Our Results

The graph below shows the live birth plus ongoing pregnancy rates per retrieval for both fresh and frozen transfers in four broad categories of patients treated in our program:

• women under the age of 38 years using their own eggs (2006-08)
• women aged 38-42 years using their own eggs (2006-08)
• women aged 43-46 years using their own eggs (2001-08)
• donor egg recipients of all ages (2006-08)

Underlying Principles

All success rates are fractions, which consist of a numerator and a denominator. For convenience the success rates are reported as percentages, i.e. fractions with a fixed denominator of 100.

Live births are the only outcome (numerator) which matters to consumers. However, in order to include recent results, we have expanded the numerator to include both live births and ongoing pregnancies that have progressed beyond the first trimester (13 weeks). In our experience the vast majority (95%) of such pregnancies end in live births. Thus early "biochemical" pregnancies, miscarriages, ectopic pregnancies and elective pregnancy terminations are all excluded from the numerator.

In departure from the traditional reports compiled by SART and CDC, live births and ongoing pregnancies achieved from one egg retrieval procedure are included in the numerator regardless of whether the pregnancy resulted from transfer of fresh or frozen embryos. If more than one pregnancy has been established from one egg retrieval procedure, only the first pregnancy is included. Live births and ongoing pregnancies initiated during 2006-2008 through transfer of frozen embryos generated from retrievals before January 1, 2006 are excluded.  

Egg retrieval is the defining step of ART and thus the best denominator for reporting ART results. Patients whose treatment is cancelled before the egg retrieval, have not actually undergone ART. For completeness sake, the table below also lists cycle starts as an alternate denominator.

Why do we favor this novel format?

ART treatments produce a huge quantity of data which can be presented in a variety of ways. We aim here to give you a simple but accurate way of assessing your chance of a live birth in our program because such information is most likely to help you choose the right treatment for you.

Traditionally, pregnancies resulting from transfer of fresh embryos have been tabulated separately from pregnancies resulting from frozen transfers even though they both arise from the same retrieval procedure. As our program has evolved, achieving a high pregnancy rate is no longer our only goal. We strive to combine satisfactory live birth rates with avoidance of high order multi-fetal deliveries (The Dilemma of Twins) and complete elimination of severe OHSS (How to Avoid OHSS).

Frozen embryo transfers play a key role in allowing us to achieve these disparate goals. Consequently, a growing number of our patients choose to freeze all their embryos and rely entirely on transfers of frozen-thawed embryos. Therefore, it makes sense to combine the pregnancies from transfers of fresh and frozen embryos generated in the same retrieval cycle. While this novel way of reporting results requires more effort to analyze data than the traditional approach, it provides consumers with more useable information.

Why do we report results for more than one year?

In a relatively small program, such as ours, there is high random variation from year to year due to differences in patient mix and other factors. This variation is further exaggerated by dividing results into 4 to 5 age groups. While such subgroups make sense for reporting national results, for individual clinics it becomes harder to detect important trends and patients often may not “see the forest for the trees.” For instance, the CDC annual report for 2006 includes no fewer than 30 different “success rates” for each one of the 426 reporting clinics.

The Actual Data and Calculations

How do our results compare with other clinics?

We simply do not know. Clinics report results in different ways and their patient populations are also different. While it is widely assumed that outcome differences between clinics are due to different levels of skill, differences in patient selection play a major role.

Both SART and CDC emphatically state that a direct comparison of success rates between individual clinics may not be meaningful and may be misleading because patient medical characteristics and treatment approaches vary from clinic to clinic.

Nevertheless, such comparisons, like other forbidden fruit, are hard to resist, so following are a few helpful hints. When examining results from other clinics, it is advisable to pay close attention to what type of pregnancy is being reported (the numerator of the fraction). In general, overall or clinical pregnancy rates are higher than live birth rates because the former include miscarriages and other unfavorable outcomes. For example, in the 2006 CDC national report for women aged 38-40 the pregnancy rate per cycle start was 27.7% but the live birth rate per cycle start was only 20.6%.

In addition to making sure that the numerators are comparable, the denominators can also vary and include in descending order: cycle starts, retrievals and transfers. In general, live birth rates per cycle start are lower than live birth rates per transfer because the former denominator includes cycles which do not reach retrieval and cycles in which there is no transfer. For example, in the 2006 CDC report for women aged 38-40 years, the percentage of cycle starts which resulted in live birth was 20.6%, the percentage of egg retrievals which led to live birth was 24.1% and the percentage of embryo transfers which resulted in live birth was 26.7%.

Obviously the highest success rate is obtained by combining the largest numerator (any pregnancy) with the smallest denominator (embryo transfer). Conversely the lowest success rate results from using the smallest numerator (live births) with the largest denominator (cycle starts).

National Clinic-specific Reports

SART (Society for Assisted Reproductive Technologies) is a professional society under the aegis of the American Society for Reproductive Medicine which has been reporting clinic-specific results collected by member clinics, such as the Alta Bates IVF Program, since 1989. Thus the results for Alta Bates from 1989 through 2007 are in the public domain.

CDC (Centers for Disease Control and Prevention) is a branch of the federal government which has been publishing U.S. national clinic-specific reports since 1995. The most recent CDC report is for 2006.