Preauthorisation delays: where the clock actually goes between admission and approval

Most hospital preauthorization delay reasons sit inside the hospital, before submission. Four timestamps that show where the clock goes, and how to fix it.

Ask a hospital owner for hospital preauthorization delay reasons and the answer almost always points outward. The TPA sits on files. The insurer raises the same query twice. The desk closes at six. All of that happens. But when a hospital finally instruments the full journey from the decision to admit to the authorisation letter, the largest recoverable block of time is usually not at the payer at all. It is inside the hospital, between the moment admission is decided and the moment a complete request is actually submitted.

That is the short answer. The common preauthorisation delay reasons, in the order they cost the most: the file is opened late because nobody owns it; the documentation pack is assembled after the fact rather than at admission; the diagnosis, procedure coding and plan of treatment do not match each other; the tariff or package on the estimate does not match the payer's agreed rate; and when a query comes back, it is answered by whoever is on the desk rather than by the person who knows the case. Only after all of that does the payer's own turnaround start.

Why the hospital's half of the clock is invisible

Most hospitals report preauthorisation turnaround as a single average, and that average is almost always measured from submission. It has to be, because submission is the first event the system timestamps. The decision to admit is a conversation. The request for documents is a phone call. The consultant signing the plan of treatment is a signature on paper. None of those produce a row in a database, so none of them appear in the number the CFO sees on Monday morning.

The result is a measurement system that can only ever blame the payer. If your clock starts at submission, every hour before submission is free. And hours that are free get spent.

You cannot manage the half of the delay your reporting is structurally unable to see.

The regulatory environment has moved on this. IRDAI has been issuing circulars setting expectations for cashless authorisation on admission and for final authorisation at discharge, and the current position is published in the health department section of IRDAI's circulars for health insurance. The practical consequence for a hospital is worth sitting with. As the payer side of the clock gets shorter and more scrutinised, the hospital side becomes a larger proportion of the total. A hospital that has never measured its own half will find that its turnaround problem does not improve even when the insurer's does.

Four timestamps worth instrumenting this month

None of this requires software you do not already own. It requires four events to be recorded, per case, with a name attached:

• Admission decided. The moment the treating consultant confirms inpatient admission, not the moment the IP number is generated.
• File opened. The moment a named person takes ownership of the preauthorisation for that case.
• Pack complete. The moment every document the payer requires for that procedure exists, signed, in one place.
• Submitted. The moment the request actually leaves the hospital.

Then record two more on the return leg: query received, and query answered. Six timestamps, in a spreadsheet if necessary. Within a fortnight you will know whether your problem is ownership, documentation, coding, or the payer. Most hospitals are surprised by which one wins.

What to fix before buying anything

Three changes cost nothing and usually move the number.

• Write the document pack per payer, per procedure. Not a general checklist. A specific list for the twenty procedures that make up most of your cashless volume, with the exact fields each payer rejects on. Keep it where the admission desk can see it at the point of admission, not in a folder.
• Name an owner at admission, not at submission. One person per case, visible on the board. Preauthorisation files stall most often in the gap where everyone assumes someone else has it.
• Run a five-minute stalled-file review every morning. Every case where the pack is not complete within a defined window, and every unanswered query older than one working day. Read the list out. The list shortens once it is read out loud.

If after a month the delay is genuinely at the payer, you now have per-payer, per-procedure evidence to take into your next tariff conversation. That evidence is worth more than the complaint it replaces.

Where Ospia fits

Ospia HOS is an AI-native hospital operating system rather than another hospital management suite. Instead of more screens for your staff to remember to fill, it runs a workforce of twenty-one AI agents, each with a defined identity, scope, permissions, tools and escalation path. Two of them are relevant here.

Eligibility, bed allocation policy and deposit rules resolve together at admission, and Front Office keeps the registration verified before the visit — so the documentation pack is being assembled from the moment admission is decided rather than the afternoon after. Documentation and coding gaps are flagged before submission, and Insurance Watch chases preauthorisations that expired or aged while a patient is still admitted and moves document-complete claims to the payer before they age. Because both act on one patient record, the clinical note, the coding and the estimate are drawn from the same source rather than reconciled later.

Autonomy is a dial the hospital holds, set per task from L0 to L4. The illustrative starting point published for Insurance Watch is L3; that is a starting point, not a fixed setting, and raising or lowering it is an audited decision with a name attached. Any execution above L2 requires a named human. No agent can act outside the task registry or above the level the hospital configured.

For a multi-speciality hospital with mixed cashless and cash volume, this usually shows up first as a shorter gap between admission and submission, because the gap is finally visible. We are not publishing a number for it. Ospia runs hospitals in production today, but we publish measured outcomes only as each hospital completes hypercare and agrees to be named, and until then there is no case study to point you at. If you are comparing options, our comparison pages say where the other product is genuinely stronger for your situation.

Ospia's stated intent is that the platform pays for itself inside the first year through billing leakage recovered, claim rejections avoided, and admission and discharge hours released. That is a target we hold ourselves to, not a result we have published.

Start with the arithmetic

Before a demonstration, do the sum yourself. Our revenue leakage calculator is published open — no email address, no form. Put your own admission volume and cashless mix into it and see what the delay between admission and submission is worth to you in a year. If the number is small, you have saved yourself a meeting. If it is not, book the executive demonstration and bring your six timestamps.