Nivedita "Nive" Baliga
High Level Review of Medicare Data
The Medicare-Medicaid Coordination Office, a Department of Federal CMS, makes Medicare Parts A, B, and D data available to states for care coordination purposes.
Illinois completed the six month application about a year ago and has been working to prepare the data for use by our partners for the past nine months.
The planning for sharing Medicare claims data through CCCD occurred at the same time as that for Medicaid claims. Since the inception of CCCD, the plan has been to include Medicare data as part of CCCD once it was available.
The goal is to provide Medicare data in the March CCCD run (the fifth working day of March is 3/7/2014). There are multiple challenges related to this March goal.
There was a separate application process for historical Parts A & B, a different one for current Parts A&B, and yet another for Part D.
Similar to differences in application processes, there are significant differences in how data is transported to states. These differences include:
Different contractors for delivery and support for each Medicare data "type".
Different delivery schedules.
Different supporting documentation.
Different file formats that varied by data type.
Two majors steps to sharing Medicare data:
Upload the data to HFS data warehouse
Merge the data together and format into CCCD
The uploading of data to the data warehouse is finished. The focus is now formatting it into CCCD. Again, the goal is to start to deliver Medicare data in March. Things may develop over the next few weeks that complicate our ability to meet this goal.
All supporting CCCD documentation will be updated to include Medicare by the mid February. Please check the HFS CCCD webpage for updates.
The current version of the CCCD Data Dictionary now includes which fields will and will not have Medicare data. Again, there may be updates to this as we finalize things over the next few weeks.
Please submit any Medicare questions you have to Paul Stieber.
Let's now move to outstanding questions. . .
#1 Why does dx code have a length of 8?
Question: Why does the field diagnosis-cd have a length of 8? Can you please provide few examples for such diag codes.
Answer: The length is because there are older claims in the HFS data warehouse (pre-HIPAA) that are 8 digits. In order to maintain the integrity of the older data, the field length is always kept at 8. Since the oldest claims in CCCD are from 2007 (7 years old), there are no claims that are 8 digit codes.
#2 Can Amount Fields be negative?
Question: Do you think amount fields could be negative, if yes what would be the position of sign?
Answer: The only amount fields that can be negative are Long Term Care (LTC) claims. They can be negative at the line level. However, if you add up all line of an LTC claim for one DCN, the total should not be negative.
Question: We see same POS_CD is mapped to multiple standard POS_Cd. Can you please tell us how to use this xref and pick one out of many? Eg: Current 'K' is mapped to Unconverted (12,13,14)
Answer: Place of Service is converted from 2 bytes to 1 byte in order to fit into our old system. Therefore, the HFS PlaceOfServiceCd will not map 1:1 to HIPAA standard codes. We've translated HFS PlaceOfServiceCds to the HIPAA standard codes as best as we can in the crosswalk posted on the HFS CCCD website.
#4 Control File Layout
Question: We need the begin and end position for each field in control file for MMAI and ICP.
Answer: All data is shared through the single CCCD dataset. There is not a different one for MMAI and ICP. The file begin and end positions for this one control file are as follows:
FileType CHAR (25) Positions 1-25
LoadDt Date Format 'YYYY-MM-DD' Positions 26-35
RecordCount Integer Positions 36-45
Future CCCD Partner Meetings
Wednesday, 2/19/2014; 10:00-11:00 AM Central
Wednesday, 3/12/2014; 10:00-11:00 AM Central
Need for future meetings after March is being assessed
Please send feedback. . .
Complete survey at the end of this webinar and/or E-mail Paul Stieber.