Prenatal screening & probability results for Down syndrome

Odds are good that Dice open sourceyou don’t understand your prenatal screening probability results for Down syndrome … and neither does your doctor.

Prenatal screening provides a reassessment of the chances of having a child with Down syndrome. Any prenatal screen has the chance of a false positive or a false negative, no matter the level of stated accuracy. This applies to traditional screening tests, like AFP, quad, and the nuchal translucency-combined test, and it applies to non-invasive prenatal screening (NIPS), like MaterniT21, verifi, Panorama, and Harmony.

Since 2007, professional guidelines have called for all women to be offered prenatal screening and diagnostic testing for Down syndrome. These guidelines, and prenatal genetic testing in general, are premised on respecting a woman’s autonomy by giving her information on which she can decide whether and how to continue her pregnancy or to terminate it. Except, one of the fundamental challenges to respecting a woman’s right to information is that humans, in general, have difficulty understanding probabilities.

Historically poor understanding of risk scores

In the wake of the change in practice guidelines, researchers studied how best to present the probability results of prenatal screening. They did so, because study upon study had reported that both patients and medical professionals have a poor understanding of risk scores.

Going back to 1982, a study found that doctors, nurses, and midwives “have difficulties understanding, communicating, and correctly solving statistical problems.” One of the factors was in the way risk scores were reported and shared with patients.

Probability presentation formats

The researchers considered three types of probability presentation formats:

  • Frequencies, i.e. “one out of every 724 fetuses of women your age will be diagnosed with Down syndrome”
  • 1-inN format, i.e. “the probability of giving birth to a baby with Down syndrome for a woman with normal results is 1:724.”
  • Visual format, i.e. one white dot amongst 724 black dots, with the white dot symbolizing the chance of having a child with Down syndrome

They considered these formats due to several studies reporting challenges in understanding probability results:

  • A study in 1985 found that “[t]he majority of women misunderstood their risk levels even though they received [probability results] in both verbal (e.g., very low) and probability formats (e.g. 1 in 800).”
  • Another study found that “[o]nly 9.5% of undergraduates, 42% of the graduate students, and 46% of the genetic counselors answered” correctly when trying to identify the probability for Down syndrome; in the study it was 0.18% probability, but answers ranged from 0% to 900%.
  • A study of obstetricians, midwives, and patients were asked to identify the correct probability when presented as a percentage or frequency: when presented as a percentage, only 5% of the obstetricians answered correctly; when presented as a frequency, 63% of the obstetricians answered correctly.

Frequency better format, but still problematic

Given the change in recommendations such that all women were to be offered prenatal screening, the researchers wanted to find out which format allowed for the best level of comprehension. They conducted a study of graduate students who were presented a screen result of 1:181 and asked what was the correct probability for having a child with Down syndrome, the answer being 1/181 or 0.55%. Regardless of format, half of the participants couldn’t answer correctly. From their research, presenting the results as a frequency, i.e. one out of every 181 pregnancies would be positive for Down syndrome, had the best level of comprehension, but not by much over a 1-in-format.

What do prenatal screening results mean?

Based on the survey of studies, what prenatal screening results mean will depend on how they are presented, who you ask, and still you may not get the right answer.

In the same professional recommendations, women are to be referred to a genetic counselor after a prenatal screen result, which will have been delivered by their obstetrician. Based on the studies above, four out of ten obstetricians will incorrectly state the probability from a prenatal screen result, even when presented in the format they understand best. And, only four out of ten genetic counselors they are referred to will interpret the probability correctly.

What’s more, with the newest tests, NIPS, most of the labs do not report their test results in a 1-in-N format. Sequenom even touts as though it’s a good thing that it reports a “positive” or “negative,” which tells nothing about the true probability of having a child with Down syndrome.

Remember, prenatal screening for Down syndrome is premised on respecting a woman’s right to know. But odds are good that she will be told the wrong odds after having a prenatal screening test.

Leave a Reply

%d bloggers like this: