Labs
The most dangerous object in a clinic is an abnormal result nobody opened.
The test was ordered, the sample went, the lab reported, and the report landed somewhere — a printout in a tray, an email nobody owns, a folder on the shared drive. The patient assumes no news is good news. Often it is. Sometimes it is the opposite.
Every result here is tracked until somebody has actually opened it. Not delivered, not filed — read, by a named person, with a time against it.
The screen
34
9
2
118
2Potassium 6.4 — critical, nobody notifiedheld 41 min
9Outside reference range, unreviewedpending sign-off
5Sample collected, no result 26h onchase lab
3Ordered panel, sample never drawnre-collect
12Released and read by the patientclosed
Abnormal values are held, not released. A result a clinician has not seen is not a result the patient should read first.
The gap that matters
The result nobody opened.
A lab result is not finished when it is ready. It is finished when a clinician has read it and acted — and the gap between those two is where a missed abnormal lives.
Illustrative widths. Drapto tracks the third bar, not the second — an unread abnormal stays on a worklist until somebody opens it.
Unread results
“Delivered” is not the same thing as “somebody read it”.
Most systems treat a result as done when it arrives. That is the wrong finish line. The failure that harms a patient is not a report going missing in transit; it is a report arriving correctly and sitting unopened while everyone assumes somebody else has seen it.
The report arrives. It is filed. If it was abnormal, that fact is discovered whenever the patient next comes in — or when they come in with something worse.
The result stays on a list marked unread until a clinician opens it. Who opened it, and when, is recorded. Nobody has to remember to check.
What this is not. Drapto does not interpret results, does not flag them as clinically urgent, and does not decide what is important. It tracks whether a human has looked. Software that triaged results by severity would be making a clinical judgement, and it would be wrong at exactly the moments that matter most.
Reference ranges
A “normal” range that ignores sex and age is normal for nobody.
Haemoglobin, creatinine, alkaline phosphatase and a long list of others have genuinely different reference intervals by sex, and several shift substantially with age. A single printed range applied to everybody produces two errors, and they point in opposite directions: false reassurance for one group, false alarm for another.
Ranges here are held per test, per sex and per age band, so the flag against a value is the flag for that patient rather than for an average of everybody.
The flag follows the patient
Same value, different patient, different reading. That is the point of a reference interval and it is routinely lost.
Paediatric ranges are not adult ranges scaled down
Several analytes move sharply through childhood. Applying the adult band to a child is not conservative, it is wrong in both directions.
The range travels with the result
Into the record, and onto the copy the patient reads — so a number they can see is a number they can place.
Ordering
Ordered inside the consultation, back inside the record.
- Ordered from the chartThe doctor picks the test while writing, not on a separate slip that has to be carried, transcribed and matched back later.
- The patient knows what and whyThe order goes onto the same page as their prescription, in plain words. Nobody arrives at the counter unsure what they are here for.
- The result returns to the chartAttached to the patient and to the order that produced it, rather than to a name that has to be matched by hand.
- Unread until readIt sits on the unread list until a clinician opens it, and the record shows who did.
- The patient keeps itAt their own
/result/<token>, and in their record at/my/<token>with their values over time and the direction they are moving.
The commercial side
Margin per test, because a catalogue without it is a price list.
A clinic running its own lab, or sending out to one, usually knows the price charged and rarely knows the margin on each test after reagents, send-out cost and rework. The catalogue here carries both, so the tests that lose money are visible rather than assumed.
We publish no benchmark margins. Reagent costs, send-out rates and volumes vary enormously between clinics, and a figure from somebody else's lab is not a target for yours. The number that matters is your own, and the point of the catalogue is that you can see it.
What it will not do
The limits, plainly.
It does not interpret results. No severity flags, no triage, no suggestion of what a value means. It shows the value against the right reference range and records who read it.
It is not a diagnostic device and does not perform testing. It is the ordering, tracking and record layer around whichever lab does the work.
It will not release a result to a patient before a clinician has seen it where the clinic has set it that way. Who sees what first is a clinical decision, not ours.
The failure this exists to stop
From order to acknowledgement
A haemoglobin of 12.5
| Patient | Reference range | Flag |
|---|---|---|
| Woman, 34 | 12.0 – 15.5 | Normal |
| Man, 34 | 13.5 – 17.5 | Low |
| Child, 7 | 11.5 – 15.5 | Normal |
One column of ranges gets one of these three wrong, every time.
What we will never do
Things we have promised not to do
- Take a cut of a consultation fee
- Charge per appointment
- Sell a position in the directory
- Host star ratings
- Change your price mid-contract
How we are actually paid
- Hospitals, for revenue integrity
- Companies, for corporate health
- Labs and pharmacies, for their modules
- Implementation, because it is somebody’s week
- Nothing from a patient, ever