Patient no-shows in Indian OPDs: why they happen and what reduces them

A no-show costs an OPD twice. The doctor’s slot goes empty, and the patient who could have had it was told to come another day. For the patient who stayed away, a missed review can mean a problem caught late. This guide sets out what is actually known about no-shows in Indian OPDs, which fixes have evidence behind them, and where common advice is weaker than it sounds.

How common are no-shows?

A systematic review of 105 studies found an average no-show rate of about 23%, highest in Africa (43.0%) and lowest in Oceania (13.2%). There is no peer-reviewed India-wide figure. The 30% or so that some software sellers quote for India comes without a source, so don’t plan around it. Measure your own instead: booked patients who neither came nor cancelled, divided by all bookings, by doctor and session, and separately for new and follow-up patients.

One large Indian dataset comes from AIIMS New Delhi, a free public hospital with extreme demand, where only 23% to 47% of new-patient bookings turned up between 2015 and June 2017, depending on how they were booked. That isn’t typical of a private hospital. The useful part is the comparison between booking channels, further down.

Why do patients miss OPD appointments?

The most recent Indian evidence comes from a 2026 study at a tertiary teaching hospital with online booking, where researchers phoned 100 patients within days of a missed visit. Asked to rate each reason from 1 to 5, they scored these highest:

The same patients also rated the online system as easy to use (4.02), so read the second line with care. Distance from the hospital made no significant difference. Two findings matter more than the ranking. First, 76% of these patients said they had received a reminder, so a reminder alone is not enough. Second, ease of cancelling or rescheduling scored 2.74 out of 5, the lowest of anything asked, and only 18% had tried to reschedule.

The global review adds two warning signs: a long gap between booking and appointment, and a history of missing. A patient booked three weeks out, who missed last time, needs more than one message.

Do reminders work?

Yes, within limits. A Cochrane review of eight randomised trials (6,615 patients, none in India) found that text reminders raised attendance compared with no reminder (risk ratio 1.14). Attendance was 67.8% with no reminder, 78.6% with text reminders and 80.3% with phone-call reminders. Texts and calls worked about equally, and texts cost less.

Indian trials point both ways. At a dental college OPD in Ghaziabad, patients sent an SMS a day before and again on the day kept 79.2% of appointments on time, against 35.5% of those sent nothing. But in a trial at two public sub-district hospitals in Punjab, WhatsApp reminders sent 3 days and 1 day before a hypertension follow-up, with short messages correcting common myths about blood pressure, made no significant difference: 21.8% came, against 19.6% without. Reminders fix forgetting. They do little for a patient who feels fine and sees no reason to return; that needs the doctor to say, at the visit, why the review matters.

When to send reminders, and what to say

The trials above sent reminders anywhere from four days ahead to the day itself, and none compared timings head to head. For an Indian OPD, two messages cover most of what goes wrong:

1. The evening before. The family is home and can plan leave and transport, and there is still time to give the slot to someone else if the patient cancels.
2. The morning of the visit, with the time to come. This catches anyone who forgot overnight and keeps people from arriving at 8 for an 11 o’clock turn.

A booking made weeks ahead deserves a third reminder a few days before. Write in the patient’s language: in the 2026 study, 78 of the 100 patients agreed or strongly agreed that they would prefer reminders in their own language. Say who, when and where, what to bring, and how to cancel. A template, with made-up names:

Make it easy to cancel, move or come late

A cancellation is far better than a no-show, because the slot can go to someone else. Yet in the 2026 study, only 18% of the patients who missed had tried to reschedule. Three things help:

• One tap or one call. Put “can’t come” and “change date” in the reminder, and make sure the phone is answered. If cancelling means ringing a line that’s always busy, most patients won’t bother.
• A waitlist for full days. Note who wanted a full day, and call the first in line when a slot frees up.
• Room for the late. A patient 20 minutes late is still a patient. Fit them in after the next one rather than sending them home.

A cancellation call can be this short. It’s an example written for this guide:

Does the booking channel matter?

AIIMS New Delhi published turn-up rates for each way of booking, from 2015 to June 2017 (AIIMS white paper, Figs 5 and 6):

Among the ways to book from home, the call centre stood out: new patients who booked by talking to a person came about twice as often as those who booked through the IVRS (46% against 23%). The paper doesn’t say why, and different people use different channels, so this is an association, not proof. Our reading: a time agreed in conversation feels like a promise, while a slot handed out by a machine is easier to forget. AIIMS’s IVRS didn’t even let patients choose a doctor.

Should you overbook?

Overbooking, booking more patients than slots to cover expected no-shows, does fill the doctor’s day, but it lengthens waits. In an Italian outpatient centre, overbooking raised productivity but slightly increased patients’ waiting time and staff overtime, and it worked because it was built on the centre’s own no-show data. AIIMS did something similar with the Indian Statistical Institute, predicting turn-up from past data to set each OPD’s new-patient limit, to within 10% of the patients who actually came.

The cost lands on the days everyone comes. Our illustration, for a session of 30 six-minute slots and a 20% no-show rate:

The arithmetic: patients on an average day = bookings × 0.8; overrun = extra bookings × 6 minutes. So overbook last, after reminders and easy cancelling have brought the rate down, and by your own measured rate for each doctor and session, never a guess. The waits it adds are covered in our guide to reducing OPD waiting time.

Consent comes first

Reminders on WhatsApp work only if they’re allowed. Three rules:

1. Ask first, clearly. WhatsApp’s Business Messaging Policy lets a business message people only if they have given their number and opted in, and it must honour any request to stop. India’s Digital Personal Data Protection Act, 2023 says consent must be “free, specific, informed, unconditional and unambiguous with a clear affirmative action” (section 6(1)), and that withdrawing it must be as easy as giving it (section 6(4)). These duties apply once the DPDP Rules’ 18-month phase-in ends in May 2027 (PIB, 2025), but there is no reason to wait. A box ticked on the patient’s behalf is not a yes.
2. Keep it to the appointment: confirmations, reminders and changes. No offers.
3. Never medicines. WhatsApp’s policy bars businesses from using it to buy, sell or promote prescription drugs or medical products. Medicine orders belong with your pharmacy, not in a chat.

Where DrMate fits

DrMate builds these steps into the front desk. WhatsApp messages go only to patients who said a clear yes, on the call, at the counter or on the booking page, and “stop” is honoured. Patients get a booking confirmation and a reminder from 6 pm the evening before, and their taps act on the booking: “I’ll come” confirms it and “Cancel” frees the slot. Callers can move or cancel by phone with the AI receptionist, day or night, after a clear yes. Full days have a waitlist, and when a place frees up, reception calls the first person waiting. The front desk report shows no-shows for the last 7 or 30 days. DrMate never sends medicines, offers or orders on WhatsApp.

You can hear the AI receptionist move a booking on our demo line, 080 6553 5931, where it answers as Sunrise Hospital, a sample hospital we made up.