Thirteen hours a week, per physician
Prior authorisation is the paperwork that eats the day. It does not stop when the clinic is seeing patients.
See the surveyAn agent that checks eligibility, assembles the prior-authorisation packet from the record, logs the decision and tracks the appeal — and never decides what is medically necessary.
Illustrative scenario — not a real client result
Authorisation queue · 07:50
9 packets assembled
07:51 · Eligibility, codes, clinical notes attached
2 denials flagged
07:54 · Appeal path and deadline identified
“Can we skip the scan?”
07:56 · Clinical decision · with the clinician
Medical necessity is never the agent's call
Eligibility and prior-authorisation agent · Healthcare administration
What healthcare administration owners tell us goes wrong in an ordinary week.
Prior authorisation is the paperwork that eats the day. It does not stop when the clinic is seeing patients.
See the surveyMost appealed denials are overturned. The ones that are not appealed are the ones nobody had time to track.
See the volumesA system denies hundreds of requests without opening a file. The clinic inherits the consequence at the front desk.
See the documented casesEach figure opens its source, year, sample and limits. Click a badge to check it.
American Medical Association, prior-authorisation physician survey
2025
Method and sample are recorded in the source registry.
Read the source for its scope and limits.
Open the source39
prior-authorisation requests per physician per week
AMA survey of 1,000 physicians, fielded late 2024. Self-reported; recruitment method not published in the summary.
American Medical Association · 2025American Medical Association, prior-authorisation physician survey
2025
Method and sample are recorded in the source registry.
Read the source for its scope and limits.
Open the source13 hrs
spent per physician per week on prior authorisation
The same AMA survey — physicians and their staff together. A self-reported estimate.
American Medical Association · 2025KFF, Medicare Advantage prior-authorisation determinations in 2024
2026
Method and sample are recorded in the source registry.
Read the source for its scope and limits.
Open the source80.7%
of appealed Medicare Advantage denials were overturned
KFF analysis of CMS data: 52.8 million determinations in 2024, 7.7% denied. An appeal outcome, not a clinical judgement.
KFF · 2026The agent checks eligibility and benefits against the plan, assembles the packet from the record — codes, notes, prior treatment — and submits it through the channel the payer accepts.
It logs the decision, and if a request is denied it identifies the appeal route and the deadline, drafts the cover letter from the clinician's own words, and puts it in front of a person.
It never states that a treatment is medically necessary, never changes a clinical code to fit a payer rule, and never tells a patient what to do next clinically.
6 steps · 2 decisions you own
Reads from
Use the buttons or the left and right arrow keys. Nothing here is a real customer.
Northgate Family Practice, four physicians
Payer portal and practice inbox · 07:50
A four-physician practice. The agent assembles authorisation packets before the clinic opens. Every message is simulated; no real patient data is used.
Northgate Practice's authorisation assistant. I prepare and track prior authorisations. Clinical questions go to the clinician. Reply HUMAN for the billing team.
The agent moves the paperwork. A clinician owns the medicine.
Justifications are quoted from the note and signed by the treating clinician; the agent never writes one.
New packet types are held for human review until the practice signs them off.
A denial cannot be closed by the agent; it needs an appeal, a clinical decision or a documented stop.
Pause for a payer, a procedure type or the whole practice; the queue falls back to the billing team.
Without your approval · No medical-necessity statement, no code changed to fit a payer rule, no clinical advice to a patient — and every denial stays on a list until someone closes it.
Three documented investigations with their sources. Two are reporting rather than findings of liability; all three are reasons the clinical decision stays with a person.
OIG reviewed 250 prior-authorisation denials from 15 of the largest Medicare Advantage organisations, issued in June 2019. It found 13% met Medicare coverage rules, with some payment denials caused by overlooked documents or system-processing errors.
The agent keeps the submitted packet and the denial reason side by side, so a denial based on a document that was sent is visible rather than lost — and appealable inside the deadline.
13%
sampled denials that met coverage rules
ProPublica reported that a Maryland patient's $350 vitamin-D test was denied as not medically necessary, and that a system enabled batch denials without opening patient files — more than 300,000 requests over two months. Cigna disputed the description of the system.
The agent prepares packets for clinical review; it never issues a determination. Nothing about a patient's care is decided by a rule that has not seen the record.
300,000+
payment requests denied in two months, as reported
ProPublica reported that an algorithm flagged therapists and patients for review based on treatment patterns, after which coverage could be cut, and that officials in three states had found the programme illegal by the end of 2021.
Pattern-based flags are not a decision this workflow may make. Escalations are made from the record by a clinician, and every escalation states its reason in words a patient could read.
3 states
officials who had found the programme illegal, as reported
Every figure resolves to a source with its sample and limits. No figure without one.
39
prior-authorisation requests per physician each week
AMA, 1,000 physicians, late 2024. Self-reported.
American Medical Association, prior-authorisation physician survey
2025
Method and sample are recorded in the source registry.
Read the source for its scope and limits.
Open the source13 hrs
a week per physician on authorisation paperwork
Same survey. Physicians plus their staff.
American Medical Association, prior-authorisation physician survey
2025
Method and sample are recorded in the source registry.
Read the source for its scope and limits.
Open the source52.8M
Medicare Advantage authorisation determinations in 2024
KFF analysis of CMS administrative data. 4.1 million (7.7%) fully or partly denied.
KFF, Medicare Advantage prior-authorisation determinations in 2024
2026
Method and sample are recorded in the source registry.
Read the source for its scope and limits.
Open the sourceDrag the handle, or use the left and right arrow keys when it is focused.
Mon
Forty requests in the queue; eligibility checked one screen at a time.
Tue
A packet is sent without the notes the payer asked for; it comes back denied.
Wed
Nobody notices the denial until the patient calls about the scan.
Thu
An appeal deadline passes while the letter waits for a signature.
Mon 07:51
Packets assembled from the record, missing items flagged before submission.
Tue
Every submission has a channel, timestamp and reference.
Wed
Denials appear on a list with the reason and the deadline.
Thu
Appeal drafted in the clinician's words, waiting only for a signature.
Illustrative example · not a measured result
Keep your answers. They are the brief for the first version we would build for you.
How many eligibility checks and prior authorisations does your team handle in a week?
Which fields come from the record, and which does a person always confirm?
How do you log a decision today, and where does the paper trail live?
What must never be decided by anything other than a clinician?
Next
Put your own figures into the calculator below. It runs in your browser, and nothing is sent or stored.
Put your week into the calculatorBring the four answers to the free one-hour call. We will use them to sketch the first version, and you keep the written plan either way.
Talk about your businessAn estimate, not a promise. The model and its assumptions are printed below.
Your inputs
All payers, all procedure types.
Checking eligibility, assembling, submitting, chasing.
From your own payer reports.
Billing staff time.
Pulling notes, writing the letter, chasing the clinician.
Example defaults are typical for a small business. Replace them with yours. Nothing is stored, sent or shared.
Estimated result
IllustrativeStaff hours back each month
0–0
Routine work handled by the helper: 70% of the share you marked routine.
Time saved per item: 80% of the minutes you entered.
Monthly volume: weekly figures × 4.33; monthly figures used as entered.
Cost: hours saved × your staff cost per hour, in pounds as an illustration.
Reply time: routine items move within the hour while the helper runs. Complex work still waits for a person.
No. It quotes the treating clinician's note, verbatim, and the clinician signs every justification and appeal.
No. Codes come from the record. A mismatch with the payer's rule is flagged to the billing team, never fixed by altering the claim.
It goes on a list with the stated reason and the appeal deadline. The agent cannot close it; a person resolves or documents it.
Only about paperwork and timing. Clinical questions are routed to the clinician, and the patient is told who will reply.
Not at first. New packet types are reviewed until the practice signs them off; review is then removed per type, not wholesale.
It stays inside the practice's systems and the payer's portal. Access is role-limited, logged, and retained on the practice's schedule.
Three to four weeks: payer list, packet templates, eligibility connection, and a parallel run with the billing team.
Start here
One hour with last week's queue and one denial letter. We will show you what the agent would assemble, what it would flag, and where a clinician must sign.