AMGA MIPS Collaborative June 21, 2017
Calculating the MIPS score The MIPS composite performance score will include four weighted categories: MIPS Composite Performance Score Quality Cost Improvement activities Advancing care information 2 Source: The Centers for Medicare & Medicaid Services
MIPS Cost (Resource use) Measure Good news No reporting requirement or impact in 2017 (CMS will still provide feedback though) Bad news It rapidly grows to 30% of scoring weight in 2019 3
What is a Cost Measure? Represents the Medicare payments for the medical care furnished to a patient during an episode of care Based on episode groups that: Identify items and services furnished in addressing a condition Serve as a unit of comparison For example, an episode group for meniscus repair identifies care services furnished for this procedure and enables comparison of clinicians providing these services Informs clinicians on the costs of their patients care for which they are responsible Can be aligned with quality of care assessment so that patient outcomes and smarter spending can be pursued together Calculated using Medicare claims data so no additional data submission is required (i.e., no additional clinician burden) 4
Components of a Cost Measure 1 Defining an episode group Assigning costs to the episode group Attributing episode groups to clinicians Risk adjusting episode groups Aligning cost with quality 5
Component 1: Defining an Episode Group An episode group focuses on clinical conditions requiring treatment (the condition itself or procedures to treat the condition) Example: a procedural episode group that is surgical in nature could include: pre-operative services, surgical procedure, anesthesia, follow-up care, services related to complications, readmissions An episode is a specific instance of an episode group for a given patient and clinician Example: A clinician might be attributed 20 episodes (instances of the episode group) from the episode group for heart failure in a year Can vary in scope (e.g., narrow and precise or broad and general) Example: An episode group for cataract removal with insertion of intraocular lens prosthesis has a narrow scope. In comparison, an episode group for gastrointestinal hemorrhage has a broad scope. Can be divided into sub-groups to define more homogeneous patient cohorts Example: Gastrointestinal hemorrhage may be divided into sub-groups for upper and lower gastrointestinal hemorrhage Three types of episode groups in December posting: Acute Inpatient Medical Condition Chronic Condition Procedural 6
Component 2: Assigning Costs to the Episode Group Assignment of items and services determines what is included in episode costs and depends on role of attributed clinician Episode window determines the period of time during which claims are eligible to be assigned to the episode Items and Services that Are Assigned to the Episode Group Direct Services Provided by the attributed clinician Indirect Services Provided or ordered by other clinicians in the same clinical context Post-acute care Ancillary care Consequences of care (e.g., complications) Items and Services that Are Not Assigned to the Episode Group Unrelated Services Unrelated to the clinical management of the patient s condition or procedure that is the focus of the episode group 7
Component 3: Attributing Episode Groups to Clinicians Attribution is the assignment of responsibility for an episode of care to a principal (or managing) clinician Attribution should be transparent to clinicians and only hold them responsible for outcomes they can reasonably be expected to influence Patient relationship categories and codes being developed under MACRA can be used in conjunction with claims-based rules for attribution Development Timeline for Patient Relationship Categories and Codes April 2016 Posted for public comment December 2016 Revised Posting for public comment April 2017 Operational List 8
Component 4: Risk Adjusting Episode Adjusts for factors outside the clinician s control that can influence cost - Age, comorbidities, illness stage/severity, other aspects of patient s clinical history Aims to avoid penalizing clinicians who treat unhealthy or complex patients Selection of risk adjustment method will be informed by analyses, technical expert panels, clinical committees, and public comment 9
Component 5: Aligning Cost with Quality Alignment with indicators of quality is necessary to compensate for information not adequately captured by episode costs Quality assessments might include: Complications, rehospitalizations, unplanned care and other consequences Outcomes of care Overuse, underuse, misuse Processes of care Functional status of patient Patient experience 10
What is the QRUR CMS developed report to provide cost and quality feedback Report s a Tax Identification Number (TIN) performance against national benchmarks QRUR s are generated mid year for informational purposes, and annually for payment adjustments. You must request this report from CMS at https://www.cms.gov/medicare/medicare-fee-for-servicepayment/physicianfeedbackprogram/obtain-2013-qrur.html. 11
What does the QRUR have to do with the MIPS Cost Measure? MIPS Quality and Cost measures are carry-overs from the PQRS and VBM programs provides directionally correct information. The first MACRA performance report is due to be released in July 2017 and is expected to be the QRUR since no other information is available Becoming familiar with the QRUR will lessen the learning curve since they are likely to be part of the initial CMS releases surrounding a TIN s quality and cost performance. 12
QRUR Attribution Methodology Based on TIN Two Step Process Excluded: Beneficiary had Part A only or Part B only for any month For per capita cost measures, must be enrolled in part A and B each month (unless death) Medicare Advantage or other private Medicare Plan for any month Resided outside U.S. 13
Step 1 More primary care services (charges) from PCPs, NPs, PAs, and CNSs in TIN than any other TIN 14
15
Step 2 No primary care services during the period Most primary care services from specialists within the TIN 16
2015 ANNUAL QUA L ITY AND RESOURCE USE REPORT AND THE 2017 VALUE-BASED PAYMENT MODIFIER MERCY HEALTH PHYSICIANS CINCINNATI LLC LAST FOUR DIGITS OF YOUR MEDICARE-ENROLLED TAXPAYER IDENTIFICATION NUMBER (TIN): 7881 PERFORMANCE PERIOD: 01/01/2015-12/31/2015 ABOUT THIS REPORT FROM MEDICARE The 2015 Annual Quality and Resource Use Report (QRUR) shows how your group or solo practice, as identified by its Medicare-enrolled Taxpayer Identification Number (TIN), performed in calendar year 2015 on the quality and cost measures used to calculate the Value-Based Payment Modifier (Value Modifier) for 2017. In 2017,the Value Modifier will apply to all physicians in groups with two or more eligible professionals and to physicians who are solo practitioners who bill under the Medicare Physician Fee Schedule. It will not apply to eligible professionals who are not physicians. As a participant in a Medicare Shared Savings Program Accountable Care Organization (ACO) in 2015, your TlN's 2017 Value Modifier is based on the ACO's quality performance in 2015. The information contained in this report is believed to be accurate at the time of production. The information may be subject to change at the discretion of the Centers for Medicare & Medicaid Services (CMS), including, but not limited to, circumstances in which an error is discovered. YOUR TIN'S 2017 VALUE MODIFIER Average Quality,Average Cost = NeutralAdjustment (0.0%) I Your ACO's performance was determined to be average on quality measures. As a participant in a Shared Savings Program ACO in 2015, your TlN's cost composite is classified as Average Cost. This means that the Value Modifier applied to payments foritems and services under the Medicare Physician Fee Schedule for physicians bi ll ng under your TINIn 2017 will result In a neutral adjustment, meaning no adjustment (O.Oo/o). The scatter plot below shows how your TIN ("You" diamond) compares to a representative sample of other TINs on the Quality Composite scores used to calculate the 2017 Value Modifier. 17
Exhibit 3-ECC. Effective Clinical Care Domain Quality Indicator Performance Domain Score You 0.44 + -2.0-1.0 0.0 1.0 2.0,<;;-4.0-3.0 Standard deviations from the mean (positive scores are better) 3.0 2: 4.0 Measure Identification Number(s) 1* (GPRO DM-2, CMS122v3) 8 (GPRO HF- 6, CMS144v3) 112 (GPRO Prev-5, CMS125v3) 113 (GPRO Prev-6, CMS130v3) 117 (GPRO DM- 7, CMS131v3) Measure Name Number of Eligible Cases Performance Rate Your ACO Standardized Performance Score Included in Domain Score? All TINs and ACOs in Peer Group Benchmark Standard (National Deviation Mean) Diabetes: Hemoglobin A1c Poor Control 296 14.86% 0.59 No 28.42% 22.83 Heart Failure (HF): Beta-Blocker Therapy for Left 169 84.62% -0.11 Yes 86.35% 16.04 Ventricular Systolic Dysfunction (LVSD) Breast Cancer Screening 326 72.70% 0.62 Yes 54.61% 29.31 Colorectal Cancer Screening 248 71.37% 0.76 Yes 47.55% 31.31 Diabetes: Eye Exam 296 36.49% -1.80 No 85.70% 27.29 118 (GPRO CAD- 7) 204 (GPRO IVD- 2, CMS164v3) Coronary Artery Disease (CAD): Angiotensin- Converting Enzyme (ACE) Inhibitor or Angiotensin Receptor Blocker (ARB) Therapy -- Diabetes or Left Ventricular Systolic Dysfunction (LVEF < 40%) lschemic Vascular Disease (IVD): Use of Aspirin or Another Antithrombotic 251 84.06% 0.59 Yes 72.74% 19.16 337 87.54% 0.32 Yes 79.60% 25.03 236 (GPRO HTN- 2, CMS165v3) 370 (GPRO MH- 1, CMS159v3) Controlling High Blood Pressure 579 76.17% 0.48 Yes 69.03% 14.78 Depression Remission at Twelve Months 24 0.00% - No - - - Diabetes Mellitus (DM): Composite (All or Nothing Scoring) (includes GPRO DM-2 and GPRO DM-7) 296 31.08% - No - - 18
Note: If an asterisk (*) appears after the measure identification number,it indicates that the measure is an inverse (negative) measure, and a lower performance rate for this measure reflects better performance. This is taken into account when calculating the quality domain score, such that a positive (+) domain score indicates better performance and negative (-) domain score indicates worse performance. Only those measures for which benchmarks are available and for which your ACO had at least 20 eligible cases are included in the domain score. The benchmark for a quality measure is the case-weighted national mean performance rate among all TINs in the measure's peer group during calendar year 2014. The peer group is defined as all TINs nationwide that reported the measure and had at least 20 eligible cases during calendar year 2014. If a dash (-) appears in the Benchmark column, this indicates that no benchmark is available for this measure. For TINs or ACOs that reported quality data to the PQRS via the GPRO Web Interface, GPRO DM-2 (measure #1) and GPRO DM-7 (measure #117) are components of the "Diabetes Mellitus: Composite (All or Nothing Scoring)" measure and are not included in the calculation of the domain score as individual measures. Exhibit 3-PCE. Person and Caregiver-Centered Experience and Outcomes Domain Quality Indicator Performance Domain Score No domain score was calculated because your AGO did not have at least one measure that had the minimum number of eligible cases to be included in the domain score. Exhibit 3-PCE is not displayed because your AGO did not have at least one eligible case for at least one measure in this domain. 19
2015 ANNUAL QUA L ITY AND RESOURCE USE REPORT AND THE 2017 VALUE-BASED PAYMENT MODIFIER MERCY HEALTH PHYSICIANS LAST FOUR DIGITS OF YOUR MEDICARE-ENROLLED TAXPAYER IDENTIFICATION NUMBER (TIN): PERFORMANCE PERIOD: 01/01/2015-12/31/2015 ABOUT THIS REPORT FROM MEDICARE The 2015 Annual Quality and Resource Use Report (QRUR) shows how your group or solo practice, as identified by its Medicareenrolled Taxpayer Identification Number (TIN), performed in calendar year 2015 on the quality and cost measures used to calculate the Value-Based Payment Modifier (Value Modifier) for 2017. In 2017,the Value Modifier will apply to all physicians in groups with two or more eligible professionals and to physicians who are solo practitioners who bill under the Medicare Physician Fee Schedule. It will not apply to eligible professionals who are not physicians. As a participant in a Medicare Shared Savings Program Accountable Care Organization (ACO) in 2015, your TlN's 2017 Value Modifier is based on the ACO's quality performance in 2015. The information contained in this report is believed to be accurate at the time of production. The information may be subject to change at the discretion of the Centers for Medicare & Medicaid Services (CMS), including, but not limited to, circumstances in which an error is discovered. YOUR TIN'S 2017 VALUE MODIFIER Average Quality,Average Cost = NeutralAdjustment (0.0%) I Your ACO's performance was determined to be average on quality measures. As a participant in a Shared Savings Program ACO in 2015, your TlN's cost composite is classified as Average Cost. This means that the Value Modifier applied to payments foritems and services under the Medicare Physician Fee Schedule for physicians bi ll ng under your TINIn 2017 will result In a neutral adjustment, meaning no adjustment (O.Oo/o). The scatter plot below shows how your TIN ("You" diamond) compares to a representative sample of other TINs on the Quality Composite scores used to calculate the 2017 Value Modifier. 20
21
Exhibits 3-CCC A and B. Communication and Care Coordination Domain Domain Score You 0.72 + -4.0-3.0-2.0-1.0 0.0 1.0 2.0 3.0 4.0 Standard deviations from the mean (positive scores are better) A. Communication and Care Coordination Domain Quality Indicator Performance Exhibit 3-CCC-A is not displayed because your ACO did not have at least one eligible case for at least one measure in this domain. B. Communication and Care Coordination Domain CMS-Calculated Quality Outcome Measures Exhibit 3-CCC-B provides information on the quality outcome measure calculated from Medicare Part A and Part B claims data. Your ACO All TINs and ACOs in Peer Group Performance Measure Measure Name Number of Performance Standardized Included Benchmark Standard Identification (National Category Number(s) Eligible Rate Performance in Domain Mean) Deviation Cases Score Score? Hospital CMS-3 All-Cause Hospital 17,357 14.28% 0.72 Yes 15.32% 1.43 Note: Readmission CMS-3 calculated by the Centers for Readmission Medicare & Medicaid Services using Medicare Part A and Part B claims data. Lower performance rates for this measure indicates better performance. This is taken into account when calculating the quality domain score, such that a positive (+) domain score indicates better performance and a negative (-) domain score indicates worse performance.the benchmark for this measure is the case-weighted national mean performance rate among all TINs and ACOs in the measure's peer group during calendar year 2014. The peer group for CMS-3 is defined as all TINs nationwide with 10 or more eligible professionals that had at least 200 eligible cases and all ACOs in the Shared Savings Program with at least 1 eligible case. Exhibit 3-ECR. Efficiency and Cost Reduction Domain Quality Indicator Performance Domain Score No domain score was calculated because your AGO did not have at least one measure that had the minimum number of eligible cases to be included in the domain score. Exhibit 3-ECR is not displayed because your ACO did not have at least one eligible case for at least one measure in this domain. 22
PERFORMANCE ON COST MEASURES Your TIN's Cost Tier: Average The Cost Composite Score for particpants of a Shared Savings Program ACO is provided in this report for informational purposes only and is based on the TIN's cost performance, not the ACO's cost performance. Exhibit 4. Your TIN's Cost Composite Score -+- Low Cost Average Cost High Cost... You 0.08 S -4.0-3.5-3.0-2.5-2.0-1.5-1.0-2.5 3.5 2: 4.0 0.5 0.0 0.5 1.0 1.5 2.0 3. 0 Standard Deviations from the Peer Group Mean (Negative Scores Are Better) Your TIN's Cost Composite Score (Exhibit 4) indicates that your TIN's overall performance on cost measures is 0.08 standard deviation from the mean for your TlN's peer group. As a participant in a Shared Savings Program ACO in 2015, your TlN's cost performance is classified as Average Cost under quality-tiering. The Cost Composite Score is provided for informational purposes only. The Cost Composite Score and Quality Composite Score are the two summary scores used to calculate the Value Modifier under quality-tiering. The Cost Composite Score standardizes a TlN's cost performance relative to the mean for the TlN's peer group, such that O represents the peer group mean and the TlN's Cost Composite Score indicates how many standard deviations a TlN's performance is from the mean. Your TIN's peer group includes all TINs subject to the 2017 Value Modifier for which a Cost Composite Score could be calculated, with the exception of TINs that participated in the Shared Savings Program in 2015. A TlN's Cost Composite Score is classified into one of three cost tiers (high, average, or low), based on how the score compares to the mean for the TIN's peer group. To be considered either High Cost or Low Cost, a TIN's score must be at least one standard deviation from the peer group mean and statistically significantly different from the mean at the five percent level of significance. That is, a TIN with a statistically significant positive Cost Composite Score of one (+1.0) or higher would be classified as High Cost, and a TIN with a statistically significant negative score of one (-1.0) or lower would be classified as Low Cost. A TIN with any other Cost Composite Score would be classified as Average Cost. That is, a TIN with a Cost Composite Score in the range between (but not including) negative one (-1.0) and positive one (+1.0) would be classified as Average Cost, because its score is less than one standard deviation from the mean. A TIN with a score of negative one (-1.0) or lower or positive one (+1.0) or higher that is NOT statistically significantly different from the mean would also be classified as Average Cost. Glossary Terms Cost Composite Score Quality-tiering Standard deviation Statistical significance Value Modifier (Value-Based Payment Modifier) 23
24
25
PERFORMANCE ON QUALITY MEASURES Your TIN's Quality Tier: Average As a participant in a Shared Savings Program ACO, your TIN's Quality Composite Score is based on the quality performance of the ACO in 2015. Exhibit 2. Your TIN's Quality Composite Score Low Quality Average Quality You 0.49 + High Quality -+ -4.0-3.5-3.0-2.5-2.0-1.5-1.0-0.5 0.0 0.5 1.0 1.5 2.0 Standard Deviations from the Peer Group Mean (Positive Scores Are Better) 2.5 3.0 3.5 2:4.0 Your TlN's Quality Composite Score (Exhibit 2) indicates that your ACO's overall performance on quality measures is 0.49 standard deviation from the mean for your TlN's peer group. Because your TIN's Quality Composite Score is less than one standard deviation from the mean, your TlN's quality performance is classified as Average Quality under quality-tiering. The Quality Composite Score and Cost Composite Score are the two summary scores used to calculate the Value Modifier under quality-tiering. The Quality Composite Score standardizes a TIN's quality performance relative to the mean for the TIN's peer group, such that O represents the peer group mean and the TIN's Quality Composite Score indicates how many standard deviations a TIN's performance is from the mean. Your TlN's peer group includes all TINs subject to the 2017 Value Modifier for which a Quality Composite Score could be calculated. A TlN's Quality Composite Score is classified into one of three quality tiers (high, average, or low), based on how the score compares to the mean for the TIN's peer group. To be considered either High Quality or Low Quality, a TlN's score must be at least one standard deviation from the peer group mean and statistically significantly different from the mean at the five percent level of significance. That is, a TIN with a statistically significant positive Quality Composite Score of one (+1.0) or higher would be classified as High Quality, and a TIN with a statistically significant negative score of one (-1.0) or lower would be classified as Low Quality. A TIN with any other Quality Composite Score would be classified as Average Quality. That is, a TIN with a Quality Composite Score in the range between (but not including) negative one (- 1.0) and positive one (+1.0) would be classified as Average Quality, because its score is less than one standard deviation from the mean. A TIN with a score of negative one (-1.0) or lower or positive one (+1.0) or higher that is NOT statistically significantly different from the mean would also be classified as Average Quality. Glossary Terms Quality Composite Score Quality-tiering Standard deviation Statistical significance Value Modifier (Value-Based Payment Modifier) 26
Example of CPC Plus Report The next two slides are copies of the CPC Plus reports that CMS is sending to participating practices. These reports show Average Cost of Care trends over time. 27
28
29
Next Steps Continue to monitor CMS updates on MACRA performance report which is expected to be released in July. Emphasize accuracy in your current RAF scores for your Medicare patients since the cost measure is Risk Adjusted to insure proper reporting. 30
Questions?? 31