Arbay M. Osman & a. v. Wen Lin & a., 147 A.3d 864

Background

Seventeen Somali Bantu refugee children (and three U.S.-born children of Somali Bantu refugees) lived in Manchester apartments in 2005–2006 allegedly contaminated with lead; plaintiffs have elevated blood-lead levels and sued landlords for lead-related neurological injury.
Plaintiffs retained clinical neuropsychologist Peter Isquith, Ph.D., who evaluated 20 children using the RIAS and NEPSY-II and concluded 17 had neurological deficits more likely than not caused by lead exposure.
RIAS and NEPSY-II normative samples were developed for U.S.-born, English-speaking populations; NEPSY-II expressly excluded children for whom English is a second language, and neither test was validated for Somali Bantu refugees.
At a six-day evidentiary hearing, defendants moved to exclude Isquith’s testimony under Rule 702 and RSA 516:29-a (Daubert framework), arguing the tests and Isquith’s application were unreliable for this population and purpose.
The superior court excluded Isquith’s testimony, finding (1) the instruments lacked known validity/reliability for this population and (2) Isquith’s application (including inconsistent cutoff percentiles, reliance on unvalidated cross-cultural interpretation, and failure to control for other risk factors) rendered his methodology unreliable.
On interlocutory appeal, the New Hampshire Supreme Court reviewed whether the trial court unsustainably exercised its discretion in excluding the expert; it affirmed, holding the exclusion was within the trial court’s gatekeeping role.

Issues

Issue	Plaintiff's Argument	Defendant's Argument	Held
Admissibility of Isquith’s expert testimony under RSA 516:29-a / Rule 702	Isquith used accepted assessment tools and sound cross-cultural methods; exclusion oversteps gatekeeper role	Tests not normed/validated for Somali Bantu population; Isquith’s administration and interpretation were unreliable and unvalidated	Affirmed exclusion: trial court reasonably found methodology not reliably applied to facts
Whether use of RIAS/NEPSY-II can establish lead-caused neurodeficits in this population	Valid clinical approach; cautious multi-test pattern method yields reliable conclusions	Instruments were developed for English-speaking U.S. norms; cannot infer neurodeficits or causation without normative baseline	Court: reasonable to conclude tests not validated for this purpose; inferences speculative
Whether lack of peer-reviewed validation for Isquith’s specific interpretive method is fatal	Method reflects clinical practice and accepted principles; should be tested via adversary process	No publications validating his specific cutoff/requirement of two low scores in a domain; methodology is ipse dixit	Court: absence of validation/support and conflicting testimony justified exclusion
Whether inconsistent cutoff thresholds (10th v. 15th percentile) defeated reliability	Cutoff variation was part of cautious, conservative approach	Inconsistency undermines reproducibility and reliability of diagnostic conclusions	Court: inconsistency supported finding that application skewed methodology and warranted exclusion

Key Cases Cited

Daubert v. Merrell Dow Pharm., 509 U.S. 579 (general reliability gatekeeping for scientific expert testimony)
Kumho Tire Co. v. Carmichael, 526 U.S. 137 (gatekeeper may exclude expert methods not shown reasonable for particular application)
State v. Langill, 157 N.H. 77 (New Hampshire adoption of Daubert factors and standard for excluding unreliable expert testimony)
Baxter v. Temple, 157 N.H. 280 (discussion of NEPSY and neuropsychological testing in NH context)
Blue Dane Simmental v. Am. Simmental Ass’n, 178 F.3d 1035 (upholding exclusion where expert’s method lacked industry use and supporting literature)
Baker Valley Lumber v. Ingersoll-Rand, 148 N.H. 609 (application of Daubert framework in NH)