home  /  causation & health effects  /  general causation
causation & health effects · toxic tort science

General causation and epidemiology.

A population-level question decided on a body of literature that is almost never unanimous. What matters is how the whole of it is weighed, not which studies are quoted.

get started

What was the exposure?

Start a conversation with our AI Research Concierge, already scoped to general causation. Pick a starting point, or describe your situation directly.

AI Research Conciergegeneral causation · triage, not a substitute for an expert
Tell me the substance and the disease at issue, and roughly what exposure levels are alleged. I'll help scope what the general causation showing would require. I won't tell you whether causation exists — that is the expert's work.

General causation asks whether a substance can cause a given disease in humans at exposure levels of the kind alleged. It is answered from the scientific literature as a body — epidemiological studies of exposed populations, animal toxicology, mechanistic and genotoxicity data, and any authoritative assessments by bodies that have reviewed the same evidence. The literature is rarely one-directional, so the work is less about locating supportive studies than about explaining defensibly why the body of evidence as a whole supports or does not support the inference, including the studies that cut the other way. Opinions that read as advocacy through selective citation are the ones that draw successful challenges.

mechanisms

What the evidence base consists of.

Different study types answer different parts of the question, and each carries characteristic limitations that an opponent will press.

Epidemiological studies

Cohort, case-control, and cross-sectional studies of exposed human populations — the most directly probative and the most subject to confounding.

Animal toxicology

Controlled dose studies establishing effects and dose-response, with the standing question of interspecies and high-to-low dose extrapolation.

Mechanistic evidence

Genotoxicity, mode of action, and biological plausibility — evidence that the substance does something capable of producing the effect.

Authoritative reviews

Assessments by bodies that have systematically reviewed the same literature, useful as corroboration but not a substitute for the analysis.

Dose-response relationship

Whether effect tracks exposure — among the most persuasive considerations when present, and conspicuous when absent.

Consistency & replication

Whether the association appears across populations, designs, and investigators, or rests on a small number of studies.

methodology

What the evidence shows — and what we examine.

How a general causation analysis is built.

Systematic search protocolA documented, reproducible search strategy — so the analysis cannot be characterised as having started from the conclusion.
Study quality appraisalAssessing design, power, exposure characterisation, and confounding control, applied evenly to studies on both sides.
Structured weight of evidenceReasoning across the considerations the field uses — strength, consistency, dose-response, temporality, plausibility, coherence.
Contrary evidence treatmentExplicit engagement with studies that do not support the inference. Silence about them is the most common avoidable weakness.
what's at stake

What general causation decides.

It is usually dispositive: failure here ends the matter regardless of the individual facts.

whether the case survives at all viability of an entire MDL or class bellwether selection and outcome settlement posture across a docket years of litigation on either side the shape of every downstream opinion

The dose the literature actually addresses.

A frequent basis for exclusion is reliance on studies at exposure levels far above those alleged. General causation established at occupational concentrations does not automatically extend to environmental ones, and the gap has to be argued rather than assumed.

common questions

General causation — practical questions.

What happens when the epidemiology is mixed?

That is the usual condition, and it is what the weight-of-evidence approach exists to handle. The analysis has to explain why the studies supporting the inference are more probative than those that do not — on design quality, exposure characterisation, statistical power, or confounding control — rather than simply counting studies on each side. Mixed literature does not preclude a general causation opinion, but it raises the standard of explanation considerably, and an expert who does not engage with the contrary studies at all will be asked about them under oath.

Can general causation be established for a rare disease with no epidemiology?

Sometimes, and it is a recurring problem because rare diseases are precisely the ones epidemiology struggles to study — insufficient cases for statistical power, and the rarity itself can be the signal. Courts have accepted opinions grounded in toxicology, mechanism, and case evidence where human studies do not exist, particularly where the mechanistic case is strong and coherent. The analysis has to be candid that human epidemiology is absent and explain why the remaining evidence still supports the inference, rather than obscuring the gap.

How much do agency classifications carry?

They carry real persuasive weight and are not determinative, and the distinction matters because the two are frequently conflated in argument. Regulatory and advisory classifications are made for protective purposes, often deliberately conservative, and use decision rules that are not the tort standard — a hazard classification says a substance can cause harm under some conditions, not that it more likely than not caused a particular outcome at a particular dose. They are properly used as corroboration and as evidence that a serious body reviewed the literature, not as a substitute for the causation analysis.

Does the substance need to be the only possible cause?

No, and this is a common misunderstanding of what general causation requires. Most diseases in these matters have multiple known causes, and the existence of others does not defeat the showing — smoking does not preclude an asbestos claim. General causation asks whether the substance is capable of causing the disease. Alternative causes belong to specific causation, where they must be identified and addressed for the individual. Conflating the two is a frequent error in briefing on both sides.

related

Related specialization areas & resources.

Assessing general causation?

Describe the substance and the disease. We will scope the evidence question and connect you with the right expert.

contamination assistanttriage · not a causation opinion
Tell me the substance and the disease at issue, and roughly what exposure levels are alleged. I'll help scope what the general causation showing would require. I won't tell you whether causation exists — that is the expert's work.