The Appointment Nobody Booked

The scans I worry about are not the ones on my worklist. They are the ones that are not, the studies that were ordered but never happened, because the appointment that would have made them real was never booked. A clinician requests an ultrasound or a follow-up scan, tells the patient to call and arrange it, and then the patient, who is anxious, or busy, or simply could not get through, does not. Weeks later the referring doctor assumes it is done, the patient assumes it is in hand, and the study that would have answered a question sits in a gap that no worklist shows, because a study that was never booked never appears on any list at all.

As a radiologist I see the scans that happen. What I have come to understand is how many do not, and that the reason is almost never clinical. It is that the one action standing between a referral and a diagnosis, a phone call that has to be answered and turned into a booking, is exactly where the process is weakest.

This is a composite drawn from the referrals an imaging service handles, not a single identifiable patient.

An imaging referral has two lives. First it is ordered, a clinician decides the scan is needed. Then, and only then, it has to be scheduled, which in most services means the patient calls to arrange it. Those are two different events with a fragile bridge between them, and the bridge is a phone call. If the call is not answered, or ends in a message, or the patient never makes it, the referral stalls in the space between ordered and booked, where nothing is watching it.

That gap is invisible by design. A booked appointment is on a schedule someone reviews; an ordered-but-never -booked scan is on no schedule at all. It does not show as a no-show, because there was never an appointment to miss. It simply does not exist, and its non-existence is silent until a symptom worsens or a referring doctor asks why the result never came. In imaging, where the whole point is to answer a clinical question in time, that silence is the most expensive failure there is.

Where the referral fell through

When we traced one such referral, the failure was mundane at every step.

Stage	What happened	State of the scan
Scan ordered	Patient told to call to book	Exists on paper only
Patient calls	Line engaged at lunchtime	Not booked
Patient calls again	Voicemail, leaves a number	Not booked
Callback attempted	Patient at work, misses it	Not booked
Weeks pass	Everyone assumes it is done	Never booked

Not one stage is a clinical error, and not one is anyone being careless. It is the same structural weakness each time: the referral depended on a phone call landing, and the phone call never did. The scan was never refused; it was simply never booked.

The call that closes the gap

The fix is a phone that turns every scheduling call into a booking, on the call, the first time. When the patient rings, 2care answers in about 2 seconds, at any hour, with no engaged tone. It confirms who they are against the record, takes the ordered study, reads the live imaging schedule, offers a slot, and writes the appointment into the practice system across 95 or more platforms, confirming it before the patient hangs up. It gives the preparation instructions in the same call, the fasting, the timing, the what-to-bring, so the scan is not lost later to a patient who arrived unprepared and had to rebook.

Because the whole voice pipeline is native and owned end to end, the reply lands in about 480 milliseconds a turn, so an anxious patient is not rushed. It answers 1,000 or more calls at once at 99.9 per cent uptime, so a busy Monday of referrals does not meet a busy tone, and no scheduling call is lost to a queue. See how that booking reaches the record on the platform and the integrations page.

The safety questions a scan call carries

An imaging booking is not purely clerical, and this is the part I was most careful about. A scheduling call often carries a safety question that a booking flow must not answer on its own. "Can I still have the scan, I might be pregnant?" or "last time the contrast dye gave me a reaction" or "the lump has actually got bigger since the referral" are not scheduling details; they are clinical, and each needs a person. The agent does not decide contrast safety, rule on a possible pregnancy, or judge a changing symptom. It recognises that the call has crossed into clinical territory, leaves the booking flow, and routes it to a person in about 3 seconds, a clinical question to a clinician in about 4, each carrying the transcript and the resolved patient.

Every escalation records the phrase that triggered it, so the boundary is auditable and can be tuned for our protocols. That is what makes it safe to put in front of an imaging referral: it is quick where speed is harmless, booking the scan, and it stops dead where a safety judgement is required, which is exactly the line a booking system must respect.

Access is a diagnosis on time, or not

It is easy to file scheduling under administration. In imaging it is closer to diagnosis, because a scan that is not booked is a question that is not answered, and a question not answered on time is sometimes a finding caught too late. Phone access is a stated priority for practice leaders, named among the top patient-access focuses for 2026 in an MGMA poll of practice leaders, and for an imaging service that access is the difference between a referral that becomes a report and one that evaporates in the gap.

A line that answers every scheduling call and finishes the booking does more for diagnostic yield than it appears to, because the studies it saves are the ones that would otherwise never have been counted, the appointments nobody booked, now booked.

The scan that never shows on any list

What makes this failure so persistent is that nothing in a normal imaging service is built to catch it. A no-show is logged, a cancelled scan is logged, a reported study is logged. But a scan that was ordered and never booked leaves no trace in any of those systems, because it never became an appointment in the first place. There is no row for it, no missed slot, no gap on a worklist, only an intention that quietly expired. When a service reviews its numbers, the referrals that fell through are the ones it cannot see, so they are never counted and never chased. The only way to close a leak that leaves no trace is to stop it at the source: to make the one call that would have booked the scan land every time, so the referral never enters the invisible gap at all. A booking that happens on the call is a scan that exists on a schedule, where it can be seen, prepared for and reported.

Where 2care is right for an imaging service

2care is right for a service that lives or dies on referrals converting into booked studies: a radiology department, a standalone imaging centre, any practice where an ordered scan depends on the patient calling to arrange it. It is right because it closes the gap between ordered and booked in one call, captures the preparation so the scan is not lost to a rebook, and pulls the safety questions out of a scheduling call and puts them in front of a person. A service whose referrals are all booked internally, with the patient never in the loop, may feel this less; a service that asks patients to call and book will feel it in the referrals that stop disappearing.

Frequently asked questions

How does it stop a referral from falling through?

By turning the scheduling call into a booking on the spot. It answers in about 2 seconds, confirms the patient, reads the live schedule, books the ordered study into the record across 95 or more systems, and gives the prep, all in one call. There is no message and no callback, so the referral cannot stall in the gap where scans get lost.

What if the patient asks a safety question when booking?

It is escalated, not answered. A possible pregnancy, a contrast reaction, or a worsening symptom is routed to a person in about 3 seconds with the transcript attached. The agent never rules on contrast safety or a symptom itself; it recognises that a clinician is needed for the question and gets one, while still holding the booking.

Does it handle scan preparation instructions?

Yes. It gives the preparation for the specific study in the same call, the fasting, the timing, what to bring or avoid, so a patient does not arrive unprepared and have to rebook. A rebooked scan is another chance for the referral to fall through, which is exactly what capturing prep on the first call prevents.

Can it cope with a heavy referral day?

Yes. It answers 1,000 or more calls at once at 99.9 per cent uptime, so a Monday backlog of patients all ringing to book after a weekend of referrals each reaches an answered line in about 2 seconds, rather than a busy tone that sends the referral back into the gap.

How does the booking reach our system?

It is written into the practice system across 95 or more platforms as a first-class appointment against the resolved patient, confirmed only after the record accepts it. Nothing waits as a note for staff to enter, so the scan the referral asked for is on the real schedule, not in a callback pile, before the patient hangs up.

What a report depends on

I used to think a report depended on the scan and the referral. I have learned it depends, first, on whether the appointment was ever booked, and that whether it was booked depends on whether a phone call was answered and finished. The appointment nobody booked is the one I never get to report, and the phone is where it is won or lost.

Hear a scan booked and a safety question escalated, on your own system, when you book a demo.