S02 - Attachment 15 - FY21 EPRP Sampling Cohorts v1.0.docx

Attachment 15 FY 2021 EPRP Sampling Cohorts The External Peer Review Program (EPRP) samples Veterans by cohort. A cohort is a group of Veterans sharing a defining characteristic, such as diabetes. There are 18 cohorts in the EPRP sampling process for FY2021. The Cohorts are numbered for ease of reference. In the past, these numbers were referred to as Category Numbers or CATNUMS. Moving forward, they will be referred to simply as Cohort Numbers.

Outpatient Cohorts A Veteran who meets the following criteria is eligible for selection in the outpatient random sample:

Veteran who was not already reviewed during the current fiscal year.

Veteran had an outpatient encounter[footnoteRef:2] anywhere in VHA recorded in the Corporate Data Warehouse (CDW) at least once during the 13-24 months prior to the study month. This encounter is considered an “anchoring” visit which identifies an “established” Veteran. [2: An outpatient encounter is an outpatient workload visit in a “count” clinic where a provider sees a patient, writes diagnoses (ICD-10) and procedures (CPT), and documents the encounter in a progress note. A detailed explanation for workload can be found on the BISL CDW website: https://vaww.cdw.va.gov. EPRP further narrows an outpatient encounter to Veterans who were not seen as inpatients (i.e., the appointment status is not “Inpatient Appointment”).]

Veteran had an encounter in the study month in any Nexus designated clinic (the clinic stop code may be in either primary or secondary position).

Veteran must have had at least one encounter from 12 months prior to the beginning of the study month up to the end of the study month with a practitioner in any of the following categories: Allopathic and Osteopathic Physician, M.D. or D.O. Physician, Physician Assistant, Advanced Practice Nursing, or Psychologist.

A Nexus can be defined as a connected group. For purposes of EPRP, Nexus refers to a designated group of clinics having a specific stop code (primary or secondary) as shown in the tables below.

Approximately 17% of the EPRP Nexus sample represents Veterans whose encounters were at more than one VAMC in the year prior to the sample month. In these cases, the VAMC assignment is based on the VAMC where the majority of Nexus encounters occurred in the 12 months prior to and including the study month. Should there be an equal number of encounters across facilities, accountability is assigned to the facility where the Veteran had his or her latest Nexus encounter. The accountable division within a facility is assigned based on where the most primary care visits occurred. Accountability is randomly assigned should there be an equal number of visits across divisions.

A listing of clinic stops excluded from Nexus sampling can be found in Appendix A.

Non-Mental Health Nexus Clinics (Cohort 50)

Mental Health Nexus Clinics (Cohort 51)

533[footnoteRef:3] [3: For use by MH clinicians who provide individual care, primary diagnosis is medical rather than psychiatric, e.g. Chronic pain, Essential Hypertension, Low Back Pain, Migraine Headache, Obesity, etc.]

Cohort 50/Nexus – Non-Mental Health subgroup: Veterans who had a workload encounter in the study month in any of the Nexus clinics and who are not in the Mental Health (MH) subgroup. Cases are flagged for Advanced Illness and/or Frailty as exclusion criterion for the following measures: C9H, CVRM1, CVRM2, DMG23H, DMG27H, DMG34H, IHD20H, IHD53H, P32H, P33, P61H Cohort 51/Nexus – Mental Health subgroup: Veterans who had any mental health (MH) diagnosis as either an inpatient (primary or secondary position) or outpatient, at least one time in the previous 12 months.

Cases with documentation of an ICD-10-CM mental health diagnosis within the past year are designated by a MH flag. This includes SCI Outpatient. A full list of codes can be found in the FY21 EPRP Sampling Cohorts Appendix, located on the EPRP Resources page. These cases may be excluded from designated outpatient measures if there is medical record documentation that only Behavioral Health Care was received in the VHA with explicit documentation of refusal of VHA Primary Care, and that primary care was received in a non-VHA setting (enables question OTHRCARE located in CGPI Validation module). For a list of measures associated with this cohort, see Table 3 below.

Cohort 16/Outpatient AMI: Includes Veterans with an outpatient AMI ICD-10-CM diagnosis code I252 recorded during the sampling month, including outpatient SCI. Cases are flagged to indicate previous AMI. This is an augmented sample. In an augmented sample, an additional number of cases are added to the initial cohort sample to ensure that enough records are allocated to accurately represent a facility’s Veteran population. The cases are then weighted to adjust for any facility bias. The weights are unique to each case at the time of sampling and can be viewed in the Combined Measure Master (CMM). Final scores incorporate weights to correct for this oversampling.

Cohort 36/SCI Unit subgroup: Includes Veterans who are randomly selected from both non-SCI center hospitals (with a minimum 1 case selected per facility) and SCI center hospitals. If there are not enough cases to achieve the minimum facility target in a given sample month, then those cases available that were not previously selected in the prior two months are used. Veterans can only be selected once in a fiscal year (Veteran-based denominator). Veterans sampled in prior months for the current fiscal year (i.e., within the last 2-3 months of the previous fiscal year or an outpatient cohort) are not included in any given monthly sample.

All Nexus outpatient rules apply to SCI cases (i.e., must be an established Veteran, seen by provider). The outpatient sample selects Veterans with:

Outpatient encounter in the sample month and ICD-10-CM code found in the FY21 EPRP Sampling Cohorts Appendix, during the sample month and Match with master list of SCI Veterans maintained by the SCI Program Office and 210 or 215 clinics stop codes for visit during the past year The above algorithm is used to identify true SCI Veterans known to the SCI Program. It excludes Veterans with paraplegia and quadriplegia that do not have a spinal cord injury. The Master list of SCI Veterans is updated periodically by the SCI Program Office.

Cohort 48/Women: Includes women ages 20 – 69 who have been seen in the study month in any of the Nexus clinics and who had an anchoring visit 13-24 months prior to the study interval. This cohort augments the random Nexus sample for gender-related indicators. This is an augmented sample. In an augmented sample, an additional number of cases are added to the initial cohort sample to ensure that enough records are allocated to accurately represent a facility’s Veteran population. The cases are then weighted to adjust for any facility bias. The weights are unique to each case at the time of sampling and can be viewed in the Combined Measure Master (CMM). Final scores incorporate weights to correct for this oversampling.

Cohort 54/Frail Elderly: Includes Veterans age 75 and older. Cases are flagged Cohort 54 (Frail Elderly) for full CGPI review. This is an augmented sample. In an augmented sample, an additional number of cases are added to the initial cohort sample to ensure that enough records are allocated to accurately represent a facility’s Veteran population. The cases are then weighted to adjust for any facility bias. The weights are unique to each case at the time of sampling and can be viewed in the Combined Measure Master (CMM). Final scores incorporate weights to correct for this oversampling. For a list of measures related to this cohort, see Table 3.

Cohort 60/Diabetes Mellitus: Includes Veterans with a Diabetes Mellitus (DM) diagnosis recorded during the sample month. This is an augmented sample and includes inpatient and outpatient SCI. In an augmented sample, an additional number of cases are added to the initial cohort sample to ensure that enough records are allocated to accurately represent a facility’s Veteran population. The cases are then weighted to adjust for any facility bias. The weights are unique to each case at the time of sampling and can be viewed in the Combined Measure Master (CMM). Final scores incorporate weights to correct for this oversampling. The encounter during the sample month may be an inpatient admission or outpatient encounter. If the encounter during the sample month is an outpatient encounter, there must be documentation of another outpatient encounter or an inpatient admission with an ICD-10-CM code of diabetes in the past two years. See Table 3 for a list of measures associated with this cohort. A list of DM ICD-10-CM codes can be found in the FY21 EPRP Sampling Cohorts Appendix.

Cohort 61/Inpatient Sample for Spinal Cord Injury and Disorder (SCI&D): Only Veterans in SCI center hospitals are included in this sample. The inpatient sample selects Veterans discharged from the SCI bed section (#22). SCI Veterans in LTC setting are not sampled. Veterans are eligible for selection only once in a year within an SCI Center (Veteran-based denominator). Veterans sampled in prior months in the previous FY are not included in any given monthly sample. If there are not enough cases to achieve the monthly facility target, then cases available but not previously selected in the prior month are used. A fixed number of cases in each SCI Center are sampled each sample month (eight per month per SCI center, total 75 cases per year). Almost all sampled cases are eligible for inclusion in the inpatient measures (i.e. approximately 95% of cases are eligible for inclusion). All cases sampled from the SCI Centers for the inpatient measures are also included in the denominator of the SCI outpatient measures.

New Number, same cohort Cohort 69//Home Care: Includes Veterans who show evidence of active Home-Based Primary Care (HBPC) status during the month of study. Active status is determined by a documented encounter in a HBPC stop code (primary position) of 170 through 177 during the month of study. The list for selection is created by searching 30 to 60 days prior to the most recent stop code used. The eligibility is verified by the clinical reviewer. Veterans are sampled only once in a fiscal year, which may limit the sample for small facilities. An attempt to sample 10 Veterans is made for most facilities. Facilities with 40 or fewer unique Veteran encounters in the study month have a sample size of 6, if possible. Facilities with more than 200 unique Veterans have a sample size greater than 10 with a maximum number of 15. Veterans with a SCI diagnosis (see SCI cohort) used in the study month are excluded from the sample. Evidence that HBPC services were discontinued or that the Veteran expired are not included for review.

New Number, same cohort Cohort 80/ Communication of Test Results (CTR): Quarterly Random sample of Veterans that had one of the following outpatient tests reported during the study month that was ordered by a VA provider: Fecal Occult Blood Test/Fecal Immunochemical Test (FOBT/FIT), Hepatitis-C Virus (HCV), Human Immunodeficiency Virus (HIV), Alpha-fetoprotein (AFP), Dual Energy X-ray absorptiometry (DEXA) Scan, Mammogram, Pap/Human Papillomavirus (HPV), Computerized Tomography (CT) Scan, and Chest X-ray. The sample excludes tests performed in the community. The sample includes normal and abnormal test results. Positive FOBT/FIT test results are excluded from the sample if the test result was positive within the past 6 months. These cases can be flagged for the following:

MHV enrollment indicating a My Healthy Vet account and used to auto fill test results communicated question, method, date of communication of test results;

Result (value or pos/neg) identifying test results and auto fill software;

Category test identifies the test and is used to identify test for auto-filling date / result of test.

This Cohort is sampled quarterly.

Table 1: Outpatient Cohorts and EPRP Data Set Question (DSQ) Modules* for FY21 Cohorts and EPRP Data Set Question Modules Outpt AMI SCI Clinics Women 50, 51 Nexus Frail Elderly

DM

Inpt SCI

CBOC

HBPC

CTR

CGPI Validation[footnoteRef:4] [4: ]

HBPC

x

Med Rec[footnoteRef:5] [5: ]

CTR

x

*EPRP Data Set Question Modules (DSQ)-question sets are used in the abstraction process and provide specific guidance and clarification of data elements; individual questions are used to score measures, but some questions are informational only. DSQs are grouped together by measure set.

§ CGPI: The Validation, Core, Mental Health (MH), PI and Shared Modules are enabled. Specific disease modules are enabled based on Cohort and/or documentation of condition in record.

Medication Reconciliation is enabled based on Cohort 50, 51, and 54.

Inpatient Population Cohorts As of FY20, VHA began submitting electronic clinical quality measures (eCQMs) to The Joint Commission (JC) for ORYX measure requirements. VHA must select four eCQMs which may include: eSTK-2, eSTK-3, eSTK-5, eSTK-6, eED-2, eVTE-1, or eVTE-2. The Office of Analytics and Performance Integration (API, formerly RAPID) coordinates VHA’s annual selection of ORYX measures with VISN Quality Management Officers (QMOs) and submits final selections and data to Joint Commission (JC). Inpatients who are 18 years of age and older who have a length of stay < (less than or equal to) 120 days are eligible.

Please note that for HOP (72, 73, 74, and 76) and Sepsis cohorts (81), records for hospitalized Veterans can take some time to complete, resulting in delays in their inclusion into agency databases. Because the process of selecting samples comes shortly after the end of each month, the list of Veterans with discharges for these diagnoses is likely to be incomplete. For this reason, the sampling process continues to search for Veteran discharges that occurred in prior months but have only recently been reported to the database. Any newly reported discharges are included in the next sample, if it is still possible to include the results obtained for that Veteran in the next submission of accreditation data. Since JC has processes to evaluate the completeness of a facility’s data submissions, it is highly recommended that coding of these cases be completed as soon as possible after discharge.

The Joint Commission manual can be found here: Joint Commission Manual Inpatient Cohorts Cohort 70/Global: Note: This cohort may be collected and used for internal accountability, public reporting (e.g. CMS-Hospital Compare), or eCQM validation. It includes all Veterans admitted to the inpatient acute care hospital setting with a length of stay < 120 days, regardless of diagnoses are eligible. Sample sizes are based on the 2020 Joint Commission Global Inpatient Population specifications. Samples are contingent on the average monthly initial Veteran population as follows:

Delirium Risk questions apply to Global cases and are flagged for review of measure FE81. The Electronic Delirium Risk Assessment (e-NICE) is applied to the cases in each sample and a flag added to the pull list for cases that meet the denominator (high risk for delirium).

This cohort is also used for IMED consent measures and are pulled based on the surgical codes found in the tables below (see FY21 EPRP Sampling Cohorts Appendix). These cases are flagged for iMED review for the following measures: SC1ALL, SC2ALL, SC2CT, SC2GEN, SC2NEU, SC2OBG, SC2ORTH, SC2URO, SC2VAS. For FY21, IMED will be included in the Global cohort, but we anticipate moving to electronic measurement in the future.

Cohort 72/ Hospital Outpatient Population Acute Myocardial Infarction (HOP- AMI)): Note: This cohort may be collected and used for internal accountability, public reporting (e.g. CMS-Hospital Compare) or eCQM validation; however, it was not used for ORYX reporting as of 1/1/17. The CPT and ICD-10-CM codes for this population are included in the FY21 EPRP Sampling Cohorts Appendix, OP Table 1.0 and 1.1.

Cohort 73/ Hospital Outpatient Population Chest Pain (HOP- CP): Note: This cohort may be collected and used for internal accountability, public reporting (e.g. CMS-Hospital Compare) or eCQM validation; however, it was not used for ORYX reporting beginning with discharges on/after 1/1/17. Chest Pain Population: CPT codes found in FY21 EPRP Sampling Cohorts Appendix OP Table 1.0 and ICD-10-CM Principal Diagnosis in the FY21 EPRP Sampling Cohorts Appendix OP Table 1.1a OR Other Diagnosis on OP Table 1.1a together with a Principal Diagnosis that is not on OP Table 1.1 N. Cohort 74/ Hospital Outpatient Population Emergency Department (HOP ED): Note: This cohort may be collected and used for internal accountability, public reporting (e.g. CMS-Hospital Compare) or eCQM validation; however, it was not used for ORYX reporting beginning with discharges on/after 1/1/17. This cohort includes Veterans seen in a hospital Emergency Department. The CPT codes for this population can be found in the FY21 EPRP Sampling Cohorts Appendix OP Table 1.0. There are no ICD-10-CM requirements for HOP-ED.

O. Cohort 76/ Hospital Outpatient Population Stroke (HOP- Stroke): Note: This cohort may be collected and used for internal accountability, public reporting (e.g. CMS-Hospital Compare) or eCQM validation; however, it was not used for ORYX reporting beginning with discharges on/after 1/1/17. This cohort includes all Veterans seen in a hospital Emergency Department with a CPT code found in FY21 EPRP Sampling Cohorts Appendix, OP Table 1.0 and an ICD-10-CM Principal Diagnosis Code for Acute Ischemic or Hemorrhagic Stroke found in FY21 EPRP Sampling Cohorts Appendix, HOP Table 8.1).

P. New Cohort/Measure OP29 Cohort 77/ Hospital Outpatient Population (HOP- Appropriate Follow-up Interval for Normal Colonoscopy in Average Risk Veterans) Note: This cohort may be collected and used for internal accountability, public reporting (e.g. CMS Hospital Compare) or eCQM validation. The CPT and ICD-10-CM codes for this population are derived from the CMS Hospital Outpatient Quality Reporting Specifications Manual, V13.0.

Inclusion Criteria:

Age 50 to 75 when encounter occurred ICD Z12.11 as one of the Dx codes assigned to the encounter, AND at least one CPT code identified below:

44388: Colonoscopy through stoma;

45378: Diagnostic colonoscopy/screening colonoscopy for non-Medicare Veterans GO121: Screening colonoscopy for other Medicare Veterans Exclusion criteria:

Have at least one of the CPT modifiers identified below assigned to any of the CPT inclusion criteria #3:

52: Reduced Services- under certain circumstances a service or procedure is partially reduced or eliminated at the physician’s discretion 53: Discontinued Procedure- under certain circumstances the physician may elect to terminate a surgical or diagnostic procedure 73: Discontinued Out-Patient Hospital/Ambulatory Surgical Center (ASC) Procedure Prior to the Administration of Anesthesia- due to extenuating circumstances or those that threaten the well-being of the Veteran 74: Discontinued Out-Patient Hospital/Ambulatory Surgical Center (ASC) Procedure After the Administration of Anesthesia- due to extenuating circumstances or those that threaten the well-being of the Veteran

OR

Have at least one of the ICD codes identified below assigned to the encounter:

Z83.71: Family history of colonic polyps Z86.010: Persona; history of colonic polyps Z80.0: Family history of malignant neoplasm of GI tract Z85.038: Personal history of malignant neoplasm of large intestine Q. New Cohort/Measure Cat 191-192 Cohort 78/ Cataracts: CAT191- 20/40 or Better Visual Acuity within 90 days Following Cataract Surgery. Percentage of patients aged 18 years and older with a diagnosis of uncomplicated cataract who had cataract surgery and no significant ocular conditions impacting the visual outcome of surgery and had best-corrected visual acuity of 20/40 or better (distance or near) achieved within 90 days following the cataract surgery.

CAT192 - Complications within 30 Days Following Cataract Surgery Requiring Additional Surgical Procedures Inclusion for both measures:

All Veterans aged 18 years and older who had cataract surgery and no significant ocular conditions impacting the visual outcome of surgery at a VHA center.

Veterans aged ≥ 18 years on date of encounter.

CPT 66840; 66850; 66852; 66920; 66930; 66940; 66982; 66983; 66984 without modifiers 55. 56.

R. New Cohort/Measure Transition of Care TRC 1, 2, 3, 4 Cohort 79/TRC All Veterans 18 years or older as of September 30 of the measurement year who were discharged from an acute or nonacute admission during the measurement year. This is a new sample protocol connecting inpatient stay and outpatient providers. The numerator is documentation of evidence of notification of inpatient admission on the day of admission or the following day. This cohort is sampled in January and July only.

S. New Cohort/Measure Sepsis Bundle SEP-1 (sep90, sep1a-1d) Cohort 81/ Sepsis Inpatients age 18 and over with an ICD-10-CM Principal or Other Diagnosis Code of Sepsis, Severe Sepsis, or Septic Shock.

This cohort may be collected and used for internal accountability and/or public reporting (e.g. CMS-Hospital Compare).Included Populations: Discharges age 18 and over with an ICD-10-CM Principal or Other Diagnosis Code of Sepsis, Severe Sepsis, or Septic Shock as defined in the FY21 EPRP Sampling Cohorts Appendix, Inpat Table 4.01.

Excluded Populations:

Directive for Comfort Care or Palliative Care within 6 hours of presentation of severe sepsis Directive for Comfort Care or Palliative Care within 6 hours of presentation of septic shock Administrative contraindication to care within 6 hours of presentation of severe sepsis Administrative contraindication to care within 6 hours of presentation of septic shock Length of Stay >120 days Transfer in from another acute care facility Veterans enrolled in a clinical trial for sepsis, severe sepsis or septic shock treatment or intervention Veterans with severe sepsis who are discharged within 6 hours of presentation Veterans with septic shock who are discharged within 6 hours of presentation Veterans receiving IV antibiotics for more than 24 hours prior to presentation of severe sepsis Sample sizes are based on National Hospital Inpatient Quality Measures specifications produced by the Centers for Medicare & Medicaid Services (CMS). Sample cases for the sepsis measure must ensure that the population and quarterly sample size meets the following conditions:

Delirium Risk questions apply to Sepsis cases and are flagged for review of measure FE81. The Electronic Delirium Risk Assessment (e-NICE) is applied to the cases in each sample and a flag added to the pull list for cases that meet the denominator (high risk for delirium).

Table 2: Inpatient Cohorts and EPRP Data Set Question Modules FY21 Cohorts and EPRP Data Set Question Modules Global

72, 73, 74, 76, 77

HOP

Cataract

(CAT)

Transition of Care

(TRC)

Sepsis

(SEP)

HOP

x

Delirium Risk[footnoteRef:6] [6: ] x x

Cataract x

Transition of Care x

Sepsis x

** Delirium Risk questions apply to Global and Sepsis cases. The Electronic Delirium Risk Assessment (e- NICE) is applied to the cases in each sample and a flag added to the pull list for cases that meet the denominator (high risk for delirium). EPRP collection of the numerator will continue for the cases meeting the denominator criteria

Table 3 Cohort Scoring by Measure

Additionally, the frail elderly cases may be flagged by the statistician for review in the following CGPI measures: c5, c6, c7n, c9h, c12, chf14, chl1, cvrm1, cvrm2, dmg23h, dmg27h, dmg31h, dmg34h, ihd6, ihd20h, ihd53h, mdd40, mov6, mrec43, mrec54, mrec55, mrec56, mrec57, mrec58, mrec59, mrec60, mrec61, mrec62, ostp10, p7, p10, p24, p25h, p26h, p27, p32h, p33, p41h, p42, p43h, p61h, ptsd51, pvc12, sa7, sa17, smg8, smg9, smg10

Appendix A: FY 2021 Outpatient Stop Code Exclusions