HHS I.G. Audit: 'Medicare Hospice Provider Compliance Audit – Professional Healthcare at Home, LLC'

WASHINGTON, June 19 -- The Health and Human Services Inspector General issued the following audit report (No. A-09-18-03028) entitled "Medicare Hospice Provider Compliance Audit: Professional Healthcare at Home, LLC" filed under the Centers for Medicare and Medicaid Services:

* * *

Here are excerpts:

Why OIG Did This Audit

The Medicare hospice benefit allows providers to claim Medicare reimbursement for hospice services provided to individuals with a life expectancy of 6 months or less who have elected hospice care.

Previous OIG audits and evaluations found that Medicare inappropriately paid for hospice services that did not meet certain Medicare requirements.

Our objective was to determine whether hospice services provided by Professional Healthcare at Home, LLC (Professional Healthcare), complied with Medicare requirements.

How OIG Did This Audit

Our audit covered 3,458 claims for which Professional Healthcare (located in Fairfield, California) received Medicare reimbursement of $20.3 million for hospice services provided from April 1, 2016, through March 31, 2018. We reviewed a random sample of 100 claims. We evaluated compliance with selected Medicare billing requirements and submitted these sampled claims and the associated medical records to an independent medical review contractor to determine whether the services met coverage, medical necessity, and coding requirements.

What OIG Found

Professional Healthcare received Medicare reimbursement for hospice services that did not comply with Medicare requirements. Of the 100 hospice claims in our sample, 79 claims complied with Medicare requirements. However, for the remaining 21 claims, the clinical record did not support the beneficiary's terminal prognosis. In addition, for 1 of these 21 claims, there was no documentation that a hospice physician or hospice nurse practitioner had a required face-to-face encounter with the beneficiary. Improper payment of these claims occurred because Professional Healthcare's policies and procedures were not effective in ensuring that the clinical documentation it maintained supported the terminal illness prognosis. On the basis of our sample results, we estimated that Professional Healthcare received at least $3.3 million in unallowable Medicare reimbursement for hospice services.

What OIG Recommends and Professional Healthcare Comments

We recommend that Professional Healthcare: (1) refund to the Federal Government the portion of the estimated $3.3 million in Medicare overpayments that are within the 4 year claims reopening period; (2) based upon the results of this audit, exercise reasonable diligence to identify, report, and return overpayments in accordance with the 60-day rule; and (3) strengthen its policies and procedures to ensure that hospice services comply with Medicare requirements.

Professional Healthcare, through its attorney, stated that it disputed nearly all of our findings and did not concur with our recommendations. Professional Healthcare disagreed with our determinations for all 21 questioned sampled claims but agreed to return any overpayment for 1 claim for which the beneficiary's clinical record lacked documentation of a required face-to-face encounter. Professional Healthcare stated that our independent medical review contractor erred by consistently relying on only a limited portion of the clinical record to assess the certifying physician's terminal prognosis. In addition, Professional Healthcare's statistical expert challenged the validity of our statistical sampling methodology and the resulting extrapolation.

After reviewing Professional Healthcare's comments, we maintain that our finding and recommendations are valid. We also reviewed Professional Healthcare's statistical expert's comments and maintain that our sampling methodology and extrapolation were statistically valid and resulted in a legally valid and reasonably conservative estimate of the amount that Medicare overpaid to Professional Healthcare.

* * *

TABLE OF CONTENTS

INTRODUCTION ... 1

Why We Did This Audit ... 1

Objective ... 1

Background ... 1

The Medicare Program ... 1

The Medicare Hospice Benefit ... 1

Medicare Requirements To Identify and Return Overpayments ... 3

Professional Healthcare at Home, LLC ... 4

How We Conducted This Audit ... 4

FINDING ... 5

Medicare Requirements ... 5

Terminal Prognosis Not Supported ... 6

RECOMMENDATIONS ... 6

PROFESSIONAL HEALTHCARE COMMENTS AND OFFICE OF INSPECTOR GENERAL RESPONSE ... 7

Nonconcurrence With Recommendations ... 8

Professional Healthcare Comments ... 8

Office of Inspector General Response ... 8

Concerns Related to Audit Process ... 9

Professional Healthcare Comments ... 9

Office of Inspector General Response ... 9

Clinical Judgment and Support for Terminal Prognosis ... 10

Professional Healthcare Comments ... 10

Office of Inspector General Response ... 10

Office of Inspector General Sampling Methodology ... 11

Professional Healthcare Comments ... 11

Office of Inspector General Response ... 11

APPENDICES

A: Audit Scope and Methodology ... 14

B: Related Office of Inspector General Reports ... 16

C: Statistical Sampling Methodology ... 17

D: Sample Results and Estimates ... 18

E: Professional Healthcare Comments ... 19

* * *

INTRODUCTION

WHY WE DID THIS AUDIT

The Medicare hospice benefit allows providers to claim Medicare reimbursement for hospice services provided to individuals with a life expectancy of 6 months or less who have elected hospice care. Previous Office of Inspector General (OIG) audits and evaluations found that Medicare inappropriately paid for hospice services that did not meet certain Medicare requirements./1

OBJECTIVE

Our objective was to determine whether hospice services provided by Professional Healthcare at Home, LLC (Professional Healthcare), complied with Medicare requirements.

BACKGROUND

The Medicare Program

Title XVIII of the Social Security Act (the Act) established the Medicare program, which provides health insurance coverage to people aged 65 and over, people with disabilities, and people with end-stage renal disease. The Centers for Medicare & Medicaid Services (CMS) administers the Medicare program.

Medicare Part A, also known as hospital insurance, provides for the coverage of various types of services, including hospice services./2

CMS contracts with Medicare Administrative Contractors (MACs) to process and pay Medicare hospice claims in four home health and hospice jurisdictions.

The Medicare Hospice Benefit

To be eligible to elect Medicare hospice care, a beneficiary must be entitled to Medicare Part A and certified by a physician as being terminally ill (i.e., as having a medical prognosis with a life expectancy of 6 months or less if the illness runs its normal course)./3

Hospice care is palliative (supportive), rather than curative, and includes, among other things, nursing care, medical social services, hospice aide services, medical supplies, and physician services. The Medicare hospice benefit has four levels of care: (1) routine home care, (2) general inpatient care, (3) inpatient respite care, and (4) continuous home care. Medicare provides an all-inclusive daily payment based on the level of care./4

Beneficiaries eligible for the Medicare hospice benefit may elect hospice care by filing a signed election statement with a hospice./5

Upon election, the hospice assumes the responsibility for medical care of the beneficiary's terminal illness, and the beneficiary waives all rights to Medicare payment for services that are related to the treatment of the terminal condition or related conditions for the duration of the election, except for services provided by the designated hospice directly or under arrangements or services of the beneficiary's attending physician if the physician is not employed by or receiving compensation from the designated hospice./6

The hospice must submit a notice of election (NOE) to its MAC within 5 calendar days after the effective date of election. If the hospice does not submit the NOE to its MAC within the required timeframe, Medicare will not cover and pay for days of hospice care from the effective date of election to the date that the NOE was submitted to the MAC./7

Beneficiaries are entitled to receive hospice care for two 90-day benefit periods, followed by an unlimited number of 60-day benefit periods./8

At the start of the initial 90-day benefit period of care, the hospice must obtain written certification of the beneficiary's terminal illness from the hospice medical director or the physician member of the hospice interdisciplinary group/9 and the beneficiary's attending physician, if any. For subsequent benefit periods, a written certification by only the hospice medical director or the physician member of the hospice interdisciplinary group is required./10

The initial certification and all subsequent recertifications must include a brief narrative explanation of the clinical findings that supports a life expectancy of 6 months or less./11

The written certification may be completed no more than 15 calendar days before the effective date of election or the start of the subsequent benefit period./12

A hospice physician or hospice nurse practitioner must have a face-to-face encounter with each hospice beneficiary whose total stay across all hospices is anticipated to reach a third benefit period./13

The physician or nurse practitioner conducting the face-to-face encounter must gather and document clinical findings to support a life expectancy of 6 months or less./14

Hospice providers must establish and maintain a clinical record for each hospice patient./15

The record must include all services, whether furnished directly or under arrangements made by the hospice. Clinical information and other documentation that support the medical prognosis of a life expectancy of 6 months or less if the terminal illness runs its normal course must be filed in the medical record with the written certification of terminal illness./16

Medicare Requirements To Identify and Return Overpayments

OIG believes that this audit report constitutes credible information of potential overpayments. Upon receiving credible information of potential overpayments, providers must exercise reasonable diligence to identify overpayments (i.e., determine receipt of and quantify any overpayments) during a 6-year lookback period. Providers must report and return any identified overpayments by the later of: (1) 60 days after identifying those overpayments or (2) the date that any corresponding cost report is due (if applicable). This is known as the 60-day rule./17

The 6-year lookback period is not limited by OIG's audit period or restrictions on the Government's ability to reopen claims or cost reports. To report and return overpayments under the 60-day rule, providers can request the reopening of initial claims determinations, submit amended cost reports, or use any other appropriate reporting process./18

Professional Healthcare at Home, LLC

Professional Healthcare, doing business as Kindred Hospice, is a for-profit provider located in Fairfield, California, that furnishes hospice care to beneficiaries who live in California. From April 1, 2016, through March 31, 2018 (audit period), Professional Healthcare provided hospice services to approximately 1,000 beneficiaries and received Medicare reimbursement of about $20.5 million./19

National Government Services, Inc. (NGS), serves as the MAC for Professional Healthcare.

HOW WE CONDUCTED THIS AUDIT

Professional Healthcare received Medicare Part A reimbursement of $20,583,610 for hospice services provided during our audit period, representing 3,680 paid claims. After we excluded 222 claims, totaling $195,935, our audit covered 3,458 claims totaling $20,387,675.20 We reviewed a random sample of 100 of these claims, totaling $602,411, to determine whether hospice services complied with Medicare requirements. Specifically, we evaluated compliance with selected billing requirements and submitted these sampled claims and the associated medical records to an independent medical review contractor to determine whether the services met coverage, medical necessity, and coding requirements.

We conducted this performance audit in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives.

Appendix A describes our audit scope and methodology, Appendix C describes our statistical sampling methodology, and Appendix D contains our sample results and estimates.

* * *

FINDING

Professional Healthcare received Medicare reimbursement for hospice services that did not comply with Medicare requirements. Of the 100 hospice claims in our sample, 79 claims complied with Medicare requirements. However, for the remaining 21 claims, the clinical record did not support the beneficiary's terminal prognosis. In addition, for 1 of these 21 claims, there was no documentation that a hospice physician or hospice nurse practitioner had a required face-to-face encounter with the beneficiary. Improper payment of these claims occurred because Professional Healthcare's policies and procedures were not effective in ensuring that the clinical documentation it maintained supported the terminal illness prognosis.

On the basis of our sample results, we estimated that Professional Healthcare received at least $3.3 million in unallowable Medicare reimbursement for hospice services./21

As of the publication of this report, these overpayments include claims outside of the 4-year reopening period./22

Notwithstanding, Professional Healthcare can request that a Medicare contractor reopen the initial determinations for those claims for the purpose of reporting and returning overpayments under the 60-day rule without being limited by the 4-year reopening period./23

MEDICARE REQUIREMENTS

To be eligible for the Medicare hospice benefit, a beneficiary must be certified as being terminally ill. Beneficiaries are entitled to receive hospice care for two 90-day benefit periods, followed by an unlimited number of 60-day benefit periods. At the start of the initial 90-day benefit period of care, the hospice must obtain written certification of the beneficiary's terminal illness from the hospice medical director or the physician member of the hospice interdisciplinary group and the individual's attending physician, if any. For subsequent benefit periods, a written certification from the hospice medical director or the physician member of the hospice interdisciplinary group is required. Clinical information and other documentation that support the beneficiary's medical prognosis must accompany the physician's certification and be filed in the medical record with the written certification of terminal illness./24

A hospice physician or hospice nurse practitioner must have a face-to-face encounter with each hospice beneficiary whose total stay across all hospices is anticipated to reach a third benefit period. The face-to-face encounter must occur before, but no more than 30 calendar days before, the third benefit period recertification, and every benefit period recertification thereafter, to gather clinical findings to determine continued eligibility for hospice care. The narrative associated with the third benefit period recertification and every subsequent recertification must include an explanation of why the clinical findings of the face-to-face encounter support a life expectancy of 6 months or less. The physician or nurse practitioner who performs the face-to-face encounter must attest in writing that such an encounter occurred./25

TERMINAL PROGNOSIS NOT SUPPORTED

For 21 of the 100 sampled claims, the clinical record provided by Professional Healthcare did not support the associated beneficiary's terminal prognosis. Specifically, the independent medical review contractor determined that the records for these claims did not contain sufficient clinical information and other documentation to support the medical prognosis of a life expectancy of 6 months or less if the terminal illness ran its normal course. In addition, for 1 of these 21 claims, there was no documentation that a hospice physician or hospice nurse practitioner had a required face-to-face encounter with the beneficiary.

RECOMMENDATIONS

We recommend that Professional Healthcare at Home, LLC:

* refund to the Federal Government the portion of the estimated $3,358,906 for hospice services that did not comply with Medicare requirements and that are within the 4-year reopening period;/26

* based upon the results of this audit, exercise reasonable diligence to identify, report, and return any overpayments in accordance with the 60-day rule/27 and identify any of those returned overpayments as having been made in accordance with this recommendation; and

* strengthen its policies and procedures to ensure that hospice services comply with Medicare requirements.

PROFESSIONAL HEALTHCARE COMMENTS AND OFFICE OF INSPECTOR GENERAL RESPONSE

In written comments on our draft report, Professional Healthcare, through its attorney, stated that it disputed all of our findings and did not concur with our recommendations. Professional Healthcare disagreed with our determinations for all 21 sampled claims questioned in our draft report and provided specific responses for each of the 21 claims. However, Professional Healthcare agreed to refund or repay any overpayment for the one claim for which the beneficiary's clinical record lacked documentation of a required face-to-face encounter.

Professional Healthcare stated that courts have recognized that a difference in two physicians' clinical judgments cannot render the certifying physician's judgment invalid. In addition, Professional Healthcare stated that our independent medical review contractor erred by consistently relying on only a limited portion of a patient's clinical record to assess the certifying physician's terminal prognosis, which was based on a full assessment of the patient's complete medical condition. Furthermore, Professional Healthcare stated that our

independent medical review contractor repeatedly found that documentation was insufficient because it did not satisfy Local Coverage Determination (LCD) criteria. Professional Healthcare stated that LCD guidelines are not mandatory, and failure to meet those guidelines cannot support a claim denial.

Professional Healthcare engaged a statistical expert, who analyzed our statistical sampling methodology and, based on that analysis, stated that our methodology is not statistically valid and should not be used as a basis to calculate an extrapolated overpayment. Professional Healthcare's comments are included as Appendix E./28

After reviewing Professional Healthcare's comments, we maintain that our finding and recommendations are valid. We also reviewed the report prepared by Professional

Healthcare's statistical expert and maintain that our statistical sampling methodology and extrapolation were statistically valid and resulted in a legally valid and reasonably conservative estimate of the amount that Medicare overpaid to Professional Healthcare. The following sections summarize Professional Healthcare's comments and our responses.

Professional Healthcare Comments

Professional Healthcare did not concur with our three recommendations as follows:

* Regarding our first recommendation, Professional Healthcare stated that it had been unable to locate documentation for one sampled claim for which the beneficiary's clinical record lacked documentation of a required face-to-face encounter. Professional Healthcare stated that it will refund or repay any overpayment associated with this claim. Nonetheless, Professional Healthcare stated that based on its own clinical review of the beneficiaries' medical records, all 21 sampled claims that OIG found to be improper were supported by the patient's clinical record and billed appropriately. In addition, Professional Healthcare stated that our sampling methodology was not statistically valid and should not be used as a basis to calculate an extrapolated overpayment. Professional Healthcare stated that it intends to vigorously challenge our findings for the 21 sampled claims and any sampling methodology used to calculate and extrapolate overpayments by exercising its rights to appeal any adverse findings through the Medicare administrative appeals process.

* Regarding our second recommendation, Professional Healthcare acknowledged "its legal obligation to exercise reasonable diligence to identify potential overpayments within the preceding six years based on receipt of credible information that an overpayment may exist." However, Professional Healthcare stated that it disagreed with our findings and believes that the sampled claims are supported by the patients' clinical records and were billed appropriately.

* Regarding our third recommendation, Professional Healthcare disagreed that its policies and procedures allowed any systemic issues to occur. Professional Healthcare stated that OIG has not identified any particular policies or procedures that it believes to be lacking or insufficient and that the findings reflect a largely effective compliance program.

Office of Inspector General Response

We clarified in the footnote to our first recommendation that OIG audit recommendations do not represent final determinations by Medicare. Action officials at CMS, acting through a MAC or other contractor, will determine whether a potential overpayment exists and will recoup any overpayments consistent with CMS's policies and procedures. If a disallowance is taken, a provider has the right to appeal the determination that a payment for a claim was improper (42 CFR Sec. 405.904(a)(2)). An overpayment based on extrapolation is re-estimated depending on the result of the appeal.

We maintain that our findings and recommendations are valid and that improper payment of the 21 sampled claims occurred because Professional Healthcare's policies and procedures were not effective in ensuring that the clinical documentation it maintained supported the terminal illness prognosis.

CONCERNS RELATED TO AUDIT PROCESS

Professional Healthcare Comments

Professional Healthcare stated that it has numerous concerns with our audit process.

Professional Healthcare also stated that the draft report did not provide a single reason why Professional Healthcare was selected for audit.

Professional Healthcare stated that it has serious concerns about the qualifications of our independent medical review contractor and that OIG has not provided any substantive information by which Professional Healthcare can assess the contractor. In addition, Professional Healthcare stated that the medical review determinations contain the same vague statement that the reviewer is a physician and holds a board certification. Professional Healthcare stated that without receiving any information about the reviewer, it can assess the reviewer only through his or her individual medical determinations of the sampled claims.

Professional Healthcare stated that our independent medical review contractor repeatedly found that documentation was insufficient either because it did not satisfy LCD criteria or because of the patient's score according to the Advanced Dementia Prognostic Tool (ADEPT).

Professional Healthcare stated that LCD guidelines are not mandatory and that failure to meet those guidelines cannot support a claim denial. Finally, Professional Healthcare stated that the ADEPT score is not even part of the LCD guidelines for patients with Alzheimer's disease or dementia, and it is not an accurate means of predicting a dementia patient's prognosis.

Office of Inspector General Response

We selected Professional Healthcare for a compliance audit through the use of computer matching, data mining, and data analysis techniques that identified hospice claims that were at risk for noncompliance with Medicare billing requirements.

We used an independent medical review contractor that is a licensed physician who specializes in hospice and palliative medicine and is familiar with Medicare hospice guidelines and protocols. Although our independent medical review contractor referenced the ADEPT score in conducting the medical review, the contractor properly used the appropriate statutory and regulatory hospice criteria, as well as applicable LCD guidelines, as the framework for determining terminal status. Specifically, our independent medical review contractor applied standards set out in 42 CFR Sec. 418.22(b)(2), which requires that clinical information and other certification of terminal illness and be filed in the medical record./29

We acknowledge that some beneficiaries who did not meet the guidelines in the hospice LCDs may still be appropriate for hospice care based upon an individual assessment of the beneficiary's health status. Accordingly, our independent medical review contractor merely used LCD guidelines as a tool to evaluate the terminal prognosis. In conclusion, it was the opinion of our contractor that the documentation in the clinical records did not support the terminal prognosis. Therefore, we maintain that our independent medical review contractor consistently and appropriately applied Medicare hospice eligibility requirements when it determined whether the certified terminal prognosis was supported.

CLINICAL JUDGMENT AND SUPPORT FOR TERMINAL PROGNOSIS

Professional Healthcare Comments

Professional Healthcare stated that the findings in our draft report are based entirely on a subjective difference in clinical opinion and that our independent medical review contractor determined in his or her own medical opinion that the portion of the patient's clinical record assessed did not support the terminal prognosis. Professional Healthcare cited several court cases and stated that a difference in clinical judgment cannot render the physician's certification false or invalid for billing purposes.

Professional Healthcare disagreed with our determinations for the 21 sampled claims in our draft report for which our independent medical review contractor found that the associated beneficiaries' clinical records did not support the terminal illness prognosis. Professional Healthcare stated that our contractor consistently failed to apply the appropriate standard for assessing whether the clinical record supported the terminal prognosis. Professional Healthcare also stated that our independent medical review contractor failed to consider all of the relevant factors and information related to the patient's life expectancy and based the findings on a limited "snapshot" portion of the patient's clinical record.

Office of Inspector General Response

As previously mentioned, we used an independent medical review contractor that is a licensed physician who specializes in hospice and palliative medicine and is familiar with Medicare hospice guidelines and protocols. In conducting the medical review, our contractor properly used the appropriate statutory and regulatory hospice criteria, as well as applicable LCD guidelines, as the framework for its determinations. Our contractor acknowledged the physician's terminal diagnosis and evaluated the clinical records provided by the hospice for each sampled claim (including necessary historical clinical records), guided by questions rooted in the Medicare requirements, to determine whether the certified terminal prognosis was supported. When the clinical records and other available clinical information supported the physician's medical prognosis of a life expectancy of 6 months or less if the terminal illness runs its normal course, a determination that hospice eligibility criteria were met was made. In addition, the decisions in the court cases that Professional Healthcare referenced addressed whether a difference in clinical judgment can render a physician certification false for purposes of False Claims Act liability and therefore are inapplicable to OIG audit recommendations and CMS recoveries arising from OIG audits.

Based on our review of Professional Healthcare's comments, we maintain that the clinical records for each of the 21 sampled claims did not support the associated beneficiary's terminal prognosis. For the reasons stated above, we disagree with Professional Healthcare's statement that our independent medical review contractor failed to apply the appropriate standard for assessing whether the clinical record supported the terminal prognosis. We also disagree that our contractor considered only a limited "snapshot" portion of patient records in making determinations on the claims.

OFFICE OF INSPECTOR GENERAL SAMPLING METHODOLOGY

Professional Healthcare Comments

Professional Healthcare challenged the validity of our statistical sampling methodology, engaged a statistical expert to review our sampling methodology, and provided a copy of the statistical expert's report. The statistical expert stated that our sample and extrapolation are not statistically valid and should not be used as a basis to calculate an extrapolated overpayment because: (1) the audit findings did not meet the high-error-rate criteria in the Social Security Act and CMS's Medicare Program Integrity Manual (MPIM) to justify the use of extrapolation, (2) the audit findings did not meet the 5-percent error rate criteria in OIG's Corporate Integrity Agreement (CIA) to justify the use of extrapolation, (3) OIG ignored statistical principles by excluding underpayments or unpaid (i.e., zero-paid) claims from the universe of claims, (4) OIG's sample is not sufficient to achieve the standard precision and confidence level for this type of statistical estimate, (5) OIG did not provide information sufficient to re-create the sampling frame and sample or OIG's overpayment estimate, (6) OIG did not state the sort order of the sampling frame, and (7) OIG failed to provide the random number seed that was used to initialize the random number generator.

Office of Inspector General Response

After reviewing the statistical expert's report, we maintain that our sampling methodology and extrapolation are statistically valid. The legal standard for use of sampling and extrapolation is that it must be based on a statistically valid methodology, not the most precise methodology./30

We properly executed our statistical sampling methodology in that we defined our sampling frame and sample unit, randomly selected our sample, applied relevant criteria in evaluating the sample, and used statistical sampling software (i.e., the OIG, Office of Audit Services (OAS), statistical software RAT-STATS) to apply the correct formulas for the extrapolation.

The statutory and manual requirement that a determination of a sustained or high level of payment errors must be made before extrapolation can be used applies only to Medicare contractors--not OIG./31

In addition, OIG no longer uses the 5-percent error-rate threshold in its CIAs. Moreover, even in prior CIAs that used the 5-percent error-rate threshold, the threshold was used to determine when an additional claims sample (referred to as a "full sample") needed to be selected and reviewed based on the results of a probe sample (referred to as a "discovery sample"). The entity under the CIA was required to extrapolate the results of the full sample, regardless of the error rate./32

Professional Healthcare relies heavily on the MPIM in its arguments that the removal of zero-paid claims ignored statistical principles. The MPIM does not apply to OIG. Even if this manual applied to OIG, it expressly allows for the removal of "claims/claim lines [that] are attributed to sample units for which there was no payment."/33

More generally, OIG may perform a statistical or non-statistical review of a provider without covering all claims from that provider.

Furthermore, OIG's statistical estimates are applied only to the sampling frame from which the sample was drawn.

To account for the precision of our estimate, we recommend recovery at the statistical lower limit of a two-sided 90-percent confidence interval. Lower limits calculated in this manner are designed to be less than the actual overpayment total in the sampling frame 95 percent of the time. The use of the lower limit accounts for the precision of our estimate in a manner that generally favors the auditee./34

Professional Healthcare focuses on the 5 percent of cases where the provider may have to pay more to the Government; however, these cases are inherently rare, and when they arise, the amount the provider may have to over-reimburse to the Government tends to be small. If we had selected a larger sample size, the average effect and the most likely effect would have been that we would have recommended that Professional

Healthcare refund a larger amount to the Government.

We provided Professional Healthcare with sufficient information to re-create the statistical sample and to calculate our estimate given the overpayment amounts in our sample. We also provided Professional Healthcare with the medical review determinations underlying the errors identified in our audit. Because Professional Healthcare stated that it does not have sufficient information to connect the sample overpayment amounts to the medical review determinations, we will work with Professional Healthcare to ensure that it has the necessary information to make this connection. The sampling frame was sorted using a field in OIG's copy of CMS's National Claims History (NCH) file that uniquely identifies claims. After being sorted by this field, the frame was numbered before we generated the random numbers for the sample. We also provided Professional Healthcare with the random-number seed that was used to generate the random numbers.

* * *

View full report at https://oig.hhs.gov/oas/reports/region9/91803028.pdf