On January 20, 2015, the US Supreme Court rendered its precedent-setting decision in Teva[1] that reversed the Federal Circuit’s practice of reviewing all District Court claim constructions de novo on appeal. Instead the Supreme Court found that some decisions are entitled to deference as a consequence of certain factual findings that require Courts of Appeal to apply a “clear error” standard of review, and consequently, bringing the review of US claims constructions methodology more in line with the Canadian approach.

The phrase within the claim at issue in Teva was “molecular weight of 5 to 8 kilodaltons”, which Sandoz argued was “fatally indefinite” as the claim did not establish which method of calculation should be used. After considering the expert evidence, the District Court concluded the claim was sufficiently definite as the skilled person would understand “molecular weight” to be calculated by “peak average molecular weight”. The Federal Circuit disagreed, finding, on a de novo review, that the term “molecular weight” was indefinite and thus, the patent invalid. In Canada, interference with the construction on appeal would be doubtful, as patents are construed with a mind willing to understand from the perspective of the skilled person[2].

In reversing the Federal Circuit’s decision, the US Supreme Court held that a Court of Appeals must not set aside a District Court’s findings of fact unless it is “clearly erroneous”. While the Supreme Court identified that claims construction is a question of law, sometimes construction may have “evidentiary underpinnings” that require a District Court to resolve subsidiary factual disputes, which like all factual determinations, must be reviewed for clear error[3]. While the Dissent in Teva argued that claims construction does not involve factual findings, the Majority held that this reasoning overlooked instances, as was present in Teva, where parties present competing fact-related claims by different experts, which must be resolved by the court[4].

In describing how to apply a “clear error” standard to claims construction, the Supreme Court held that often times, construction requires only an examination of the claim language without underlying factual disputes, and in these instances, construction amounts solely to a determination of law that a Court of Appeals will review de novo. However, where the District Court must consult extrinsic evidence in order to understand, for example, the science or meaning of a term, the District Court will need to make a subsidiary factual finding, which must be reviewed for clear error on appeal[5]. The Supreme Court confirmed however that after resolving any factual dispute, the ultimate interpretation is still a legal conclusion, that the Court of Appeal must review de novo. It is only where the Court of Appeal seeks to overturn the Judge’s finding on an underlying factual dispute that a “clear error” standard applies.

The decision in Teva brings US claims construction methodology closer in line with the Canadian approach. While the use of expert evidence to understand the terms and underlying science in a claim[6] is routinely used by courts in Canada, construction is purely a question of law[7], reviewable on a “correctness” standard[8]. However, as concluded in Teva, a Canadian judge’s assessment of expert evidence and his or her factual determinations as to the state of the art are considered factual findings, that may only be reversed on appeal in light of a “palpable and overriding error”[9]. Consequently, both Canada and now the US permit judicial deference on review for factual determinations during a claim construction.