Product liability & mass torts
Where it started. Causation experts, Rule 702 rulings, and how each court ruled.
Industries
The work started with physicians. The same method — sourced findings, a named identity anchor, nothing unconfirmed — now runs across engineering, finance, environmental science, forensics, intellectual property and regulatory practice.
Coverage by discipline
There is no universal identifier for an expert. Each discipline has a different strongest identifier, and that determines how confidently a finding can be stated.
| Industry | Identity anchor | What the dossier pulls |
|---|---|---|
| Medicine, pharma & life science | NPI — NPPES hard anchor |
Prior rulings, publications, guideline authorship, industry payments, trial investigator roles |
| Academic & scientific, any field | Researcher ID (ORCID) hard anchor |
Publication record, citation counts, retractions, grant funding, co-author networks |
| Finance, securities & damages | CRD — FINRA BrokerCheck hard anchor |
Disclosure and discipline events, filings that name the expert, comparables for valuation |
| Engineering & accident reconstruction | Inventor + assignee medium anchor |
Design-defect literature, recall and complaint history, federal funding, standards work |
| Environmental & toxicology | Researcher ID (ORCID) medium anchor |
Exposure-science literature, enforcement actions, chemical hazard record, funding sources |
| Regulatory affairs & standards | Name + agency weak anchor |
Rulemaking comments, advisory roles, congressional testimony, agency contracts |
A hard anchor is a registry number that resolves to exactly one person. A medium anchor needs one corroborating fact. A weak anchor never links automatically — two independent facts must agree, or it goes to a human.
Why anchors decide everything
An NPI, ORCID or CRD resolves to one person. Two different NPIs mean two different people — always. These findings can be stated plainly.
An inventor name plus an assignee, or an author plus a canonical institution ID, needs one more agreeing fact before the dossier asserts it.
Where no registry exists, a name match is a lead, not a finding. It is labelled unconfirmed, excluded from scoring, and routed to a reviewer.
A real example from our own testing
Searching the case law for one expert’s name returned a 1991 federal case bearing that surname — in which the person was a party to the lawsuit, not the expert witness.
A tool that counts that as testimony history is wrong in a way you would only discover in a deposition. Uncommon Context attaches the name, marks it unconfirmed, scores it zero, and tells you a human has to look.
The same rule killed three other false findings during development: a grant search that credited 50 NIH projects to the wrong person with the same surname, a funding search that returned a different researcher entirely, and a payments record that could not be attributed because five candidates shared one name.
Beyond MDLs
Where it started. Causation experts, Rule 702 rulings, and how each court ruled.
Damages and valuation experts, checked against filings, disclosure events and their own published models.
Engineering and reconstruction experts, checked against recall data, standards work and the literature they cite.
Exposure and epidemiology experts, checked against enforcement records, chemical hazard data and funding sources.
Coverage depth varies by discipline because the public record does. Where a source does not exist — state licensing boards, deposition transcripts, jury verdicts — the dossier says so instead of implying coverage.