court allows some claims based on allegedly misleading statistical claims for pregnancy test: Bayes' Theorem in the courts

In re Natera Prenatal Testing Litig., No. 22-cv-00985-JST, 2023 WL 3370737, -- F. Supp. 3d – (N.D. Cal. Mar. 28, 2023)

Natera sells Panorama, a noninvasive prenatal testing (“NIPT”) product which screens for an array of fetal chromosomal and genetic conditions:

NIPTs are screening tests, not diagnostic tests; while an NIPT can screen patients for a “high risk” of the presence of a particular fetal condition, a patient who receives a positive NIPT result should follow up with diagnostic testing to confirm the presence of that condition.… Diagnostic tests, though more accurate than NIPTs, are also more invasive, are associated with a risk of miscarriage, and must be conducted later in the pregnancy term. In 2020, the American College of Obstetricians and Gynecologists (“ACOG”) changed its guidance to recommend that all pregnant persons “be offered both screening and diagnostic testing options.”

Natera’s advertising touted “fewer false positives and fewer false negatives” and offered patients the ability to “[d]iscover more about your baby’s health.” It included a testimonial from a patient “now planning a birth at a regular maternity ward instead of closer to the children’s hospital thanks to this painless test!” Its website described Panorama as “the most reliable way of non-invasively assessing a baby’s health,” having been tested with “the largest prospective NIPT study” with outcomes of “~90% samples with genetic truth.” It stated that a “high risk” finding for a particular condition “indicates a very high probability that your baby may have [the] condition.” It highlighted Panorama’s positive predictive value (PPV) for Down syndrome of over 90% and describes Panorama as “[n]on-invasive and highly accurate, ... identif[ying] more than 99% of pregnancies affected with Down syndrome [with] the lowest reported false positive rate of any prenatal screening test for the commonly screened chromosomal abnormalities: [Down syndrome], trisomy 18, and trisomy 13.”

The importance of PPV is easy to overlook. (I’m working on a paper with Chris Buccafusco now about this problem in music copyright cases.) If there is any false positive rate at all, and if the base rate of the tested-for condition is low, then a positive result can be much more likely than not to be a false positive, even if the test has a very low false positive rate. And most people find this very difficult to understand—how could a 99% reliable test be wrong more often than not? This is sometimes known as “base rate neglect.” Down syndrome is the best case for PPV because its prevalence in the population is much higher than most of the other tested-for conditions.

As the court explained,

Panorama’s accuracy varies widely across conditions. While Panorama is very effective in screening for Down syndrome, it is much less effective in screening for rare genetic conditions, including those caused by microdeletions. … Natera knew, but did not disclose, that Panorama has a high rate of false positives for microdeletion-related and other rare conditions. In 2016, Natera acknowledged the results of a published study which found that Panorama had an 18% PPV for DiGeorge syndrome, a rare genetic condition. In other words, 82% of Panorama’s positive results for DiGeorge syndrome were false positives. Panorama’s PPVs for other rare conditions – which similarly were not disclosed in marketing materials – are as low as 2-5%, such that up to 98% of positive results for those conditions are false positives.

Yet Natera allegedly advertised Panorama “as reliable overall, prominently emphasizing the accuracy rates of the tests for more common conditions like Down [s]yndrome [which] Panorama can reasonably detect.” Natera’s brochure for consumers didn’t include any information about PPV rates, and the website listed only the PPV for Down syndrome.

The putative class plaintiffs all received Panorama results requiring further evaluation and subsequently underwent subsequent monitoring by medical specialists or invasive diagnostic testing, incurring additional expenses. Each of their Panorama results turned out to be a false positive. They brought the usual California statutory claims, warranty and common-law claims, and claims under the consumer protection laws of other states (Maryland, Illinois, Florida, and New Jersey).

Claims for misrepresentation by partial omission with particularity, plaintiffs needed to identify the allegedly misleading representations that each of them saw and relied upon in deciding to purchase Panorama, but they didn’t.

However, the court rejected the learned intermediary doctrine as applied to the fraud-based claims. “Under California law, failure-to-warn claims brought in products liability actions are subject to the learned intermediary doctrine, which holds that a manufacturer of prescription drugs or certain medical devices satisfies its duty to warn by providing adequate warning to the prescribing physician, rather than the patient.” Natera didn’t identify any California courts that applied this doctrine to consumer protection claims. Even if that was sometimes appropriate, Natera didn’t explain why these claims were “essentially” or “disguised” failure-to-warn claims, as would be required to do so; plaintiffs were arguing that the products’ accuracy was misrepresented, not that they posed undisclosed safety risks. The court reasoned similarly as to other states’ laws.

Did Natera owe a duty to disclose? Under California law, “to be actionable[,] the omission must be contrary to a representation actually made by the defendant, or an omission of a fact the defendant was obliged to disclose.” A plaintiff sufficiently pleads a duty to disclose where: (1) the plaintiff alleges the omission was material; (2) the alleged defect was central to the product’s function; and (3) the defendant (a) is plaintiff’s fiduciary, (b) has “exclusive knowledge” of material facts, (c) “actively conceals” a material fact, or (d) makes misleading partial representations. Plaintiffs failed to identify a relevant “defect” in the tests. In another false advertising case challenging the undisclosed presence of child and slave labor in the supply chain, the Ninth Circuit rejected the existence of any duty to disclose because “the labor practices in question ... are not physical defects that affect the central function of the chocolate products.” Thus, the omission-based California claims were dismissed with leave to amend. (This seems to me to be slicing the salami a bit too fine, but I admit I’m uncomfortable with product liability analogies anyway. The “defect” here is that the product doesn’t provide the information it claims, which is central to its function of providing medical information. If it were a physical monitor that flashed red when it shouldn’t, it seems to me that should count, so why would it be different here? Also, whether something is a “defect” may depend entirely on how it’s advertised: if I advertise paint as paint, it’s not defective, but if I advertise it as a cancer cure it is.)

The court also found that the plaintiffs didn’t need not plead an independent duty to disclose for omission-based claims brought under the Maryland, Illinois, Florida, and New Jersey laws.

Implied warranty of merchantability: this requires that consumer goods, among other things, “are fit for the ordinary purposes for which such goods are used.” “Plaintiffs plausibly allege that the basic function of a test which screens for a specified set of genetic conditions is to accurately detect the risk of such conditions, and that Panorama was unfit to do so.” But there was no privity as required for an implied warranty of merchantability, even though plaintiffs alleged reliance on Natera’s advertising. However, there was a split of district court authority about whether there’s a third-party beneficiary exception to the vertical privity requirement, which the court resolved in plaintiffs’ favor.  Plaintiffs alleged they “are the intended third-party beneficiaries of agreements between Natera and their physicians and health insurers”; that these agreements to use Panorama “were designed and intended for the benefit of Plaintiffs ... to make health care decisions,” and that Natera “understood that Plaintiffs[ ] ... would require that [Panorama] provide reliable and accurate information regarding genetic abnormalities that may affect their pregnancy and the well-being of their bab[ies].”