YOU ARE HERE: LAT HomeCollections

BOOSTER SHOTS: Oddities, musings and news from the
health world

Is IVF success in your future? If you don't have a crystal ball, this prediction method may help.

January 05, 2011|By Karen Kaplan, Los Angeles Times
  • An egg is prepared for fertilization at an IVF clinic in Pasadena.
An egg is prepared for fertilization at an IVF clinic in Pasadena. (B??atrice de G??a/Los Angeles…)

Before embarking on a medically invasive, expensive and emotionally taxing effort to get pregnant through in vitro fertilization, it sure would be nice to get a good sense of whether it’s likely to work. After all, only about 1 in 4 attempts resulted in a live birth as recently as 2007.

So researchers from England and Scotland scoured data from more than 144,000 IVF cycles in the United Kingdom and looked for factors that might predict which couples stood the best chance of having a baby with assisted reproduction and which faced long-shot odds.  And indeed, a few patterns emerged.

Couples were more likely to have a “successful live birth” (defined as delivering at least one baby that survived for at least one month) if the woman had given birth previously – those who had succeeded with IVF were 58% more likely to succeed in their next IVF attempt, and those who conceived naturally were 19% more likely to succeed with IVF. In addition, couples had better odds of success if the woman used gonadotrophin or hormone replacement (instead of antiestrogens) and if doctors used intracytoplasmic sperm injection, also known as ICSI.

RELATED: Worry can affect IVF success

Couples were 23% less likely to succeed in IVF once the woman reached the age of 35, and the odds continued to fall along with rising age (once the woman reached 45, the chances of success fell by 85%). Other detrimental factors included being infertile for a longer period; having a greater number of prior failed IVF attempts; being on one’s second or third treatment cycle (instead of the first); and using the woman’s own egg instead of a donor egg. Finally, the chances of success were lower when infertility was due to tubal, anovulatory or cervical disease or a male cause.

The model, described in Wednesday's edition of the journal PLoS Medicine, represents the first comprehensive update to the so-called Templeton model, which was published in 1996. The new model had more predictive value (though it’s hard to say how much, since the amount of improvement was reported as an increase in a statistical measure called “area under the curve of receiver operator characteristics”).

The researchers condensed their model into a program designed for clinical use, which is available online at There’s also an app for that – called IVFpredict – for use on iPhones and Android phones.

RELATED: Two embryos aren't twice as nice for fertility patients

Los Angeles Times Articles