Two sets of highly credentialed experts. CVs a mile long. Department chairs, extensive publication records.
Both sides testified on whether a probe could have caused a specific perforation injury. One side said absolutely not. The other said yes, it could.
Neither side cited any literature. In depositions, both said they weren’t aware of any.
A systematic search later turned up six case studies and series — same procedure, same technology, nearly identical perforation. Some even attributed the injury to human error.
That evidence could have tipped the case.
In Episode 22 of the Expert Witness Playbook, Dr. Brian Woodruff talks with Michael Graham, a medical and scientific research consultant with more than twenty years as a medical librarian, about what most attorneys and experts miss when they rely on a PubMed search and call it done.
We cover the evidence pyramid, common misuses of literature in Daubert challenges, why some of the best databases are off-limits to most clinicians, and where AI tools like OpenEvidence and ChatGPT genuinely help — and where they quietly fail, including fabricated citations that look completely real until you try to open them.
If you’re an expert witness, attorney, or LNC building a case on medical evidence, this one is worth your time.
To learn more about Brian Woodruff, go to his personal website, Child Neurology Expert.
