										
Methodology 										
Overview										
										
										
DATA SOURCES										
"
We use MedPAR FY2023 and merge it with mdprin to get the claims used for rate setting.
We also use the following datasets:
- MBISG data to identify the race of the beneficiary. 
- EDB : To identify LIS, ESRD, Dual enrolled, disabled beneficiary. 
- MedPAR : To identify the beneficiaries with SDOH and behavioral health related diagnosis 
- RUCA : To identify rural beneficiaries.
- ADI and CME : To identify beneficiary living in areas at 85th percentile or above. 

"										
DATA RESTRICTIONS										
"Removed Maryland (ST=""21"") and IHS(PTYPE=8) providers."										
REPORT ASSEMBLY										
										
"
In this workbook, we show the beneficiary share for each provider based on beneficiary characteristics. 

First we identify beneficiary characteristics using the following methodology :
Race: Using MBISG data we identify 6 racial groups. 
We sum up the MBISG assigned Race probabilities by provider. We then divide the total probability by race of a provider by  sum of all the race probabilities of a provider.

Using EDB we identify :
Dual Eligibility: The bene is fully dual or partial dual. Using EDB data, we check if the beneficiary had a dual status during the month of discharge. The dual status is indicated by the following variables : ""B"", ""D"", ""G"" , ""A"", ""C"", ""E"", ""F"",""H"".
LIS status:  We checked the LIS status of beneficiary during the month of discharge. The LIS status is indicated by :'A', 'B',  'C',  'D',  'E',  'F',  'G',  'H',  'I ', 'J', 'K', 'L',  'M',  'N',  'O',  'P',  'Q',  'R', 'S',  'T ', 'U'.
ESRD:  We checked the ESRD status of beneficiary during the month of discharge. The ESRD status is indicated by  ""1"" .
Dual or LIS : Dual or LIS status during the month of discharge.
Disability: orig_mscd :
We checked if the orig_mscd is 20 or 21. 20 is for disabled without ESRD and 21 is for Disabled with ESRD

Using MedPAR we identify :
Social Determinants of Health : The number of beneficiaries billed with diagnosis code starting from Z55 - Z65.
Behavioral Health : The number of beneficiaries billed with diagnosis code starting from F01 - F99.
  
Using RUCA:
The number of beneficiary billed with zip codes that have a RUCA code between 4 - 10 are identified as living in a rural area. For the purpose of this exercise, rural area includes micropolitan, small town , and rural (where census tract has zero population and no rural-urban identifier information) areas only. Not coded areas (ones with a RUCA code = 99) are not included. 




"										
"For ADI:
We use Common Medicare Enrollment Research File (CME_RF1) to get beneficiary's 9 digit zip code. We merge MedPAR with the 9 digit zip code using acumen created beneficiary id. We then merge the MedPAR and ADI information using the 9 digit zip code. Any beneficiary that has ADI greater than or equal to 85 are identified."										
"
Once the beneficiary characteristic is identified, we take the following steps:

Step 1: Using the MedPAR FY2023 claims data and EDB, we calculate the total proportion of discharges for beneficiaries based on their characteristics at the provider level. We do so by calculating total discharges for beneficiaries by their characteristics of a provider. For example, total discharges for LIS beneficiaries of a provider / total discharges for a provider. 





"										
										
										
										
										
										
