IVF Clinic Rates

IVF success rates by age, and how far US clinics differ

By Berkcan K., editor · Published 2026-10-24

Nationally, the CDC’s 2022 data show that 49.7% of intended egg retrievals for patients under 35 using their own eggs led to a live birth, falling to 36.3% at 35 to 37, 23.1% at 38 to 40 and 7.7% over 40. Clinic by clinic, the spread inside each band is wide: among the 381 clinics with at least 20 intended retrievals under 35, the middle 80% reported between 34.2% and 61.4%. Age moves the rate more than anything else in the report, but within one band the distance between a clinic near the bottom of the range and one near the top is large enough that it pays to know where a quoted figure sits.

The tables below are computed from the CDC 2022 Final clinic files for the 449 clinics with a page on this site (clinics that reported 2022 cycles and are open or reorganized). The rates themselves are CDC-reported; the percentiles are calculated here across clinics, one clinic counting once regardless of size.

How far apart are clinics in each age band?

Live births per intended egg retrieval, patients using their own eggs, CDC 2022 Final. Only clinics whose rate is printed as a percentage are included; the CDC prints a fraction instead when a band has fewer than 20 intended retrievals.

Age bandNational rate (CDC)Clinics with 20+ intended retrievals10th percentile clinicMedian clinic90th percentile clinic
Under 3549.7%38134.2%49.1%61.4%
35 to 3736.3%34522.5%36.1%47.5%
38 to 4023.1%34511.9%22.7%32.7%
Over 407.7%2942.9%8.2%16.0%

Three patterns stand out. The median clinic sits within a point of the national rate in every band, so the national figure is a fair description of a typical clinic, not just of the largest ones. The spread in percentage points narrows with age, from 27 points between the 10th and 90th percentile under 35 to 13 points over 40. In relative terms it does the opposite: under 35 the 90th-percentile clinic reports less than twice the rate of the 10th, while over 40 it reports more than five times as much (16.0% against 2.9%).

Is the spread the same for new patients?

The CDC also follows patients with no prior ART through every intended retrieval they had in the year, fresh and frozen transfers included. That cumulative rate is the closest the report comes to “if I start here, how often does it end in a birth?”, and it spreads in much the same way.

Age bandNational rate, new patients after all retrievals (CDC)Clinics with a percentage10th percentile clinicMedian clinic90th percentile clinic
Under 3560.8%35644.5%60.5%71.2%
35 to 3749.9%29931.7%49.1%62.2%
38 to 4035.2%26920.0%34.4%47.5%
Over 4013.4%1894.3%12.5%23.8%

The drop in clinic counts tells its own story. Only 189 of the 449 clinics had enough new patients over 40 for the CDC to print a percentage at all. For the rest the cell is a fraction, a dash for a suppressed value, or “no cycles reported”, which is why so many clinic pages have gaps in the oldest band.

Why do clinics in the same band report different rates?

The CDC is direct about this: “Clinics may have different rates because they have different patient populations or ART treatment methods.” It lists patient factors, “their age, the quality of their eggs and sperm, the cause of their infertility, and genetic factors”, alongside the training and experience of clinic and laboratory staff. It also names selection: “Some clinics may be more willing than others to accept patients with low chances of success”, while others “may encourage them to use donor eggs, which can result in higher success rates among some patients.” The CDC adds that it “does not collect information on clinic-specific patient selection practices”, so nothing in the file separates a strong laboratory from a favourable patient mix.

Size matters too. “The more cycles that a clinic carries out, the less the rate is likely to vary,” the CDC notes, and the over-40 band is where clinics are smallest. A clinic with 25 intended retrievals over 40 moves four points for every birth, so the top and bottom of that band include a fair amount of chance. The band is also open-ended: a clinic whose older patients are mostly 41 is not treating the same group as one whose patients are mostly 44.

Do some states run higher than others?

State medians sit close together and every large state spans most of the national range. For patients under 35, among clinics with 20 or more intended retrievals:

With ten or a few dozen clinics per state (the clinic count for every state is in a separate guide), a state median is a description of who reported, not a ranking, and the spread inside each state is far wider than the gap between states. The state pages list every clinic alphabetically with its under-35 rate and the number of retrievals behind it.

How should one clinic’s figure be read against this range?

  1. Start with the band and the count. Find the rate for your age band and check that it is a percentage on 20 or more cycles, not a fraction. The success rate lookup prints the count beside every rate.
  2. Place it in the range. A rate near the median is typical for that band. A rate near the 10th or 90th percentile is a reason to ask about the clinic’s patients and methods, not a verdict either way.
  3. Prefer the cumulative new-patient rate when you are starting treatment, and compare it with the second table rather than the first.
  4. Compare like with like. The side-by-side comparison aligns two or three clinics band by band with their denominators.

None of these figures is a prediction for one person. The CDC’s own wording is that “clinic success rates do not reflect any one patient’s chance of success” and that patients “should talk with a doctor to understand their own chance of success based on their characteristics.”

Sources