Somewhere in every imaging department, someone is refreshing a worklist that has not changed since the last time they refreshed it. Somewhere else, a front desk is answering the same question for the third time today: yes, it was sent, no, nobody can say why it has not shown up yet.
Neither person is doing anything wrong. They are doing the only thing the software left them, which is checking. That habit, checking instead of being told, is the quiet tax underneath a lot of what gets called radiology workflow automation, and it is the specific tax OmniPACS set out to remove when it rebuilt how Condor handles notifications.
Good sharing design and good notification design turn out to be the same problem wearing two faces: how does the right person find out something happened without going looking for it. One half of that problem is getting a study out the door cleanly. The other half, the one this article is about, is what happens the moment after: who finds out, how fast, and whether they had to pick up a phone to learn it.
The Old Model Was Polling
Before a system can tell you something, it has to know that telling you is its job. Most PACS platforms, including older versions of OmniPACS, were built around the opposite assumption: the system holds the information, and the person goes and gets it. That is polling, in the plainest sense.
A tech refreshes a worklist to see if a new study landed. A referring office calls to ask if a report is final. A patient’s family calls back because nobody called them first.
None of that is a training problem or a discipline problem. It is what a system built to be checked, rather than a system built to speak up, produces by default. Every one of those checks is also a small interruption to whatever the person doing the checking was actually supposed to be doing that minute.
What Radiology Workflow Automation Looks Like Here
Condor’s approach to imaging notifications starts from a different premise: the event should reach the person who needs it, not wait for them to come find it. A study arrives, and the tech who owns that worklist finds out. A share gets opened, and the sender knows without calling to ask. A report goes final, and the referring side hears about it instead of guessing.
None of that is a bigger list of alerts. It is the same handful of moments that already mattered, delivered as a signal instead of left as something to discover.
That shift is not a feature bolted onto the side of the platform. It follows directly from the same rebuild that replaced OmniPACS’s legacy PHP application end to end, because a system built to push events in real time needs a different foundation than one built to answer a page request when someone happens to load it.
Which Events Matter and Which Are Noise
The easy version of this feature is also the wrong one: notify on everything and call it thorough. A system that pings for every status change teaches people to stop reading its pings, and once that happens, the one alert that actually mattered gets ignored along with the ninety that did not.
What Alert Fatigue Actually Costs a Department
This is a documented pattern well outside radiology, not a hunch. Left unchecked, the volume of low-priority signals trains a person to tune out the channel altogether, a response known as alert fatigue, and the danger is not the noise itself. It is that a real signal, buried in the same channel, stops getting the reaction it deserves. A department that gets paged for everything is, functionally, a department that gets paged for nothing.
A Simple Test Before an Event Becomes an Alert
Good notification design is closer to editing than engineering: deciding what is worth a person’s attention and cutting the rest, on purpose. A useful test is whether an event changes what someone should do in the next few minutes. A study arriving changes what a tech does next. A share being opened changes whether a sender needs to follow up.
A routine status update that nobody acts on does not clear that bar, no matter how easy it would be to fire it off. That instinct holds up in research on how clinicians respond to clinical alerts: reminders people trust get acted on, and reminders that show up too often get filtered out by habit, whether or not they were ever turned off on purpose.
Where Sharing and Notifications Meet
A share that goes out without a notification attached leaves the sender in exactly the position sharing was supposed to fix: they did their part, and now they wait, and eventually they call to check. Tying the notification to the event closes that loop.
The sender finds out a share was opened the same way a tech finds out a study arrived, as a signal instead of a guess, and the same real-time delivery that makes that possible is what real-time image collaboration between care teams depends on more broadly. The result is not a louder inbox. It is a shorter list of reasons anyone needs to pick up a phone.
Who Feels This Every Day
A practice manager feels this as fewer staff-hours spent confirming things the software already knows. A referring-provider relationship owner feels it as one less category of call that starts with an apology. A tech feels it as a worklist that changes because something actually happened, not because they hit refresh on a hunch. None of these are dramatic wins on their own.
Stacked across a week, in a department already dealing with the everyday bottlenecks that slow a radiology workflow down, removing a routine phone call is exactly the kind of unglamorous fix that a real throughput framework is built to surface: not one big change, several small frictions that stop happening at once.
The Foundation for What Comes Next
An event stream built to reach the right person the moment something happens does not have to stop at a browser tab left open on a desk. The same underlying signal is also the foundation for notifications that follow someone onto a device they actually carry, which is its own story and one OmniPACS is not ready to tell in full yet. What is true today is that the plumbing for it already exists, because it had to exist before any of the notifications described here could work at all.
Fewer Calls, Not More Noise
Existing customers with questions about which events trigger a notification on their own site do not have to guess: that is what support@omnipacs.com is for. Anyone still deciding whether “the system tells you” is more than a slogan can see the new Condor experience and then run the only test that actually matters afterward: count how many times the phone still rings that week asking a question the worklist already answered. A number trending toward zero is what closing the loop was supposed to look like all along.
Frequently Asked Questions
What is radiology workflow automation?
Radiology workflow automation refers to software that handles routine coordination tasks, such as routing studies, updating worklists, and signaling status changes, without a person manually checking or triggering each step. Notifications are one piece of it: the part that tells a person something happened instead of leaving them to look for it.
What is alert fatigue?
Alert fatigue is the desensitization that sets in when someone receives too many notifications, especially low-priority or repetitive ones, to meaningfully respond to each. Over time, people start tuning out the entire channel, including the alerts that actually mattered. Well-designed notification systems limit volume deliberately to avoid causing it.
How soon should a referring provider be notified when a report is ready?
There is no single fixed answer since it depends on the finding and the workflow, but the general principle is that a routine report should reach the referring side promptly once finalized, without requiring a follow-up call to confirm it exists. Genuinely urgent findings typically follow separate, faster communication paths.
What counts as a critical finding in radiology?
A critical finding is a result significant enough that a delay in communicating it could affect patient care, distinct from routine findings that can wait for standard reporting channels. Departments generally define this category explicitly so that critical findings get faster, more direct notification than the rest of the worklist.