Appointment wait times and no-shows are the same problem

Waiting lists and missed appointments are usually owned by different people, but they are one number seen from two sides: demand that cannot get in, and capacity that is already booked and never used. In the United States the average wait for a physician appointment reached 31 days in 2025; in Australia, 26.0% of people said they waited longer than they felt acceptable for a GP appointment in 2024-25. Meanwhile, no-show rates reported in the outpatient literature run between 12% and 42% — capacity that was scheduled, staffed and then lost.

No front office can hire its way out of a workforce shortage this quarter. It can stop losing the slots it already has. This is what the current data supports, and what it does not.

In the US the queue is now measured in weeks

AMN Healthcare’s 2025 survey of appointment wait times found an average of 31 days to a physician visit. As reported by Advisory Board, the survey covered 1,391 medical offices across 15 large metropolitan markets, with fieldwork between 15 January and 24 February 2025. The same measure was 26 days in 2022 and 21 days in 2004.

The national average is the least useful number in the report. The spread is where the operational decisions live: the survey put the average across all surveyed specialties at 65 days in Boston and 12 days in Atlanta. A practice does not compete with a national average; it competes with the next available appointment across the street.

Family medicine, at 23.5 days, is 14% longer than in 2022 and 16% longer than in 2009. OB-GYN, at 41.8 days, is 33% longer than in 2022. Orthopedic surgery moved the other way, down 29% since 2022. A specialty that gets faster while the rest slow down is worth reading closely rather than explaining away: whatever those offices are doing with their schedules, it is happening inside the same workforce market as everyone else.

Australia measures the wait people feel, not the days on the schedule

Australia does not publish a national days-to-appointment benchmark for general practice. It publishes something arguably more useful: what people say about the wait. The Australian Bureau of Statistics released its Patient Experiences results for 2024-25 on 18 November 2025, and the headline was that fewer Australians are delaying care — 26.6% delayed or did not see a GP when needed, down from 29.2% the year before.

Two of those numbers move in opposite directions, and that matters for a clinic manager. Routine access improved. Urgent access did not: among people who saw a GP for urgent medical care, 41.6% were seen within four hours and 47.0% waited 24 hours or more, slightly worse than the previous year. A clinic that judges itself only on average booking lead time will miss the segment where the wait actually hurts.

The two national numbers do not measure the same thing

It is tempting to line the figures up and declare one market worse than the other. They are not comparable, and treating them as if they were is how boards end up chasing the wrong metric.

One is a supply measurement taken by calling offices. The other is a self-reported experience collected from households. A supply number tells you whether you have capacity. An experience number tells you whether that capacity was reachable. If your reporting only carries the first, the second can deteriorate for a year without a single line on your dashboard changing.

Insurance acceptance sits under the US supply number and rarely appears on the same slide. AMN’s survey, as reported by Advisory Board, found 82% of physician offices accepted Medicare and only 53% accepted Medicaid — with Medicaid acceptance as low as 28% in New York and as high as 85% in Detroit. A short average wait in a metro market does not mean a short wait for every payer. Australia’s ABS series, by contrast, already embeds cost delay as its own line (7.7% delayed or did not see a GP because of cost in 2024-25) rather than folding payer acceptance into days-to-slot.

Every no-show is a queue place that was already paid for

Here is the part that connects the two. While people wait weeks for a slot, a meaningful share of the slots that were successfully booked go unused.

A 2024 systematic review in Health Science Reports, which screened 23,403 records and included 16 studies, reports that no-show rates “in some studies” fall between 12% and 42%, and that in general outpatient clinics the rate “can even reach around 50%”. Those studies are drawn from mixed international outpatient settings, not from a US or Australian primary-care benchmark, so treat the range as a reason to measure your own rate rather than as a target to compare against.

The mechanics are unforgiving. A missed appointment on a schedule with a multi-week queue is not a lighter day. The room, the staffing and the wait were all paid for, nothing was delivered, and someone who would have taken that slot is still in the queue. Recovering unused capacity is the only lever that adds appointments this month without adding clinicians.

The same review found that open-access scheduling — keeping part of the day unbooked for same-day demand — reduced no-show rates significantly in 10 of the 16 studies, produced a non-significant reduction in 4, and changed nothing in 2. That is a real effect with real exceptions, not a guaranteed fix.

Reminders work, and the size of the effect is now measured

The evidence on reminders got sharper this year. A systematic review and meta-analysis published in the Journal of Hospital Management and Health Policy in March 2026 pooled 10 studies covering 8,236 participants and found that appointment reminders improved attendance with a risk ratio of 1.11 (95% CI 1.05 to 1.19) compared with usual care — roughly an 11% relative improvement.

Two details in that analysis are worth more than the headline. First, telephone reminders showed a statistically significant effect (RR 1.11, 95% CI 1.04 to 1.19, across 5 studies and 3,369 participants), while SMS reminders trended in the right direction but did not reach conventional significance (RR 1.14, 95% CI 0.99 to 1.31, across 4 studies and 4,636 participants). Second, heterogeneity was substantial (I² = 83%), meaning results varied widely between settings. The pooled studies were hospital-attendance studies, not general practice, which is another reason to treat the number as an order of magnitude rather than a forecast.

That maps onto what people say they want. In the outpatient review above, “about 97.2% of patients chose phone call (50.5%) and SMS (short message service) (46.7%)” as their preferred reminder method — an almost even split between the two channels, which is an argument for asking each person rather than standardizing on the cheapest channel.

What your practice or clinic can do this week

• Measure the two numbers separately. Time to the next available appointment for a new booking, and the share of booked slots that go unused. Most schedules report the first and quietly absorb the second.
• Reconcile them in the same meeting. Every recovered slot is a queue place freed the same week. If those two numbers are owned by different people, nobody is managing access.
• Make canceling as easy as booking. A slot released 48 hours out can be refilled. The same slot released at 9 a.m. usually cannot. Cancellation friction does not protect attendance; it converts cancellations into no-shows.
• Keep a working waitlist, not a list. A released slot needs a named person who can be contacted and confirmed quickly, or the release is theoretical.
• Ask which channel each person wants, and record it. Preference splits almost evenly between a phone call and a text message, and the pooled evidence for telephone reminders is currently the stronger of the two.
• Count the contacts you never answered. Calls abandoned in the queue and messages left unanswered after hours never appear as a no-show or as a wait — they appear as someone who booked somewhere else.

Where the operational detail differs by market, we keep it separate — one version for US practices, one for Australian general practice.

None of this shortens a 41.8-day OB-GYN queue. It does mean that the capacity you already staffed gets used, which is the only part of the access equation that responds to a decision made this week.