ClinicOps  /  Briefings  /  Prior authorization

How to stop prior authorizations from expiring before the visit

By Jareer Ali  ·  9 min read  ·  Updated July 2026

An approved authorization that lapses before the patient is seen is the quietest way a practice loses money. The work was done. The payer said yes. And then a date passed in a folder nobody was watching.

Every practice manager knows the feeling. A prior auth comes back approved, everyone exhales, and it goes into the pile of things that are handled. Six weeks later the patient finally comes in for the MRI, and the auth window closed eleven days ago. Now it is a rework, a delay, an annoyed patient, and sometimes a write-off. Nobody did anything wrong, exactly. That is the problem. There was no system, so there was no one to blame and nothing to catch it.

This guide walks through the fix. It is not software you have to buy, and it is not a new person you have to hire. It is a small operating pattern you can build this week inside the tools you probably already own.

Why approved auths still expire

An expiring auth is almost never a knowledge problem. Your team knows auths expire. It is a tracking problem, and it usually has the same three roots:

• The auth lives in an email, a fax, or a note in the practice-management system, but not in any list that someone reviews on purpose.
• No single person owns it after approval. Getting the approval had an owner. Watching the approval did not.
• There is no alarm. The only thing standing between the auth and its expiry date is someone happening to remember.

Fix those three and the leak closes. Here is how.

An auth without an owner and a due date is not tracked. It is just stored.

The system, in four moves

1. One list, every auth

Create a single place where every prior auth lives from the moment it is requested to the moment the service is delivered. One list. Not one per provider, not one per payer. Each auth is a row with a handful of fields: patient (use a matter or chart number, never clinical detail), service, payer, date approved, date it expires, and status.

The single most important field is the expiry date, because it is the one that costs you money when it is missing.

2. An owner on every row

Every auth gets one named owner: the person responsible for making sure that service happens inside the window, or that the auth is renewed if it will not. This is not necessarily the person who filed it. It is whoever will notice. When an auth is nobody's job, it is everybody's blind spot.

3. Alarms before the edge, not at it

This is the piece almost every practice is missing. Set automatic reminders that fire well before the expiry date, not on it:

1. An alert to the owner when the service is not yet scheduled and the auth is 14 days from expiring.
2. A second alert at 7 days that also copies a second person, so a single sick day cannot sink it.
3. A final flag at 2 days that escalates to the practice manager.

The specific days matter less than the principle: the reminder reaches a human with time to act, and it reaches a second human if the first one is out.

4. A weekly two-minute review

Once a week, someone opens the list sorted by expiry date and looks at the top. That is the whole ritual. Auths approaching their window get scheduled or renewed. It takes two minutes because the alarms have already done the watching. The review just confirms nothing fell through the automation.

How to know it is working

You will feel it before you can measure it, but two numbers tell the story:

• Expired-auth rework per month. This should fall to near zero within the first full cycle.
• Auths renewed on time versus let lapse. When the alarms are doing their job, renewals stop being emergencies.

If your practice does a meaningful volume of authorized services, closing this one gap often pays for the entire effort of building the system, several times over, in the first quarter.

The honest catch

None of the four moves is hard. Building them so your team actually runs them, week after week, once the initial enthusiasm fades, is the hard part. A tracker everyone abandons in a month is worse than no tracker, because it looks handled while it quietly is not. The work is in the design of the ownership and the alarms, and in the training that makes it stick. That is exactly the part most practices get wrong on their own, and exactly the part we build.