The nine numbers I want from every surgical session

01

The nine

Session date and start and end time. Not the booked times. The real ones. The gap between booked and actual, tracked over months, is the most reliable early signal of a site under strain that I know of.

Lead surgeon, and every other clinician who performed part of the procedure. Every one. No number matters more in a review, and none is more often collapsed into a single name. If three people worked on a case, the record says three people.

Technicians present, by name, for the whole session or part. The same reasoning. When an outcome is being examined two years on, "the usual team" is not an answer.

Volume, in whatever unit the procedure counts in. Grafts, units, sites treated. One number, defined once across the network so it means the same thing everywhere. Two clinics counting differently is worse than neither counting.

Planned volume against delivered volume. Two numbers, not one. A session that consistently delivers under plan is telling you something about assessment. One that consistently delivers over is telling you something more worrying.

Any substitution. Product, instrument, consumable, technique. Most sessions have none, and the field is blank, which is fine. The value is entirely in the rare session where it is not blank, and in a network without this field nobody ever finds those.

Adverse events and near misses, including none. The word "none" written by a person is a different thing from a blank field. A blank means nobody was asked. Ask.

Patient tolerance, as a single judged rating with a free line for why. Soft, and worth having anyway. The free line is where a technician writes the sentence that turns up in a review eighteen months later and explains everything.

Photographs taken, yes or no, and by whom. Not the photographs, the fact of them. At network scale, this field tells you which sites have quietly stopped documenting.

02

What is deliberately not on it

Anything the clinical note already carries properly. Diagnosis, consent, plan, medication. Duplicating those creates two sources that will disagree, and when they disagree you will not know which was written first.

Anything requiring a calculation. If the form asks for a rate, somebody works it out at the end of a long day and gets it wrong. Collect the raw figures and compute later, where the arithmetic can be checked.

Anything with an opinion baked into the options. "Session quality: good, acceptable, poor" produces "good" forever. If you want a judgement, ask for it in words and accept fewer responses.

A field that is always filled in the same way is not collecting anything. It is producing reassurance.

03

Making it survive contact with a real day

Test how long the form takes under ordinary working conditions. Remove duplication and make required evidence easy to record. A short form is useful only if the clinical and operating owners agree that the necessary information remains.

It has to be completed in the room. Not at the desk, not in the car. The reason is not discipline, it is that the technician who is about to leave still remembers whether the second surgeon came in for twenty minutes, and by the car park that detail has gone.

And it has to be seen by somebody within the week. A form nobody reads gets filled in worse every month, and the decline is invisible until you go looking and find a year of identical entries. If nobody is going to read it, do not collect it. Collecting a record you ignore is worse than having none, because it lets you believe you have oversight you do not have.

04

The nine serve questions nobody has asked yet

Not performance management. I want to say that plainly, because the moment a session record becomes an input to somebody's review it starts describing a better day than the one that happened.

They are for answering a question you cannot predict. A supplier issue two years from now, a pattern in outcomes at one site, an inspection that asks who performed a specific case. Every one of those is answerable in an afternoon if the nine exist and unanswerable if they do not.

05

Control, and its limits

You do not control what a review will ask, or when, or about which session. The exercise exists for exactly that reason: you are recording against questions that have not been thought of yet.