Guides · Mass tort

General Causation in Mass Torts: Bradford Hill Under Rule 702

In most pharmaceutical and toxic tort litigation, general causation decides the case: without admissible testimony that the product can cause the disease, every individual claim falls. The appellate decisions agree that the Bradford Hill considerations and the weight-of-the-evidence approach are reliable methods. The fights are over how a particular expert applied them. This guide covers what the courts examine, how they police selective use of the literature and dose-response, and what each side's epidemiologist has to show.

Guide 10 of 10 · October 20, 2026 · 8 min read
On this page
  1. The method is
  2. Association first,
  3. Weight of the
  4. Cherry-picking and
  5. Dose-response and
  6. What each
  7. What to prepare

The method is accepted; the application is the case

The Sixth Circuit put the starting point bluntly: "Bradford Hill is undeniably a reliable methodology." In re Onglyza (Saxagliptin) & Kombiglyze (Saxagliptin & Metformin) Products Liability Litigation, 93 F.4th 339 (6th Cir. 2024). The Fourth Circuit describes the considerations as "a series of factors used by epidemiologists to determine whether an observed association between two variables is causal," including temporal relationship, strength of the association, dose-response relationship, replication, biological plausibility, consideration of alternative explanations, cessation of exposure, specificity, and consistency with other knowledge. In re Lipitor (Atorvastatin Calcium) Marketing, Sales Practices & Products Liability Litigation, 892 F.3d 624 (4th Cir. 2018). It added that "[n]o single criterion is dispositive." The Second Circuit, in the acetaminophen litigation, said the same: "No single Bradford Hill factor is required to infer causation." Rutledge v. Walgreen Co., No. 24-916 (2d Cir. July 13, 2026).

That acceptance does not end the inquiry. The Third Circuit explained that "[f]lexible methodologies, such as the 'weight of the evidence,' can be implemented in multiple ways; despite the fact that the methodology is generally reliable, each application is distinct and should be analyzed for reliability." In re Zoloft (Sertraline Hydrochloride) Products Liability Litigation, 858 F.3d 787 (3d Cir. 2017). The 2023 amendment to Rule 702(d), which asks whether the opinion "reflects a reliable application" of the method, points courts at exactly this question. Most decisions below applied the earlier text, but they turn on application, where the amendment directs attention.

Association first, then causation

Bradford Hill is a tool for deciding whether an association is causal. It is not a substitute for finding an association. In Lipitor, the Fourth Circuit quoted the Reference Manual on Scientific Evidence that it is proper to employ the considerations "only after a study finds an association to determine whether that association reflects a true causal relationship." The district court there excluded the plaintiffs' general causation opinion at the lowest dose because the association found at the first step was too weak to support the second.

Statistical significance is part of that first step, but courts resist making it a mechanical gate. The Third Circuit in Zoloft declined "to state a bright-line rule" and cautioned that "[d]iscussions of statistical significance should thus not understate or overstate its importance." It also called significance "an important metric to distinguish between results supporting a true association and those resulting from mere chance." The plaintiffs' expert in Zoloft lost in part because, as the court put it, "[b]y applying different techniques to subsets of the data and inconsistently discussing statistical significance, [the expert] does not reliably analyze the weight of the evidence."

Weight of the evidence: synthesis, not a pile of studies

The weight-of-the-evidence approach asks the expert to reason across several lines of evidence: epidemiology, toxicology, animal studies, and mechanism. The leading defense of the approach is the First Circuit's decision in Milward v. Acuity Specialty Products Group, Inc., 639 F.3d 11 (1st Cir. 2011), which reversed the exclusion of a benzene causation opinion. The court held that "[n]o serious argument can be made that the weight of the evidence approach is inherently unreliable. Rather, admissibility must turn on the particular facts of the case." It described "[t]he hallmark of the weight of the evidence approach" as "reasoning to the best explanation for all of the available evidence," and faulted the district court for treating "the separate evidentiary components of [the expert's] analysis atomistically."

Courts after Milward have insisted that the synthesis itself be disciplined. The Third Circuit in Zoloft held that "the 'techniques' used to implement the analysis must be 1) reliable and 2) reliably applied," and that an expert "must explain 1) how conclusions are drawn for each Bradford Hill criterion and 2) how the criteria are weighed relative to one another." Rutledge applied the same idea from the other direction: the Second Circuit affirmed exclusion of one of the plaintiffs' experts because the district court was entitled to conclude that the opinion "was unreliable because it did not weigh or properly synthesize the factors under the so-called Bradford Hill methodology that he employed."

Checking off the considerations is not a weight-of-the-evidence analysis. The weighting has to be visible, explained, and applied evenly.

Cherry-picking and inconsistent standards

Selective use of the literature is the most common ground for exclusion. In Onglyza, the plaintiffs' expert, addressing the analogy consideration, looked to a different class of diabetes drugs rather than the class to which the product belonged, because the chosen drugs had been shown to worsen heart failure. The Sixth Circuit held that "[t]his cherry-picking 'undermines principles of the scientific method and is a quintessential example of applying methodologies (valid or otherwise) in an unreliable fashion,'" quoting the Fourth Circuit's Lipitor decision. It summarized three independent reasons for exclusion: "(1) improper reliance on SAVOR to the exclusion of all other human studies, (2) unqualified analysis of animal studies, and (3) cherry-picking and inconsistent consideration of the Bradford Hill factors."

Zoloft shows the same problem in statistical form. The court found that the expert understated the importance of a larger study that found no significant association and should have "sufficiently discredit[ed] other studies that found no association or a negative association with much more precise confidence intervals." The lesson for both sides is symmetry: whatever standard the expert applies to a study that helps must be applied to the studies that hurt.

Not every methodological choice is cherry-picking, though. In Rutledge, the district court excluded the plaintiffs' lead epidemiologist for performing a single Bradford Hill analysis across two related neurodevelopmental conditions and including studies with symptom-based endpoints. The Second Circuit reversed, holding that "[t]hat ruling was erroneous because it penalized [the expert] for using a methodology that epidemiologists routinely use." The question is whether the field does it, not whether the court would have done it differently.

Dose-response and exposure thresholds

Biological gradient is one of the considerations, and dose often becomes a separate requirement. In Lipitor, the Fourth Circuit endorsed the principle that "the dose makes the poison" and affirmed the district court's demand for a dose-by-dose analysis. It did not adopt a universal rule: "We do not suggest that every case involving a claim of injury resulting from a pharmaceutical drug will require a dose-by-dose analysis, and an expert witness will not necessarily need to define the precise lower bound of exposure risk." But where "each plaintiff took one of only several commercially available doses, clinical data exist that enable an expert to perform a causation analysis at each dose, and experts (including plaintiffs' own) acknowledge that there is some relationship between dosage and harm," the court may require one.

Rutledge again shows both outcomes in one appeal. The Second Circuit affirmed exclusion of another of the plaintiffs' experts because the district court was entitled to conclude that the opinion "did not reliably address the dose and duration exposure threshold for risk." Measured pharmaceutical exposure invites dose-specific opinions; diffuse environmental exposure may not allow them, and the expert should say which situation applies and why.

What each side's epidemiologist must show

The plaintiffs' epidemiologist carries the burden. In practice that means identifying the relevant body of literature with a stated search and inclusion method; showing an association before applying Bradford Hill; addressing chance, bias, and confounding; explaining how each consideration was assessed and how they were weighed against each other; confronting null and contrary studies on the same terms as supportive ones; and tying the opinion to the doses and exposure durations at issue. The Supreme Court's warning in General Electric Co. v. Joiner, 522 U.S. 136 (1997), frames the risk: nothing in Daubert or the rules "requires a district court to admit opinion evidence that is connected to existing data only by the ipse dixit of the expert," and "[a] court may conclude that there is simply too great an analytical gap between the data and the opinion proffered."

The defense epidemiologist has no burden of proof but faces the same scrutiny of method. In Onglyza, the court noted that the defendants' experts used the same Bradford Hill framework. A defense expert who applies the considerations more strictly than the literature supports, or who dismisses every positive study while accepting every null one, invites the same cherry-picking critique. The strongest defense opinions explain why the association is not established at the first step, or why specific considerations fail, using the same rules the plaintiffs' expert was obliged to follow.

What to prepare

  • Document the literature search, inclusion criteria, and every study excluded, with the reason.
  • Show the association at the first step, with its strength, confidence interval, and treatment of chance, before any causal analysis.
  • Write out how each Bradford Hill consideration was assessed and how the considerations were weighed against one another.
  • Apply the same quality and significance standards to supportive and contrary studies, and explain any departure.
  • For analogy and mechanism, use the product's own class or justify the comparison in terms the field accepts.
  • Address dose and duration directly, and state whether the exposure data permit a dose-specific opinion.
  • When attacking, identify the specific step where the method was not followed rather than disagreeing with the conclusion.
  • Confirm that methodological choices challenged as unusual are ones epidemiologists actually use, with published examples.

Supreme Mind builds Expert Witness AI Agents for litigation, for any expert on any matter, from case intake to settlement. For a general causation dispute, a lawyer can run the plaintiffs' or the defense epidemiologist as a class of expert, and one run returns a brief in four sections: the expert's likely opinion, ranked cross-examination weaknesses, the Daubert / Rule 702 attack surface, and a directional settlement range. Experts are de-identified composites of a class, never a named individual; supporting quotations are checked verbatim against the public record, and nothing it produces is filed or testifies. A complete sample brief is free, and the method explains how it works.

Sources

This guide is general information, not legal advice.

How each court of appeals applies amended Rule 702: see the Rule 702 tracker.

Run this analysis against the expert on your own matter.Start freeRead a sample brief →

Experts in this guide's practice

General-causation epidemiologistSpecific-causation physicianProduct design engineerToxicologist (dose-response)Warnings and human factors expertFDA regulatory expertCumulative exposure causation expert (asbestos)Pharmacovigilance and signal detection expertBiostatistician (meta-analysis)Medical device engineer

The series

10Mass tort · October 20, 2026General Causation in Mass Torts: Bradford Hill Under Rule 70209Securities · October 19, 2026Section 11 Damages and Negative Causation08Antitrust · October 19, 2026Overcharge Regressions in Price-Fixing Cases07Antitrust · October 19, 2026How Often Courts Exclude Antitrust Economists, and Why06Securities · October 13, 2026Loss Causation, Damages and the Inflation Ribbon05Securities · October 6, 2026Market Efficiency at Class Certification: Cammer and Krogman04Any expert · September 29, 2026Amended Rule 702 After Nearly Three Years: What Courts Exclude03Antitrust · September 22, 2026Common Impact and Pass-Through at Class Certification02Securities · September 15, 2026Price Impact at Class Certification After Goldman01Securities · September 8, 2026Daubert Challenges to Event Studies in Securities Cases

Related: the expert agent library, a complete sample brief, and how it works.