The Genetics Clinic That Cut Its Wait Time and Couldn't Prove It

Ask the head of a clinical genetics service how long a family waits for a first appointment and you'll usually get an honest answer, which is "too long". Ask for the actual figure (the median, or the longest wait sitting on the list this week) and the conversation tends to slow down. Someone offers to check the spreadsheet. Someone else points out that the referral date is on the paper form, which is in the tray, or has been scanned into the record, or is still in the shared mailbox. Everyone in the room knows the wait is a problem. What nobody can do, at least not quickly, is put a number on it.

This isn't a failing of any particular department, and it's worth saying so up front. Three recent peer-reviewed papers, read together, suggest it's close to the normal state of genetics services, that the waits behind it are getting longer, and that the gap usually sits in the referral record rather than in the clinic itself. That last part is the useful bit, because a wait you can't measure is also one you can't take to a budget meeting.

Waits are long, and getting longer

In 2026 the American College of Medical Genetics and Genomics published a survey of the directors of its member clinics. For routine general paediatric and adult genetics appointments, 49% of programs (21 of 43) reported waits of more than six months. Cancer genetics did better at 26%, biochemical genetics at 7%, and prenatal at 0%, which makes sense given a pregnancy brings its own deadline with it. Urgent visits were generally seen within two weeks. The authors' conclusion was that wait times "continue to increase compared with surveys from previous years".¹

The survey is American, and the Australian workforce picture differs in the details, but the shape of it will be familiar to anyone who has sat through an HGSA meeting: more referrals every year as genomic testing moves into mainstream specialties, against a counselling and medical genetics workforce that grows a good deal more slowly. Six months is also, in our experience, about the point where referrers stop trusting the service and start phoning.

The clinic that cut its wait and could not prove it

The second paper is the one worth reading in full, because it describes a service doing more or less the right thing. In 2018 a medical centre in Nebraska opened a genetic-counsellor-led clinic to take general referrals off the geneticists' list and bring down the time from referral to first visit. Three years later the team published an evaluation of the first 109 patients.²

The model worked. Mean referral-to-visit time came out at 49.8 days, against six to nine months at the comparison clinic across town, and 88% of patients didn't need to see a medical geneticist afterwards at all. On the clinical side there's not much to argue with.

The denominator is the interesting part, though. That 49.8-day figure rests on 32 patients, because a referral date was available for only 32 of the 109. For the other 77, nobody had recorded when the referral arrived. Where the referring provider was missing, the authors fell back to the primary care provider on file and, to their credit, flagged that as a possible bias. The data came from the electronic record, "supplemented with paper referral records", and the counsellors cross-checked by hand to fill in what they could.

None of that is a criticism of the clinic; it's just what happens when the referral lives on a form and the status lives in someone's head. The service genuinely improved access for its patients, and then couldn't produce the one number that would have made the case for a second counsellor. A wait time you can only report for 29% of your patients is hard to put in front of a finance committee.

A shared system does not fix a missing field

It's tempting to assume the answer is a bigger system. NSW and the ACT have had one for well over a decade: a state-wide genetic information system that links all 31 public clinical genetics services into what the authors of a 2021 study describe as a "virtual state-wide service". The study interviewed twelve genetic counsellors and ran focus groups with fourteen past patients and carers.³

Counsellors valued the shared family information, and patients were reassured by how carefully it was looked after. The counsellors' list of challenges is the part that matters here: inconsistent data entry, the effort of keeping the system current, and a lack of integration with other health systems. A shared database is only as good as what each service puts into it, and that depends on whether the referral was captured properly at the front door in the first place. The state-wide system holds the family's pedigree. It doesn't tell the clinic coordinator which of this week's forty referrals has been sitting untriaged for three weeks, and it was never meant to.

What the three studies have in common

Read together, the three papers point at the same gap from different angles. The directors' survey measures the problem at a national level. The Nebraska evaluation shows a service fixing it and then struggling to measure its own success. The NSW study shows that even a mature, shared, state-wide system depends on the quality of what's entered at the start. In each case the weak point is the same record: the referral, in the stretch between the day it arrives and the day it turns into an appointment.

That stretch is where a paper form, a scanned PDF or a spreadsheet row quietly causes the trouble. The referral exists, but it has no date a report can read, no status that separates "received" from "triaged" from "waiting on information" from "ready to book", and no history of who moved it or when. The department can see its patients once they're booked. What it can't see is its queue.

What a measurable referral looks like

The fix is a good deal smaller than a new EMR module and a good deal faster than a workforce plan. A referral becomes measurable when four things are true of its record:

1. It has a received date that is captured, not transcribed. The day the referral is entered is the day the clock starts. Nobody types it later from a fax header.
2. It has a status, and the status list is the department's own. Received, in triage, more information requested, ready to book, booked, redirected. Each status answers the coordinator's question: what do I do with this now?
3. Every status change is a dated, attributed line in a history. Who moved it, when, from what, to what, and optionally why. This is the audit trail that accreditation asks for and that the Nebraska team had to reconstruct by hand.
4. The list can be counted. How many are in each status, how long the oldest has waited, what the average wait is for the bucket you care about. Every number leads to the records behind it.

With those four in place, the Nebraska clinic's 49.8 days would have rested on 109 patients rather than 32, the directors' survey could have been answered from a dashboard rather than from memory, and the NSW counsellors' "inconsistent data entry" becomes a mandatory field with an inline error rather than a theme in a qualitative study.

None of this asks software to make a clinical decision. The triage category is recorded exactly as the clinician made it. What changes is that the record carries its own timeline, so the service can see where families are waiting and for how long, and say so with a number when asked.

A question to ask on Monday

If you run or coordinate a genetics service, there's a reasonably quick test. Ask how many referrals are waiting right now and how long the oldest has been there. If the answer takes more than a minute, or depends on who happens to be in the office, the problem probably isn't the clinicians. It's that the referral is a form rather than a record. Of everything on the list, that's the cheapest to fix, and it's the one that makes every other improvement measurable.