Going Further
1. Discuss the regulation of prenatal tests.
Read the following section from the article, which quotes Alberto Gutierrez, the former director of the F.D.A. office that oversees many medical tests:
“These numbers are meaningless,” said Mr. Gutierrez, the former F.D.A. official, after reviewing an advertisement for the Quest Diagnostics QNatal Advanced Test.
The test is advertised as getting positive microdeletion results right 75 percent of the time. But that figure comes from a single study that included nine confirmed cases of microdeletions, for a test that screens for seven such disorders. The company doesn’t specify how the tests perform individually, and declined to provide that data. (In a statement, Quest said its test has “excellent performance.”)
The F.D.A. considered regulating these tests a decade ago, but backed away. If the agency had oversight, Mr. Gutierrez said, Quest would be required to publish a brochure, but “it would not look like this.”
Nonetheless, companies are charging ahead, viewing microdeletions as a major business opportunity — especially if they can persuade more doctors to order them and more insurers to cover them.
Discuss as a class: Should the F.D.A. regulate these tests and their marketing materials? If so, what guidance and rules should the agency provide? If not, why not?
2. False positives and criminal justice.
Law can be framed in terms of positive and negative test results. In the courts, you are innocent (negative — no guilt detected) until proven guilty (positive — guilt was detected). False positives (convicting an innocent defendant) are often seen as worse than false negatives (failing to convict a true criminal). Discuss as a class:
Why do you think the U.S. court system assumes “innocent until proven guilty,” rather than “guilty until proven innocent?”
How do you think the assumption of innocence affects the reliability of court decisions? Does it raise or lower the probability of a false positive (convicting someone who’s actually innocent)?
Sometimes, courts accept evidence based on controversial software that can make errors in identifying criminals. Discuss this episode of “The Daily” about a facial recognition program’s false positive in the Michigan court system. Should this type of software be allowed as evidence in the courts? If so, what false positive rate would be acceptable?
3. Test accuracy and Covid-19. Analyze the benefits and detriments of quicker but slightly-less-accurate rapid Covid tests by reading this article. Discuss as a class:
Rapid tests return results very quickly and can be deployed to test many more people in the population (especially in the event of an outbreak). However, they may produce a higher false negative rate (failing to detect real Covid infections) than slower tests. Is the speed of rapid tests worth the higher rate of false negatives? Why or why not?
Answer Key: Here is a link to this lesson plan with all of the answers included.
¹ Technical Note for Technical Folks: We can calculate P(Positive Test) using the law of total probability:
P(Positive Test) = P(Positive Test | Has Disorder) P(Has Disorder) + P(Positive Test | No Disorder) P(No Disorder)
P(Positive Test) = 0.999(0.00025) + 0.001(0.99975)
P(Positive Test) = 0.0012495
How did we find P(Positive Test | No Disorder)? In the lesson, we stated that Test A had 99.9% reliability. We defined this as meaning that P(Positive Test | Has Disorder) = 0.999 and P(Negative Test | No Disorder) = 0.999. So, we can use the following:
P(Positive Test | No Disorder) = 1 – P(Negative Test | No Disorder) = 1 – 0.999 = 0.001
Want more Lessons of the Day? You can find them all here.
