How to reduce OPD waiting time: what Indian studies show and what works

Ask patients what they dislike about an OPD and few mention the doctor. They mention the wait: the queue for the पर्ची (the OPD slip), the bench outside the room, the second queue at the pharmacy. This guide puts Indian numbers on that wait, shows where the time goes, and sets out the fixes that have evidence behind them. Most cost little, and few need another doctor.

How long do patients wait in Indian OPDs?

Published Indian studies keep finding the same pattern: a long wait for a short consultation. Most were done in government teaching hospitals, where crowds are biggest, so private OPD waits are likely to be shorter. But these are the waits patients remember, and they compare every hospital against them.

At the North Indian hospital, the whole visit, from arrival to collecting medicines, took 318.6 minutes on average, and the consultation was under 5% of it. At a teaching hospital in Ranchi, 58.3% of 400 patients reported waiting more than three hours between registering and seeing the doctor, and the longer people waited, the less willing they were to come back.

Where does the time go?

Three things explain most of it.

Everyone arrives at once. AIIMS New Delhi found that 79% of its OPD patients arrived before 10 am, but only 19% of them had been seen by 10 am. Patients often come early because they were told “come in the morning”, or because they fear losing their turn. In the Ranchi study, the time patients arrived was linked to their total wait; their department was not.

The first counter is a bottleneck. At PGIMER, patients queued 45 minutes just to register. At the North Indian hospital, 2 to 3 of its 13 registration counters took most of the patients while the rest ran nearly empty. A CAG audit of Delhi government hospitals found registration counters open from 7:30 to 11:30 am; at one hospital’s orthopaedic OPD, each counter averaged up to 628 patients a day: about 23 seconds per patient (our arithmetic: 240 minutes ÷ 628).

Patients make too many stops. At the North Indian hospital, the pharmacy bill, the payment and the medicines were three counters in three places, and the pharmacy stage alone took 119.4 minutes.

What works to reduce OPD waiting time

These are the fixes with the best evidence, roughly in the order we’d try them.

1. Give every patient a time window, not “come at 9”

After its arrival study, AIIMS split morning OPD reporting into slots of 8 to 9, 9 to 10 and 10 to 10:30 am, and printed the reporting time on each appointment slip. It didn’t publish a before-and-after figure, but the arithmetic is plain. Our illustration: a doctor sees one patient every 6 minutes from 9 am. If all 60 of the day’s patients arrive at 9, the average wait is 177 minutes and the last patient waits nearly six hours. Give the same 60 patients windows of 10 an hour, and the average wait falls to 27 minutes, with no extra doctor. Real days are messier, but the gap is that big.

Size each window to the doctor’s real pace, not the hoped-for one, and leave room for walk-ins. Here is how a booking call can set the window. It’s an example written for this guide, and the names are made up:

2. Let patients book by phone

The PGIMER authors recommended an appointment system by phone or internet, to cut the registration queue. The phone matters most: AIIMS started a call centre in April 2015 because most of its patients could not use the web portal or the IVRS. At a tertiary hospital in Uttarakhand, 99.5% of 400 OPD patients could use a smartphone and 71% knew online registration existed, yet only 26% used it; most didn’t know how. A booking line that answers, and gives a time window, does more for the morning crush than an app few patients will open.

3. Take patients’ details before they arrive

If the details are taken on the call or by a QR scan, the patient can skip the registration queue. The government says an IIHMR study found that ABDM’s Scan and Share service, where patients scan a QR code to get a digital queue token, cut the wait at registration from about an hour to 2 to 5 minutes; 23.21 crore such tokens had been issued by 18 June 2026 (PIB, 2026). A small time-and-motion study at a medical college in Agra found ABHA online registration took 4.15 minutes against 10.37 at the counter. AIIMS sends patients who already have an appointment to separate fast-track counters.

This shortens the first queue only. If everyone still arrives at 9, they now wait outside the doctor’s room instead.

4. Staff counters to the rush, and split the queues

Count arrivals by the half-hour for a week and staff counters to that curve, moving people to the busy counters rather than leaving some empty. The Indian Public Health Standards 2022 for district hospitals set the number of registration and dispensing counters by bed strength, ask for dedicated counters for women, the elderly and people with disabilities, and ask for a token number or an electronic queue. Two more splits help almost everywhere: keep booked patients out of the walk-in queue, and put the pharmacy bill, payment and medicines at one window instead of three.

5. Fast-track the very sick, the elderly and people with disabilities

A queue in token order is fair only if everyone in it is equally well. NABH’s 6th-edition standards ask that access to care be prioritised by clinical need (AAC.2.e). The Maintenance and Welfare of Parents and Senior Citizens Act, 2007 directs state governments to ensure separate queues for senior citizens (section 20), and the Rights of Persons with Disabilities Act, 2016 asks governments to give persons with disabilities “priority in attendance and treatment” (section 25(1)(c)). These duties fall on governments, not directly on private hospitals, but patients notice which hospitals follow them.

The clinical part matters most. Put someone at the front, usually a nurse, who looks out for chest pain, breathlessness, a very sick child or a pregnant woman in pain, and sends them to the emergency instead of the token queue. AIIMS posted nursing staff in its registration area to screen every patient and steer them to the right OPD.

6. Tell patients their real wait, and say when the doctor is late

David Maister’s essay on the psychology of waiting lines explains why: uncertain waits feel longer than known ones, and unexplained waits feel longer than explained ones. A patient told “about 40 minutes, you’re fourth” waits more calmly than one told “बैठ जाइए” (please sit down). At a government medical college in Thiruvananthapuram, putting televisions with local-language health videos in a crowded OPD waiting area cut reported arguments by about 81%, from 39 a week to 7.25, while waiting times stayed the same (Kandachamkulam and colleagues, 2026).

When a doctor is running late, say so early, to everyone waiting or on the way, with a new expected time. A message can be this short (a template; the name is made up):

7. Take steps out of the visit

Lean and Six Sigma methods map every step of the visit and remove the ones that add nothing for the patient. At an Indian super-speciality hospital attached to a manufacturing company, a Lean Six Sigma project cut the average OPD wait from 57 to 24.5 minutes (Gijo and Antony, 2014). The North Indian study’s suggestions were cheap: reassign registration counters to demand, add signs and guides, show the queue on a screen, and dispense common generic medicines from counters near the clinics (an idea it did not test).

Will new software cut waiting time?

Not on its own. At two government hospitals in Hyderabad, waits of two hours or more were reported by 40.2% of patients at the hospital with a health management information system and 23.4% at the one without (Gowda and colleagues, 2026). The authors pointed to patient load and technical problems. Software helps when it changes the flow, with time windows, registration before arrival and a live queue, not when it puts the same crowd on a screen.

How to measure OPD waiting time

You can’t improve a number nobody measures. NABH’s 6th-edition standards list “waiting time for outpatient consultation” as a key performance indicator, reported monthly in minutes:

• Formula: total minutes waited for consultation ÷ total number of outpatients.
• The clock starts when the patient reaches the OPD, which may not be when they registered. For a patient with an appointment, it starts at the appointment time.
• The clock stops when the consultant, not a junior doctor or resident, begins the assessment. A booked patient seen before their time counts as zero.
• No sampling: every outpatient counts.

The National Quality Assurance Standards for public hospitals also list waiting time among the service-quality indicators to measure every month (standard H4).

The simplest method is a pen: write the time on each patient’s slip when they arrive, and again when they go in to the consultant. A worked example for five patients:

The NABH average is 28.2 minutes (141 ÷ 5). Measured from arrival it is 58.8 minutes (294 ÷ 5), and three of the five waited more than an hour. Report both, with the share of patients waiting over an hour. The NABH figure shows how well you keep appointments; the arrival figure shows how early your patients still come.

Where DrMate fits

DrMate is an AI front desk for Indian hospitals, and several of these fixes are built into it. Its AI receptionist answers your OPD line day and night in Hindi and Hinglish and books each caller into a genuinely free time with the right doctor. The front desk gives every booking a time or an arrival window and a queue number, and each patient gets a private link showing their place and expected wait. When a doctor is delayed, reception presses “Running late”, every later patient’s time shifts, and those who said yes to WhatsApp updates are told. A 7- and 30-day report shows the average wait and no-shows, so you can see whether a change worked.

It won’t fix a pharmacy split across three counters or a registration desk with too few staff. Those need the process changes above, and our walk-through of the OPD process flow shows where to start.