United States v. Hakop Gambaryan

654 F. App'x 888

• Court: Court of Appeals for the Ninth Circuit
• Decided: July 5, 2016
• Docket: 15-50258
• Status: Unpublished

Opinion

MEMORANDUM **

Hakop Gambaryan (“Gambaryan”) was convicted after trial of billing Medicare for power wheelchairs he provided to Medicare beneficiaries for whom the devices were not medically necessary, in violation of 18 U.S.C. § 1347(a). At sentencing, the district judge found that over $2.5 million of Gambaryan’s total Medicare billings were fraudulent. The application of the then-applicable Sentencing Guidelines for cases in which the intended loss exceeds $2.5 million, U.S.S.G. § 2Bl.l(b)(l)(J) (2014), along with an additional enhancement we discuss below, id. § 2Bl.l(b)(7), resulted in a Guidelines range of sixty-three to seventy-eight months.

The district judge imposed a sentence of eighty-four months, after considering the factors set out in 18 U.S.C. § 3553(a) and “evaluating] the various kinds of sentences available as well as the Guidelines sentencing range.” While Gambaryan does not challenge the substantive reasonableness of the sentence, he does argue that the district judge erred in calculating the appropriate Guidelines range. Because the Guidelines must be correctly calculated—if they are to be taken into account, as they must be, in determining an appropriate sentence under § 3553(a)—and because we hold that the district judge did not do so, we vacate the sentence and remand for resentencing on an open record.

1. Gambaryan argues that the enhancement was not supported by clear and convincing evidence of the amount of loss. Considering only the loss relating to the instances of fraud for which Gambaryan was convicted would have led to a Guidelines range of six to twelve months. See U.S.S.G. § 2Bl.l(b)(l)(C) (2014). The disputed sentencing enhancements increased this range to sixty-three to seventy-eight months. See id. § 2Bl.l(b)(l)(J). We have held that “facts found in support of Guidelines enhancements that turn out to have a disproportionate impact on the ultimate sentence imposed [must] be established by clear and convincing evidence,” rather than the otherwise applicable preponderance of the evidence standard. United States v. Staten, 466 F.3d 708, 720 (9th Cir. 2006). Indeed, “where a severe sentencing enhancement is imposed on the basis of uncharged or acquitted conduct, due process may require clear and convincing evidence of that conduct.” United States v. Treadwell, 593 F.3d 990, 1000 (9th Cir. 2010). Based on this consideration and others applicable in these circumstances, see id. we hold that the loss enhancement was required to rest on clear and convincing evidence.

This standard has not been met. First, although Gambaryan billed Medicare for $3,370,200, only $2,541,805 of those claims were for power wheelchairs. Nevertheless, the record is unclear as to the manner in which the district judge arrived at the loss calculation of over $2.5 million. Indeed, the PSR calculated the loss as between $1 million and $2.5 million and, when granting bail pending appeal, the district judge observed that he was understandably troubled by the fact that “we could [not] quantify how many of the total claims submitted were fraudulent.” There was an exceedingly narrow margin dividing a sentencing enhancement for intended loss exceeding $1 million and the enhancement for loss exceeding $2.5 million. The $41,805 difference works out to be about eight or nine wheelchairs out of approximately 300, or about three percent of the power wheelchair claims.

More significantly, the evidence at trial was not sufficiently clear and convincing to justify the assumption that nearly the entire amount billed for power wheelchairs was fraudulent. To cite just one of a number of examples, of the 123 doctors whose prescriptions for power wheelchairs Gam-baryan filled, only thirty-four were listed in a Medicare database as having been compromised—meaning that they were “identified as ... fraudulent providers ... [for] billing abusively or having reported that their [National Provider Identification] numbers were stolen.” While this evidence may have been sufficient to support the inference that all of the prescriptions written by these thirty-four compromised doctors were fraudulent, there was con-cededly no evidence that any of the other eighty-nine prescribing doctors were compromised. Significantly, even as to the compromised doctors, the record does not reveal the number of power wheelchair claims that were predicated on prescriptions written by these doctors.

Nor does our holding in United States v. Popov, 742 F.3d 911 (9th Cir. 2014), compel a contrary result. There, it was assumed that all of the claims submitted to Medicare were fraudulent. The defendants argued, however, that, “[b]ecause it is well known that Medicare routinely pays much less than the billed amount, ... the district court should have calculated the intended loss based on the amounts actually paid by Medicare.” Id. at 915. We rejected this argument for reasons upon which we need not dwell here. Id. at 915-16. Unlike Popov, in this case the threshold issue is whether all of the claims submitted. to Medicare for power wheelchairs were fraudulent, and not whether, if they were fraudulent, the intended loss was the amount Medicare paid.

In sum, we have held that a district judge “need not make [his or her] loss calculation with absolute precision [and] need only make a reasonable estimate of the loss based on the available information.” United States v. Zolp, 479 F.3d 715, 719 (9th Cir. 2007) (citing U.S.S.G. § 2B1.1 cmt. n.3(C)). Nevertheless, where the difference between a sentencing enhancement rests on $41,805 out of $2,541,805 in total claims (or approximately eight or nine wheelchairs), and where the available information in the record is insufficient to establish that all (or almost all) the claims filed were fraudulent, we are unable to conclude that the estimated loss was reasonable. We remand on an open record so that, if the Department of Justice wishes to press for the higher enhancement based on a loss of over $2.5 million—rather than an enhancement based on loss exceeding $1 million, which the trial record supports—it will have the opportunity to come forward with additional evidence, perhaps by finally opening the boxes of files Gam-baryan produced pursuant to a subpoena on the first day of trial. Moreover, contrary to Gambaryan’s suggestion, the district judge is not limited to the counts of conviction. Nor do we suggest that the factors set out in 18 U.S.C. § 3553(a), upon which the district judge previously relied, would not warrant a sentence greater than the applicable Guidelines range. Nevertheless, before such a sentence can be imposed, the applicable Guidelines range must be accurately calculated.

2. Gambaryan also argues that increasing his Guidelines range for defrauding a federal health care program more than $1 million violated the Ex Post Facto Clause. Gambaryan’s base offense level was increased by two steps pursuant to U.S.S.G. § 2B1.1(b)(7) (2014). This specific offense characteristic was added to the Sentencing Guidelines in November 2011, after Gambaryan committed the offenses charged in the indictment. See U.S.S.G. app. C vol. Ill amend. 749 (2011). We agree with the Department of Justice that this requires the sentence to be vacated. See Peugh v. United States, — U.S. -, 133 S.Ct. 2072, 2084, 186 L.Ed.2d 84 (2013). Nevertheless, Gambaryan’s success on this issue may possibly be canceled out by United States v. Adebimpe, 819 F.3d 1212 (9th Cir. 2016), which was decided while this appeal was pending and which held that a two-step increase for abusing a position of trust could apply in these circumstances, id. at 1214—an increase the district judge initially rejected.

VACATED AND REMANDED.

** This disposition is not appropriate for publication and is not precedent except as provided by 9th Cir. R. 36-3.