NABH 6th edition documentation checklist: what your system has to produce

An NABH 6th edition documentation checklist is not a folder to assemble. It is evidence your systems should produce daily. What that means, with the official source.

Ninety days before an assessment, somebody in your hospital starts a folder. By day sixty it is a shared drive with forty subfolders. By day thirty it is three people photocopying, and asking the ICU in-charge whether anyone actually recorded the daily review of the ventilated patient in bed 4. The assessment goes fine. Nobody enjoys it. And everybody knows that what was demonstrated that week is not how the hospital runs the other fifty-one.

So when a medical director searches for an NABH 6th edition documentation checklist, the honest answer is not a list of files to collect. It is this: the checklist is already written, it is published by NABH, and your job is not to assemble it in the last quarter. Your job is to make sure the hospital's systems produce the evidence as a by-product of the work, so that the checklist becomes a report you run rather than a project you staff. This article is about which classes of evidence a hospital information system must generate on its own, what you can fix without buying anything, and where software actually helps.

Read the standard, not somebody's summary of it

There is a genre of downloadable checklist doing the rounds in Indian hospital WhatsApp groups: a spreadsheet of objective elements, lifted from an earlier edition, with a vendor logo in the corner. Some of it is wrong. Most of it is stale. And all of it is unnecessary, because the source is public.

Work from the NABH Hospital Accreditation Standard, 6th Edition, published January 2025, and from NABH's own listing of its accreditation standards and programmes. The standard states its chapters, its objective elements and their commitment levels itself. Nothing this article says replaces reading it, and any vendor who hands you a clause-by-clause checklist without pointing you at the published document first is selling you a shortcut that will not survive an assessor's follow-up question.

What a vendor can usefully tell you is different: given that the standard asks for this evidence, what must a system do so the evidence exists before anyone goes looking for it.

Why the evidence goes missing in the first place

The gap is almost never ignorance of the standard. Your quality team knows the standard better than your consultant does. The gap is structural, and it has three causes.

Documentation happens after the work, not during it. A nurse completes a procedure and writes it up at the end of the shift. A CAPA is raised in a meeting and recorded in minutes. A risk is identified in a huddle and lives in somebody's notebook. Every one of these is a real record, and every one is written at a moment when the system that could have captured it automatically had already moved on.

The hospital runs on three or four systems that do not agree. Billing knows the patient was in ICU for six days. The nursing chart says five. The discharge summary, typed in Word, says something else again. When an assessor traces one patient journey end to end, which is what a good assessor does, the discrepancy is the finding. Not the care. The discrepancy.

Evidence is treated as a document class rather than an operational output. A hand hygiene audit becomes a PDF. An HAI surveillance return becomes a monthly email attachment. An antimicrobial stewardship review becomes a committee file. Once evidence is a document, it can be produced late, produced selectively, or produced by one person who then leaves.

The question an assessor is really asking is not "do you have the record?" It is "could this record have been made any other way than by the work actually happening?"

The four classes of evidence your system must produce

Rather than a clause list, think in classes. Every requirement that touches documentation falls into one of four, and each has a different systems implication.

• Transactional evidence, generated per patient, per event. Registration and consent, assessment and reassessment timings, medication orders, transfers, discharge summaries. This must be captured at the point of work, timestamped, attributed to a named clinician, and impossible to backfill silently. If your system allows a retrospective entry without marking it as one, it is producing documents, not evidence.
• Surveillance evidence, generated continuously, reported periodically. Infection surveillance, antimicrobial use, incident reporting, indicator data. The defining property is denominators. A surveillance rate is only credible if the system knows device days and patient days without anyone counting them by hand.
• Governance evidence, generated by decisions. CAPA with owners and closure dates, a risk register that is dated and reviewed, committee decisions traceable to the data that prompted them. This is where most hospitals are weakest, because these artefacts usually live outside the clinical system entirely.
• Statutory evidence, generated by law rather than by NABH. Prescriber identity and registration on every document, generic-name prescribing, serialised certificates, the record-release clock, consent for data sharing, GST-compliant invoicing. Assessors look at these because non-compliance with Indian law is not a quality question, it is a licence question.

Audit your current stack against those four classes and you will usually find transactional evidence is adequate, statutory evidence is partial, surveillance evidence depends on one dedicated person, and governance evidence exists only in minutes.

What to fix before you consider buying anything

Four of these cost nothing but discipline, and they move you further than most software purchases.

• Trace three patients end to end, this month. One ICU case, one day-care, one long-stay medical. Follow every document that patient generated across every system. Time how long it takes and write down every discrepancy. That exercise is your real gap analysis, and it is more honest than any readiness scorecard.
• Move the risk register and CAPA log out of minutes and into a dated, owned, reviewable list. Even a shared spreadsheet with owner, due date and closure evidence beats a committee file, because it can be shown at any hour without preparation.
• Fix denominators before numerators. If device days and patient days are counted manually, your surveillance evidence has a single point of failure with a notice period.
• Stop preparing in quarters. Run one internal tracer per month against the published standard, on a random day, with no notice. The purpose is not the score. The purpose is to find out what the hospital can produce without warning.

Where Ospia fits

Ospia HOS is an AI-native hospital operating system rather than another hospital management system, and the reason it is relevant here is architectural, not promotional: one patient record across the hospital's domains means the assessor's end-to-end trace has nothing to reconcile.

On compliance, Ospia is explicit about what two words mean. On our compliance page, "supported" means implemented and demonstrable, and "by design" means the platform produces the required behaviour or evidence as a by-product of normal operation. NABH evidence sits in the second category: CAPA, a live risk register, HAI surveillance and antimicrobial stewardship are produced by daily work rather than assembled for an assessment, and NABH indicator packs are part of the compliance surface alongside FHIR R4, HL7 v2, ABDM, DPDP and GST e-invoicing.

The statutory layer is enforced at write time rather than audited after the fact. NMC rules 1.3.2, 1.4.2 and 1.5 are applied as the document is created: generic-name prescribing, prescriber statutory identity on every document, serialised certificates, and the 72-hour record-release clock escalated before it lapses. Registration is ABHA-ready, and every use of personal data traces to recorded consent held in a consent ledger that is consulted before use.

Among the agents running in the system are quality, infection control and NMC compliance watchers, and the ICU assistant surfaces documentation gaps to the clinical team rather than leaving them to be found at discharge. Every agent runs on an autonomy dial the hospital sets per task, from L0, where a person does the work and the system records it, through to L4, where the system acts inside hospital-written boundaries and every run still leaves a replayable ledger entry, an event trail and a reversal path. For a medical director, that dial is the adoption argument: nothing has to be autonomous for the record-keeping to improve.

What we will not claim

Ospia HOS is built and running hospitals in production today, and the clinical core is live. We do not publish a count of those hospitals, and we publish names only as each one completes hypercare and agrees to be named. We have no NABH case studies, no published accreditation outcomes and no measured audit-preparation savings, and we will not estimate them. Ospia is not ISO 27001 certified; we list it as a roadmap item. Our workflows are reviewed with practising doctors, nurses and administrators, and the clinical advisory panel has seats open.

Accreditation is awarded to your hospital by NABH assessors, not by a software vendor. No system can produce an outcome there. What a system can do is make sure that when someone asks for the record, it already exists.

If you want to see the evidence trail against your own chapters rather than a slide deck, the useful next step is an executive demonstration: a working system, traced through one patient journey, with the gaps shown as well as the strengths. It is also worth reading how this differs from a conventional HMS deployment in our comparison with Napier, and, if you run a multi-speciality hospital, what the compliance surface looks like for multi-speciality hospitals specifically.

Frequently asked questions

Where do I get the official NABH 6th edition documentation requirements? From NABH directly. The 6th edition standard published in January 2025 states its own chapters and objective elements, and the NABH standards page lists which accreditation programme applies to your hospital. Treat any third-party checklist as a reading aid, never as the source.

Can a hospital information system make us NABH accredited? No. Accreditation is assessed by NABH against your practice. A system can reduce the manual assembly of evidence and remove discrepancies between departments, which is a different and more modest claim.

What is the most common documentation gap? In our experience of reviewing workflows with practising doctors, nurses and administrators, governance evidence, meaning CAPA closure and a dated, reviewed risk register, is more often weak than clinical documentation, because it usually lives outside the clinical system.

Does Ospia publish NABH outcomes from its customers? Not yet. Ospia publishes no case studies; the case-study slots on the site are unfilled placeholders describing what will be measured. Hospital names are published as each one completes hypercare and agrees to be named.