How to improve turnaround time (TAT) in a diagnostic centre

Turnaround time, or TAT, is the elapsed time between two defined points in the life of a sample, usually from collection or receipt to the authorised report. NABL lists it among the quality indicators a medical laboratory must monitor, and patients and referring doctors judge a lab by it more than by anything else. Improving it starts with measuring it per test, from timestamps rather than memory.

How should a lab measure TAT?

NABL's specific criteria for medical laboratories, NABL 112A, define TAT as the elapsed time between two specified points through the pre-examination, examination and post-examination processes. The important words are two specified points. A lab has to decide which two, and then record them for every sample.

The useful timestamps, in order, are registration, collection, receipt at the lab, result entry or analyser result, authorisation, and delivery. With all six recorded, TAT can be reported for any pair. Registration to delivery is what the patient feels. Receipt to authorisation is what the bench controls. Collection to receipt is the courier.

• Record timestamps by scan, not by typing. A barcode scan at each step is the only way the times are honest.
• Report per test, not per lab. A CBC and a culture have different clocks, and an average across them means nothing.
• Look at the distribution, not the mean. The question a doctor asks is what fraction of reports were out by the promised time, so track the share of samples within target and the worst cases, not the average.
• Separate stat, routine and outsourced samples, because they carry different promises.

Pathixio records these timestamps from barcode scans and analyser results and shows TAT per test and per stage on the dashboard. Whatever software you use, the first week of measuring usually surprises the owner more than any later improvement.

Where does the time go?

Once the numbers are per stage, the bottleneck is visible. The table lists the usual suspects.

In most standalone labs we see, the largest single gap is result to authorisation, because reports wait for a pathologist who visits at a fixed time. The second is the courier from collection centres. Neither is a bench problem, and neither is solved by buying a faster analyser.

What targets should a lab set per test?

Targets belong to the lab, set from its own data and its own promises to referrers. A few principles help. Set a target per test group rather than one for the lab. Set it as a promise the front desk can repeat to the patient, in hours or working days. Set it from where you are, then tighten it. And write down the exceptions. NABL 112A states limits only for specific cases: it says TAT for frozen sections and squash smears should not exceed twenty minutes, and that cytology specimens other than intra-operative ones shall not exceed three working days, and it asks labs to identify and document situations where a limit may be exceeded.

For routine haematology and biochemistry most labs promise same-day or next-morning reports; for cultures, days; for histopathology, several working days depending on the tissue. Whatever the number, the software should print it on the receipt and start the clock at collection. A promise printed on the receipt is the strongest driver of internal discipline we know.

Ten fixes that usually work

1. Print the promised report time on the receipt and send it by WhatsApp at registration.
2. Scan at every hand-off so the pending list is real. No scan, no next step.
3. Interface the analysers, starting with the one producing the most results, so results land without typing. See what analyser integration should mean.
4. Sort the pathologist's queue by time remaining to target, with abnormal and stat reports at the top.
5. Let the pathologist sign from a phone or tablet between visits, with the same audit trail as at the desk.
6. Fix dispatch times for each collection centre and show transport time per centre on the dashboard.
7. Write rerun criteria per test so a technician does not repeat a run out of habit.
8. Batch by target, not by convenience. Run stat samples as they arrive.
9. Alert the supervisor when a sample crosses eighty percent of its target time, not after it is late.
10. Review the ten worst TATs every week, one by one, and fix the cause rather than the symptom.

None of these requires new instruments. Most require the software to be set up to enforce a rule that everyone already agrees with.

How to keep it improved

TAT drifts back when nobody looks. Put the per-test TAT report on the owner's dashboard and review it in the weekly meeting. Keep the target list in the software, not on a notice board, so the alerts follow it. Include TAT in the quality indicators file for the internal audit, which NABL 112A asks to be done at least once in twelve months, and in the corrective action log when a target is missed repeatedly. ICMR's Good Clinical Laboratory Practices guideline, on its guidelines page, is a useful companion for the process side.

A lab that publishes its report time and meets it is rare enough in India that referrers notice. That, more than any marketing, is what fills the register.

Frequently asked questions

What is a good TAT for a routine blood test?

There is no single national figure, and we would not invent one. NABL asks each lab to define and monitor its own turnaround time, and states limits only for specific cases such as frozen sections and routine cytology. Most standalone labs promise routine haematology and biochemistry the same day or next morning. Set your target from your own measured data.

Should TAT be measured from collection or from receipt?

Measure both, and report both. Collection to report is the patient's experience and includes transport, so it is the number to promise. Receipt to report is what the bench controls, so it is the number to manage. When the two diverge for a collection centre, the courier is the problem, not the lab.

Does analyser interfacing really change TAT?

It removes the typing step and the queue of printouts waiting to be entered, which in a busy lab can be an hour or more at peak. It also removes the reruns caused by typing errors. The larger gains usually come from the authorisation queue and delivery, but interfacing is the fix that costs least in staff habit.