UNITED STATES of America, Plaintiff-Appellee, v. Dr. Philip ROBINSON, Defendant-Appellant, 705 F. App'x 458

GRIFFIN, Circuit Judge. In this False Claims Act case, the jury returned a verdict in the government’s favor. Defendant moved for a new trial. The district court denied his motion, and he now challenges that denial. We affirm. I. From January 1, 2007, to January 31, 2012, Dr. Philip Robinson submitted more than 25,000 claims to Medicare seeking reimbursement for optometry services he provided to nursing home patients. Medicare reimbursed more than $1.4 million for those claims. The government then sued Robinson under the False Claims Act, alleging that many of the claims were for medically unnecessary services.

At trial, the government relied on statistical sampling and extrapolation to establish liability and to prove damages. A statistical expert testified that she used computer software to select a random 30-claim sample representative of the larger universe. A medical expert then identified 25 of those 30 claims as medically unnecessary. And the statistical expert testified that she could estimate the government’s damages by multiplying the average overpayment on each sample claim by the total claims.

Because this method produced estimates, the statistical expert testified that she could conclude with a 90% degree of confidence (rather than with certainty) that the government’s damages were between $974,000 and $1.1 million. The jury ultimately concluded that Robinson had submitted 11,085 false claims and awarded the government $419,075.81 in damages. Robinson moved for a new trial, challenging the jury’s damages award. Specifically, he argued that for the jury to conclude he submitted 11,085 false claims, it must have found that only 13 of the 30 sample claims were false.

With fewer than 25 false claims in the sample, he contended, the average overpayment would have differed significantly from the figure the statistical expert used to estimate damages. He emphasized that the government had provided the average amount it thought Medicare had overpaid on each sample claim, but never gave the jury the individual reimbursement amounts. Thus, Robinson asserted, the jury lacked the numbers it needed to duplicate the government’s damages calculation in accordance with its own conclusion that fewer than 25 of the sample claims were false. The district court denied the motion.

It noted that the jury had been instructed— without objection from Robinson—that damages need not be proven with mathematical certainty, but could be based on evidence supporting a reasonable estimate of loss. And it observed that Robinson failed to rebut the government’s expert testimony on damages and presented no damages evidence of his own. The district court concluded that the jury determined the percentage of false claims in the universe—43%—and awarded the government that proportion of the damages it sought: an approach the district court deemed reasonably supported by the evidence.

Robinson now appeals, arguing the district court erred in denying his new trial motion. II. Robinson argues the district court erred for two reasons. He contends that calculating damages through statistical sampling and extrapolation is improper in FCA cases. And he claims that the jury’s damages calculation was unreasonable. We review a district court’s denial of a new trial motion for an abuse of discretion, and reverse only if we have “a definite and firm conviction that the trial court committed a clear error of judgment.” Armisted v. State Farm Mut. Auto. Ins. Co., 675 F.3d 989, 995 (6th Cir. 2012).

As a preliminary matter, we will not consider Robinson’s first argument because he raises it for the first time on appeal. We “review, the case presented to the district court,” not “a better case fashioned after a district court’s unfavorable order.” Barney v. PNC Bank, 714 F.3d 920, 925 (6th Cir. 2013). Robinson raised a similar argument in his

summary judgment motion, in which he challenged the use of statistical sampling and extrapolation to establish FCA liability. But in his new trial motion, he never challenged the validity of using statistical sampling and extrapolation to calculate FCA damages.

Instead, he argued only that “the jury did not know how to calculate a finding for damages from the evidence and produced a damage finding not based upon anytesti-mony in the record.” Turning to Robinson’s second argument, our “review of a jury’s damage award is extremely deferential”; we will not order a new trial “unless the award is contrary to all reason.” Advance Sign Group, LLC v. Optec Displays, Inc., 722 F.3d 778, 787 (6th Cir. 2013). Here, Robinson has not shown that the award was unreasonable.

He simply reasserts the argument he made to the district court: he walks through the statistical expert’s calculation, alleges the government never provided the jury with the individual reimbursement amounts, and contends the jury therefore lacked the .numbers needed to calculate damages in accordance with its own conclusion that fewer than 25 of the sample claims were false. - But Robinson is incorrect; the government provided the jury with the individual reimbursement amounts.

And, in any event, this argument sidesteps the real issue on appeal: whether the district court abused its discretion by concluding. .the jury’s actual damages calculation was reasonable.

Robinson does not address the district court’s finding that “the jury’s award of damages in this case was logically consistent with its determination of liability and corresponded to the mathematical calculations about which evidence was presented.” And he does not attack the jury’s actual calculation other than by stating: “the jury cannot determine a damage finding by simply using a ratio of false sample claims over total sample claims.” Because Robinson provides neither an explanation of his position nor any legal support for it, he has not shown an abuse of discretion. III.

For these reasons, we affirm the district court’s judgment.