Skip to main content

Polyclinic Walk-In Queues: Assign Patients Without Losing Ownership

A shared walk-in queue works when each patient has a clear eligibility group, one accepted consultation owner and a visible route for reassignment. The next available doctor can take an eligible patient who agrees to that arrangement; availability alone does not establish clinical suitability. Keep urgency assessment with trained clinical staff and explain any change in doctor or queue order.

By Bractis · Updated · 8 min read

For Indian polyclinic owners, reception teams and doctors sharing walk-in patients across rooms or split OPD sessions.

Decide which patients can genuinely share a queue

A common queue can help two general physicians share demand, but a cardiologist, dermatologist and paediatrician are not interchangeable simply because they use the same reception desk. Define the service group before discussing automatic allocation. For each group, identify the participating doctors, session availability, relevant restrictions and how a patient can request continuity with a particular clinician.

At arrival, ask whether the patient is comfortable seeing an available doctor in that group. Someone returning to discuss a previous plan may prefer the original doctor even when another room is free. Explain the expected arrangement without promising an exact waiting time. Record the preference where staff assigning the visit can see it; a note buried in a previous consultation will not guide today's queue.

Separate administrative eligibility from urgency. A routine-looking token number is not a clinical assessment. Staff need a clear way to request clinical attention when a concern is raised, including when a patient declines to describe symptoms publicly. Software should not require the patient to finish every registration field before staff can alert the appropriate clinician.

Separate queue position, assignment and consultation

Use a small number of states with meanings staff can explain. Waiting means the patient is available for assignment. Assigned means a doctor has accepted responsibility for the next consultation, while consultation started records that the encounter has actually begun. A screen opened for a quick preview should not silently imply that a doctor has seen the patient.

FHIR R4 represents encounter status and participants separately; it also notes that local business rules determine status transitions. That distinction is a useful design reference, not a requirement to adopt these exact queue labels.

Give each transition an actor and time. If reception assigns a patient to an unavailable doctor, preserve the failed or reversed assignment rather than changing the record to imply continuous ownership. The patient-facing display can remain simple while the internal history explains why a transfer happened.

On a narrow screen, scroll the table horizontally to read all columns.

Shared walk-in states and the decision each one supports
StateWhat it meansNext action and responsibility
Eligible, waitingPatient is present and suitable for this service poolQueue coordinator offers the next eligible assignment
Assigned, not startedOne doctor has accepted the visitReceiving room calls the patient; coordinator can resolve absence
In consultationClinical interaction has begunTreating doctor controls any onward clinical handover
Return to waitingAssignment ended before consultation, with reasonCoordinator restores an explained position without creating another arrival
Left or unavailablePatient is not currently ready to attendRecord contact or return arrangement; do not mark consultation complete

Source: HL7 FHIR R4: Encounter definitions

Test what happens when two doctors select the same patient

The difficult moment is not an ordinary click. It is two rooms seeing the same waiting patient and selecting that row before either screen refreshes. A useful demonstration uses two staff accounts and two devices. Both attempt to accept the same fictional visit, then inspect the accepted owner on each device and at reception.

The expected outcome is one accepted assignment, with the other doctor receiving an understandable update. Two green success messages are a failure even if the database eventually contains only one owner: both doctors may call the patient. A silent switch from Doctor A to Doctor B is also inadequate because the first room may already be preparing the consultation.

Ask what happens if the acceptance reaches the server but the reply is lost. Staff should be able to inspect the existing assignment before trying again. If the product cannot safely coordinate simultaneous selection, nominate one dispatcher during the pilot. That is a deliberate staffing arrangement to evaluate, not a reason to claim that simultaneous allocation has been solved.

Move an assignment without losing the patient

An assigned doctor may be called away, reach the end of a session or discover that another clinician is needed. Distinguish a pre-consultation reassignment from a clinical handover. The former changes the next available room; the latter transfers information and responsibilities after clinical work has started.

Before moving a waiting patient, explain the proposed clinician, any change in fees and the option to wait where that is feasible. Reception should not overwrite the preference simply to make a queue metric improve. Preserve the original arrival time and record the new assignment separately so a transfer does not restart the measured wait.

A task model can distinguish the person requesting work from the person currently responsible for it. FHIR R4's Task definitions provide that distinction. In the clinic's own process, make acceptance observable: sending a notification to the next room is not enough evidence that someone has taken over.

Source: HL7 FHIR R4: Task definitions

Measure waiting without hiding transfers or exclusions

Define arrival-to-start time using the original arrival and the actual start of consultation. Assignment time is a different milestone. Report patients still waiting at the observation cutoff separately; otherwise a session with the longest waits unfinished can appear deceptively quick.

In a fictional five-patient sample, completed waits are 10, 15, 20, 25 and 40 minutes. Their mean is 22 minutes and median is 20. A sixth patient has already waited 50 minutes and has not started. Do not report that person's wait as zero or quietly omit their existence; describe the five completed waits plus one ongoing wait of at least 50 minutes.

Review the reasons for outliers before changing the rule. A named-doctor preference, a clinical escalation, an unavailable interpreter and a missed call have different remedies. Avoid rewarding clinicians solely for short consultations or assigning every complex patient to one room to protect another doctor's figures.

Pilot the policy with the staff who will explain it

Start with one service group and a defined session, using fictional records for the exception rehearsal. Include a patient who requests a particular doctor, a same-name pair, a simultaneous claim, a temporarily absent patient and a doctor leaving early. Reception should explain each outcome without a developer translating the screen.

During a real operational pilot, use the clinic's authorised systems and ordinary access controls. Keep diagnoses off public queue displays and use a discreet calling method for people who cannot hear the announcement or need assistance reaching the room. A shared queue does not require sharing every clinician's private notes with every front-desk role.

At the review, retain the cases where the process failed: duplicate calls, unclear ownership and unaccepted reassignment. Compare those with actual staffing and patient preferences. Adopt wider pooling only when the team can explain who owns each waiting visit, including during disruption.

Fictional clinic example: two rooms, one accepted visit

At 18:10, Doctors A and B both select visit V-084 in a general-medicine pool. The system accepts A's assignment and shows B that the visit is already assigned. Before consultation starts, A is called away. Reception explains the change; the patient agrees to B, who accepts at 18:14. The original 18:00 arrival remains unchanged. Consultation begins at 18:18, so arrival-to-start is 18 minutes, not four minutes from reassignment. These times illustrate the record logic; they are not a performance target.

Shared walk-in assignment worksheet

Use these prompts in your clinic’s own notes. Record the result, responsible person and next action for each item.

Pool boundaries
Which doctors can share which visits, and when does patient preference or clinician assessment require a different route?
Simultaneous selection
What does each room see when two users accept one fictional visit, including a lost response?
Transfer acceptance
Who remains responsible before the receiving doctor accepts, and how is the patient informed?
Measurement cutoff
Which start event defines waiting time, and how will patients still waiting appear in the review?

Questions doctors ask

Can every walk-in be allocated to the next free doctor?

Only within a clinically appropriate service group and the patient's agreed arrangement. A free room does not establish the doctor's suitability for every complaint, age group or continuity need. Clinical urgency follows the clinic's assessment pathway.

Should opening a chart automatically claim the patient?

The clinic should distinguish previewing information from accepting the consultation. If opening a chart does claim the visit, that behaviour must be explicit, reversible through an accountable process and visible to the other rooms.

Does reassignment restart the waiting time?

Keep the original arrival and record reassignment as a separate event. A second waiting measure can describe time after reassignment, but it should not replace total arrival-to-consultation time.

What if the assigned doctor leaves before seeing the patient?

Record why the assignment ended, explain the alternatives and obtain acceptance from the receiving doctor. Until that is settled, someone must own the unresolved waiting visit rather than leaving it in an unattended inbox.

What should we demonstrate before buying shared-queue software?

Use two accounts to attempt one assignment, then rehearse absence, named-doctor preference and a lost network response. Ask to see the final owner, preserved arrival time and readable history. Confirm these behaviours in the configured product.

Sources and scope

Updated on . An original operational design for Indian polyclinics, supported by narrowly cited technical definitions. It does not prescribe triage rules or claim automatic allocation is available in every Bractis configuration. No independent clinical or legal review is claimed.

  • HL7 FHIR R4: Encounter definitions

    Version 4.0.1, a fixed technical reference rather than the latest release or an Indian legal standard. Checked 25 October 2026; actual implementations differ.

  • HL7 FHIR R4: Task definitions

    Version 4.0.1. Separates task status, requester and owner. Used as a data-model illustration, not proof that any clinic product implements it. Checked 25 October 2026.

Try the workflow
with Bractis.

Bring your service groups and a fictional two-doctor assignment conflict. Use the walkthrough to evaluate ownership, reassignment and patient communication in the available configuration.

Request a clinic demo