The scenario below is based on a real systems problem. Identifying and operational details have been changed to protect the privacy of the people and organization involved. Although the events described here are fictionalized, the systems questions they raise are common across many healthcare settings.
I have an occupational hazard.
Spend enough years studying systems, measurement, and information, and your imagination starts getting ideas of its own.
Whenever someone tells me about an operational problem, my brain has an annoying habit of turning it into a science fiction movie.
In my head, warning lights begin flashing.
🚨 CRITICAL WARNING. INFORMATION TRANSFER FAILURE DETECTED. PROCESS INTEGRITY COMPROMISED. THE RELIABILITY OF ALL DOWNSTREAM OPERATIONS IS NOW UNKNOWN. EVERY DECISION MADE FROM THIS POINT FORWARD MAY BE BASED ON CORRUPTED, MISSING, OR STALE INFORMATION. IMMEDIATE INVESTIGATION REQUIRED.
The actual conversation is usually much less dramatic.
“I think we’re having billing problems.”
Recently I was talking with a clinician about an operational issue in a healthcare organization. To protect that clinician’s privacy, I’ve changed the identifying and operational details. The scenario that follows didn’t happen exactly this way, but it illustrates the same kind of systems problem.
Imagine this.
A client updates their address at the front desk.
Reception records the change.
The appointment goes ahead.
Everyone believes the client’s information has been updated.
Weeks later, claims submitted to a government funder begin coming back unpaid because the billing information no longer matches the client’s record.
Most people would describe this as a billing problem.
That description immediately raises questions in my mind.
Is it really billing?
Or is billing simply where the organization first noticed something had already gone wrong?
What information had to move before that claim could ever be submitted?
How many people touched it?
How many systems?
How many handoffs?
How was each handoff confirmed?
What did every rejected claim have in common?
I wasn’t looking at billing anymore.
I was looking for the leak.
Somewhere between the moment the client shared updated information and the moment the claim was submitted, the information failed to reach the next place it was needed. The rejected claims weren’t the failure. They were the alarm. The investigation begins. Eventually, the leak is found. The broken handoff is repaired. The claims begin processing again. The incident is declared fixed. A chorus of jubilation resounds. Most investigations end there.
Mine usually doesn’t.
Instead, another question appears. How long had it been leaking? If no one can answer that, more questions follow:
- How many claims passed through the leak?
- Can every affected record be identified?
- Can the organization account for every rejected claim?
- How confident are we that we’ve found them all?
- At that point, I’m no longer trying to understand the leak.
- I’m trying to understand the flood.
Now step back from clinical care. Let’s bring this scenario home. Like literally. Imagine discovering a leaking pipe in your home. Repairing the pipe is important. But no one repairs the pipe, walks away, and assumes the rest of the house is fine. The first question isn’t whether the pipe has been fixed. It’s how long it had been leaking. Then the sleuthing begins. You inspect the flooring. You look behind the walls. You check the insulation. You search for water damage in places that were never supposed to get wet. You hope you don’t find black mould growing where nobody could see it. Because repairing the pipe doesn’t undo everything the leak touched.
Operational failures aren’t very different.
Repair restores the process.
Recovery means reconstructing what happened while the process wasn’t working.
Professionals in data management have been thinking about these challenges for decades. The DAMA Data Management Body of Knowledge (DAMA-DMBOK®) treats information as an organizational asset that must be governed throughout its lifecycle. Data governance, data quality, systems integration, master data management, and related practices exist because organizations depend on information arriving where and when it is needed, and in a form that can be trusted (DAMA International, 2024).
Healthcare is no exception. The billing software isn’t the system. Reception isn’t the system. The government funder isn’t the system. The system is everything required to move information from one point to the next, verify that it arrived, detect when it didn’t, and recover when something goes wrong. That’s why my imagination insists on rolling the disaster movie whenever someone describes an operational problem. Not because every rejected claim is catastrophic. Because the imaginary spaceship computer always asks the question I hope organizations will ask.
Not:
“How do we fix billing?”
But:
“Where’s the leak?”
And once that’s answered...
“How big was the flood?”
Behind every rejected claim is time spent reconstructing history instead of supporting care.
Behind every failed handoff is information that didn’t reach the next decision.
Behind every information pathway is a person depending on the organization to get it right.
The symptom sounds the alarm.
The investigation finds the leak.
Recovery measures the flood.
This was never a billing problem.
Billing was simply where the organization first noticed something was wrong.
The real issue was that information was failing to reach where it was needed, and no one realized until claims started coming back.
Billing wasn’t the problem.
The leak in the system was.
Billing was just the water stain on the ceiling.
