CMS Pub. 100-04, ch. 3, § 140.1.3

Rehabilitation Facility Met The Classification Criteria

(Rev. 12575; Issued:04-11-24; Effective: 07-12-24; Implementation: 07-12-24)

A. - Determination of the Compliance Review Time Period

This section provides an overview of the guideline to determine the compliance review

period. In general, the OPOLE, and A/B MAC (A) will use data from the most recent,

consecutive, and appropriate 12-month time period (as defined by CMS) that starts on or

after July 1, 2004, to determine if a facility is in compliance with all of the criteria used to

classify a facility as an IRF. OPOLE and A/B MAC (A) will notify the facility of the time

period that will be utilized. OPOLE and A/B MAC (A) will begin reviewing data 4 months

prior to the start of the facility’s next cost reporting period.

For Cost Reporting Periods Starting from July 1, 2004 to July 1, 2006 Only

The compliance review periods are determined based on the following:

1. Guidelines for Determining Compliance Review Periods For IRFs With

Cost Reporting Periods That Start Between July 1, 2004 and October 31,

2004.

Data prior to July 1, 2004 will not be used to determine an IRF’s compliance with the

requirements in §140.1.1B-D. Thus, for IRFs with cost reporting periods beginning on

or after July 1, 2004 and before November 1, 2004, less than 12 months of data will be

used in their first compliance review period after July 1, 2004. Refer to the first 5 rows

of the Table of Compliance Review Periods (below) for an illustration of this.

2. Guidelines for Determining an IRF’s Compliance Percentage When the

Required Compliance Percentage Threshold Differs Across Two Cost

Reporting Periods

When a cost reporting period starts on or after July 1, 2005, but not later than June 30,

2006, and the compliance review period spans two cost reporting periods, the

compliance percentage is calculated using either of the following two methods. The

IRF must have a patient population in each of the two portions of time in order to use

either of the two methods described below.

(A) The IRF must meet the applicable compliance percentage threshold in each of

the two portions of the compliance review period separately, as illustrated in the

example below.

The following is an example of how this first method would be applied:

The compliance review period for an IRF that has a cost reporting period from July

1, 2005 through June 30, 2006 is March 1, 2005 to February 28, 2006.

The IRF must meet a compliance threshold of 50 percent for the cost reporting

period of July 1, 2004 to June 30, 2005.

The IRF must meet a compliance threshold of 60 percent for the cost reporting

period of July 1, 2005 to June 30, 2006.

In this example, the first portion of the compliance review period (from March 1,

2005 to June 30, 2005) is part of the IRF’s cost reporting period that started on

July 1, 2004 and ends on June 30, 2005. The second portion of the compliance

review period (from July 1, 2005 to February 28, 2006) is part of the IRF’s cost

reporting period that starts on July 1, 2005 and ends on June 30, 2006.

Therefore, for the portion of the compliance review period from March 1, 2005 to

June 30, 2005, the compliance percentage threshold that the IRF must meet is 50

percent.

For the portion of the compliance review period from July 1, 2005 to February 28,

2006, the compliance percentage threshold that the IRF must meet is 60 percent.

If the IRF does not meet the compliance percentage threshold of 50 percent for the

March 1, 2005 to June 30, 2005 portion of the compliance review time period, or

the compliance percentage threshold of 60 percent for the July 1, 2005 to February

28, 2006 portion of the compliance review time period, it will be determined that

the IRF failed to meet the compliance percentage threshold for the entire

compliance review period consisting of March 1, 2005 to February 28, 2006.

(B) The A/B MAC (A) computes one weighted average compliance percentage for

the entire 12-month compliance review period. The resulting weighted average

compliance percentage will be used to determine if the facility met the compliance

threshold requirements in §140.1.1B-D.

The following is an example of how this second method would be applied: The

compliance review period for an IRF that has a cost reporting period from

August 1, 2005 to July 31, 2006 is April 1, 2005 to March 31, 2006. However, the

compliance review period is divided into two portions: April 1, 2005 to July 31,

2005 and August 1, 2005 to March 31, 2006.

In the following hypothetical example, 45 percent of the cases met at least one of

the medical conditions listed above in §140.1.1C from April 1, 2005 to July 31,

2005, and 80 percent of the cases met at least one of the medical conditions listed in

§140.1.1C from August 1, 2005 to March 31, 2006. The weighted average

compliance percentage from the two portions of time must be calculated as follows

for compliance review periods beginning on or after January 1, 2013.

4/12 = 0.333 which is rounded to 0.33

8/12 = 0.666 which is rounded to 0.67

0.33 X 45% = 0.1485

0.67 X 80% = 0.5360

0.1485 + 0.5360 = 0.6845 which is rounded to 68%

Based on this result of 68 percent from the weighted average calculation, it will be

determined that the IRF met the compliance percentage threshold for the

compliance review period starting on April 1, 2005.

The table below entitled “Examples of Compliance Review Periods” provides

examples of compliance review periods associated with various cost reporting

periods.

Examples of Compliance Review Periods. For a facility that has been classified as

an IRF, but is not a “new” IRF as defined below in 140.1.4, the following table

provides examples of the compliance review periods associated with different cost

reporting periods.

Examples of Compliance Review Periods

Start Date of the Cost

Reporting Period for

Which a Facility Will

(or Will Not) be

Classified (or Retain

Classification) as an

IRF

Compliance

Review

Period:

(Admissions

or Discharges

During)

# of

Months

in

Review

Period

Compliance Percentage

Threshold

07/01/2005 07/01/2004 -

02/28/2005

8 50%

08/01/2005 07/01/2004 -

03/31/2005

9 50%

09/01/2005 07/01/2004 -

04/30/2005

10 50%

10/01/2005 07/01/2004 -

05/31/2005

11 50%

11/01/2005 07/01/2004 -

06/30/2005

12 50%

07/01/2006 03/01/2005 -

02/28/2006

12 03/01/2005 to 06/30/2005: 50 %

07/01/2005 to 02/28/2006: 60 %

08/01/2006 04/01/2005-

03/31/2006

12 04/01/2005 to 07/31/2005: 50 %

08/01/2005 to 03/31/2006: 60 %

09/01/2006 05/01/2005-

04/30/2006

12 05/01/2005 to 08/31/2005: 50 %

09/01/2005 to 04/30/2006: 60 %

10/01/2006 06/01/2005-

05/31/2006

12 06/01/2005 to 09/30/2005: 50 %

10/01/2005 to 05/31/2006: 60 %

11/01/2006 07/01/2005-

06/30/2006

12 07/01/2005 to 10/31/2005: 50 %

11/01/2005 to 06/30/2006: 60 %

12/01/2006 08/01/2005-

07/31/2006

12 08/01/2005 to 11/30/2005: 50%

12/01/2005 to 07/31/2006: 60%

For Cost Reporting Periods Starting After July 1, 2006

The compliance review periods are determined based on the following:

1. Guidelines for Determining an IRF’s Compliance Percentage When the Required

Compliance Percentage Threshold Is the Same for the Entire Compliance Review

Period

To minimize the level of effort required by Medicare contractors and IRFs, contractors

must review one continuous 12-month period if the compliance percentage threshold is

the same throughout the entire compliance review period for all compliance review

periods beginning on or after January 1, 2013.

2. Guidelines for Determining the Compliance Review Period of a Facility Classified

as a New IRF. According to the regulations in §412.25(c), a new IRF can only

begin being paid under the IRF PPS at the start of a cost reporting period. If the IRF

begins treating patients prior to the start of a cost reporting period, it may receive

payment under the IPPS until the start of the next cost reporting period, at which

point it can begin receiving payment under the IRF PPS if it meets all of the

applicable requirements in §412.25 and §412.29. A new IRF will have a compliance

review period that starts immediately when its cost reporting period starts, and ends

four months before the start of its next cost reporting period. For example, if a

facility has a cost reporting period that starts on July 1, 2022 and is a new IRF, its

compliance review period would start on July 1, 2022 and end on February 28,

2023. Thus, a facility classified as a new IRF will have an initial compliance review

period that is 8 months in length, in order to allow OPOLE and A/B MAC (A) a 4-month time period to make and administer a compliance determination.

3. Guidelines for Determining an IRF’s Compliance When the IRF Expands its Bed

Capacity. Effective October 1, 2011, as long as an IRF meets all of the applicable

requirements in §412.25(b) and 412.29(c)(2), it may add new beds one time, at any

time, during a cost reporting period. The IRF must provide written certification that

the inpatient population it intends to serve (including the patients served in the new

beds) meets the requirements in §412.29(b). In addition, the new IRF beds will be

included in the compliance review calculations under§412.29(b) from the time that

they are added to the IRF.

4. Guidelines for Determining the Compliance Review Period of a Facility That

Changes Its Cost Reporting Period. A facility that changes its cost reporting period

will have a new compliance review period that is based on its new cost reporting

period. For example, if an IRF changes the start of its cost reporting period from

July 1, 2011 to October 1, 2011, then the start date of its compliance review period

will also change from March 1,2011 to June 1, 2011. Excessive changes to cost

reporting periods are not permitted.

NOTE:

For cost reporting periods beginning on or after July 1, 2006, the compliance threshold

that must be met is 60 percent as discussed in the FY 2009 IRF PPS proposed rule (73

FR 22674, 22687 through 22688) and finalized in the FY 2009 IRF PPS Final Rule (

73 46388). Thus, for all compliance review periods beginning on or after January 1, 2013

(except in the case of new IRFs, as described in section 140.3.4 above), the compliance

review period will be one continuous 12-month time period beginning 4 months before the

start of a cost reporting period and ending 4 months before the beginning of the next cost

reporting period.

Patient comorbidities that satisfy the criteria specified in 42 CFR 412.23(b)(2)(i) shall be

included in the calculations used to determine whether an IRF meets the 60 percent

compliance percentage for cost reporting periods beginning on or after July 1, 2007.

B. - Types of Data Used to Determine Compliance with the Classification

Criteria

1. Starting on July 1, 2004, the A/B MAC (A) will use the verification procedures

specified below in subsection C which is entitled “Verification of the Medical Condition

Criterion Using the Inpatient Rehabilitation Facility-Patient Assessment Instrument

(IRF-PAI) Data Records” or subsection D which is entitled “Verification of the Medical

Condition Criterion Using the Inpatient Rehabilitation Facility’s Total Inpatient

Population” to verify that an IRF has complied with the requirements specified above in

§140.1.1B-D.

2. The verification procedure specified below in subsection C (that is, verification using

the IRF-PAI data) will only be used if the A/B MAC (A) has verified that the IRF’s

Medicare Part A fee-for-service inpatient population is at least 50 percent of the IRF’s

total inpatient population. Effective for compliance review periods beginning on or

after October 1, 2009, A/B MACs (A) must include the IRF’s Medicare Part C

(Medicare Advantage) inpatient population, along with the IRF’s Medicare Part A fee-for-service inpatient population, in determining whether at least 50 percent of the IRF’s

total inpatient population is made up of Medicare patients.

3. General Guideline Regarding Submission of a List of the Inpatients in Each IRF: In

order to verify that an IRF’s Medicare Part A fee-for-service and Medicare Part C

(Medicare Advantage) inpatient populations (combined) reflect the IRF’s total inpatient

population, the A/B MAC (A) in writing will instruct the IRF to send the A/B MAC (A),

by a specific date, a list showing the hospital patient number of each inpatient IRF

admission during the IRF’s 12-month compliance review period. Note that the term

“hospital patient number” used throughout this section refers to a unique patient

identifier used internally within the hospital for patient identification and record-keeping purposes. For each inpatient on the list, the IRF must include the payer the IRF

can bill, or has billed, for treatment and services furnished to the inpatient. If an

inpatient on the list has multiple payers that the IRF can bill, or has billed, the IRF must

include and specify each type of payer. In addition, for each inpatient on the list, the

IRF must include the IRF admission and discharge dates.

Exception to the General Guideline: The Secretary of Health and Human Services can

declare a Public Health Emergency under section 319 of the Public Health Service Act

or another appropriate statute, and the President can declare either a National

Emergency under the National Emergencies Act or a Major Disaster under the RobertT.

Stafford Disaster Relief and Emergency Assistance Act, or other appropriate law. In

accordance with such declarations, certain regulations or operational policies may be

waived in specific geographic areas for limited and defined periods of time. If

applicable, in accordance with the waiver provisions, the IRF may be permitted to admit

patients (referred to in this section as national emergency or disaster inpatients) who

otherwise would be admitted to another inpatient setting. The national emergency or

disaster inpatients will not be included as part of the IRF’s total inpatient population

when the IRF’s compliance with the requirements specified in§140.1.1B-D is

determined by the A/B MAC (A) reading a sample of medical records. Therefore, when

the IRF submits the list of hospital patient numbers stipulated above in section

140.1.3B3, the IRF will identify each national emergency or disaster inpatient by

placing either the capital letter “E” or “D” after the patient’s unique internal hospital

identification number. The A/B MAC (A) will verify the information and, if appropriate,

exclude these patients from the list of inpatients used to select a sample of medical

records. The IRF should appropriately document in the medical record sufficient

information to identify an inpatient as a national emergency or disaster inpatient.

4. The A/B MAC (A) will use the list of hospital patient numbers to determine the IRF’s

total inpatient population during the IRF’s compliance review period. In addition to

the above processes, the A/B MAC (A) has the discretion to sample and compare other

parameters (that is, diagnoses, procedures, length-of-stay, or any other relevant

parameter) to determine that the Medicare Part A fee-for-service and Medicare Part C

(Medicare Advantage) population (beginning on or after October 1, 2009) is

representative of the IRF’s total inpatient population.

A determination by the A/B MAC (A), in accordance with the preceding

methodologies, that the IRF’s inpatient population for the compliance review period

consisted of at least 50 percent Medicare Part A fee-for-service and Medicare Part C

(Medicare Advantage) patients (beginning on or after October 1, 2009) means that the

A/B MAC (A) can use the procedure stipulated below in subsection C to

presumptively determine if the IRF met the compliance threshold as specified above in

§140.1.1B-D.

5. The A/B MAC (A) will inform OPOLE if an IRF fails to send the list showing the

hospital patient number associated with each inpatient IRF admission during the most

recent, consecutive, and appropriate 12-month period, as defined by CMS. Further, the

A/B MAC (A) will inform OPOLE if the list of hospital patient numbers does not show

the payer or payers or the admission and discharge dates for each hospital patient

number on the list. OPOLE will notify the IRF that failure to send the A/B MAC (A)

the list within an additional 10 calendar days will result in a determination by OPOLE

that the IRF has not met the requirements specified above in §140.1.1B-D and the

facility will no longer be eligible for payment under the IRF PPS.

C. - Verification of the Medical Condition Criteria Using the

Inpatient Rehabilitation Facility-Patient Assessment Instrument

(IRF-PAI) Data Records (The Presumptive Methodology)

1. To determine if a facility has presumptively complied with the criteria specified

above in §140.1.1B-D, the CMS will enable the A/B MAC (A) to access the CMS' IRF-PAI data records. Specifically, each A/B MAC (A) will be allowed to access only the

IRF-PAI information submitted by IRFs that submit claims to that A/B MAC (A).

In order to ensure that the software that matches each IRF to a particular A/B MAC (A)

is constantly updated, the A/B MAC (A) must electronically send OPOLE a table that

has at least the following title and column headings:

A/B MAC (A) List Of IRF Provider Numbers (Specify The A/B MAC (A)’s Name)

The Name of

Each IRF

That Submits

Claims To

This A/B

MAC (A)

IRF Provider

Number

IRF Cost

Reporting

Period

After checking the A/B MAC (A)’s list of IRFs for completeness and, as necessary,

communicating with the A/B MAC (A) to ensure the accuracy of the information,

OPOLE will forward the A/B MAC (A)’s list of IRFs to the CMS designated mailbox

IRFACTIVEPROVIDER@cms.hhs.gov. The CMS contractor that maintains the IRF-PAI database will then, if necessary, update the IRF-PAI database software used to

presumptively verify compliance with the requirements specified in §140.1.1B-D. The

A/B MAC (A) must coordinate with their CMS OPOLE to obtain access to the software

system.

The A/B MAC (A) will provide OPOLE with user information from all A/B MAC (A)

staff that are required to access the IRF-PAI data records.

When the A/B MAC (A) accesses the IRF-PAI data records, the A/B MAC (A) will be

able to generate an IRF compliance review report using the IRF-PAI information from

the IRFs on the A/B MAC (A)’s list. The Internet Quality Improvement and Evaluation

System (iQIES) system software used to generate the IRF compliance review report

will automatically use the specific diagnosis codes from the appropriate files listed in

“Presumptive Compliance” relative to the appropriate fiscal year regulation which are

available for download from the IRF PPS website at :

https://www.cms.gov/medicare/payment/prospective-payment-systems/inpatient-rehabilitation/rules-related-files to determine if a particular IRF is presumptively in

compliance with the requirements specified in §140.1.1B-D. Prior to generating the IRF

compliance review report, the A/B MAC (A) must allow the IRF to decide whether the

IRF compliance review report will be generated using the IRF- PAI data records of

patients who were admitted during the IRF’s compliance review period (even if they

were discharged outside of the compliance review period), or the IRF-PAI data records

of patients who were discharged during the IRF’s compliance review period (even if

they were admitted outside of the compliance review period).

Below are the sections of the IRF compliance review report with example data

required for submission:

IRF Compliance Review Report

State Provider

Number

Provider

Name

Cost Report

Start Date

Compliance Review

Period

Any

State

IRF

Number

Best

Rehab

08/01/2023 04/01/2022 To

03/31/2023

Submitted Assessments Eligible Assessments Percent

100 60 60%

The submitted assessments section identifies all the IRF-PAI data records that the IRF submitted

to the IRF-PAI database during the compliance review period. The eligible assessments are the

assessments submitted during the compliance review period that match one of the codes in the

associated files described as “Presumptive Compliance” relative to the appropriate fiscal year

regulation, which can be downloaded from the IRF PPS website at

https://www.cms.gov/medicare/payment/prospective-payment-systems/inpatient

rehabilitation/rules-related-files. The cost report start date shown is the start of the facility’s

next cost reporting period.

2. If an IRF’s inpatient Medicare Part A fee-for-service and Medicare Part C (Medicare

Advantage) populations (combined) (beginning on or after October 1, 2009) are at least 50 percent

of its total inpatient population and the presumptive methodology (described above) indicates that

the IRF met or exceeded the requirements specified in §140.1.1B-D, then the IRF is presumed to

have met the requirements specified above in §140.1.1B-D. However, even when an IRF is

presumed to have met the requirements specified above in §140.1.1B-D, OPOLE and A/B MAC

(A) still have the discretion to instruct the IRF to send to OPOLE or A/B MAC (A) specific

sections of the medical records of a random sample of inpatients, or specific sections of the

medical records of inpatients identified by other means by the CMS or the A/B MAC (A).

3. Each A/B MAC (A) must submit a report to the appropriate CMS OPOLE (with a copy to the

CMS Central Office) on at least a quarterly basis that shows each IRF’s status with respect to

compliance with the requirements specified above in §140.1.1B-D.

The associated files described as “Presumptive Compliance” relative to the appropriate fiscal year

regulation, which can be downloaded from the IRF PPS website at

https://www.cms.gov/medicare/payment/prospective-payment-systems/inpatient-rehabilitation/rules-related-files will be used to determine presumptive compliance with the

requirements specified above in §140.1.1B-D.

The files listed in “Presumptive Methodology Files – Implementation of Changes” that is attached

to the IRF Compliance Rule Specification Files, which can be downloaded from the IRF PPS

website at http://www.cms.gov/Medicare/Medicare-Fee-for- Service-

Payment/InpatientRehabFacPPS/Criteria.html, will be used to determine presumptive compliance

with the requirements specified above in §140.1.1B-D.

D. – Additional Verification of Arthritis Condition Criteria

Compliance with the regulatory requirements for the arthritis conditions specified above in§140.1.1 B-D cannot be determined by the presence of an impairment group code or diagnosis code alone, but can

only be verified through review of the IRF medical record. For this reason, we removed arthritis

impairment group codes and diagnosis codes from the list of codes used to determine presumptive

compliance for compliance review periods beginning on or after October 1, 2015

However, beginning on or after October 1, 2015, we also provided for an additional item on the IRF-PAI (item #24A) to enable IRFs to indicate whether the patient’s arthritis condition(s) meets all of the

relevant regulatory requirements specified in §140.1.1 B-D. Using the process described below, the

A/B MAC (A) must verify through medical review whether the IRF cases that would not otherwise

meet the compliance criteria, and that have a “1 – Yes” marked in item #24A, meet the severity and

prior treatment requirements in §140.1.1B-D. If so, then the A/B MAC (A) must add the appropriate

number of these cases to the cases that meet the presumptive compliance criteria.

The A/B MAC (A) shall use the following process for compliance review periods beginning on or after

October 1, 2015:

1. If the A/B MAC (A) has determined, using the process outlined in subsection C above, that the

IRF does not presumptively meet the requirements specified above in §140.1.1B-D, then the A/B

MAC (A) must access an IRF-PAI data report called the IRF Arthritis Verification Report through the

iQIES system .

Below are the sections of the IRF Arthritis Verification Report with example data:

IRF Arthritis Verification Report

Provider

Number

Provider

Name

Patient

Name

Patient ID IRF-PAI ID Admission

Date

Discharge

Date

Best

Rehab

A. Smith

B. Jones

Z. Honey

12345678A

22345678A

32345678A

987654321

987654322

987654323

10/1/22

10/2/22

10/3/22

10/15/22

10/14/22

10/13/22

2. The A/B MAC (A) must determine whether or not adding all of the cases listed on the IRF

Arthritis Verification Report for that IRF would be enough to increase the IRF’s compliance

percentage to equal or exceed 60 percent.

3. If adding all of the cases listed on the IRF Arthritis Verification Report for that IRF would be

enough to increase the IRF’s compliance percentage to equal or exceed 60 percent, then the

A/B MAC (A) must use generally accepted statistical sampling techniques to obtain a

statistically valid random sample of those patients listed for the IRF on the IRF Arthritis

Verification Report. If the total number of patients listed for the IRF on the IRF Arthritis

Verification Report is less than 10, then the A/B MAC (A) will review all patients listed for

the IRF on the IRF Arthritis Verification Report. (Note that if adding all the cases listed in the

IRF Arthritis Verification Report for that IRF would not be enough to increase the IRF’s

compliance percentage to equal or exceed 60 percent, then the A/B MAC

(A) will not proceed further with the IRF arthritis verification process and will use the

presumptive compliance percentage generated from section C above.)

4. The A/B MAC (A) will obtain and examine the medical record sections and any other

pertinent information submitted by the IRF to determine if the patients from the random

sample obtained in step 2 meet all of the severity and prior treatment requirements specified

in §140.1.1B-D.

5. The percentage of patients from the list that the A/B MAC (A) determines to have met the

severity and prior treatment requirements specified in §140.1.1B-D will be extrapolated to the

complete list of patients for the IRF on the IRF Arthritis Verification Report.

6. The A/B MAC (A) shall then add the appropriate number of cases (based on the percentage

in step 4) from the IRF Arthritis Verification Report to the cases that meet the presumptive

compliance criteria, and re-calculate the IRF’s presumptive compliance percentage.

For example:

• IRF A submitted 545 IRF-PAIs in the compliance review period.

• IRF A’s presumptive compliance percentage (determined by the A/B MAC

(A) using the steps outlined in subsection C above) was less than 60 percent.

• In this case, the A/B MAC (A) must access the IRF Arthritis Verification Report.

• The IRF Arthritis Verification Report shows 100 patients listed for IRF A.

• The A/B MAC (A) determines that inclusion of all 100 patients listed for IRF A would

increase IRF A’s presumptive compliance percentage enough to meet or exceed the 60

percent threshold.

• The A/B MAC (A) uses generally accepted statistical sampling techniques to

randomly select 10 patients from that list for medical review, and based on the

medical review determines that 7 of the 10 patients meet all of the requirements

specified in §140.1.1B-D.

• The A/B MAC (A) will then extrapolate this percentage (7/10 = 70 percent) to the full list

of patients shown on the IRF Eligibility Arthritis Verification Report for IRF A. Thus,

the A/B MAC (A) will add 70 patients (70 percent of 100) listed for IRF A on the IRF

Arthritis Verification Report to the total number of IRF A’s patients that meet the

presumptive compliance criteria.

• The A/B MAC (A) will then recalculate IRF A’s presumptive compliance percentage

with the addition of the 70 cases.

• The A/B MAC (A) will base the determination of the IRF’s presumptive compliance

with the requirements specified in §140.1.1B-D on the updated calculation of the IRF’s

presumptive compliance percentage.

• The A/B/ MAC (A) will report this updated presumptive compliance percentage,

instead of the presumptive compliance percentage from subsection C above, on the

quarterly report that the A/B MAC (A) sends to the appropriate CMS OPOLE (with a

copy to the CMS Central Office).

NOTE: Even when an IRF is presumed to have met the requirements specified above in§140.1.1B-D

using the updated presumptive methodology calculation, OPOLE and A/BMAC (A) still have the

discretion to use the medical review methodology described in subsection E below to verify the IRF’s

compliance with the requirements in §140.1.1B-D.

E. - Verification of the Medical Condition Criteria Using the Inpatient

Rehabilitation Facility’s Total Inpatient Population (Medical Review

Methodology)

1. The A/B MAC (A) must use the IRF’s total inpatient population to verify that the IRF has

met the requirements specified above in §140.1.1B-D if:

(i) the IRF’s Medicare population (including Medicare Part A fee -for-service and Medicare

Part C (Medicare Advantage) patients, effective October 1, 2009) is not at least 50 percent of

its total inpatient population; or

(ii) the A/B MAC (A) is unable to generate a valid IRF compliance review report

using the IRF-PAI database methodology specified previously; or

(iii) the A/B MAC (A) generates an IRF compliance review report, based on the use of the

presumptive methodology, which demonstrates that the IRF has not met the requirements

specified above in §140.1.1B-D.

If the IRF’s Medicare Part A fee-for-service and Medicare Part C (Medicare Advantage)

populations (combined, effective October 1, 2009) comprise less than 50 percent of the IRF’s

total inpatient population, or the A/B MAC (A) otherwise determines that the Medicare Part A

fee-for-service and Medicare Part C (Medicare Advantage) populations (combined, effective

October 1, 2009) are not representative of the overall IRF inpatient population, or the A/B MAC

(A) is unable to generate a valid report using the presumptive methodology, the presumptive

determination is that the IRF did not meet the requirements specified above in §140.1.1B-D.

2. As previously stated above, the A/B MAC (A) will instruct the IRF to send the A/B MAC (A)

a list showing the hospital patient number of each inpatient that the IRF admitted during the most

recent, consecutive, and appropriate 12-month period, as defined by CMS. The list of hospital

patient numbers must include the payer(s) and admission and discharge dates that correspond

with the inpatients whose hospital patient numbers are shown on the list. The A/B MAC (A) will

then use generally accepted statistical sampling techniques to obtain a random sample of

inpatients from the list. The random sample of inpatients drawn from the list must be sufficiently

large to ensure that the A/B MAC (A) can determine, with at least 95 percent confidence,

whether the IRF’s compliance percentage is below the required compliance threshold (i.e., not in

compliance) or at or above the required compliance threshold (i.e., in compliance).

For example, suppose that the required compliance threshold for an IRF to be in compliance

with the requirements specified above in §140.1.1B-D is 60 percent. The A/B MAC (A) reviews

a random sample of claims from IRF A and estimates that IRF A’s compliance percentage is 58

percent. Suppose that the standard deviation that the A/B MAC (A) calculates for IRF A’s

random sample of IRF claims is plus or minus 4 percentage points, so that the 95 percent

confidence interval in this particular example is between 54 percent and 62 percent (with 58

percent as the midpoint). In this case, the IRF is considered to be in compliance with the 60

percent rule, since 60 percent is within the 95 percent confidence interval. To verify whether the

IRF is in fact in compliance with the requirements specified above in §140.1.1B-D, the A/B

MAC (A) may need to draw a larger random sample of the IRF’s inpatients. For example, a

larger random sample of IRF A’s inpatients might have reduced the standard deviation to plus or

minus 1 percentage point, which would have led the 95 percent confidence interval to be

between 57 percent and 59 percent. This would have demonstrated with 95 percent confidence

that the IRF was not in compliance with the requirements specified above in §140.1.1B-D

(because the entire 95 percent confidence interval was below the required compliance threshold

of 60 percent).

If the compliance percentage threshold differs within the compliance review period (i.e., is 50

percent for a portion of the compliance review period and 60 percent for the other portion of the

period), then a random sample of inpatients will be drawn from each of the two time periods

separately.

The use of generally recognized statistical sampling principles may result in a determination that

it would be inappropriate to use a sample to determine the facility’s compliance percentage. If a

random sample is not appropriate in a particular case, then the A/B MAC (A) will use the IRF’s

entire inpatient population to determine the IRF’s compliance percentage. In addition, if the IRF

had 100 or fewer inpatients during the compliance review period, then the A/B MAC (A) must

use the IRF’s total inpatient population (consisting of both Medicare and non- Medicare

inpatients) to determine the IRF’s compliance percentage.

Prior to selecting the random sample of inpatients, the A/B MAC (A) must allow the IRF to

decide if the IRF wants the sample to contain either the patients who were admitted during the

IRF’s compliance review period (even if some of those patients were discharged outside of the

compliance review period) or the patients discharged during the IRF’s compliance review

period (even if some of those patients were admitted outside of the compliance review period).

If the A/B MAC (A) uses a random sample of the IRF’s inpatient population (rather than the

IRF’s total inpatient population) to determine the IRF’s compliance percentage, then the A/B

MAC (A) must ensure that an adequate sample size is used to determine (with at least a 95 percent

statistical level of confidence) whether or not the IRF has met the requirements in §140.1.1B-D. In

some cases, this will require the A/B MAC (A) to expand the size of the random sample of

inpatients selected from a particular IRF.

The A/B MAC (A) will instruct the IRF to send it copies of specific sections of the medical records

for all of the inpatients to be used in the compliance review. The A/B MAC (A) has the discretion

to decide which specific sections of the medical records to obtain, provided that the requested

medical record sections contain enough information to allow the A/B MAC (A)’s reviewers to

determine the medical condition(s) for which each inpatient received treatment in the IRF. In

addition to submitting the requested sections of the medical records, the IRF has the discretion to

send the A/B MAC (A) other clinical information regarding these same inpatients.

3. The A/B MAC (A) will examine the medical record sections and any other information

submitted by the IRF to determine if the IRF meets the requirements specified above in

§140.1.1B-D. To determine if a specific inpatient matches one of the medical conditions

specified in §140.1.1C, the A/B MAC (A) may use the diagnosis and impairment group codes

specified in the appropriate files listed in “Presumptive Methodology Files – Implementation of

Changes” that is attached to the IRF Compliance Rule Specification Files, which can be

downloaded from the IRF PPS website at https://www.cms.gov/medicare/payment/prospective-payment-systems/inpatient-rehabilitation/rules-related-files for general guidance. The A/B

MAC (A) is not permitted to use these codes to make a final determination as to whether or not

the specific inpatient required intensive rehabilitation services for treatment of one or more of

the medical conditions specified in §140.1.1C. The determination of whether a specific

inpatient required intensive rehabilitation services for treatment of a condition can only be

determined through careful review of that inpatient’s unique clinical characteristics and

circumstances, as reflected in the inpatient’s medical record.

4. In general, when the A/B MAC (A) is using a sample of medical records to determine

compliance with the requirements in §140.1.1B-D, the A/B MAC (A) always has the discretion to

determine if a patient meets or does not meet any of the medical conditions listed in §140.1.1C

based upon a review of the clinical record, regardless of the results of the presumptive

methodology described previously. In other words, the compliance percentage that is determined

using the medical review methodology described in this section will supersede the compliance

percentage that was determined for the same compliance review period using the presumptive

methodology. To ensure that the compliance review process is similar for all IRFs, the A/B MAC

(A) must have written policies that describe the reasons for using a random sample of medical

records to determine an IRF’s compliance percentage when the presumptive methodology has

shown that the IRF met the compliance threshold.

5. The A/B MAC (A) will inform OPOLE if an IRF fails to provide information in accordance

with the requirements specified above in subsection D2. OPOLE will notify the IRF that failure

to provide the A/B MAC (A) with the information in accordance with the requirements

specified above in subsection D2 will result in a determination by OPOLE that the IRF has not

met the requirements specified above in §140.1.1B-D.

F. – Submission of Compliance Data Requirements

By the 15th day of each month, the A/B MAC (A) responsible for determining the

compliance percentage for each IRF using either of the methods specified above in

§§140.1.3C or 140.1.3D shall submit a report to CMS via e- mail to

cmscompliancedata@cms.hhs.gov. Instructions regarding the format of the report, how to

complete the report, and where to send it are specified on the IRF PPS website at:

https://www.cms.gov/medicare/payment/prospective-payment-systems/inpatient-rehabilitation/rules-related-files.

The submitted report should indicate the IRF data for the specific month and year, as well as the MAC

name. The data elements for the report must include the following:

• MAC Number,

• Name of Provider,

• City,

• State,

• Zip Code,

• Provider Number,

• Cost Report Begin Date (mm/dd/y

• Review Period From and To Dates (mm/dd/yyy),

• Percent of the Medicare Population,

• Percent of Compliance using the Presumptive Method (if the presumptive method is used),

• Percent of Compliance using Medical Review Sample (if the medical review method is used

following a failed presumptive method or at the discretion of the MAC), and

• Comments (indicating any additional relevant information).

The following demonstrates how a typical report should look. All data provided in the example report

below are fictitious.

IRF Data for the month and year of: May 2022

MAC Name: BCBS of Healthy OH

MAC

Number

Name of

Provider City State Zip

Provider

Number

01234

Acme

Rehabilitation

Unit Healthy OH 12345 04T123

01234

Zenith

Rehabilitation

Unit

Healthy OH 12345 04T321

Cost Report

Begin Date

(mm/dd/yyyy)

Review

Period

From and

To Dates

(mm/dd/yyyy

-

mm/dd/yyyy)

Percent of

the

Medicare

Population

Percent of

Compliance

using the

Presumptive

Method

Percent of

Compliance

using

Medical

Review

Sample Comments

07/01/2022

01/01/2021 –

02/29/2022 50.00% 60.00%

10/01/2022

06/01/2021 –

05/31/2022 20.00% NA 65.00%

1. The A/B MAC (A) must verify that the requirements specified above in

§140.1.1B-E and §140.1.2 G-K were met.

2. The State Agency will determine whether the criteria specified above in

§140.1.1F-K and §140.1.2 Q were met.