DuFour v. BP

Almost three years on the beach at Pascagoula cleaning crude oil and decontaminating boom, a worker who says it made his asthma worse, two experts, and a court that found neither of them could carry the question from the worker to the jury.

The rule

Some injuries explain themselves. A broken leg after a crash needs no scientist to connect it to the crash. A disease that could have several causes is different, and the court said so at the outset: for “non-readily observable injuries, or those with multiple possible etiologies,” the law has traditionally required expert testimony on causation, because “[a] plaintiff in such a case cannot expect lay factfinders to understand medical causation”. Chemical exposure and asthma sit on that side of the line; the court found the link between exposure and the worsening of a preexisting asthma “more akin to” developing cancer after years of toxic exposure than to breaking a leg after being hit by a car. The expert must then prove two things. The court repeated the rule, in a footnote quoting an earlier spill case, that a plaintiff “must prove both general and specific causation”: that the substance is capable of causing the condition in people generally, and that it caused this plaintiff’s. The framework, the court noted, “derives from Texas law,” and it applied it here as part of the general maritime law the parties agreed governs. The gate the expert must pass is Rule 702 and Daubert, which “provides the analytical framework for determining whether expert testimony is admissible,” under which the trial judge must “determine whether the expert testimony is both reliable and relevant,” and the appeals court reverses the judge’s call only when it is “manifestly erroneous” (DuFour v. BP, slip op. at 5 to 8, 14 to 15).

The facts

The worker spent almost three years after the spill doing shoreline cleanup and boom decontamination in Pascagoula, Mississippi, exposed to crude oil and chemical dispersants by inhalation, airborne contact and direct contact. He alleged the substances caused a chronic worsening of his asthma, and he sued under the medical settlement that followed the spill, which requires a worker bringing such a claim to prove his injuries were “legally caused by” the exposure. He offered two experts on causation: an expert retained for the case, and the pulmonologist who treated him. The district court declined to hold an evidentiary hearing, excluded both, and granted summary judgment to the defendants. On appeal the worker conceded that “[w]ithout an expert,” he “could not prove the central issue in this litigation,” “the level, extent, and duration of his exposure to harmful substances” (slip op. at 1 to 4).

The holding

Affirmed, as to both experts. The retained expert rested on three studies, and the court found that none addressed the worsening of preexisting asthma; the one the fight centered on, which the opinion calls the Lawrence Study, had excluded people with pre-existing asthma, which the worker had, so he was not on common footing with the people it studied, and when self-reported wheezing was taken out of its definition of asthma, the study itself concluded, and the expert admitted, that the results were “largely attenuated.” The district court had found “simply too great of an analytical gap between the data and the conclusions in the Original Report,” and, of the expert’s own say-so, that it “leaves too great of an analytical gap between the data and his causation opinion”; the Fifth Circuit held that was no abuse of discretion, and that the study’s statistical significance did not cure the mismatch. It also noted the benchmark the toxic-tort cases use for the strength of an association, a relative risk above 2.0 as “a common benchmark for reliability under Daubert,” and noted that even if Mississippi law applied instead of maritime law, that state’s high court has held that “a dose-response ratio is needed to indicate the level of exposure” to benzene (slip op. at 14 to 19). The treating pulmonologist fared no better: he had written no report, the worker did not rely on his medical records, and the only evidence of his opinion was his deposition, in which, the district court found, he “does not attempt to relate” the worker’s “purported worsening chronic asthma to any sort of chemical exposure.” The court left for another day whether a treating physician whose opinion was formed for litigation must file a written report (slip op. at 19 to 21). With no admissible expert, the worker “was required to present reliable expert evidence that his asthma was aggravated by the chemicals,” had not, did not contend that more time would change the record, so that “his claim rises or falls on the record presently before us,” and the court held that the district court “properly granted summary judgment to BP.” “For the foregoing reasons, we AFFIRM” (slip op. at 10 to 12, 21).

The dissent

Judge Oldham would not have ended the case. His opening sentence says he is unconvinced the worker “should be poured out of court.” On the science, he reads the study as having “showed a linear relationship between exposure and asthma,” and he answers the majority’s distinction between causing asthma and worsening it in one line: “If X demonstrably and linearly causes Y, then a factfinder could reasonably find that X exacerbates Y.” On the law, he writes that “Rule 702 and Daubert require a ‘fit’ between the expert’s opinion and the relevant dispute in the case,” that the dispute is set by the substantive law, and that “The operative complaint does not tell us if we are supposed to apply general maritime law or state tort law,” a question he illustrates with a hypothetical spider bite in two states with two causation rules. He objects to the district court’s reliance on the Reference Manual on Scientific Evidence, and to striking the report as the expert’s own say-so, which, he writes, “makes as much sense as striking the report for using a disfavored font.” “I respectfully dissent” (slip op. at 22 to 31).

Why it matters

Four things for a worker, or a family, whose case is a refinery release, a plant exposure, a well-site chemical or a spill. First, in these cases the expert is the case, and the report is the expert. A diagnosis from a treating doctor is not a causation opinion unless the doctor says, in writing, that the exposure caused the condition and why; this worker’s own treating pulmonologist was still not a causation witness. Expert explains the gate, and this decision shows what falls at it. Second, general and specific causation are two proofs, not one. The study has to show that the substance can cause this condition, and the expert then has to show that it did so in this person, at the dose this person got; a study of a different condition, or of people unlike the plaintiff, is a gap the court will not let a jury leap. Third, the dose is a fact to be gathered, not assumed: how much, how long, how close, through what route, which is why the first letter in an exposure case asks for the air monitoring, the badge data, the safety data sheets and the medical surveillance records before they are gone, which is Spoliation. Fourth, the dissent is a map. Which law governs the causation question is not a formality; the dissent would have asked whether state tort law, with its own causation rule, applied, and a plaintiff who pleads that question early controls the standard the expert is measured against. The clock on these cases is its own trap, the subject of Repose and Limitations: a disease that shows up late is exactly the case the latent-disease door was written for, and only if the dates are proved.

One caution. This was a maritime case under a settlement agreement, decided on its record, and the worker lost because of what his experts did not do, not because the law doubts that chemicals can hurt lungs. The lesson is about proof. The court said what it needed and did not get, and the next case can bring it.

DuFour v. BP Exploration & Production, Inc., No. 23-60361 (5th Cir. Oct. 27, 2026) (Richman, J., joined by Ramirez, J.; Oldham, J., dissenting; published), from the Southern District of Mississippi, No. 1:19-CV-591. Every quotation above is from the slip opinion at the pages given; the worker and the two experts are named in the opinion and referred to here by role. The statutes of repose and limitations: Tex. Civ. Prac. & Rem. Code § 16.012 (subsection (d)) and § 16.003. This is general information, not legal advice.