Wait times
For serious mental illness, how often is there no treatment at all, who gets left out, and where are the supply deserts?
The average wait for a new-patient physician appointment reached 31 days in 2025, up 48% since 2004. Waits vary sharply by specialty (41.8 days for OB/GYN, 12 for orthopedics) and by metro (65 days in Boston, 12 in Atlanta).
The problem
Across the U.S. hospital and ambulatory-care landscape, access failure is no longer just a scheduling issue. Patients face different bottlenecks depending on service line, market, payer, and acuity: routine specialty visits back up in high-demand metros, Medicaid patients encounter provider-acceptance barriers, and people with serious mental illness can hit a complete capacity wall when beds or treatment slots do not exist locally.
The recommendation
Treat access as an integrated capacity portfolio, not a single wait-time metric. The recommended operating model is to separate delay, payer acceptance, and hard capacity deserts, then assign owners for psychiatric capacity, high-wait specialties, Medicaid network performance, and market-level demand management.
How long you wait
From the 31-day headline and its 20-year climb to which specialties and metros drive it, and the extreme tail.
Average new-patient wait, survey over survey (2004 to 2025)
Specialty count rose from 4 to 6 over time, which affects strict comparability. Gastroenterology, the longest-wait specialty, was added in 2025.
Read it this way The line only climbs, and 2022-to-2025 is its steepest segment, so the 31-day average is part of a worsening multi-year trend, not a plateau. Because the survey added gastroenterology as a sixth specialty in 2025, part of that final jump reflects a change in what's being measured, not only longer waits for the same mix of doctors. Use this chart to identify which access bottleneck it is showing, delay, payer acceptance, geography, or capacity, and to decide which part of the recommended access portfolio should own the fix.
⊞ data table⬇ CSV
| Survey year | Average wait (days) | Specialties counted |
|---|---|---|
| 2004 | 20.9 | 4 |
| 2009 | 20.5 | 5 |
| 2014 | 18.5 | 5 |
| 2017 | 24.1 | 5 |
| 2022 | 26 | 5 |
| 2025 | 31 | 6 |
AMN Healthcare (Merritt Hawkins), 2025 Survey of Physician Appointment Wait Times · 2025 · source
Average days to a new-patient appointment by specialty (2025)
Across 15 major metros. Reference line is the six-specialty average of 31 days.
Read it this way Bars above the 31-day reference line, OB/GYN, gastroenterology, dermatology, and cardiology, are pulling the average up, while orthopedic surgery and family medicine sit well below it. The table shows most specialties got worse since 2022, but orthopedic surgery actually improved, from 16.9 to 12.0 days. Use this chart to identify which access bottleneck it is showing, delay, payer acceptance, geography, or capacity, and to decide which part of the recommended access portfolio should own the fix.
⊞ data table⬇ CSV
| Specialty | Wait 2025 (days) | Wait 2022 (days) |
|---|---|---|
| Obstetrics-Gynecology | 41.8 | 31.4 |
| Gastroenterology | 40 | n/a |
| Dermatology | 36.5 | 34.5 |
| Cardiology | 32.7 | 26.6 |
| Family Medicine | 23.5 | 20.6 |
| Orthopedic Surgery | 12 | 16.9 |
AMN Healthcare (Merritt Hawkins), 2025 Survey of Physician Appointment Wait Times · 2025 · source
How much the wait varies by metro, within each specialty (2025)
The box spans the lowest to highest of the 15 metro values for that specialty. The dot marks the metro median.
Read it this way The width of each box shows how much a patient's wait depends on which of the 15 metros they live in, not just their specialty: dermatology spans 6 to 103 days. A tight box, like orthopedic surgery's 4-to-22-day range, means the wait is fairly predictable regardless of metro. A wide one means metro matters as much as specialty. Use this chart to identify which access bottleneck it is showing, delay, payer acceptance, geography, or capacity, and to decide which part of the recommended access portfolio should own the fix.
⊞ data table⬇ CSV
| Specialty | Lowest metro (days) | Median (days) | Highest metro (days) |
|---|---|---|---|
| Obstetrics-Gynecology | 13 | 42 | 84 |
| Gastroenterology | 7 | 40 | 86 |
| Cardiology | 15 | 24 | 72 |
| Dermatology | 6 | 21 | 103 |
| Family Medicine | 4 | 15 | 69 |
| Orthopedic Surgery | 4 | 12 | 22 |
AMN Healthcare (Merritt Hawkins), 2025 Survey of Physician Appointment Wait Times · 2025 · source
Average new-patient wait by metro (2025)
Composite across all six surveyed specialties. Reference line is the national average of 31 days.
Read it this way Metros to the right of the reference line, led by Boston at 65 days and Portland at 57, wait roughly double the national average, while Atlanta and New York sit near a third of it. This is a composite across six specialties per metro, so it can't show whether one specific specialty is driving any single metro's position. Use this chart to identify which access bottleneck it is showing, delay, payer acceptance, geography, or capacity, and to decide which part of the recommended access portfolio should own the fix.
⊞ data table⬇ CSV
| Metro | Average wait 2025 (days) |
|---|---|
| Boston | 65 |
| Portland | 57 |
| Seattle | 45 |
| Minneapolis | 38 |
| Detroit | 38 |
| Philadelphia | 36 |
| San Diego | 30.5 |
| Dallas | 26 |
| Denver | 24 |
| Los Angeles | 23 |
| Washington, D.C. | 22 |
| Miami | 21 |
| Houston | 17 |
| New York | 13 |
| Atlanta | 12 |
AMN Healthcare (Merritt Hawkins), 2025 Survey of Physician Appointment Wait Times · 2025 · source
The extreme tail: longest single-specialty waits recorded (2025)
Longest single new-patient wait recorded for one specialty in one metro.
Read it this way These four cases, up to 291 days for dermatology in Portland, are the single longest waits recorded anywhere in the survey, not typical waits for those metro-specialty pairs. They show how bad the worst case can get, but say nothing about how often waits that extreme actually occur. Use this chart to identify which access bottleneck it is showing, delay, payer acceptance, geography, or capacity, and to decide which part of the recommended access portfolio should own the fix.
⊞ data table⬇ CSV
| Metro and specialty | Longest wait (days) |
|---|---|
| Dermatology in Portland | 291 |
| OB/GYN in Boston | 231 |
| Gastroenterology in Detroit | 208 |
| Cardiology in Washington, D.C. | 175 |
AMN Healthcare (Merritt Hawkins), 2025 Survey of Physician Appointment Wait Times · 2025 · source
Whether you're seen at all
Access denied, not just delayed. The Medicaid acceptance cliff and the mental-health treatment gap, including psychiatric bed deserts.
New-patient acceptance: Medicare vs Medicaid by specialty (2025)
Share of offices accepting new patients with that coverage. Medicaid trails Medicare in every specialty.
Read it this way In every specialty the Medicaid bar is shorter than the Medicare bar, and the gap is widest in dermatology (75% vs 26%). This measures new-patient acceptance, not actual wait times or care quality once a patient is seen, so it can't tell you how the visit itself compares. Use this chart to identify which access bottleneck it is showing, delay, payer acceptance, geography, or capacity, and to decide which part of the recommended access portfolio should own the fix.
⊞ data table⬇ CSV
| Specialty | Medicare accepting (%) | Medicaid accepting (%) |
|---|---|---|
| Cardiology | 91 | 72 |
| Gastroenterology | 85 | 57 |
| Family Medicine | 84 | 55 |
| Obstetrics-Gynecology | 76 | 62 |
| Orthopedic Surgery | 78 | 46 |
| Dermatology | 75 | 26 |
AMN Healthcare (Merritt Hawkins), 2025 Survey of Physician Appointment Wait Times · 2025 · source
Medicaid acceptance by metro (2025)
Share of offices in each metro accepting new Medicaid patients. Reference line is the national average of 53%.
Read it this way Metros left of the 53% reference line, especially New York (28%), Dallas (30%), and Houston (31%), are where a Medicaid card is least likely to get a new patient in the door, while Detroit and Boston sit well above average. This is a composite across specialties, so it can mask a specialty like dermatology, where Medicaid acceptance runs even lower nationally. Use this chart to identify which access bottleneck it is showing, delay, payer acceptance, geography, or capacity, and to decide which part of the recommended access portfolio should own the fix.
⊞ data table⬇ CSV
| Metro | Medicaid acceptance (%) |
|---|---|
| Detroit | 85 |
| Boston | 80 |
| Minneapolis | 75 |
| Denver | 61 |
| Seattle | 61 |
| Los Angeles | 58 |
| Portland | 57 |
| Atlanta | 51 |
| Philadelphia | 49 |
| San Diego | 45 |
| Washington, D.C. | 42 |
| Miami | 42 |
| Houston | 31 |
| Dallas | 30 |
| New York | 28 |
AMN Healthcare (Merritt Hawkins), 2025 Survey of Physician Appointment Wait Times · 2025 · source
Adults with serious mental illness who received no treatment, 2019 to 2022
NSDUH. Share of adults with serious mental illness who got no mental health treatment in the past year.
Read it this way The share is essentially flat across four years, easing only slightly to 33.3% by 2022, so roughly a third of adults with serious mental illness go untreated in any given year with no clear improvement. The line alone can't say why: cost, provider shortage, and stigma can all contribute, and this chart can't separate them. Use this chart to identify which access bottleneck it is showing, delay, payer acceptance, geography, or capacity, and to decide which part of the recommended access portfolio should own the fix.
⊞ data table⬇ CSV
| Year | No treatment (%) |
|---|---|
| 2019 | 34.5 |
| 2020 | 35.5 |
| 2021 | 34.6 |
| 2022 | 33.3 |
SAMHSA, Results from the 2022 National Survey on Drug Use and Health · 2022 · source
Mental-health treatment received, by insurance coverage (2022)
Share of adults with any mental illness who received treatment. Uninsured adults lag sharply.
Read it this way The three insured categories cluster tightly around 53 to 55%, but the uninsured bar drops to 29.4%, roughly half. That gap licenses the conclusion that having any coverage matters far more for getting mental-health treatment than which type of coverage a person has. Use this chart to identify which access bottleneck it is showing, delay, payer acceptance, geography, or capacity, and to decide which part of the recommended access portfolio should own the fix.
⊞ data table⬇ CSV
| Coverage | Received treatment (%) |
|---|---|
| Medicaid or CHIP | 55.1 |
| Private | 53.5 |
| Other (Medicare, military, other) | 53.4 |
| Uninsured | 29.4 |
SAMHSA, 2022 NSDUH Detailed Tables (Table 6.22B) · 2022 · source
Geography
The same question, state by state and then county by county. Pick a state in the filter above to drill into its counties.
Frequent mental distress
County · modeled prevalence (95% CI)Each tile is a state. Pick a state in the Scope control above to drill into its counties.
CDC PLACES (model-based small-area estimates) · 2024 · source
Primary-care physicians per 100k
County · direct countEach tile is a state. Pick a state in the Scope control above to drill into its counties.
HRSA Area Health Resources File · 2023 · source
Why this matters
The wait varies as much by geography as by specialty: dermatology alone spans 6 to 103 days across the 15 surveyed metros, and the single longest recorded wait, 291 days for dermatology in Portland, shows how extreme the worst case can get even though it isn't typical. Coverage compounds the delay: Medicaid acceptance trails Medicare in every specialty, most sharply in dermatology (26% vs 75%), and mental-health treatment rates for insured adults (53 to 55%) run nearly double the uninsured rate (29.4%). Leadership should care because provider capacity and payer acceptance are two separate mechanisms, and fixing only one will not move the 31-day headline.
Recommended actions
- Target OB/GYN, gastroenterology, and dermatology capacity first, since they pull the 31-day national average up the most.
- Prioritize the highest-wait metros (Boston, Portland, Seattle) for provider recruitment or telehealth expansion, given waits running roughly double the national average.
- Monitor Medicaid acceptance by specialty as a leading indicator, since dermatology's 26% Medicaid acceptance rate signals coverage, not just provider count, is limiting access there.
- Treat the 58.9% of counties with zero inpatient psychiatric beds as a distinct, higher-priority gap from routine wait times, since it is an access wall rather than a delay.
- Track the flat 33 to 35% share of adults with serious mental illness receiving no treatment (2019 to 2022) as the KPI deciding whether any of these interventions are working.
The recommendation
Therefore, treat access as an integrated capacity portfolio, not a single wait-time metric. The recommended operating model is to separate delay, payer acceptance, and hard capacity deserts, then assign owners for psychiatric capacity, high-wait specialties, Medicaid network performance, and market-level demand management.
Demographic slice none, the survey is measured at the metro and specialty level.
Sources
- AMN Healthcare (Merritt Hawkins), 2025 Survey of Physician Appointment Wait Times · 2025
- Merritt Hawkins / AMN Healthcare, 2022 Survey of Physician Appointment Wait Times · 2022
- SAMHSA, Results from the 2022 National Survey on Drug Use and Health · 2022
- PLOS Medicine, Inpatient psychiatric bed capacity within CMS-certified US hospitals, 2011 to 2023 · 2025