Skip to main content
NewNMC’s advertising guidelines for doctors are in force from 6 Oct 2026. See what to change, and get a free compliance auditNew NMC advertising rules for doctors: what to change
AI & Automation

Keeping a Hospital Chatbot Current: Syncing OPD Timings, Doctor Rosters and Prices

A chatbot repeats whatever it last read. Here is how to make it query the roster and price master live, expire stale facts, and audit drift across your own channels.

Keeping a Hospital Chatbot Current: Syncing OPD Timings, Doctor Rosters and Prices

A hospital chatbot is only as accurate as the last time someone updated what it reads. Patients ask about three things more than anything else: when a doctor sits, whether the clinic is open, and what a procedure or health check costs. All three change constantly, and all three are the facts a bot is most likely to get wrong without anyone noticing.

This article is for hospital marketing, IT and patient-experience teams who already run, or are about to launch, a chatbot on the website or WhatsApp. It answers one narrow question: how do you design the system so the bot reads timings, rosters and prices from the place where they are actually maintained, and so that any fact that cannot be confirmed stops being stated? It assumes you know what retrieval-augmented generation (RAG) is. If you do not, start with the AI chatbot implementation guide.

The failure it prevents is ordinary and expensive. A patient travels to a Saturday OPD that was discontinued in September, or is quoted a package price that finance revised last month. The bot did not hallucinate in the usual sense. It faithfully repeated a stale document.

Why chatbots go stale

Most hospital bots are built once from a crawl of the website, a few PDFs and an FAQ spreadsheet. That content is copied into a search index. From that moment there are two versions of every fact: the one in the hospital's systems and the one in the index. Only the first is maintained.

Stale answers usually come from four places:

  • Marketing copies of operational facts. A department page says "OPD Monday to Saturday, 9 to 5". The real schedule lives in the HIS or the appointment system, and someone updates the page when they remember.
  • PDFs that outlive their validity. A price list from last financial year is still in the index because nobody removed it, and it still matches a query well.
  • Duplicates that disagree. The knowledge base holds the old and new version of a roster, and retrieval returns whichever scores higher for the wording of the question.
  • Time-relative content. "Camp on the 14th" or "offer valid this month" is correct on publication day and wrong afterwards.

Faithfulness metrics do not catch any of this. A metric such as Ragas faithfulness scores whether a response is consistent with the retrieved context. A stale document produces a perfectly faithful, perfectly wrong answer. Freshness has to be designed into the data, not tested out of the output.

Step 1: Classify facts by how fast they change

Do not treat the knowledge base as one pile. Sort every fact the bot may state into volatility classes, because each class gets a different retrieval method.

ClassExamplesChangesHow the bot should get it
StableDepartment descriptions, how to prepare for an MRI, visiting policyQuarterly or lessIndexed documents with a review date
SlowDoctor profiles, specialties, address, facility listMonthlyIndexed from the CMS, rebuilt on publish
FastOPD timings, doctor availability, holiday closuresWeekly or dailyLive lookup from the roster, never from a copied document
PricedConsultation fees, health-check packages, procedure estimatesPer revisionLive lookup from the price master, with an effective date
EventCamps, offers, webinarsPer eventIndexed with a hard expiry date

The rule that follows: fast and priced facts must not be stored in the vector index at all. The model should call a function that returns the current value, and the index should hold only the stable and slow classes. This is the single biggest change most teams can make.

Step 2: Name a source of truth for each fact

For every fast and priced fact, write down one owning system and one owning person. A simple register is enough:

FactSystem of recordOwnerUpdate trigger
OPD timings per departmentHIS or appointment schedulerFront-office headRoster change
Doctor-wise availabilityAppointment schedulerDepartment coordinatorLeave or slot change
Consultation feePrice masterFinanceAnnual or ad hoc revision
Package price and inclusionsPackage masterBilling and clinical leadPackage revision
Holiday closuresAdmin calendarAdministrationNotice issued

If a fact has no system of record, which is common for packages managed in a spreadsheet, fix that first. A chatbot cannot be more current than the least disciplined owner in the register. Where the owner is a person with a spreadsheet, at least make the spreadsheet a controlled file with a version history and a required "effective from" column.

Your records and the website should draw from the same place. If the department page says one thing and the bot says another, the patient sees an inconsistent hospital, not a technical problem.

Step 3: Build the lookup layer

The bot should not search for "Dr Rao timings". It should call something like get_opd_schedule(department, doctor_id, date_range) and receive structured data. A workable response shape has these fields:

  • value: the timings, fee or package price as structured data
  • effective_from and effective_to: when the value applies
  • source_system and record_id: where it came from
  • retrieved_at: when the lookup ran
  • status: one of confirmed, unconfirmed or unavailable

The system prompt then states the rule plainly: use only values returned by a tool for timings, availability and prices; if status is not confirmed, do not state a figure; say the front desk will confirm and offer a call-back or a booking link.

Two details matter. First, the model never does arithmetic on prices. If a package has add-ons, the lookup returns the computed total, or the bot states the base price and says extras depend on the clinician's assessment. Second, estimates for procedures should be stated as the hospital's published ranges, not as a quotation. NMC's October 2026 ethical advertising guidelines say any disclosure of charges, packages or fees must be factual, transparent and not misleading. A bot that quotes an outdated or incomplete price is a compliance risk as well as a service failure. Have your compliance reviewer read the full guidelines and decide how they apply to your communications.

Want this applied to your practice? Get a free ai & automation audit.

Book my free audit

Step 4: Put expiry rules on everything indexed

For content that stays in the index, give every chunk metadata at ingestion:

  • valid_from and valid_until (null only for genuinely timeless content)
  • review_due, set from the volatility class
  • owner
  • superseded_by, filled when a newer version is published

At retrieval time apply three filters before ranking: exclude chunks whose valid_until has passed, exclude any chunk with a superseded_by value, and exclude chunks past review_due unless the answer is flagged as stable. Event content expires automatically on its date. This makes "the camp on the 14th" disappear without a human remembering to delete it.

Add an expiry sweep to your weekly job: list every chunk within 14 days of review_due and send it to its owner. If the owner does not confirm in a week, the chunk is demoted from retrieval and the bot falls back to "please check with our team". An unanswered question is better than a wrong one.

When the bot does use dated content, make it say so: "As of the current published schedule" is honest and gives the patient a reason to verify before travelling. Do not let it say "always" or "every day" about anything that has a roster.

Step 5: Handle the conflicting-document case

Duplicates need a precedence rule. A sensible order is: live system lookup, then CMS content published within its validity window, then PDFs, then FAQ spreadsheets. When two chunks with the same topic disagree and neither is superseded, the bot should not choose. Log a conflict event and answer with the safe fallback.

Run a nightly conflict detector. For each entity (doctor, department, package), compare the structured value in the lookup layer against any numbers or times extracted from indexed text mentioning that entity. A mismatch creates a ticket for the owner of the indexed page. This is a few dozen lines of code and finds most drift before a patient does.

How to audit staleness across your own sites

The most useful first step is to measure the problem. Here is a method you can run in an afternoon on your own hospital, or on a handful of public hospital sites if you are benchmarking a market. Do not publish findings about named third parties without checking them with the provider.

  1. Pick the fast and priced facts. Choose ten: five OPD timings, three doctor availabilities, two package prices.
  2. Record the answer from the system of record on audit day, with a screenshot or export.
  3. Collect every place the fact appears: the website, PDFs, Google Business Profile, directory listings, WhatsApp auto-replies and the chatbot.
  4. Ask the chatbot each question in two phrasings, in English and one regional language, and record the response verbatim.
  5. Score each answer as correct, stale, conflicting or declined. Count a declined answer as a pass for safety but a miss for service.
  6. Date the stale ones. If you can find when the wrong value was last true, you know how long the drift lasted, which tells you the review interval you need.

Report the results as counts per class, not a single accuracy number. A bot that is right on stable facts and wrong on prices has a specific fix. Repeat the audit monthly for a quarter and you will see which owners and systems cause the drift.

Monitoring after launch

Add three signals to the dashboard:

  • Declined-because-unconfirmed rate: how often the bot used the safe fallback. A rising figure means a source system is returning errors or going unmaintained.
  • Conflict tickets opened and age: if tickets age, owners are not engaged.
  • Patient corrections: any message such as "the doctor wasn't there" or "the price was different at the counter". Tag these and review them weekly. They are your best ground truth.

If you use AiTextBotz, the WhatsApp assistant that handles the first response and hands over to your team, the handover is where an unconfirmed fact should land with the front desk. The design above applies to any bot.

Where to start

  1. List every fact the bot states about timings, availability and prices.
  2. Assign each a volatility class, a system of record and a named owner.
  3. Remove fast and priced facts from the index and replace them with lookup calls.
  4. Add valid_until, review_due and superseded_by to everything that remains.
  5. Define the safe fallback wording with your patient-experience team, and have clinical and compliance leads approve it.
  6. Run the six-step staleness audit, then repeat it monthly.

Our AI chatbot service page describes how we scope this work. The ROI of healthcare chatbots covers the commercial case, and the question of whether a clinic should use a chatbot at all is answered separately. The hospital's own clinical and compliance teams must review any answer a bot gives about care.

Filed underhospital chatbot outdated informationhospital chatbot knowledge base updatechatbot OPD timings syncchatbot doctor roster integrationchatbot package price accuracy

About the author

Co-Founder & CTO · Gurugram, India

Arush builds the systems the rest of the company runs on — the patient-acquisition pipelines, the AI chatbots, the analytics that tie a single click to a booked appointment. If a campaign reports a number, it's because something his team built is tracking it.

Free guide · 15 pages

The Patient Acquisition Blueprint

A 90-day plan from first search to booked appointment.

The Pioneers Brief — one essay, three benchmarks, one teardown, every Thursday. Subscribe free →

Free ai & automation audit

Your next chapter starts with a conversation.

Three quick questions. We come prepared with a plan for your specialty and city.

  • A 30-minute call with a healthcare specialist
  • Your top three growth opportunities, in writing
  • Honest advice — even if that means not hiring us
  • No obligation and no hard sell
You’ll speak with a senior strategist
Consultations are hosted by Arush Thapar’s team.
Step 1 of 3Takes 30 seconds

What do you want to improve first?

Details stay privateNo obligationReply within 1 working day

Client portfolio · Healthcare websites

Work relevant to keeping a hospital chatbot current syncing opd timings doctor.

Explore client work →