Proving the drug caused it

It's decided collectively rather than case by case, usually at a hearing where the court assesses whether the plaintiffs' expert methodology is scientifically reliable enough to reach a jury. That ruling frequently decides the entire litigation — if general causation experts are excluded, thousands of individual cases end at once.

Which is why coordinated litigation exists. No individual claimant could fund this.

What makes it stronger: a clear temporal relationship between starting the drug and the onset of injury; improvement after discontinuation; recurrence if the drug was restarted; the absence of other plausible causes; dose and duration consistent with the known association; and an injury pattern typical of the drug rather than atypical.

What weakens it: significant pre-existing risk factors, other medications with the same association, a long gap between use and onset, and gaps in the treatment record.

They will argue the injury would have occurred anyway. They will argue the drug's benefits outweighed the risk for someone in your condition. And they will argue the prescriber knew the risk and prescribed anyway, which is a distinct doctrine — the learned intermediary rule, under which a manufacturer's duty to warn runs to the physician rather than to you.

That last one is why the prescriber's records and testimony matter so much. Whether the physician was adequately informed, and what they would have done with better information, is often the pivotal question in a warning case.