Patient Wait Times: How to Measure Them — and What's Actually Normal

"Patient wait time" is not one number — it's three, and they get merged into one all the time. There's the lead time to get an appointment (days), the in-office wait from check-in to being seen (minutes), and third next available appointment (the access metric operators actually run on). And here's the part almost no one says out loud: there is no reliable national benchmark for in-office wait. The famous "18 minutes" is a survey, not a measurement — the largest real dataset puts the median closer to four. So the move isn't to chase someone else's average — it's to measure your own trend.

Below: the three metrics untangled, what the data actually says about each, the evidence on what a wait really costs you — and the two numbers worth tracking every week.

The three metrics people merge into one

Almost every "average patient wait time is X" article you'll read quietly blends three different measurements into a single scary number. They answer different questions and have different fixes, so pull them apart first.

1. Appointment lead time (days to get in)

How long a patient waits between calling and being seen. This is what AMN Healthcare's Survey of Physician Appointment Wait Times measures — and note the word survey. Its widely quoted figure is a new-patient average across a handful of specialties in a set of large metro markets, not a national number for every practice. It tells you about market access; it says nothing about how your own front desk runs.

2. In-office wait (minutes from check-in to seen)

The minutes a patient sits in your waiting room and exam room before the provider walks in. This is the number patients feel most sharply — and, as we'll see, the number with the shakiest benchmark of the three. Most figures you've heard for it come from consumer surveys run by rating vendors, not from measured data.

3. Third next available appointment

The operator's metric. It's the average number of days until the third next open slot, drawn from the advanced-access model that Murray and Berwick introduced in JAMA in 2003. Why the third and not the very next opening? Because — as a VA evidence review of access measures puts it — "the first and second available appointments may reflect openings created by patients canceling appointments and thus does not accurately measure true accessibility." The next open slot might just be a fluke cancellation; the third is a truer read on your real backlog. You can measure it per provider, weekly, from your own schedule.

So what's actually normal?

This is where it pays to look twice, because the popular numbers fall apart on a second read.

Appointment lead time: rising, but read the survey carefully

AMN Healthcare's 2025 survey reported an average new-patient wait of 31 days across six specialties in 15 large metro markets — up 19% since 2022 and 48% since 2004. The per-specialty spread is wide: OB/GYN 42 days, gastroenterology 40, dermatology 36.5, cardiology 33, family medicine 23.5, and orthopedic surgery just 12.

But be careful before you repeat "31 days" as gospel. Two caveats the headlines skip. First, the composite jump is partly a composition change: 2025 added gastroenterology — a high-wait specialty — that wasn't in the 2022 survey, which mechanically lifts the average. Look at like-for-like specialties and the picture is mixed; orthopedic surgery actually fell 29%. Second, the survey's own method inflates the mean: when a physician was "booked out" and not taking new patients, that office was recorded as a 365-day wait. That's defensible for a survey, but it isn't the wait a typical patient experiences. For reference, the 2022 edition's composite was a calmer 26 days, built from 1,034 physician offices called across those same 15 metros. To its credit, AMN's method is mystery-shopper — research staff phone real offices as prospective patients — which makes it sturdier than a consumer-recall poll like Vitals. It's still a survey of big metros, though, not a measurement of your town.

In-office wait: the benchmark almost everyone cites doesn't exist

This is the crux of the whole topic. You've seen "the average office wait is about 18 minutes" everywhere. That figure traces to the Vitals Physician Wait Time Report — a rating vendor's consumer survey whose series ended in 2018, built partly on a few hundred online respondents. It's self-reported perception, not measured time.

The largest measurement tells a different story. A 2017 Health Affairs study by Oostrom and colleagues analysed 21.4 million EHR-timestamped visits across 2,581 practices — one software vendor's panel, using 2013 data — and found a median in-office wait of about four minutes (or about 9.5 minutes if you count only the visits that had any wait at all), with roughly 17% of visits running past 20 minutes and 10% past 30. That's one vendor's practices, not a national sample, so it isn't a benchmark either — the useful part is the shape. Most visits are quick, but the tail is real, and it's the tail patients remember.

So the verdict: no government body or professional association we could find publishes a national in-office wait benchmark. Even MGMA, when it wrote about wait times in 2025, cited AMN's appointment figure rather than any in-office number of its own, and reported a small poll in which "66% of groups reported wait times stayed the same (40%) or shortened (26%)." What exists are surveys and internal targets — not a standard to measure against.

Why wait time is worth fixing (what the evidence shows)

You don't need a benchmark to know a shorter wait is better — but it helps to know exactly how it pays, and where the evidence is strong versus thin.

Longer lead time drives no-shows (strong)

This is the best-evidenced link. A study of 46,655 appointments at the University of Virginia eye clinic found the no-show rate climbed steeply with lead time: appointments booked within 0–2 weeks had a no-show rate around 9% for residents (2.4% for faculty), rising to 38% at a six-month lead. The authors estimated no-shows would fall by nearly 60% in the best case where every appointment was booked within two weeks — a modelled ceiling, not a promise. A separate ambulatory study saw the same shape — roughly 23% no-shows within 30 days versus 47% beyond it. It's a single-clinic effect size in each case, so don't treat the exact percentages as universal — but the direction is consistent, and it's why access and no-shows belong on the same screen. We cover the downstream number in how to calculate and cut your no-show rate.

Longer in-office wait lowers satisfaction (real, but scope-limited)

A 2019 study in JBJS Open Access of 4,216 patients at an academic orthopedic clinic found the odds of a satisfied patient were nearly four times higher below a 15-minute wait than at 15 minutes or more (odds ratio 3.78; 95% CI 3.30–4.33; p < 0.01). That's a strong signal — but it's one specialty at one institution, and it's a threshold finding, not proof that satisfaction "craters at exactly 20 minutes" (a line that, again, comes from the Vitals survey framing, not a study).

"Patients leave over long waits" — softer than it sounds

You'll see confident stats that "30% of patients have walked out over a wait" or "one in five switched doctors because of it." Those come from the same Vitals consumer survey — they're self-reported intent, not a measured study of who actually left and why. Long waits plausibly cost some practices patients — but there's no clean measurement of it, so treat it as a reasonable worry rather than a proven number, and don't dress a survey up as proof of attrition. If you want the metric that actually tracks whether patients come back, that's patient retention rate, measured from your own records.

Myths worth retiring

• "The average office wait is 18–20 minutes." A rating-vendor consumer survey (series ended 2018), not a measurement. The largest measured dataset puts the median near 4 minutes.
• "31 days — that's the MGMA benchmark." Wrong on two counts: it's AMN Healthcare, and it's a survey of appointment lead time in 15 large metros — not in-office wait, and not a national figure.
• "The CDC says the average wait is 16 minutes." That's emergency-department data, a different setting entirely — and that survey ended after 2022.
• "Patients walk out after X minutes." There's no measurement behind any specific threshold or any "dollars lost per minute" figure. Treat those as marketing, not fact.

Measure your own — because there's nothing to borrow

Since no benchmark exists to grade against, the whole game is your own trend. Two numbers do the work:

Patient access is one of the twelve numbers we argue every practice should watch — see the full set in the 12 KPIs every medical practice should track.

Frequently asked questions

Written by

Olha · clinic data analyst

I build the reporting our managers open every morning at a multi-branch medical clinic — and package it so other practices don't have to start from scratch.

Every figure here is attributed to its primary source and labelled as a measurement, a survey or a target — because the difference matters. Single-institution effect sizes are named as such and shouldn't be read as universal. Where no reliable benchmark exists, we say so rather than invent one.