A federal judge certified four groups of Facebook users who allege Meta obtained protected healthcare information or affected their devices on healthcare providers’ websites or apps. He rejected Meta’s argument that differences among users defeat a class case, while leaving the alleged collection, consent and damages for later decisions.
Facebook users say Meta’s tracking pixel obtained patient-portal activity, treatment details and prescription-site communications; their expert initially examined 7.3 million buttons across 1,844 domains, an October 30 order says. In that order, Judge William H. Orrick of the U.S. District Court for the Northern District of California granted class certification, letting four groups sue as a class, and found Meta’s examples requiring individual proof “de minimis.” He also wrote that Meta’s legal bills for this case swamp any individual’s statutory or actual damages claim. The four classes cover activity from June 17, 2018, to the present. Any company whose defense to a tracking suit rests on user-by-user differences should test it against this order now: Meta’s handful of user-specific examples did not stop certification here. What Meta received, whether users consented and what any claim is worth remain contested.
Meta argued that the proposed groups contain people whose circumstances differ: someone using a device for a family member or friend, for example, or someone whose private browsing or ad blocking could affect matching to a Facebook account. It also disputed whether common evidence can identify the healthcare sites that sent protected information. Orrick held that those examples did not show a significant obstacle to hearing the claims together under the predominance test, which asks whether shared questions outweigh user-by-user ones. His ruling permits the group case to proceed; it does not decide that Meta unlawfully obtained anyone’s information.
The 3 sources behind this story: 2 court records and 1 SEC filing, each linked with what it shows. Free with the weekly Litigation Sentinel brief.
One certified group concerns Facebook users whose patient-portal, appointment, bill-payment or diagnostic-assessment button clicks were allegedly obtained by Meta from healthcare providers on the plaintiffs’ list. A second concerns alleged communications identifying a doctor, condition or treatment, excluding those button-click events. A third concerns alleged communications with a prescription-drug website about a drug the user has been prescribed. The fourth concerns users for whom Meta allegedly placed an _fbp cookie, or used their device to send data to Meta or run Core Setup logic, while they were at a healthcare provider’s site or app.
Those definitions make the proof problem concrete. A button labeled “schedule” might identify a medical appointment; a visit to a healthcare website might instead be general research. Meta argued that websites could change the pixel’s settings, that under its retention policies its oldest classwide data dates only to March 2022, and that users did different things on different sites. Plaintiffs’ experts say the pixel’s settings and sampled records can help establish what happened across the groups. Orrick allowed that contest to proceed, with Meta free to challenge the experts’ work.
The 7.3 million buttons are not 7.3 million injured people. The order describes them as an initial set of buttons associated with 1,844 domains that plaintiffs’ expert examined while developing ways to classify clicks. The expert proposed manual review, labeling patterns and an AI model followed by further manual testing. Meta argued that the expert wants the AI model to reach the scale of up to 650,000 domains at issue.
Free weekly intelligence for executives responsible for litigation exposure: the named decision-makers, dollar exposure, operating consequence, and next development.
Meta’s central objection was that common evidence could not resolve which sites sent protected information, whether users consented, Meta’s knowledge and intent, and damages. Orrick treated many of those questions as disputes that could be tested with evidence about specific sites, the pixel’s settings and Meta’s data. He wrote that Meta’s examples of people who might fall outside the groups were, on the record before him, “de minimis.”
Consent remains a live issue. Meta argued that healthcare sites gave their own disclosures, that users continued to visit after learning of the alleged practice, and that some users shared health information on Facebook themselves. Orrick wrote that if a provider’s disclosures might show consent, that consent “will be determined commonly and objectively” for that provider’s class members.
The judge also left a way to change the case’s size. Evidence about when Meta’s settings operated, which sites were covered or what reached users’ devices could narrow a class period or available damages. The court can narrow or decertify a group as the record develops.
The certified claims include privacy, contract and computer-access theories. Three groups also bring federal and California communications-privacy claims, though Orrick limited the claim under Section 631 of the California Invasion of Privacy Act and redefined the class for the state computer-access claim. P&L translation: Orrick illustrated the per-person stakes with a possible $5,000 statutory claim, expressly assuming such damages are appropriate, and called that sum “by no means insignificant” before weighing it against Meta’s legal bills. The order sets no classwide damages figure, so exposure stays unpriced; what changed is that four classes covering activity since June 17, 2018, now proceed together.
Meta’s quarterly report, its Form 10-Q for the quarter ended September 30, 2026, lists this case among several proposed class actions alleging that it improperly received people’s information from outside websites or apps through its business tools. It also describes the **separate Flo Health pixel case**, *Frasco v. Flo Health, Inc.*, where the plaintiffs seek $5,000 per member for a group they put at up to about 1.25 million people. Those are Flo Health figures, not a member count or damages demand established for this healthcare case.
The Northern District of California docket index lists agreed stipulations filed November 9 and 10, a case-management statement filed November 10 and an order on a stipulation listed November 11. The immediate contest in this case is over the evidence behind the four groups: which providers were covered, which communications reached Meta, whether disclosures amounted to consent and whether the proof holds up at summary judgment or trial. Orrick said competing evidence can be tested at summary judgment or trial and that the classes can be narrowed to fit the proof.
The readiness check has five questions about your reported litigation practices. Use the summary to choose an operating question to explore.
Take the 2-minute readiness check →Free weekly intelligence for executives responsible for litigation exposure: the named decision-makers, dollar exposure, operating consequence, and next development.