CY 2011 CMS Risk Adjustment Data Validation Overview

Uploaded by CMSHHSgov on 06.12.2012

Hello, welcome to the CY 2011 CMS Risk Adjustment Data Validation or
RADV Overview training. I want to thank everyone for participating in
this call. I am Jonathan Smith, Director of the Division of Payment
Validation (DPV) in the Medicare Plan Payment Group (MPPG) here at CMS.
DPV conducts the RADV audits. The deputy director of DPV is Tobi
Pulley. The group director of MPPG is Cheri Rice. This webinar is in
response to a request from MA contracts to learn more about RADV
I would like to take this opportunity to recognize the RADV team. The
RADV team members are: Maricruz Bonfante, Stacey Deiter, Esmail
Essajee, Ashley Franzel, Kellie Gombeski, Carolyn Kapustij, and Steve
Today's presentation is a general overview about the next round of
contract level audits. There will be a more detailed training for
contracts actually selected for audit. Keep in mind that today's
presentation is geared toward the contract level sample, which has some
differences with the national sample that CMS conducts annually to
comply with mandated error rate reporting. Although we will not be
responding to questions on the teleconference, you may continue to
submit questions to the RADV mailbox and we will respond at a later
date, via email.
The main topics for today's presentation are: A general description of
RADV and its purpose. Information about the February 24, 2012 notice
describing the Final Payment Error Calculation methodology. A
description of the various steps of a RADV audit including the appeals
process. Information about What Happens After My Contract is Selected
for an Audit. The notification process and audit operations; and
finally, the Q and A session.
Now, I will turn the presentation over to Tobi Pulley.
Thank you, Jonathan.
Tobi: There will be a few acronyms used during today's presentation and
slide 7 is a resource you can refer to follow the terminology. AROF -
Audit Report of Findings CDAT - Central Data Abstraction Tool CMS - HCC
- Centers for Medicare and Medicaid Services Hierarchical Condition
Category INV - Invalid MRD - Medical Record Dispute MRR - Medical Record
Review RADV - Risk Adjustment Data Validation RAPS - Risk Adjustment
Processing System
The purpose of RADV is to validate diagnoses submitted by contracts for
payment. Every year, CMS reports a national payment error rate for the
Part C program to comply with reporting requirements of the Improper
Payments Elimination and Recovery Act (IPERA) of 2010. The contract
level RADV audits are a corrective action to address the payment error
reported for the Part C program. It is our expectation that contract
specific RADV audits will encourage Medicare Advantage contracts to
carefully examine the diagnoses being submitted for payment, and result
in more accurate data submission practices.
I would like to mention that MA contracts are responsible for submitting
accurate data to CMS for payment and RADV is a method to evaluate the
accuracy of risk adjusted payments.
RADV verifies diagnoses that have been submitted to RAPS for payment.
The risk adjustment guidelines, and hence, the guiding principle of
RADV, states that diagnoses submitted into RAPS by Medicare Advantage
contracts must be: Documented in a medical record that was based on a
face-to-face health service encounter between a patient and a healthcare
provider Coded in accordance with the International Classification of
Diseases, Ninth Revision, or ICD-9-Guidelines for Coding Assigned based
on dates of service within the data collection period AND From an
acceptable Risk Adjustment provider type and Risk Adjustment physician
specialty This guiding principle is used as the key guideline for
medical record review." And with that, I will turn it over to Carolyn
Kapustij, who will talk about the RADV methodology notice.
Thank you, Tobi. Next we'll be discussing the Final Payment Error
Calculation Methodology.
CMS invited public comment on the proposed methodology posted on
December 21, 2010 titled "Medicare Advantage Risk Adjustment Data
Validation (RADV) Notice of Payment Error Calculation Methodology for
Part C Organizations Selected for RADV Audit - Request for Comment".
CMS carefully reviewed the more than 500 comments received on this draft
methodology and took them into consideration when finalizing the
CMS released the "Notice of Final Payment Error Calculation Methodology
for Part C Medicare Advantage Risk Adjustment Data Validation
Contract-Level Audits" on February 24, 2012. The document is available
online via CMS' Medicare Advantage Plan Payment page at the
This is the methodology that will apply to the next round of
contract-level RADV audits, which will be payment year 2011. Payment
year 2011 is the first year for which payment recovery based on
extrapolated estimates will be conducted for Medicare Advantage.
Approximately 30 MA contracts will be selected for each contract-level
RADV audit. Effective with the payment year 2011 contract-level RADV
audit, CMS will allow audited MA contracts to submit multiple medical
records for each CMS-HCC being validated. CMS will apply a
Fee-for-Service Adjuster to determine the final payment recovery amount.
Now, we will discuss the sampling aspect of the RADV Process.
The next series of slides describe the steps in the RADV process,
starting with sampling and going through appeals.
The first step in the RADV process is sampling. Not all contracts are
eligible for selection, specifically plans that do not receive risk
adjusted payments, such as cost plans, are excluded. CMS also excludes
contracts that have terminated and did not merge into another contract.
After the contracts for audit have been selected, CMS selects the
enrollees. The sampling frame is based on enrollment data for January
of the payment year. In order to be eligible for RADV an enrollee needs
to have been continuously enrolled in the same contract from January of
the data collection year, which is 2010 for the upcoming audits, through
January of the payment year, 2011 for the upcoming audits. This means
that enrollees who are new to Medicare and/or the MA program are
excluded, as well as those who switched contracts between the data
collection and payment year. CMS also excludes enrollees who had hospice
or ESRD status during that data collection year. Finally, beneficiaries
must have a CMS-HCC submitted by the contract for payment.
From the contract's eligible population, CMS then selects 201 enrollees.
The sample is a stratified, random sample. The strata are determined
by the risk scores of the contracts eligible enrollees. There are three
strata delineated by high, medium, and low risk score. 67 enrollees
will be sampled from each stratum. The sampling weights are calculated
by dividing the total number of enrollees eligible for sampling in each
stratum by 67. Now, Ashley Franzel will talk about the next step, the
medical record review process
A little later in the presentation, you will hear specifics about the
CMS process for notifying MA organizations that are selected for audit
based on the sampling specifications we just reviewed. In the next few
slides we'll share what happens with the medical records that are
collected from your MA providers and submitted to CMS in response to the
notification and distribution of sample data. The whole medical record
review process is planned to ensure that every valid submitted record is
evaluated by coders according to the ICD-9-CM Guidelines for Coding and
Reporting. It's important to point out that MA organizations are
required to follow the RADV rules and provide documentation as described
in the audit instructions. Failure to follow the instructions may
render the contract's subsequent appeals invalid and also impact the
payment recovery amount. MA organizations that have participated in
recent national RADV audits are already familiar with the Central Data
Abstraction Tool, known as CDAT. CDAT is the principle point of contact
and communication during an audit. Each audited MA contract will have
its own CDAT entry point, and will use the tool to download instructions
and data, to complete coversheets, to submit medical records, and later
to download audit reports of findings and to participate in medical
record dispute.
As announced in the February 24, 2012 Notice, CMS will allow multiple
medical records to be submitted for each CMS-HCC being audited for the
CY 2011 Contract-Level RADV audits. CDAT will be set up with a single
medical record coversheet for each audited enrollee, populated with all
of that enrollee's audited CMS-HCCs. MA organizations will submit all
medical records for an enrollee using this CDAT-based coversheet, and
will have up until the time of the deadline to replace, re-order, and
even remove medical records as needed. You may submit up to five medical
records for each audited CMS-HCC per enrollee.
So as we've discussed, CMS is prepared to accept a total of up to 5
medical records for each audited CMS-HCC per enrollee. Exceptions may
be granted in unusual circumstances.
A little earlier, we emphasized that only valid records go forward for
coding. The validity of records is determined during an intake
evaluation that is performed by senior coders*. A senior coder checks
the submitted medical record to determine that it's acceptable for RADV.
He or she also checks the CMS medical record coversheet to make sure
that all MA organization entries are consistent with the submitted
record. Failure of these checks means that no further review can be
performed because the submission is invalid. If a record is determined
to be invalid, it is checked by a second senior coder who must agree
with the first senior coder's determination. The record will not be
able to confirm any of the enrollee's CMS-HCCs. MA organizations may
submit a CMS-Generated Attestation for a valid outpatient or physician
record that is missing a signature and /or acceptable credentials.
CMS created the RADV Medical Record Checklist and Guidance to assist MA
organizations in selecting appropriate medical records. It can be found
at the web address shown on this slide. The guidance is based on issues
CMS observed with medical records submitted for previous RADV audits The
guidance addresses issues observed during intake, such as incorrect date
of service, unacceptable provider type, and also coding issues, such as
cases where a diagnosis cannot be verified using ICD-9 guidelines.
The main takeaways about medical record review are these: For each
enrollee, CMS will consider all submitted records for all audited
CMS-HCCs. All diagnoses will be abstracted from the priority medical
record that supports the CMS-HCC under review. Priority is based on the
order in which medical records are arranged by the MA contract. Remember
that instructions for the contract-level sample will include more
information about the process of requesting and submitting multiple
medical records for RADV. We'll also address questions about the
process through the CMS RADV mailbox. Following medical record review,
CMS uses the data to determine a payment error estimate. From here I'll
hand the presentation over to Carolyn Kapustij.
The medical record review results are the basis for the payment error
estimate. We will get into more details in the following slides about
the calculations, but the basic premise of the RADV calculation is the
payment error for each enrollee, multiplied by the sampling weights.
The general process to get to a contract's payment recovery amount is to
total the weighted enrollee payment errors and calculate the lower bound
of the 99% confidence interval. A FFS Adjuster will be applied to the
lower bound, resulting in a payment recovery amount. More information
on the payment recovery amount is in an upcoming slide. Note that
payment recovery will not occur until after completion of medical record
dispute. The recovery amount will be deducted from a contract's monthly
MARx payment.
To determine the payment error for each enrollee CMS takes the payment
the plan actually received and subtracts the payment the plan should
have received, based on medical record review. If RADV confirms the
diagnoses submitted for payment, there will be no payment error. Next,
the sampling weights are applied to each sampled enrollees payment
error. The weighted enrollee level errors are then totaled, resulting
in a point estimate, the payment error estimate.
The 99% confidence interval around the point estimate is calculated. CMS
then determines the preliminary payment recovery amount. There may be
situations in which the lower bound of the 99% confidence interval is
less than zero, showing a net underpayment for these plans. In these
cases, payment recovery will be constrained to zero. For cases in which
the preliminary payment recovery estimate is greater than 0, CMS will
apply the FFS adjuster to determine the final payment recovery amount.
If application of the FFS adjuster results in a net underpayment, the
final recovery amount is constrained to zero. Details on the FFS
adjuster will be shared in the future. The adjuster will be calculated
by CMS based on a RADV-like review of records submitted to support FFS
claims data.
At the conclusion of medical record review contracts will receive a
report of findings. For each audited CMS-HCC, the report will detail the
validation outcome, error type, and eligibility for the next step,
medical record dispute. The report will also show the risk score and
payment associated with each enrollee, as submitted by the contract to
RAPS and the recalculation based on medical record review results.
Instructions about medical record dispute will also be included with the
finding's report. Now, I am going to turn it over to my colleague,
Stacey Dieter.
Thank you Carolyn. Next we will be discussing the Risk Adjustment Data
Validation (RADV) process of Medical Record Dispute (MRD).
After the initial medical record review results have been released, MA
organizations have an opportunity to dispute certain types of Risk
Adjustment Data Validation errors through the process of Medical Record
This process allows your MA organization the opportunity to explain to
CMS why the medical record is valid for review and/or where the CMS-HCC
is present on the medical record you submitted. Only CMS-HCCs that you
associate with a medical record during the initial submission period
will be eligible for medical record dispute.
MA organizations must select the "one best medical record" from the
previously submitted for each audited CMS-HCC to dispute an error. If
CMS agrees with your dispute, the error associated with that CMS-HCC
will be overturned.
At the conclusion of the Medical Record Dispute process, audited MA
organizations will receive their findings through the Audit Report Post
Medical Record Review.
The Audit Report Post Medical Record Review will detail CMS-HCC
validation outcomes, error types, eligibility for appeal, as well as
payment error recovery amounts. In addition, your MA organization will
receive a written response to explain why the medical record dispute
your MA organization submitted did not overturn the original finding.
Finally, MA organizations will also receive instructions on filing
Per CMS regulation, MA organizations are afforded two types of appeals:
Medical Record Review Determination Appeal; and Payment Error
Calculation Appeal Full appeal rights are detailed in CMS regulation 42
CFR § 422.311. I'll hand off to my colleague Steve Ludwig.
Thanks Stacey. In the next few slides we'll talk about the process of
participating in a contract-level audit. It's important to emphasize
again that today's presentation is focused on contract-level audits.
The process begins when the CEO and MCO for each audited MA contract
receive an email message from CMS announcing that the contract was
selected. For this reason, it's important that the information in HPMS
is up to date. The CEO or MCO must respond to this email by confirming
the identity of the MCO, and also identifying two additional points of
contact. CMS communicates only with these three people during the
audit. In addition, these three individuals receive access to CDAT,
which we discussed earlier in today's presentation. This access begins
at the start of the audit process, and ends after the deadline for
submitting medical records. All three will receive information about
training. CMS will provide training on accessing CDAT, and also on
using it to download data and to upload medical record files.
CDAT allows CMS to meet all security requirements for transferring RADV
data. CMS provides illustrated instructions to CDAT users so that they
can easily access the system. Once they log in, they can download
complete audit instructions. They can also download a list of all of
the audited enrollees and CMS-generated attestations populated with
information for each audited enrollee. These same CDAT users come back
into the system to upload medical record files to CDAT, and also view
what they've downloaded. Later, CMS and its contractors use CDAT to
conduct medical record review.
CMS wants to make it as easy as possible for you to use the system.
When CDAT accounts are set up, we invite CDAT users to a short
teleconference on basic access to the system. We have technical folks on
the line to describe the process of logging in, using two-factor
authentication, and downloading instructions. This training happens
before any live, enrollee-specific data is available in the system.
Shortly after this training, CDAT access is temporarily closed down so
that CMS can upload enrollee-specific data into the system. Once data is
available, we invite CDAT users to a second training that focuses on the
entire contract-level audit process. This audit-specific training walks
through our instructions, and also includes more CDAT information, like
the process for uploading medical records, and also for rearranging the
priority of the records as you submit them
Following the audit-specific teleconference, CDAT users receive an email
message letting them know that data is available to download. The email
contains a password for decrypting the data. At this point, the clock
starts on the 16-week timeline for requesting medical records and
CMS-Generated Attestations from providers. The audit instructions
contain information to assist you with this process, including signed
CMS letters to accompany your requests for records. You may also want
to direct providers to an education program on Part C payment that we've
recently launched online with Medscape. Medscape is a free learning
resource for the medical community. Next, you will hear from Jonathan to
conclude this training session.
As part of our notification and invitation to this presentation, CMS
solicited questions from the participants regarding topics discussed
during this training. CMS will respond to all questions received, and
make available, where appropriate, on our MPPG DPV website Additional
questions can be sent to the RADV email box: Please
make sure that the subject of your email references this training and
that the question is a "follow-up question". We will attempt to address
additional questions and have those responses available on the MPPG/DPV
website, where appropriate. Thank you for participating in the CY 2011
CMS RADV Overview Training. Have a good day.