Appendix A - Early Consultations

APPENDIX A

EARLY INDUSTRY CONSULTATION

In this appendix, we document the extensive consultation between CMS and the industry between 1997 when the BBA first mandated risk adjustment and the PRA approval in 2002. Even though CMS documented this material in the last PRA’s Supporting Documentation, we append it here again because we believe it is important to not only document communication with industry since the last data collection approval but also to provide a listing of the agency’s efforts since the beginning of policy implementation.

Beginning in 1997, CMS communicated with the M+C (now MA) industry on a continuous, frequent basis using many different forums including the creation of technical user groups and regional training. The agency held discussions with key industry organizations such as: the American Association of Health Plans (AAHP), the Health Insurance Association of America (HIAA), the American Hospital Association (AHA), and the Blue Cross Blue Shield Association (BCBSA); Medicaid directors; the American Medical Association and specialty societies; the National Program for All-Inclusive Care for the Elderly (PACE) Association; the Practicing Physicians Advisory Committee (PPAC); and other interested parties. CMS staff also presented at large national meetings, professional society subgroups, and the Medicare Payment Advisory Commission (MedPAC). Staff presented alternative data collection plans and listened to

industry concerns, responding with revised approaches to data collection as much as possible.

We have divided the history into two sections:

 A table summarizing major forums held from 1997-2002, divided into the time periods before and after the temporary suspension of ambulatory data collection in May 2001;

 A list of key industry concerns and CMS responses when collection of ambulatory data was reinstated in 2002 using new methods that would feed data into the CMS-HCC model for 2004 payment; and

Major Forums, 1997-2002

National training sessions were held to provide the latest data collection information in overview format designed for executive level M+C organization staff. Regional training sessions were, and continue to be, designed for M+C organization technical staff responsible for collection and submission of diagnostic data to CMS. Technical user groups were designed to provide a forum for identification, discussion, and resolution of diagnostic data submission issues related to risk adjustment. User groups were conducted monthly via teleconference. In addition, a public meeting was held on January 16, 2002 at CMS headquarters in Baltimore to provide M+C organizations, providers, practitioners, and other interested parties an opportunity to ask questions and provide comments regarding the risk adjustment model selection for 2004 implementation.

The following table is divided into the period leading up to the suspension of ambulatory data collection (1997 – mid-2001) and the ensuing period leading up to its reinstatement under revised methods (mid-2001 – mid-2002).

_{Industry Consultation October 1997 - May 25 2001}

Type of Consultation	Dates	Status
Preliminary Discussions on Data Collection Approach and Risk Adjustment Methodology	October-December 1997	Complete
Monthly Conference Calls with Plans and Industry Associations (AAHP, HIAA, BCBSA, other)	Began July 1999	Complete
Public Meetings	March 1998; November 1999	Complete
Special Training (e.g. with FIs)	April 1998	Complete
National Training for M+C Organizations (at CMS Central Office)	March 2000; June 2000; September 2000	Complete
Regional Training for M+C Organizations (various sites)	June 2000; July 2000; September 2000; October 2000	Complete
Regional Risk Adjustment Training for Physicians (various sites)	August 2000; September 2000; November 2000	Complete
Technical User Groups	October-December 2000; January-May 2001	Complete
M+C Organization-Specific One-Day Site Visits to Discuss Data Collection and Data Submission	March-April 1999; May- June 2000; April-May 2001	Complete

_{Industry Consultation June 2001 – July 2002}

Type of Consultation	Dates	Status
Monthly Conference Calls with Plans and Industry Associations (AAHP, HIAA, BCBSA, other)	Began July 1999	Complete
Special Discussions with M+C Organizations and Industry Associations on Ambulatory Data Collection and Risk Adjustment Models	June-December 2001	Complete
Special Discussions with M+C Organizations on new Risk Adjustment Processing System (RAPS) Format and Data Submission	January-March 2002	Complete
Special Discussions with M+C Organizations on Physician Training Needs	February-March 2002	Complete
Public Meetings	January 2002	Complete
Regional Training for M+C Organizations (various sites)	June 2002	Complete
Technical User Groups	August 2001-July 2002	Complete
M+C Organization-Specific One-Day Site Visits to Discuss Data Collection and Data Submission	May-June 2001; March- May 2002	Complete

_{Industry Consultation October 1997 - May 25 2001}

Type of Consultation	Dates	Status
Preliminary Discussions on Data Collection Approach and Risk Adjustment Methodology	October-December 1997	Complete
Monthly Conference Calls with Plans and Industry Associations (AAHP, HIAA, BCBSA, other)	Began July 1999	Complete
Public Meetings	March 1998; November 1999	Complete
Special Training (e.g. with FIs)	April 1998	Complete
National Training for M+C Organizations (at CMS Central Office)	March 2000; June 2000; September 2000	Complete
Regional Training for M+C Organizations (various sites)	June 2000; July 2000; September 2000; October 2000	Complete
Regional Risk Adjustment Training for Physicians (various sites)	August 2000; September 2000; November 2000	Complete
Technical User Groups	October-December 2000; January-May 2001	Complete
M+C Organization-Specific One-Day Site Visits to Discuss Data Collection and Data Submission	March-April 1999; May- June 2000; April-May 2001	Complete

_{Industry Consultation June 2001 – July 2002}

Type of Consultation	Dates	Status
Monthly Conference Calls with Plans and Industry Associations (AAHP, HIAA, BCBSA, other)	Began July 1999	Complete
Special Discussions with M+C Organizations and Industry Associations on Ambulatory Data Collection and Risk Adjustment Models	June-December 2001	Complete
Special Discussions with M+C Organizations on new Risk Adjustment Processing System (RAPS) Format and Data Submission	January-March 2002	Complete
Special Discussions with M+C Organizations on Physician Training Needs	February-March 2002	Complete
Public Meetings	January 2002	Complete
Regional Training for M+C Organizations (various sites)	June 2002	Complete
Technical User Groups	August 2001-July 2002	Complete
M+C Organization-Specific One-Day Site Visits to Discuss Data Collection and Data Submission	May-June 2001; March- May 2002	Complete

_{Industry Consultation 2003}

Type of Consultation	Dates	Status
Quarterly Risk Adjustment Workshops	Fall, Winter, Spring, Summer 2003	Complete
Regional Risk Adjustment Data Training for MA organizations	April – July 2003	Complete
Risk Adjustment User Groups	August 2003 – July 2003	Complete

_{Industry Consultation 2004}

Type of Consultation	Dates	Status
Quarterly Risk Adjustment Workshops	Fall, Winter, Spring, Summer 2004	Complete
Regional Risk Adjustment Data Training for MA organizations	June – July 2004	Complete
Risk Adjustment Training for MA Organizations Special Sessions (Subjects: Risk Adjustment Methodology, Data Validation, Diagnosis Codes & Risk Adjustment, 3C’s of Risk Adjustment)	August 10, 12, 17, and 19, 2004 September 14 and 23, 2004	Complete
Risk Adjustment User Groups	August 2004 – July 2004	Complete
		Complete

_{Industry Consultation 2005}

Type of Consultation	Dates	Status
Medicare Advantage and Prescription Drug Application and Bid Training	January – February 2005	Complete
Regional Risk Adjustment Data Training for MA and MAPD organizations	June – August 2005	Complete
Prescription Drug Event (PDE) Data Regional Training for MAPD organizations	June – August 2005	Complete
Risk Adjustment User Groups	August 2005 – July 2005	Complete

_{Industry Consultation 2006 – 2008}

Type of Consultation	Dates	Status
Regional Risk Adjustment Data Training for MA organizations	February and July 2006, July – August 2007, July – August 2008	Complete
Monthly Risk Adjustment Training for MA organizations	September 2006 – August 2008	Complete
Prescription Drug Event (PDE) Data Regional Training for MAPD organizations	July 2006, July – August 2007, July – August 2008	Complete
Prescription Drug Event (PDE) Data Regional Training for PACE organizations	July 2005	Complete
Risk Adjustment User Groups for MA organizations	January – December 2006 January – December 2007 January – December 2008	Complete

_{Industry Consultation 2009 – 2010}

Type of Consultation	Dates	Status
Risk Adjustment User Groups for MA organizations	January – December 2009 January – July 2010	Complete

_{Industry Consultation 2011}

Type of Consultation	Dates	Status
Risk Adjustment Regional Training for MA Organizations	June – August 2011	Complete
Prescription Drug Event (PDE) Data Regional Training for MAPD organizations	June – August 2011	Complete

_{Industry Consultation 2012}

Type of Consultation	Dates	Status
Risk Adjustment Regional Training for MA Organizations	August 2012	Complete
Prescription Drug Event (PDE) Data Regional Training for MAPD organizations	August 2012	Complete

_{Industry Consultation 2013}

Type of Consultation	Dates	Status
Getting Started Risk Adjustment Training for MA Organizations	May 2013	Complete
Prescription Drug Event (PDE) Data Regional Training for MAPD organizations	2013	Complete
Risk Adjustment Training for MA Organizations	2013	Complete

Key Concerns and Agency Response, 2001-2002

Beginning with the January 15, 1999 announcement of the PIP-DCG methodology, CMS announced its intention to implement a comprehensive risk adjustment method and began an intensive, iterative process of consultation with the industry. In January 2000, Medicare implemented risk adjusted payments to M+C organizations basing payments in part on diagnostic information from inpatient hospital discharges. The inpatient hospital risk adjuster (PIP-DCG) was viewed as an initial step in the implementation of a more accurate risk adjustment methodology that would incorporate diagnoses received from ambulatory settings.

CMS initially implemented an encounter-based data collection system. This approach required M+C organizations to electronically submit a record of each service provided to each enrollee using standard (but abbreviated) Medicare reporting formats. Because of concerns over the burden of collecting ambulatory encounter data, CMS temporarily suspended the collection of these data on May 25, 2001 through June 30, 2002. However, BIPA still required CMS to incorporate ambulatory data with inpatient data for January 2004 risk adjusted payment. Therefore, CMS developed an improved risk adjustment methodology that incorporated ambulatory data with inpatient data while reducing data collection burden.

Summary of concerns

Primary concerns were:

 The data collection system was based on all encounters received from hospital inpatient, hospital outpatient, and physician settings.

_{ The submission requirements were based on Medicare fee-for-service (FFS) claims formats.}

 The formats were required based on the need to perform all edits within Medicare claims processing systems.

_{ M+C organizations were required to submit data, such as the Unique Physician Identification} Number (UPIN), type of bill, procedure codes, and other data for model maintenance and data verification. These and other data elements were edited and caused rejections even though they were not required for risk adjustment payment.

Summary of Response

In response to these concerns, CMS began to redefine radically the data collection and data submission process used for risk adjustment. First, after discussions with interested parties, CMS transformed the approach originally used (encounter-based reporting) to data reporting for purposes of calculating risk adjustment factors and payments only. This approach allowed CMS to reduce drastically the requirements for the amount of data submitted, the data formats used, and the data processing systems that would be utilized. CMS also decided to only require M+C organizations to submit the diagnoses required to make accurate risk adjustment payments. These two decisions allowed CMS to develop a new, more flexible and less burdensome data collection strategy and processing system. A number of other parameters of the approach were redesigned as well.

Detailed Listing

The following section provides detailed descriptions of how CMS addressed the primary concerns of M+C organizations and made substantial changes to risk adjustment data collection, data submission, and data processing.

1. Data collection for risk adjusted payments was based on all encounters received from hospital inpatient, hospital outpatient and physician settings.

CMS RESPONSE

The requirements for data collection previously established required that all encounters from the hospital inpatient, hospital outpatient, and physician settings were to be submitted by M+C organizations on a monthly basis, at a minimum. CMS addressed this issue by requiring quarterly submissions based on a 12-month data collection period. Also, M+C organizations were only required to submit each beneficiary-specific diagnosis once during a data collection period regardless of service setting. However, CMS allowed diagnoses to be submitted more frequently if the M+C organization wished to submit diagnoses based on number of encounters received.

2. The submission requirements were based on Medicare FFS claims formats. CMS RESPONSE

The previous data submission method necessitated the exclusive use of Medicare FFS claims formats such as the UB-92 (v6.0), ANSI X12 837 (v3051 or v4010), NSF (v3.0). Smaller plans were allowed to use a Medicare FFS-based software package, PC-ACE, to generate a Medicare FFS compliant form.

CMS addressed the data submission issue by allowing flexibility in use of submission formats. CMS developed the Risk Adjustment Processing System (RAPS) format specifically for M+C organizations and the collection of data for risk adjustment. This format requires M+C organizations to provide only the data required for risk adjustment by implementing a non- traditional format. Moreover, M+C organizations could use superbills to collect data for risk adjustment on periodic or encounter bases. M+C organizations could then submit these data via the RAPS format.

Specifically, CMS examined the data submission formats required for system processing and made changes to allow M+C organizations more flexibility in choice of submission format.

With RAPS, M+C organization data submission to CMS could be accomplished by one or more of the following methods:

1) full or abbreviated UB-92 Version 6.0

2) full or abbreviated National Standard Format (NSF) Version 3.0

3) ANSI X12 837 Version 30.51 (only for those submitters currently utilizing this version)

4) ANSI X12 837 Version 40.10

5) the new Risk Adjustment Processing System (RAPS) format

6) on-line direct data entry (DDE)

These changes allowed M+C organizations a number of options for submission and did not require one type of submission format. That is, each M+C organization could select the most efficient method for data submission, taking into account the unique nature of its data systems. M+C organizations could elect to utilize more than one submission method. All transactions were submitted using the same network that M+C organizations currently utilize for hospital inpatient data submission.

Regardless of the method of submission that a M+C organization selected, all transactions were made subject to the same edits. The Front-End Risk Adjustment System (FERAS) now automatically formated all DDE transactions into RAPS format. Transactions submitted in claim or encounter formats were converted to the RAPS format prior to system editing.

3. The Medicare FFS claims format necessitated the use of Medicare FFS claims processing systems.

CMS RESPONSE

Originally, CMS chose to utilize existing Medicare FFS standard processing systems to process and edit the incoming risk adjustment data. This approach to handling data became extremely burdensome to M+C organizations that were not accustomed to Medicare FFS processing systems and were not collecting many of the data elements needed to pass system edits. As mentioned above, in order to adequately address this issue, CMS created the RAPS format and processing system.

This system requires fewer data elements. The required RAPS format data elements are:

• Health Insurance Claim (HIC) Number

• Provider Type (hospital inpatient-principal diagnosis, hospital inpatient-other diagnoses, physician, and hospital outpatient )

• _{Service From Date}

• Service Through Date

• _{Diagnosis Code(s)}

This step eliminated all data elements that were not required to run the risk adjustment model, such as Unique Physician Identification Number (UPIN), procedure codes, and type of bill.

4. M+C organizations were required to submit additional data (e.g., UPIN, procedure codes, etc.) that were not required in running the risk adjustment models (currentPIP-DCG model and proposedsite neutral model). The additional data elements were edited and caused risk adjustment data to be rejected even though these data elements were not required for the risk adjustment model.

CMS RESPONSE

As discussed in issue #3 above, in order for the previous data processing approach employed by CMS to work, M+C organizations were required to submit data elements that were irrelevant for risk adjusted payments. Data elements such as UPINs and procedure codes were required for successful data processing. The data elements had to be valid and in the correct format for processing in the Medicare FFS systems. These data elements were cumbersome for the M+C organizations to collect and maintain and could delay submission and successful processing of data that was necessary for risk adjustment payment.

The number of edits required for the new processing system (RAPS) was drastically reduced from the number of edits required for the Medicare FFS processing systems that were employed for the previous ambulatory data collection. The number of edits was reduced from over 1,000 for all Medicare FFS processing systems to approximately 25 for the new RAPS system.

9