Milward v. Rust-Oleum Corp., 2016 U.S. App. LEXIS 7470

Background

Brian Milward, a long-time pipefitter, was exposed to benzene-containing products during his work and was diagnosed with acute promyelocytic leukemia (APL).
The Milwards sued Rust-Oleum alleging benzene exposure caused Brian’s APL; to prevail they needed expert proof of general and specific causation.
The district court previously excluded the Milwards’ general-causation expert; this court reversed as to general causation and remanded to address specific causation.
On remand the Milwards proffered Dr. Sheila Butler (occupational medicine/hematology) as their specific-causation expert, who offered three theories: (1) no safe level of benzene exposure; (2) a relative-risk comparison to epidemiological studies; and (3) a differential diagnosis excluding other causes.
The district court excluded Dr. Butler’s testimony under Rule 702, finding her no-safe-level theory untestable and rejecting her relative-risk and differential-diagnosis opinions because she would not weigh conflicting epidemiological studies and lacked a reliable basis to ‘‘rule in’’ benzene given the high rate of idiopathic APL.
With Dr. Butler excluded and no other specific-causation expert, the court granted summary judgment for Rust-Oleum; the First Circuit affirmed the exclusion and summary judgment.

Issues

Issue	Plaintiff's Argument (Milward)	Defendant's Argument (Rust-Oleum)	Held
Admissibility of Dr. Butler’s relative-risk opinion under Fed. R. Evid. 702	Butler relied on valid epidemiological studies showing elevated risk at exposures below Milward’s level; exclusion was improper	Butler refused to weigh conflicting studies and would not explain why she chose certain studies, so her relative-risk comparison lacked a reliable foundation	Court affirmed exclusion: expert must engage with and justify choice among conflicting studies when using relative-risk methodology; Butler’s refusal rendered her opinion unreliable
Admissibility of Dr. Butler’s differential diagnosis	Differential diagnosis is a reliable method; Butler properly ruled out other causes and ‘‘ruled in’’ benzene	Butler’s ‘‘ruling in’’ of benzene depended on her excluded methodologies; high percentage of idiopathic APL means ruling out other causes is insufficient without a reliable basis to rule in benzene	Court affirmed exclusion: differential diagnosis unreliable here because Butler had no scientifically reliable method to ‘‘rule in’’ benzene given many idiopathic APL cases
Reliance on clinical experience vs. epidemiology	Butler may rely on clinical/pathophysiologic expertise rather than act as an epidemiologist; such experience suffices for admissibility	When methodology requires assessing epidemiological literature, an expert who declines to evaluate conflicting studies cannot rely solely on clinical instinct	Court: experience alone cannot substitute for engagement with the epidemiologic record where the chosen methodology (relative risk) depends on selecting and validating studies
Effect of excluding the expert on summary judgment	Exclusion leaves other evidence sufficient for trial	Without a specific-causation expert, plaintiff cannot meet Massachusetts requirement for expert proof of causation	Court affirmed summary judgment for defendant because expert testimony was required to prove specific causation and none remained admissible

Key Cases Cited

Milward v. Acuity Specialty Prods. Grp., Inc., 639 F.3d 11 (1st Cir. 2011) (prior decision on general-causation expert admissibility)
Daubert v. Merrell Dow Pharm., 509 U.S. 579 (U.S. 1993) (trial judge gatekeeper role under Rule 702)
Kuhn v. Wyeth, Inc., 686 F.3d 618 (8th Cir. 2012) (expert must explain why relied-on studies chosen; methodology must be shown reliable)
Schultz v. Akzo Nobel Paints, LLC, 721 F.3d 426 (7th Cir. 2013) (relative-risk testimony admissible where expert explained why conflicting studies were unreliable)
Norris v. Baxter Healthcare Corp., 397 F.3d 878 (10th Cir. 2005) (failure to address contrary epidemiological evidence can render an opinion unreliable)
Ruggiero v. Warner-Lambert Co., 424 F.3d 249 (2d Cir. 2005) (‘‘ruling in’’ causation must use reliable scientific methods)
Best v. Lowe's Home Ctrs., Inc., 563 F.3d 171 (6th Cir. 2009) (differential diagnosis must reliably rule in potential cause)
Glastetter v. Novartis Pharm. Corp., 252 F.3d 986 (8th Cir. 2001) (same)
Reckis v. Johnson & Johnson, 28 N.E.3d 445 (Mass. 2015) (Massachusetts law requires expert testimony to establish medical causation)