To analyze how often American in vitro fertilization clinics transfer one embryo at a time, we relied on self-reported data that clinics provide every year to the Centers for Disease Control and Prevention.

The CDC makes this data available through a national summary that’s broken down by clinic, and as raw data.

We focused on a specific data point, known as elective single-embryo transfers. The CDC keeps this data in an attempt to encourage the transfer of one embryo at a time during IVF – which reduces the likelihood of twins, triplets or higher and the attendant health risks for mother and baby associated with multiple gestation.

An elective single-embryo transfer – or eSET in medical parlance – is the practice of using one embryo even when more are available. In other words, we didn’t include cases in which using a single embryo was the only option.

However, there were other complications standing in the way of our attempts to analyze clinic behavior.

While a high eSET rate is considered one measure of a quality clinic, there is no agreed-upon goal for an acceptable rate.

In our attempt to choose one, we consulted a number of experts. Some thought clinics should have an eSET rate of between 80 and 90 percent. Others thought 40 or 50 percent was more realistic. In the U.S., the average eSET rate is 29 percent, while in the United Kingdom, it’s 38 percent, and in Australia, it’s 76 percent.

We decided to use 50 percent as a benchmark. It falls on the realistic side of experts’ scales and offers a clean break: We could separate clinics by those that do single-embryo transfers more than half the time and those that do them less than half the time.

We also focused our analysis specifically on women under 35, because they are considered the most likely to get pregnant and the least likely to need multiple embryos.

The data we analyzed does not indicate patient prognosis, but age and eSET rate are the best available proxy to gauge clinic behavior. The CDC keeps more granular data that would offer a more complete analysis, but it keeps that data secret, saying it wants to save clinics and patients from embarrassment and harm.

The most recent data available is from 2015 and was released this spring. There were a total of 464 fertility clinics in the United States that reported their data to the CDC. However, 20 of those clinics did not report eSET data. About another three dozen did not report any data to the CDC.

Focusing on the 444 clinics that reported their eSET rates, we sorted the clinics into two categories:

  1. Clinics that had done elective single-embryo transfers more than half the time.
  2. Clinics that had done them less than half the time.

We found that nearly 80 percent – 348 clinics – did eSET less than half the time. Additionally, we learned that 68 clinics reported doing zero elective single-embryo transfers, about 1 in 7 clinics.

We then wanted to ensure that the data wasn’t being skewed by small clinics that did few IVF cycles.

So we did similar calculations for only clinics that perform more than 20 cycles annually. After excluding the 154 clinics that did very few cycles, we found that 233 clinics (or 80 percent) did not perform eSET more than half of the time. In this group of higher-volume clinics, 18 (or about 6 percent) did no single-embryo transfers by choice.

Removing the clinics that did very few cycles did not change our findings related to the frequency of multiple-embryo transfers. However, these busy clinics were more likely to do eSET.

Our reporting has found that the system of success rates currently incentivizes transferring more than one embryo at a time when it’s not necessary.

However, a clinic that does not achieve a high eSET rate is not necessarily violating best practices. Physicians say there are a number of valid medical reasons that doctors and patients may decide not to do eSET.

For example, some patients – even young women – may be appropriate candidates for using more than one embryo at a time because they have a challenging diagnosis, have failed IVF in the past or have difficulty producing quality embryos. Additionally, some patients want twins or demand more than one embryo, even after having been counseled on the risks.

And given the limited data the CDC makes public, we cannot account for some of these scenarios in our analysis. That’s why we reported the figures in broad strokes to show industry practices and did not rank individual clinics.

Additionally, industry guidelines at the time the data was collected urged clinics to offer patients a single-embryo transfer, but using two embryos still fell within recommendations. In March, the industry expanded its recommendation of using only one high-quality embryo to all women with a good prognosis, regardless of age.