What a System That Closes the Loop Actually Needs to Do

Most conversations about improving clinical systems happen one level too low.

They focus on faster documentation, fewer clicks, cleaner templates, better inboxes, more efficient ways to capture what happened in the visit. Those things matter. But a more pleasant documentation system is not automatically a more trustworthy clinical system. The deeper question is not how quickly the record can be updated. It is whether the system can maintain accountability for patient work that is not yet finished.

That distinction changes everything.

A clinical system should not be judged only by how well it stores information at a point in time. It should be judged by whether it can keep unresolved work visible across time, across handoffs, and across the ordinary delays that define Canadian primary care. Referrals wait. Results arrive out of sequence. Consult notes come back after the physician has moved through dozens of other patient concerns. A form is faxed, then re-faxed, then someone phones to confirm it was received, and the patient still has not heard anything three weeks later.

In a documentation-centred system, the work can look complete because the action was recorded. The referral was entered. The letter was generated. The chart contains the right note. From the perspective of the record, something happened.

From the perspective of patient care, the work may still be very much alive.

That is the difference physicians feel but are rarely given language for. A system designed around capturing information treats documentation as the endpoint. A system designed around follow-through treats resolution as the endpoint.

Consider a referral for a patient with worsening knee pain who has finally agreed to see orthopedics after months of trying conservative management. In one system, the referral letter is completed, attached, and sent. There may even be a note in the chart saying it was faxed on Tuesday afternoon. If nothing comes back, the responsibility quietly shifts back onto human memory: the physician remembering the patient’s concern during a later visit, the MOA noticing a missing acknowledgement, the patient calling because they have not heard anything, or someone finding the referral again while cleaning up a work queue.

Nothing in that scenario is dramatic. That is exactly why it is risky.

A resolution-centred system would understand the referral differently. Sending it would not be the end of the work. It would be one step in a longer obligation that remains unresolved until there is confidence the referral has been received, acted on, and brought to an appropriate next state. The system’s job is not simply to remember that the referral exists. Its job is to preserve the fact that this piece of patient work still requires attention.

That is what closing the loop actually means.

It is not a feature. It is a design philosophy.

A feature can be added to an existing system while leaving the underlying definition of “done” unchanged. A reminder here, a flag there, a task list bolted onto the side of the chart. Those tools may help, but they do not necessarily change what the system believes it is responsible for. If the system’s deepest assumption is that recording an action is enough, then unresolved work will always depend on people noticing what the system has stopped carrying.

A loop-closing system starts from a different assumption. It treats unfinished patient work as something that must remain visible until it has genuinely reached a safer resting point. It does not assume that silence means progress. It does not confuse transmission with completion. It does not ask physicians to mentally preserve the difference between “I sent it” and “I know it moved forward.”

Primary care is full of intervals between actions, decisions, and outcomes. A system designed around resolution does not eliminate those intervals; it remains accountable to them.

That requires a different kind of design discipline. The system must be designed around continuity rather than events, around responsibility rather than record-keeping. It must treat unresolved work as inherently important, not merely because someone remembers it exists. It must recognize that the meaningful endpoint in care is not the completion of an action but the resolution of an obligation. Most importantly, it must value follow-through as a core responsibility rather than an activity that happens outside the record.

The point is not that the system should do the physician’s thinking. It should not. Clinical judgment remains with the doctor. But the system should not require clinical judgment to be supported by constant personal surveillance of administrative and coordination gaps. There is a meaningful difference between deciding what should happen next and having to remember that something still needs to happen at all.

That difference is where mental load accumulates.

Physicians are often told they need better workflows, but what they often need is a system that shares their definition of unfinished. In real clinical work, “done” does not mean the note is signed. It does not mean the task was assigned. It does not mean the referral was sent into the world and disappeared into someone else’s process. Done means the work has reached a point where it no longer needs active concern from the clinic, or where the next responsibility is clear and visible.

Clinical systems should be designed around that standard.

This is where Aeon’s approach begins: with resolution as the endpoint, not documentation as the endpoint.

Family physicians should not have to choose between a system that documents well and a system they can trust.

Those should be the same thing.