside the rule. Knowing which of your payers are covered changes how you escalate a late decision.
How to count your own hours
Two weeks of tracking is enough. Give every person who touches prior authorization (front desk, clinical staff, the dedicated coordinator if you have one, and the physicians) a simple log: date, payer, service or drug, task (initial submission, clinical documentation, status check, peer-to-peer, appeal), and minutes. Physicians will resist; ask them to log only peer-to-peer calls and documentation requests, which are the two things that cost them time. Clinical staff should log the time gathering records for a request, which is often the largest hidden bucket.
At the end of two weeks, total the minutes by payer, by service type and by task. Then compare with the payer's share of your visits. A payer that is 15 percent of visits and 45 percent of prior authorization minutes is the one to raise at the next contract discussion, and it is the one where a gold card program or a documented service-level agreement would matter most. Our denial management engagements start with this comparison because authorization denials and authorization hours usually point at the same two or three payers.
A worked example: a four-physician orthopedic practice logged 41 hours of prior authorization work in a week, about 10 per physician, a little under the AMA figure. By payer, one Medicare Advantage plan accounted for 14 of the 41 hours while representing 11 percent of visits. By task, 11 of the 41 hours were status checks, almost all by phone. By service, MRI authorizations were 16 hours, and the practice discovered that one commercial payer had removed the MRI authorization requirement for in-network imaging eight months earlier, so about 3 hours a week were being spent obtaining approvals nobody needed. Two changes (portal-first status checks with a phone call only after 48 hours, and a quarterly review of each payer's authorization list) took roughly 12 hours a week out of the total.
Reading the pledge skepticism
Two thirds of physicians do not expect the June 2025 insurer pledge to help. We understand the skepticism, and we also think practices should track the pledge's specific commitments rather than dismissing it. The pledge included commitments to reduce the scope of services requiring authorization by January 1, 2026, to honor existing authorizations for 90 days when a patient changes plans, to give clearer denial explanations by 2026, to answer at least 80 percent of electronic requests in real time by 2027, and to have medical professionals review clinical denials.
Some of those have visibly happened. UnitedHealthcare and Cigna announced reductions to their authorization lists in 2025 and 2026. Others are hard for a practice to observe without measuring. The 90-day continuity commitment, for example, matters when a patient switches plans mid-treatment, and a practice will only know whether a payer honors it if the authorization coordinator checks. The survey's value here is as a baseline: if the next AMA survey shows the same 13 hours, the pledge has failed on its own terms, and practices will have the numbers to say so.
What to change in the practice, whatever the payers do
Three things are within a practice's control. The first is the authorization requirement list. Every payer publishes one, most update it quarterly, and the scheduling desk should be working from a current copy for each of the practice's top eight payers. Obtaining authorizations that are not required is pure waste, and submitting without one that is required is a CO-197 denial.
The second is the clinical packet. Most denials and delays come from requests that did not include what the payer's criteria ask for. For the ten services the practice authorizes most, build a checklist of the documentation the payer's medical policy names, and attach it to the request the first time. The peer-to-peer call is far shorter when the criteria were addressed in writing.
The third is tracking. Every request should have a log entry with submission date, payer, service, decision due date (seven calendar days for covered payers under the CMS rule, or the contract's stated turnaround), decision date, and outcome. That log is what makes the two-week study permanent, and it is what you bring to the payer when decisions are late.
Questions we hear
Should we hire a dedicated prior authorization coordinator?
The survey says 40 percent of practices have one. Whether you should depends on your measured hours. If the two-week log shows more than 25 to 30 hours a week spread across five people, consolidating into one trained role usually reduces total hours because one person learns the payer criteria and the portals. Below that, the work is usually better as a defined part of a clinical support role than a separate hire.
Can we bill patients or payers for prior authorization time?
Not in any general way. There is no CPT code payers accept for authorization work, and billing patients for it is prohibited by most participation agreements. The realistic lever is the contract: documented hours by payer are a legitimate point in rate negotiations and in requests for gold card status.
Does the AMA survey apply to Medicare?
Traditional Medicare historically required almost no prior authorization for physician services, so the survey burden is overwhelmingly Medicare Advantage and commercial. That is changing at the edges: the CMS WISeR model, which began January 1, 2026 in six states, introduced prior authorization for a defined list of services in traditional Medicare. Practices in those states should track WISeR requests separately in the log.
What to do this week
- Start the two-week prior authorization time log with every staff member and physician who touches the work.
- Download the current authorization requirement list from each of your top eight payers and date-stamp the copies.
- Check that your tracking log records a decision due date for every request and flags requests past it.
- List which of your payers are Medicare Advantage, Medicaid managed care or Marketplace plans, and therefore subject to the CMS seven-day and 72-hour timeframes.
- Set a date in June to total the log by payer, service and task, and to compare the result against the AMA's 13 hours.
