In re Zoloft (Sertraline Hydrochloride) Products Liability Litigation
26 F. Supp. 3d 449
E.D. Pa.2014Background
- MDL litigation alleging Zoloft (sertraline, an SSRI) taken during pregnancy causes birth defects; Plaintiffs proffer Dr. Anick Bérard as a general-causation epidemiology expert.
- Defendants (Pfizer/Greenstone) moved to exclude Dr. Bérard under Federal Rule of Evidence 702 and Daubert, challenging her methods (not qualifications).
- Dr. Bérard’s litigation opinion: SSRIs as a class, and sertraline specifically, cause a wide range of congenital malformations and other adverse pregnancy outcomes.
- Court held a Daubert hearing, reviewed her report, peer-reviewed literature, and rebuttal experts; key methodological disputes involved study selection, treatment of non‑statistically significant ‘‘trends,’’ and inference of class effects.
- The court found Dr. Bérard departed from established epidemiologic practice (relying on non‑significant trends and selectively citing studies), failed to adequately address contrary evidence (including her own prior work), and did not validate her alternative methods.
Issues
| Issue | Plaintiff's Argument | Defendant's Argument | Held |
|---|---|---|---|
| Admissibility of expert under Rule 702/Daubert | Bérard is a qualified epidemiologist whose review of the literature supports a causation opinion; trends across studies justify inference despite some non‑significant results | Methods are unreliable: she relied on non‑statistically significant trends, cherry‑picked studies, and failed to follow accepted meta‑analytic techniques | Excluded: court found her methods unreliable and not grounded in generally accepted scientific practices |
| Use of non‑statistically significant trends across studies | Trend analysis across studies can reveal associations when individual studies are underpowered | Reliance on non‑significant trends departs from accepted practice and risks false positives without objective validation | Rejected: court emphasized importance of replicated, statistically significant findings or validated meta‑analytic combining methods |
| Reliance on SSRI class effect to infer sertraline-specific causation | Class‑level associations support inferring similar teratogenic risk for sertraline | SSRIs differ chemically; epidemiology shows inconsistent, drug‑specific results; class inference is unjustified without testing | Rejected: court required drug‑specific evidence; excluded class‑effect opinion and use of class data to infer sertraline causation |
| Study selection and addressing contrary literature | Selected studies and trends represent the totality supporting causation; methodological critiques explain exclusions | Plaintiff cherry‑picked favorable studies, ignored contrary findings (including Bérard’s own prior work), and failed to account for confounding, detection bias, and temporality issues | Rejected: court found selective citation and failure to address contrary evidence undermined reliability; opinion excluded |
Key Cases Cited
- Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579 (1993) (trial court gatekeeper role for scientific expert admissibility)
- In re Paoli R.R. Yard PCB Litig., 35 F.3d 717 (3d Cir. 1994) (methodology, not conclusions, governs reliability under Rule 702)
- In re TMI Litig., 193 F.3d 613 (3d Cir. 1999) (expert qualification and reliability inquiries)
- Wade‑Greaux v. Whitehall Labs., 874 F. Supp. 1441 (D.V.I. 1994) (excluding expert who departed from teratology community’s accepted epidemiologic standards)
- Carnegie Mellon Univ. v. Hoffmann‑LaRoche, Inc., 55 F. Supp. 2d 1024 (N.D. Cal. 1999) (criticizing experts who selectively cite supportive scientific evidence)