Direct Care Data in M2

1
Direct Care Data in M2
2
Objectives

• Describe the characteristics of Direct Care
  Encounter-Level data files (Inpatient Admissions,
  Professional Encounters, Laboratory Ancillary,
  and Radiology Ancillary).
• Given a management purpose, select or avoid data
  elements of the direct care encounter data files
  in accordance with their applications and
  limitations.
• Describe appropriate use of cost and workload
  measures.

3
SIDR, SADR, and Ancillary

• Direct Care Data records
• Only completed records
• Most recent version of record
• Detail and Summary Files
• SIDR (Inpatient Admissions) One record per MTF
  disposition.
• SADR (Professional Encounters) One record per
  MTF ambulatory encounter or inpatient
  professional service.
• Lab/Rad Ancillary One record per entry into the
  CHCS system (per lab test or per radiology exam).

4
Direct Care Data Flow
Monthly
SIDR
MDR
CHCS
SADR
Weekly
At MTFs
M2
Lab/Rad
Monthly
5
Location of Direct Care Data on M2
Ancillary
SIDR
SADR
6
Records in M2
As of 2/9/07
7
Common M2 Data Elements for SIDR, SADR, and
Ancillary Lab-Rad
8
Record ID / M2 Key

• Record ID Unique number that identifies a record
  at an MTF (SIDR/SADR only).
• Inpatient Admissions Detail CHCS Patient
  Register Number (PRN)
• Professional Encounters Detail CHCS Appointment
  ID
• Use in combination with Tmt DMIS ID
• M2 Key Field in Ancillary that uniquely
  identifies a record.

9
Patient Identification

• Sponsor ID Sponsor Social Security Number
• Pseudo Sponsor ID.
• DEERS Dependent Suffix (DDS) SIDR and SADR only
• Relationship of patient to sponsor.
• DDS is unique, but may not be consistent with
  FMP.
• Family Member Prefix (FMP) SIDR, SADR,
  Ancillary
• Relationship of patient to sponsor.
• Coding scheme same as DDS, but values assigned
  differently for non-sponsors.
• FMP uniquely assigned for data coming from CHCS,
  but not necessarily unique across CHCS Hosts.

10
Person ID

• Person ID
• Electronic Data Interchange Person Number
  (EDI_PN)
• Available in M2 in data FY02-present
• Populated 95 in SIDR in all FYs
• Populated 99 in SADR for all FYs
• Person ID considered HIPAA compliant

11
Patient Demographics

• Date of Birth (MM/DD/YYYY)
• Age some have Age Group Code and Age Group
  Common
• Medicare Eligibility
• FY03 A, B, C, N
• FY02 and backwards Y, N
• Gender
• Sponsor Pay Grade Only for users with
  authorization to PHI
• Sponsor Rank Group
• Sponsor Service
• Sponsor Service Aggregate on Ancillary as well

12
Sponsor Service
13
Beneficiary Category

• Patient Category (aka, PATCAT)
• SIDR and SADR only (e.g., A11 Army AD)
• Beneficiary Category based on PATCAT
• ACT, DA, DCO, DGR, DR, DS, GRD, IDG, IGR, NAT,
  OTH, RET, UNK
• E.g., PATCAT A11 mapped to ACT
• Dependent of Guard/Reserve (DGR) is identifiable
• Ben Cat Common
• 4Active Duty and Guard
• 1Active Duty Family Member, including Guard
• 2Retirees
• 3All Others (including IDG and IGR)

14
Ben Cat Common
15
Enrollment Data

• Alternate Care Value (ACV)
• TRICARE Program in which beneficiary is enrolled
  Values
• Prime A, E (and D if using FY02 older data)
• Overseas Prime B, F, H, J
• Plus G, L
• Non-enrolled Active Duty M (10 in FY05 SADR)
• Blank or Null not enrolled
• USFHP should not be in data

16
Enrollment Data

• Enrollment Site DMISID of enrollment at time
  care delivered, or NONE
• Populated Enrollment Site does not mean
  beneficiary enrolled in Prime. (Could be Plus.)
• Enrollment Site Attributes
• If you offer TRICARE Plus, TRICARE Plus enrollees
  will be coded with your Enrollment Site as the
  DMIS ID
• Filter on ACV to limit to Prime or Plus
• PCM ID Primary Care Manager ID
• Even though ACV shows they are enrolled, there
  are instances where PCM ID is blank

17
Provider Information

• Tmt DMIS ID Attributes
• MEPRS Code of MTF department providing care
• SADR 1 MEPRS Code (1, 2, 3, or 4 places avail.)
• SIDR Up to 4 MEPRS Codes (2 3 places only)
• Admitting, Dispositioning, Up to 2 Interim
• If patient in more workcenters, info not in M2
• Lab/Rad 1 MEPRS Code (3 and 4 places avail.)
  This is the ordering MEPRS

18
MEPRS Clinical Services (SIDR)
Ex Clinical Service (MEPRS) Clinical Service (MEPRS) Clinical Service (MEPRS) Clinical Service (MEPRS)
Ex Admit 2nd 3rd Disp
1 AAA AAA
2 AAA ABC AEA AEA
3 AAA AAH ABC ABA
4 AEA ABC ABA AAA
19
Geography Related Fields

• Beneficiary Zip Code As in CHCS at time of
  service.
• Beneficiary Region (01-16, AK, etc.)
• Beneficiary HSSC Region (N, S, W, O) FY03
• HSSC Health Services and Support Contract
• Catchment Area/PRISM Area ID
• Multi-Service Market Area (MSMA) fields
• Market Area ID Specific to TRO
• TPR Flag Indicates whether beneficiary resides
  in a TRICARE Prime Remote designated area

20
Diagnosis Codes

• ICD-9 CM diagnosis codes
• 8 codes in SIDR Diagnosis 1 Diagnosis 8, where
  Diagnosis 1 is the principle diagnosis
• 4 codes in SADR Diagnosis 1 Diagnosis 4
• 4 codes in Ancillary Diagnosis 1 Diagnosis 4,
  but currently no data available (placeholders)
• No decimal points
• To find records for beneficiaries with a
  particular condition, check all diagnosis code
  positions

21
Diagnosis Filter Example
Diagnosis 1 matches pattern 250 OR Diagnosis
2 matches pattern 250 OR Etc.
22
Direct Care Inpatient Admissions
23

Inpatient Admissions Detail

• Each record represents the most recent version of
  hospital record for a patient dispositioned from
  an MTF
• Only completed records are in M2
• MTFs collect other types of inpatient records but
  they are not included in M2
• Long Stay patients included in the M2 once
  discharged from the hospital (Note inconsistent
  with Purchased Care data)

24

• Reporting lag inherent in inpatient data
  collection
• Requirement to complete SIDR w/in 30 days of
  disposition Not always met
• Lag can lead to incomplete data
• Completion Factors computed at the Tmt DMISID/FM
  level based on of SIDRs and dispositions
  reported in WWR
• Raw what has actually been received
• Total Raw estimated to completion, what is
  expected
• Total not populated if data less than 40
  complete
• Use Total measures to account for missing data,
  as most management questions require

25
SIDR Raw vs. Total Records

• FYs can look odd because of non-DHP MTF
  reporting SIDRs for which there is no WWR

As of 2/9/07
26

• Procedure 1 Procedure 8 ICD-9 CM
• Codes are left justified and decimals not
  included.
• Number of times a particular procedure was
  performed is also available
• DRG TRICARE Diagnosis Related Group
• Based on ICD-9 CM diagnosis procedure codes.
• One DRG per record
• Can change across FYs (modified, deleted, etc.)

27

• MDC Major Diagnostic Category
• 25 codes for body system being treated
• Based on primary diagnosis
• MDC is broadest grouping, retrieves records
  related to a diagnosis involving a body system.
  DRGs are more specific, ICD-9 CM
  diagnosis/procedure codes even more so

28

• MDC 14 (Pregnancy, Childbirth, and the
  Puerperium)
• DRG 371 (Cesarean Section w/o CC)
• ICD-9 Diagnosis 65421 (Previous Cesarean
  Delivery)
• What to use?
• All OB care, not just deliveries? MDC
• Just deliveries? DRG
• Specific kind of delivery? Diagnosis Code

29
Preventable Admissions

• Does not mean the particular admission could be
  prevented
• Used as an access measure or to identify
  potential case management or enrollment
• Values based on type of care (asthma, bacterial
  pneumonia, cellulitis, etc)
• Goal is low rates of these admissions for
  enrollees.
• These are not the Agency for Healthcare Research
  and Quality (AHRQ) Prevention Quality Indicators

30
Administrative Data

• Source of Admission How patient arrived in
  hospital
• Transfers into an MTF (4-8)
• Admissions from ER (0)
• Live Births (L)
• Admissions from same day surgery (S)
• Routine admission (1)

31
Source of Admission
32
Discharge Status

• Disposition Status Code
• Transfer Out (02 05)
• Left Against Medical Advice (07)
• Death in Hospital (20)
• Routine (01)

33
Disposition Status Code
34
Date Fields

• Admission Date Use to stratify admissions (Note
  that patients still in hospital not in M2.)
• Service Date Date of disposition
• FY/FM and CY/CM year and month of disposition
  date

35
Cost and Workload Measures

• Raw and Total
• Most measures have both a , Raw and ,
  Total
• Total Raw estimated to completion
• Disposition, Raw count
• Each record is a disposition
• Set to 1 for all rows in table

36
Cost and Workload Measures

• Bed Days
• Days as reported from MTF
• Some inconsistencies in reporting (e.g., some
  MTFs have 0 bed days when admission date
  disposition date)
• Bed Days, Raw (and thus Total) have 0 (zero) days
  changed to 1 bed day
• Bed Days of Record the 0 bed days have not been
  changed
• Many detailed bed day fields in M2 beyond total
  days. See Data Dictionary

37
Days by Clinical Service
Work Center (Clinical Service) Work Center (Clinical Service) Work Center (Clinical Service) Work Center (Clinical Service) Days Days Days Days Days
Adm 2nd 3rd Disp Adm 2nd 3rd Disp Total
AAA     AAA 44     44 44
AAA AFA   AFA 1 35   35 36
AAC AAB ABC AAB 2 12 3 12 17
AAA AAH ABB ABC 2 0 4 6 13
Summing across the clinical bed days does not
equal total bed days for the hospitalization
38

Cost and Workload Measures

• RWP Relative Weighted Products
• Intensity adjusted workload measure
• Based on DRG weight (i.e., relative costliness of
  that DRG, hospital costs only), LOS, and other
  factors
• RWP DRG weight for most cases although Long
  Stay Outliers receive a little more for each day
  over the threshold Short Stay Outliers receive a
  little less
• BEST WORKLOAD MEASURE
• Case Mix RWP / DISP, where DRG is not 469 or
  470 (ungroupable DRGs)

39
RWP Calculation
DRG Wgt ALOS GMLOS SST LST Per Diem
109 Coronary Bypass w/o Cardic Cath 3.9431 6.2 5.7 2 19 0.6918
40
Workload Measures
Even though San Diego had double the number of
dispositions than Walter Reed, Walter Reed
actually expended more resources (CMI)
41

Top 10 DRGs (Dispositions) - FY06
FY05 produces the same results
42

Top 10 DRGs (RWPs) FY06
FY05 produces the same results
43

Cost Data

• Based on historical MEPRS data combined with
  direct care encounter records. Inflation applied
  (Patient Level Cost Allocation or PLCA
  methodology)
• Incorporates hospital cost, as well as
  ancillaries, professional services and pharmacy
• Cost allocated to cases based on DRG weight,
  whether the patient had surgery, costliness of
  provider, and other key factors

44
Full Cost vs. Variable Cost

• Full Cost All portions of the hospitals cost.
• All MEPRS A expenses which include Rx,
  salaries, supplies, etc
• Variable Cost Cost to do one more
• All MEPRS A expenses identified as variable
  costs based on the Standard Expense Element Code
  (SEEC) Lights vs. Laundry
• FY05 and backwards based on MEPRS expense data
  from respective FY
• FY06 and forward based on FY05 MEPRS expense
  data, inflated to appropriate FY

45
Full Cost vs. Variable Cost

• Full and Variable Cost broken out by
• Direct (less salary) based on bed days
• Support based on bed days
• Laboratory based on laboratory weighted workload
  by DRG
• Radiology based on radiology weighted workload
  by DRG
• Other Ancillary based on RWP
• Clinician Salary based on professional weighted
  product
• Other Salary based on bed days
• ICU based on ICU days
• Surgical based on surgical DRG weight
• Sum of components equal Full Cost / Variable Cost
• Covered more thoroughly in other sections

46
Normative Data - Norms, Raw and Total
Civilian MTF MTF Peer
Bed Days X X X
Deaths X X X
Variable Cost NA X X
Full Cost NA X X
For example, your MTFs expected number of bed
days vs. other MTFs or Civilian data
47
Questions ?