Elena Duncan was charged by indictment with four counts of mail fraud in violation of 18 U.S.C. sec. 1341. A jury found Ms. Duncan guilty on all four counts, and she was sentenced to 52 months' imprisonment. Ms. Duncan now appeals her sentence. For the reasons set forth in the following opinion, we affirm the judgment of the district court.

* BACKGROUND

Ms. Duncan was convicted of a scheme to defraud Medicare and Medicaid through her home health care supply company, Home Care Connection ("HCC"). Ms. Duncan billed the Medicare and Medicaid programs for supplies that patients never ordered, for supplies that patients never received, and for supplies that were not ordered or prescribed by a physician. This appeal focuses on the sentencing phase of the proceedings and Ms. Duncan's overbilling for "skin barriers." A skin barrier is either a bandage or a form of lotion or powder that serves the same purpose as a bandage. The evidence submitted by the Government at trial, which the jury was entitled to accept, established that Ms. Duncan instructed her employees to add skin barriers to all patients' orders, whether or not the patients requested them. The Government introduced evidence through a registered nurse knowledgeable about the Medicare/Medicaid coding system that Medicare and Medicaid only reimburse skin barrier purchases in cases of ostomies, iliostomies or colostomies. Ms. Duncan introduced evidence from the manufacturers of skin barriers that such barriers are also appropriately used in cases of urinary incontinence, osteopetrosis, arthritis, muscular atrophy, varicose veins, diabetes, asthma, decubitus ulcers and intestinal obstruction.

After the jury found Ms. Duncan guilty of four counts of mail fraud, the district court conducted a sentencing hearing. The Government introduced the testimony of Debra LaPosa, a Medicaid fraud investigator, and Lynn Hemberger, a special agent from the United States Department of Health and Human Services, to establish the amount of loss caused by Ms. Duncan's fraud. LaPosa and Hemberger employed a three-step process to calculate the amount HCC overbilled. In the first step, the investigators calculated the amount HCC received for supplies that were never given to patients, for supplies that were not necessary for a patient's medical condition and for supplies that were never prescribed by a physician. In the second step, the investigators calculated the amount of money HCC received from Medicare and Medicaid over and above the amounts reflected in its delivery receipts. In the third step, the investigators identified diagnoses for which the use of skin barriers was not customary, then calculated the amount of money HCC received for dispensing skin barriers to treat these diagnoses. LaPosa, who calculated the loss to Medicaid, conducted the third step by hand after reviewing each of the relevant files. Hemberger, in contrast, calculated the loss to Medicare under the third step using a computer.

Hemberger testified that, in order to prevent double counting any particular overbilling, she subtracted the total amount she calculated under the first and second steps from the computer- generated amount calculated under the third step. She further testified that LaPosa did not have to account for the possibility of double counting in the same manner because she could avoid double counting as she went through the files individually by hand. Hemberger also deducted about $9,500 from her total calculation based on documents provided to her by Ms. Duncan that indicated that certain supplies may have been dispensed pursuant to physicians' orders.1 LaPosa's and Hemberger's calculations indicated that the total loss to Medicare and Medicaid was $358,272.88.

The Government also introduced the testimony of Tiawana Blackwell at the sentencing hearing. Blackwell was a former employee of Opportunity Medical ("OM"), a health care supply company started by Ms. Duncan after she stopped operating HCC. Ms. Duncan had fired Blackwell after Blackwell allegedly had stolen from OM. Blackwell originally had been called by Ms. Duncan at the guilt phase of the trial, but the district court had ruled that her testimony was irrelevant because it related to OM rather than to HCC. At the sentencing hearing, at which she was a government witness, Blackwell testified that, after her trial and before her sentencing, Ms. Duncan had used OM to commit Medicare and Medicaid fraud in much the same way she had used HCC to commit fraud. Blackwell gave investigators the names of three patients whose accounts had been used in OM's overbilling scheme. The investigators confirmed that these patients, in addition to others, had received supplies they did not want or need. The investigators also found forged physicians' orders in these patients' files. Blackwell further testified that Ms. Duncan had paid her $700 to testify falsely at the trial that skin barriers were medically necessary for all patients. Lastly, the Government introduced evidence pertaining to one of HCC's clients, Dolly Bollinger. The evidence indicated that Ms. Duncan had removed supplies from Bollinger's house after she learned that Bollinger had been contacted by an investigator.

The district court accepted the Government's calculation of the total amount of Ms. Duncan's fraud as $358,272.88. In accordance with United States Sentencing Guideline sec. 2F1.1, the court arrived at a base offense level of 15. The court added 2 levels because Ms. Duncan was a leader or supervisor under sec. 3B1.1C. The court also accepted as credible the evidence indicating that Ms. Duncan had continued to engage in Medicare and Medicaid fraud after her jury conviction, that she falsified documents in her patient files once the investigation had begun and that she went to Bollinger's home to remove unnecessary and excessive medical supplies after she learned that Bollinger had been contacted by an investigator. Based on this evidence, the court increased Ms. Duncan's offense level by 2 points for obstruction of justice. Lastly, the court concluded that Ms. Duncan's criminal history category of I did not reflect adequately the seriousness of her crime or the likelihood that she would repeat such conduct. The court stated that Ms. Duncan's post-conviction fraud would have resulted in at least one additional charge of mail fraud, which would have increased her criminal history category to III. The court therefore granted the Government's request for an upward departure and determined that Ms. Duncan ought to be sentenced as having a criminal history category of III and an offense level of 21.2 Based on this determination, the district court sentenced Ms. Duncan to 52 months' imprisonment.

Ms. Duncan first asks that we decide whether the Government's method for calculating the total amount of the fraud, which the district court accepted, overstates the total cost of HCC's fraud. She submits that the third step of the Government's methodology overstates the cost of her fraud in two respects. She first contends that LaPosa's calculation of the amount of the Medicaid fraud impermissibly counted twice some overbillings. She points out that LaPosa did not take the same precautions to avoid double counting as Hemberger did in calculating the loss to Medicare. As a second attack on the amount of loss determination, Ms. Duncan argues that LaPosa and Hemberger improperly included in their calculations billings for skin barriers that were supported by physicians' orders.

In response, the Government first submits that Ms. Duncan has waived her double-counting argument because she did not raise it in the district court. Alternatively, with respect to Ms. Duncan's first argument, the Government contends that, because LaPosa reviewed the files by hand in performing her calculation, she was able to detect and avoid any possible double counting. With respect to Ms. Duncan's second argument, the Government maintains that Ms. Duncan failed to offer evidence that some of HCC's billings for skin barriers were supported by physicians' orders.

We review a district court's calculation of loss pursuant to U.S.S.G. sec. 2F1.1(b)(1) for clear error. See United States v. Gee, 226 F.3d 885, 899 (7th Cir. Sept. 11, 2000). The Guidelines do not require that the district court compute the loss with precision; the court need only make a reasonable estimate of the loss based on the information available. U.S.S.G. sec. 2F1.1, Application Note 9. When a defendant appeals a district court's loss calculation, she must show not only that the calculation was inaccurate, but also that it was "outside the realm of permissible computations." United States v. Hassan, 211 F.3d 380, 383 (7th Cir. 2000) (quoting United States v. Jackson, 25 F.3d 327, 330 (6th Cir. 1994)).

The Government is correct that a specific double-counting objection was never presented to the district court. However, Ms. Duncan did object to the Government's method of calculation under the third step on the ground that each file should have been reviewed individually. The thrust of her objection was that, if the files had been reviewed individually, some of the billings would have been supported by physicians' orders, thereby reducing the total cost of the fraud. Had both investigators reviewed each file individually, as Ms. Duncan argued they should have, they presumably would have also uncovered any instances of double counting. Although the objection was not specifically addressed to the issue of double counting, the district court was certainly aware that Ms. Duncan objected to the Government's calculation under the third step because it was overinclusive. We shall, therefore, turn to Ms. Duncan's specific objections to the calculation accepted by the district court.

Upon examination of the record, we cannot say that the district court erred in accepting LaPosa's calculation of the Medicaid fraud without adjusting it to account for double counting, as Hemberger had done. The district court had before it Hemberger's testimony that LaPosa did not have to make the same adjustments as Hemberger because, rather than rely on a computer, LaPosa had gone through the files individually. Although both parties expended a marginal effort on producing a clear and precise record, we think that the district court was entitled to rely on this difference in methodology to conclude that no double counting had occurred with respect to the Medicaid billings.

We also see no basis for concluding that the district court erred in its evaluation of the possibility that some skin barrier billings were considered fraudulent despite a physician's authorization for their use. The Government introduced evidence before the district court that indicated that skin barriers were not covered for many of the diagnoses for which Ms. Duncan billed for them. The court also heard testimony from HCC employees that Ms. Duncan had told them to add skin barriers to patients' orders, regardless of whether the patients needed them. The only evidence Ms. Duncan offered in rebuttal was literature from skin barrier manufacturers indicating that skin barriers had a wider usage than Medicare and Medicaid allowed. The district court was permitted to credit the testimony offered by the Government over the information provided in the product literature.

We also note that Ms. Duncan turned over some boxes of HCC's files to Hemberger and LaPosa and requested that they inspect them for possible deductions from their loss calculations. Hemberger testified that she examined the files and gave Ms. Duncan "every advantage" she could when it came to physicians' orders that may have supported HCC's reimbursement requests for skin barriers. Sent. Tr. at 107. Indeed, the district court concluded that Ms. Duncan failed to provide the court with "a single instance where a skin barrier included in the Government's calculation was actually medically prescribed." Sent. Tr. at 191. We therefore conclude that the district court did not clearly err in accepting the Government's calculation of the total loss to the Medicare and Medicaid programs.

We now turn to Ms. Duncan's contention that the district court erred in increasing her criminal history category based on evidence that she had engaged in further fraud during the period between her conviction and her sentencing. The district court added three points to Ms. Duncan's criminal history category based on the Government's evidence that she had continued to defraud Medicare and Medicaid during this time.

Ms. Duncan notes that sec. 4A1.1(d) of the Guidelines instructs a sentencing court to increase a defendant's criminal history category when the defendant commits an offense "while under any criminal justice sentence." Section 4A1.1(e) of the Guidelines similarly instructs a sentencing court to increase a defendant's criminal history category if the defendant, inter alia, offends again less than 2 years after her release from prison. However, Ms. Duncan argues, there is no similar increase provided for defendants who have been tried for another offense and not yet sentenced. She relies on our decision in United States v. Connor, 950 F.2d 1267, 1276 (7th Cir. 1991), and contends that we recognized there that Congress rationally has determined that defendants who offend after previously having been sentenced are more dangerous than those who offend before they have been sentenced. Ms. Duncan concludes that, because she had not yet been sentenced at the time she allegedly used OM to commit additional fraud, she ought not be considered as dangerous as other defendants who offend again after having been sentenced.

The starting point of our analysis is U.S.S.G. sec. 4A1.3. That section reads in pertinent part

If reliable information indicates that the criminal history category does not adequately reflect the seriousness of the defendant's past criminal conduct or the likelihood that the defendant will commit other crimes, the court may consider imposing a sentence departing from the otherwise applicable guideline range.

We review a district court's decision to depart upwardly from the Guidelines under sec. 4A1.3 according to a three-part test.3 First, the district court must state adequate grounds for the departure; the adequacy of these grounds is reviewed de novo. Second, the facts cited to support the departure must actually exist in the record; the standard of review under this step is the clear error standard. Third, the degree of the departure must be linked to the structure of the Guidelines; the extent of the district court's departure is reviewed deferentially.

We employed this three-part test and sanctioned a sentencing court's decision to increase a defendant's criminal history category when the defendant has committed offenses similar to the offense of conviction during the time between conviction and sentencing in United States v. Tai, 41 F.3d 1170, 1177 (7th Cir. 1994). In that case, we held that a district court properly added two levels to the defendant's criminal history category when the defendant, convicted of extortion, continued to extort victims from prison prior to sentencing. We stated that, when a defendant commits similar offenses after conviction, it reflects the defendant's disregard for the law and the likelihood that she will commit further offenses in the future. See id. When a district court upwardly departs on this basis, it sufficiently links its departure to the structure of the Guidelines when it analogizes the post-conviction/pre-sentence conduct to a prior sentence and increases the defendant's criminal history category accordingly. See id. at 1176-77.

The district court's approach is in conformity with our holding in Tai. As in Tai, Ms. Duncan engaged in additional fraudulent conduct between the time of her conviction and her sentencing. The district court explicitly stated that, based on this additional conduct, her initial criminal history category did not reflect the seriousness of her crime or the likelihood that she would commit further crimes in the future. There is, moreover, a factual basis for the district court's determination in the record. Blackwell testified that, subsequent to her conviction and prior to her sentencing, Ms. Duncan continued to defraud Medicare and Medicaid through her new company, OM. Although Ms. Duncan argues that Blackwell's testimony is an insufficient basis on which to depart upwardly because Blackwell had reason to lie under oath, the district court found her to be a credible witness. Finally, the district court did link the extent of its departure to the Guidelines. It treated Ms. Duncan's post-conviction/pre-sentence fraud as if it had been a prior sentence and increased her criminal history category accordingly, as suggested by U.S.S.G. sec. 4A1.1A.

Ms. Duncan alternatively relies on U.S.S.G. sec. 4A1.2(a)(4). This section states in pertinent part

(4) Where a defendant has been convicted of an offense, but not yet sentenced, such conviction shall be counted as if it constituted a prior sentence under sec.4A1.1(c) if a sentence resulting from that conviction otherwise would be countable.

If this section were applicable, the district court would have enhanced Ms. Duncan's criminal history category by only one point. However, this provision applies only when the intervening criminal conduct has resulted in a conviction for which the defendant has not yet been sentenced. By its terms, therefore, this provision is not applicable to Ms. Duncan's case because she had not been convicted for the OM fraud. Therefore, unlike the situation posited in this guideline, it could not be reasonably anticipated that, in due course, an additional sentence would be imposed.4

In increasing Ms. Duncan's sentence, the district court acted in conformity with the Guidelines and our case law. We shall not disturb its determination.

We now examine whether the district court clearly erred in adding two levels to Ms. Duncan's base offense level for obstruction of justice.

The district court determined that Ms. Duncan had obstructed justice by offering to pay Blackwell $700 to testify falsely that skin barriers were medically necessary for all patients, by producing false documents to mitigate the loss to the victims of the fraud and by removing medical supplies from Bollinger's home to thwart an ongoing investigation. Ms. Duncan argues that there was an insufficient factual basis for each of the district court's findings and that any allegedly false testimony solicited from Blackwell was immaterial because the district court ruled that Blackwell's trial testimony was irrelevant. The Government responds that the factual basis for the court's conclusions is sufficient and that Blackwell's false testimony was material because, if it had been given at trial, it would have tended to influence an issue under consideration.

A district court's finding that a defendant has obstructed justice within the meaning of U.S.S.G. sec. 3C1.1 is reviewed for clear error. See United States v. Frazier, 213 F.3d 409, 416 (7th Cir. 2000). The Guidelines explicitly state that suborning or attempting to suborn perjury is a form of obstruction for which the two-level increase should be applied, see U.S.S.G. sec. 3C1.1, Application Note 4(b), as is producing false or altered documents, see id. at Application Note 4(c), and concealing evidence, see id. at Application Note 4(d).

There is factual support in the record for each of the district court's determinations regarding the obstruction enhancement. Specifically, Blackwell testified that Ms. Duncan offered her $700 to testify falsely at trial. Despite Ms. Duncan's arguments at trial that Blackwell was inherently untrustworthy and had reason to lie, the district court found her testimony credible. Also, the Government produced evidence that there were prescriptions in certain HCC patients' files that physicians denied having written. Lastly, the court heard evidence that Bollinger asked Ms. Duncan to remove the excess medical supplies from her home, but that Ms. Duncan failed to do so until she learned that Bollinger had been contacted by investigators who wanted to videotape the excess supplies. Based on this evidence, it is difficult to say that the district court clearly erred in concluding that Ms. Duncan suborned perjury from Blackwell, falsified prescriptions in patient files and tried to conceal evidence by removing supplies from Bollinger's home.

As to Ms. Duncan's claim that Blackwell's testimony was not material because the district court ruled that it was irrelevant, the Guidelines define "material" as "tend[ing] to influence or affect the issue under determination." U.S.S.G. sec. 3C1.1, Application Note 6. All that is required for an obstruction enhancement is that the suborned perjury could affect, to some reasonable probability, the outcome of the judicial process; the suborned perjury does not have to succeed in affecting the outcome. See United States v. Buckley, 192 F.3d 708, 710 (7th Cir. 1999), cert. denied, 68 U.S.L.W. 3725 (U.S. May 22, 2000) (No. 99-9189). The testimony that Ms. Duncan asked Blackwell to give certainly could have affected the outcome of her trial. If Blackwell had testified that skin barriers were medically necessary for all patients, as Ms. Duncan had asked her to do, it is possible that Ms. Duncan would not have been held responsible for overbilling for skin barriers.5 It is of no consequence that Blackwell never actually testified falsely. Obstruction of justice took place when Ms. Duncan encouraged Blackwell to testify falsely in an effort to improve her chances of acquittal.

The district court committed no reversible error in the computation of Ms. Duncan's sentence. Accordingly, the judgment of the district court must be affirmed.