NQF

Agenda Synopsis

Full Agenda

Measure Index

• Hospital-Wide, 30-Day, All-Cause Unplanned Readmission (HWR) Rate for the Merit-Based Incentive Payment Program (MIPS) Eligible Clinician Groups (Public comments received:11; MUC ID: MUC2019-27)
  
  
• Risk-standardized complication rate (RSCR) following elective primary total hip arthroplasty (THA) and/or total knee arthroplasty (TKA) for Merit-based Incentive Payment System (MIPS) Eligible Clinicians and Eligible Clinician Groups (Public comments received:5; MUC ID: MUC2019-28)
  
  
• Clinician and Clinician Group Risk-standardized Hospital Admission Rates for Patients with Multiple Chronic Conditions; in the Medicare Shared Savings Program, the score would be at the ACO level. (Public comments received:8; MUC ID: MUC2019-37)
  
  
• Hemodialysis Vascular Access: Practitioner Level Long-term Catheter Rate (Public comments received:3; MUC ID: MUC2019-66)
  
  

• Clinician and Clinician Group Risk-standardized Hospital Admission Rates for Patients with Multiple Chronic Conditions; in the Medicare Shared Savings Program, the score would be at the ACO level. (Public comments received:4; MUC ID: MUC2019-37)
  
  

• Follow-up after Emergency Department (ED) Visit for People with Multiple High-Risk Chronic Conditions (Public comments received:3; MUC ID: MUC2019-14)
  
  
• Transitions of Care between the Inpatient and Outpatient Settings including Notifications of Admissions and Discharges, Patient Engagement and Medication Reconciliation Post-Discharge (Public comments received:2; MUC ID: MUC2019-21)
  
  
• Use of Opioids at High Dosage in Persons without Cancer (OHD) (Public comments received:5; MUC ID: MUC2019-57)
  
  
• Use of Opioids from Multiple Providers in Persons without Cancer (OMP) (Public comments received:5; MUC ID: MUC2019-60)
  
  
• Use of Opioids from Multiple Providers and at a High Dosage in Persons without Cancer (OHDMP) (Public comments received:4; MUC ID: MUC2019-61)
  
  

Full Measure Information

Measure Specifications

• NQF Number (if applicable): 3495
  
  
• Description: This measure is a re-specified version of the measure, Risk-adjusted readmission rate (RARR) of unplanned readmission within 30 days of hospital discharge for any condition†(NQF 1789), which was developed for patients 65 years and older using Medicare claims. This re-specified measure attributes outcomes to MIPS participating clinician groups and assesses each group's readmission rate. The measure comprises a single summary score, derived from the results of five models, one for each of the following specialty cohorts (groups of discharge condition categories or procedure categories): medicine, surgery/gynecology, cardio-respiratory, cardiovascular, and neurology.
  
  
• Numerator: The outcome for this measure is unplanned all-cause 30-day readmission. Readmission is defined as a subsequent inpatient admission to any acute care facility which occurs within 30 days of the discharge date of an eligible index admission. Any readmission is eligible to be counted as an outcome, except those that are considered planned. To align with data years used, the planned readmission algorithm version 4.0 was used to classify readmissions as planned or unplanned.
  
  
• Denominator: Patients eligible for inclusion in the measure have an index admission hospitalization to which the readmission outcome is attributed and includes admissions for patients: Enrolled in Medicare Fee-For-Service (FFS) Part A for the 12 months prior to the date of admission; Aged 65 or over; Discharged alive from a non-federal short-term acute care hospital; and, Not transferred to another acute care facility.
  
  
• Exclusions: 1. Patients discharged against medical advice (AMA) are excluded.2. Admissions for patients to a PPS-exempt cancer hospital are excluded.3. Admissions primarily for medical treatment of cancer are excluded.4. Admissions primarily for psychiatric disease are excluded.5. Admissions for rehabilitation care; fitting of prostheses and adjustment devices (CCS 254) are excluded.6. Admissions where patient cannot be attributed to a clinician group.
  
  
• HHS NQS Priority: Communication and Care Coordination
  
  
• HHS Data Source: Administrative clinical data
  
  
• Measure Type: Outcome
  
  
• Steward: Centers for Medicare & Medicaid Services
  
  
• Endorsement Status:
  
  
• Meaningful Measure Area: Admissions and readmissions to hospitals
  
  
• Is the measure specified as an electronic clinical quality measure? No
  
  

Preliminary Analysis of Measure

• Preliminary analysis result: Conditional support pending replacement of 1789 in the program measure set and NQF CDP review of reliability performance at the physician group level in Spring 2020.
  
  
• Preliminary analysis summary
  • Contribution to program measure set:The measure addresses the priority area of communication and coordination of care: admissions and readmissions to hospitals. This measure is updated version of 1789, specified for the physician group level of analysis.
  • Impact on quality of care for patients:Physician groups have an important role to play to reduce avoidable admissions and readmissions that represent an opportunity to improve patient care transitions and prevent the unnecessary exposure of patients to adverse events in an acute care setting. This measure demonstrates a median risk adjusted readmission rate for clinician groups of 15.3%. The 10th to 90th percentile performance range spans from 13.8% to 17.1%. This distribution represents opportunity for improvement and overall less than optional performance.
  
  
  
• Does the measure address a critical quality objective not currently adequately addressed by the measures in the program set? Yes. The measure addresses the priority area of communication and coordination of care: admissions and readmissions to hospitals. Physician groups have an important role to play to reduce avoidable admissions and readmissions that represent an opportunity to improve patient care transitions and prevent the unnecessary exposure of patients to adverse events in an acute care setting. This measure is updated version of 1789, specified for the physician group level of analysis.
  
  
• Is the measure evidence-based and either strongly linked to outcomes or an outcome measure? Yes. This outcome measure is currently under review by the NQF All-Cause Admissions and Readmissions Standing Committee. Physician groups can influence this outcome by supporting appropriate discharge, medication reconciliation, reducing infection risk, and ensuring proper outpatient follow-up. Partnering with hospitals and other members of the care team can improve this outcome for patients.
  
  
• Does the measure address a quality challenge? Yes. This measure demonstrates a median risk adjusted readmission rate for clinician groups of 15.3%. The 10th to 90th percentile performance range spans from 13.8% to 17.1%. This distribution represents opportunity for improvement and overall less than optional performance.
  
  
• Does the measure contribute to efficient use of measurement resources and/or support alignment of measurement across programs? Yes. This measure is a re-specified version of 1789 for the physician group level of analysis.
  
  
• Can the measure can be feasibly reported? Yes. The measure uses claims data that can be feasibly reported; there are no fees, licensing, or requirements to use the measure.
  
  
• Is the measure applicable to and appropriately specified for the program's intended care setting(s), level(s) of analysis, and population(s)? Yes. The measure is specified for the physician group level of analysis. The NQF Readmissions and Admissions Standing Committee requested additional information from the developer on reliability performance of this measure at various case sizes for the physician group level of analysis. The NQF CDP Committee expressed support for the attribution of physician groups to improve this outcome, in coordination with hospitals and other members of the care team. The NQF CDP Committee also encouraged the developer to expand testing of SDS risk factors for this measure. The NQF CDP Committee was generally not supportive of this measure at the individual clinician level. The endorsement consideration of this measure was deferred to the Spring 2020 pending updated testing information for consideration.
  
  
• Measure development status: Fully Developed
  
  
• If the measure is in current use, have negative unintended issues to the patient been identified? n/a. The measure is not currently in use but the developers identified a potential unintended consequence that it might discourage hospitals from readmitting patients who would benefit from inpatient care. Patients with serious illnesses might re-present but be sent home in an effort to reduce readmissions. Another potential consequence is that readmissions measures unfairly penalize hospitals that serve patients with high social risk.
  
  
• MAP Rural Workgroup Finding:
  • Relative priority/utility: Rural residents are relatively more vulnerable than those in other areas (e.g., social risk, acuity, lack of access to care) and thus may be more likely to be readmitted. .
  • Data collection issues: The Workgroup noted there may be lack of resources/access in rural areas to prevent readmission (e.g., fewer urgent care options).
  • Calculation issues: The Workgroup believes this measure low case-volume will be a critical issue for rural providers.
  • Unintended consequences: The Workgroup voiced concern regarding lack of adjustment for social risk factors, which are particularly relevant for rural populations. Members believe rural providers may be unfairly penalized on this measure, given it is being proposed for a pay-for-performance program.
  • Votes: Range is 1 – 5, where higher reflects more agreement regarding suitability for the program
  • Average agreement= 2.3
  • Vote Counts: (1 – 3 votes; 2 – 5 votes; 3 – 2 votes; 4 – 1 vote; 5 – 1 vote)
  • Program gap areas: None identified.
  
  
  

Rationale for measure provided by HHS
HHS has provided additional documentation to support this measure under consideration. Hospital readmission, for any reason, is disruptive to patients and caregivers, costly to the healthcare system, and puts patients at additional risk of hospital-acquired infections and complications. Readmissions are also a major source of patient and family stress and may contribute substantially to loss of functional ability, particularly in older patients. Some readmissions are unavoidable and result from inevitable progression of disease or worsening of chronic conditions. However, readmissions may also result from poor quality of care or inadequate transitional or post-discharge care. Transitional care includes effective discharge planning, transfer of information at the time of discharge, patient assessment and education, and coordination of care and monitoring in the post-discharge period. Numerous studies have found an association between quality of inpatient or transitional care and early (typically 30-day) readmission rates for a wide range of conditions.1-8Randomized controlled trials have shown that improvement in the following areas can directly reduce readmission rates: quality of care during the initial admission; improvement in communication with patients, their caregivers, and their clinicians; patient education; pre-discharge assessment; and coordination of care after discharge.9-17 Successful randomized trials have reduced 30-day readmission rates by 20-40%.18 Widespread application of these clinical trial interventions to general practice has also been encouraging. Since 2008, Medicare Quality Improvement Organizations have been funded to focus on care transitions by applying lessons learned from clinical trials. Several have been notably successful in reducing readmissions within 30 days.19 Many of these study interventions involved enhanced clinician involvement and indicate a key role for clinicians in reducing readmissions. Further, analyses CORE performed pre-development of this measure support variation in clinician- and clinician group-level performance on 30-day readmissions for patients with acute myocardial infraction.Despite these demonstrated successful interventions, the overall national readmission rate remains high, with a 30-day readmission following over 15% of discharges. Readmission rates also vary widely across institutions.20-22 Moreover, we show below that RARRs vary from 7%-25% for clinician groups for 2015-16. Both the high baseline rate and the variability across eligible clinician groups speak to the need for a quality measure to prompt greater care improvement. Given that studies have shown readmissions within 30 days to be related to quality of care, that interventions, including those utilizing clinicians, have been able to reduce 30-day readmission rates for a variety of specific conditions, and that high and variable clinician-level readmission rates indicate opportunity for improvement, we sought to develop eligible clinician group-level measure of all-cause, all-condition 30-day unplanned readmission.1. Frankl SE, Breeling JL, L. G. Preventability of emergent hospital readmission American Journal of Medicine. Jun 1991;90(6):667-674.2. Corrigan J, Martin J. Identification of factors associated with hospital readmission and development of a predictive model. Health Services Research. Apr 1992;27(1):81-101.3. Oddone E, Weinberger M, Horner M, et al. . Classifying general medicine readmissions. Are they preventable? Veterans Affairs Cooperative Studies in Health Services Group on Primary Care and Hospital Readmissions. Journal of General Internal Medicine. Oct 1996;11(10):597-607.4. Ashton C, Del Junco DJ, Souchek J, Wray N, Mansyur C. The association between the quality of inpatient care and early readmission: a meta-analysis of the evidence. . Med Care. Oct 1997;35(10):1044-1059.5. Benbassat J, Taragin M. Hospital readmissions as a measure of quality of health care: advantages and limitations. Archives of Internal Medicine. Apr 24 2000;160(8):1074-1081.6. Courtney EDJ, Ankrett S, McCollum PT. 28-Day emergency surgical re-admission rates as a clinical indicator of performance. Annals of the Royal College of Surgeons of England. Mar 2003;85(2):75-78.7. Halfon P, Eggli Y, Pr, et al. . Validation of the potentially avoidable hospital readmission rate as a routine indicator of the quality of hospital care. Medical Care. Nov 2006;44(11):972-981.8. Hernandez AF, Greiner MA, Fonarow GC, et al. . Relationship between early physician follow-up and 30-day readmission among Medicare beneficiaries hospitalized for heart failure. JAMA. May 5 2010;303(17):1716-1722.9. Naylor M, Brooten D, Jones R, Lavizzo-Mourey R, Mezey M, Pauly M. Comprehensive discharge planning for the hospitalized elderly. A randomized clinical trial. Ann Intern Med. Jun 15 1994;120(12):999-1006.10. Naylor MD, Brooten D, Campbell R, et al. Comprehensive discharge planning and home follow-up of hospitalized elders: a randomized clinical trial. JAMA. 1999;281(7):613-620.11.Krumholz HM, Amatruda J, Smith GL, et al. Randomized trial of an education and support intervention to prevent readmission of patients with heart failure. Journal of the American College of Cardiology. . Jan 2 2002;39(1):83-89.12. van Walraven C, Seth R, Austin PC, Laupacis A. Effect of discharge summary availability during post-discharge visits on hospital readmission. Journal of General Internal Medicine. Mar 2002;;17(3):186-192.13. Conley RR, Kelly DL, Love RC, McMahon RP. Rehospitalization risk with second-generation and depot antipsychotics. . Annals of Clinical Psychiatry. Mar 2003;15(1):23-31.14. Coleman EA, Smith JD, Frank JC, Min S-J, Parry C, Kramer AM. Preparing patients and caregivers to participate in care delivered across settings: the Care Transitions Intervention. Journal of the American Geriatrics Society. Nov 2004;52(11):1817-1825.15. Phillips CO, Wright SM, Kern DE, Singa RM, Shepperd S, Rubin HR. Comprehensive discharge planning with postdischarge support for older patients with congestive heart failure: a meta-analysis. JAMA. Mar 17 2004;291(11):1358-1367.16. Jovicic A, Holroyd-Leduc JM, Straus SE. Effects of self-management intervention on health outcomes of patients with heart failure: a systematic review of randomized controlled trials. BMC Cardiovasc Disord. 2006;6:43.17. Garasen H, Windspoll R, Johnsen R. Intermediate care at a community hospital as an alternative to prolonged general hospital care for elderly patients: a randomised controlled trial. BMC Public Health. 2007;7:69.18. Leppin AL, Gionfriddo MR, Kessler M, et al. Preventing 30-day hospital readmissions: a systematic review and meta-analysis of randomized trials. JAMA Intern Med. 2014;174(7):1095-1107.19. CFMC. CFfMC. Care Transitions QIOSC. 2010; http://www.cfmc.org/caretransitions/.20. Keenan PS, Normand SL, Lin Z, et al. An administrative claims measure suitable for profiling hospital performance on the basis of 30-day all-cause readmission rates among patients with heart failure. Circ Cardiovasc Qual Outcomes. 2008;1(1):29-37.21. Krumholz HM, Lin Z, Drye EE, et al. An administrative claims measure suitable for profiling hospital performance based on 30-day all-cause readmission rates among patients with acute myocardial infarction. Circulation. Mar 1 2011;4(2):243-252.22. Lindenauer PK, Normand SL, Drye EE, et al. Development, validation, and results of a measure of 30-day readmission following hospitalization for pneumonia. J Hosp Med. 2011;6(3):142-150.

Summary of NQF Endorsement Review

• Year of Most Recent Endorsement Review: 2019
  
  
• Project for Most Recent Endorsement Review: Admissions and Readmissions
  
  
• Review for Importance: 1a. Evidence: Y-14; N-2; 1b. Performance Gap: H-0; M-15; L-1; I-0Rationale:•This is a re-specified version of the hospital-level measure, “Hospital-Wide All-Cause, Unplanned Readmission Measure” (NQF 1789). NQF 1789 was developed for patients who are 65 years or older, are enrolled in fee-for-service (FFS) Medicare, and are hospitalized in non-federal hospitals. This specified measure attributes admissions to up to three participating MIPS eligible clinicians.•The Standing Committee reviewed the logic model presented by the developer demonstrating physician group interventions that can reduce the risk of unplanned hospital visits.•The Standing Committee reviewed the range of performance for clinician groups from 13.1 in the first decile to 18.0 in the tenth decile.
  
  
• Review for Scientific Acceptability: 2a. Reliability: H-0; M-13; L-2; I-1; 2b. Validity: H-0; M-10; L-6; I-0Rationale:•The Standing Committee reviewed the input provided by the NQF Scientific Methods Panel (SMP).•The Standing Committee generally agreed with the input from the SMP that the reliability testing methodology was appropriate and that results demonstrated moderate reliability. The Standing Committee noted SMP concerns that social risk factors are excluded from the risk model given the effect size and the potential for negative consequences on access to care if this measure is not adequately risk adjusted. The Standing Committee agreed that the developer should examine other clinical variables that could underlie disparities such as frailty or functional status.•The Standing Committee reviewed the measure score empirical validity testing and face validity testing submitted at the individual physician and physician group level. The Standing Committee noted that the evidence and validity testing must be evaluated separately for the two levels of analysis. The Committee agreed to vote individually on the two levels of analysis.•The Standing Committee noted that eligible clinician groups’ risk adjusted readmission rates go down with increasing Overall Hospital Quality Star Rating and with increasing quintile of the Star Rating readmission quality score.•The Committee reviewed the Face Validity testing and results, the approach to risk adjustment and the conceptual model for socio-demographic risk adjustment.•The Standing Committee had a mixed review of the face validity of the measure, specifically, the role physician groups have in improving this outcome.
  
  
• Review for Feasibility: 3. Feasibility: H-6; M-9; L-1; I-0(3a. Clinical data generated during care delivery; 3b. Electronic sources; 3c. Susceptibility to inaccuracies/ unintended consequences identified; 3d. Data collection strategy can be implemented)Rationale:•The Standing Committee agreed that the measure uses claims data that can be operationalized; however, the measure is not yet in use. There are no fees, licensing, or requirements to use the measure.
  
  
• Review for Usability: 4a. Use: Pass-14; No Pass-2; 4b. Usability: H-1; M-11; L-4; I-0Rationale:•The Standing Committee acknowledged that this measure is planned for use in the CMS MIPS program.•The Standing Committee noted that this is a new measure and there is no information available on performance improvement. This measure is not currently used in a program, but a primary goal of the measure is to provide information necessary to implement focused quality improvement efforts. Once the measure is implemented, the developer plans to examine trends in improvements by comparing RSRR over time.
  
  
• Review for Related and Competing Measures: No related or competing measures noted.
  
  
• Endorsement Committee Recommendation: Standing Committee Recommendation for Endorsement: Yes-11; No-5Rationale•The Committee noted that the measure passed each of the criteria and is suitable for continued endorsement.
  
  

Measure Specifications

• NQF Number (if applicable): 3493
  
  
• Description: This measure is a re-specified version of the measure, Hospital-level Risk-standardized Complication rate (RSCR) following Elective Primary Total Hip Arthroplasty (THA) and/or Total Knee Arthroplasty (TKA) (National Quality Forum 1550), which was developed for patients 65 years and older using Medicare claims. This re-specified measure attributes outcomes to Merit-based Incentive Payment System participating clinicians and/or clinician groups (provider) and assesses each provider's complication rate, defined as any one of the specified complications occurring from the date of index admission to up to 90 days post date of the index procedure.
  
  
• Numerator: The outcome for this measure is complication defined as acute myocardial infarction (AMI), pneumonia, and sepsis/septicemia/shock complications within seven days from the index admission date; death, surgical site bleeding, and pulmonary embolism within 30 days from the index admission; mechanical complications and periprosthetic joint infection/wound infection within 90 days of the index admission. The complication outcome is a dichotomous (yes/no) outcome. If a patient experiences one or more of these complications in the applicable time period, the complication outcome for that patient is counted in the measure as a "yes".
  
  
• Denominator: Patients eligible for inclusion in the measure are those age 65 years and older admitted to non-federal acute care hospitals. An index admission is the hospitalization during which an elective Total Hip Arthroplasty (THA) and/or Total Knee Arthroplasty (TKA) procedure was performed and to which the complication outcome is attributed. Eligible index admissions are identified using International Classification of Diseases-Tenth Revision-Procedure Coding System (ICD-10-PCS) procedure codes in Medicare inpatient claims data. For risk adjustment and outcome assessment, patients must have continuous enrollment in Medicare fee-for-service (FFS) for 12 months prior to the procedure and 90 days after it. The measure cohort is fully harmonized with the existing hospital-level measure.
  
  
• Exclusions: This measure excludes from the denominator admissions for patients:1. With a femur, hip or pelvic fracture coded in the principal discharge diagnosis field for the index admission.2. Undergoing partial hip arthroplasty (PHA) procedures (with a concurrent Total Hip Arthroplasty or Total Knee Arthroplasty [THA/TKA]).3. Undergoing revision procedures (with a concurrent THA/TKA).4. Undergoing resurfacing procedures (with a concurrent THA/TKA). 5. With a mechanical complication coded in the principal discharge diagnosis field for the index admission.6. With a malignant neoplasm of the pelvis, sacrum, coccyx, lower limbs, or bone/bone marrow or a disseminated malignant neoplasm coded in the principal discharge diagnosis field for the index admission.7. With a procedure code for removal of implanted devices/prostheses.After excluding the above admissions to identify elective primary THA/TKA procedures, the measure also excludes admissions for patients:8. Who were transferred to the index hospital.9. Who leave the hospital against medical advice (AMA).10. With more than two THA/TKA procedure codes during the index hospitalization.Note: The measure does not count complications that occur in the outpatient setting and do not require a readmission.
  
  
• HHS NQS Priority: Communication and Care Coordination
  
  
• HHS Data Source: Claims
  
  
• Measure Type: Outcome
  
  
• Steward: Centers for Medicare & Medicaid Services
  
  
• Endorsement Status:
  
  
• Meaningful Measure Area: Preventable healthcare harm
  
  
• Is the measure specified as an electronic clinical quality measure? No
  
  

Preliminary Analysis of Measure

• Preliminary analysis result: Support for Rulemaking
  
  
• Preliminary analysis summary
  • Contribution to program measure set:The addition of this measure supplements a set of surgical measures in MIPS related to TKA and THA, but with a stronger outcome measure capturing complications stemming from these procedures.
  • Impact on quality of care for patients:This measure could potentially improve the quality of surgical care delivery and follow-up care for a common and costly surgical procedure performed for Medicare beneficiaries.
  
  
  
• Does the measure address a critical quality objective not currently adequately addressed by the measures in the program set? Yes. Currently MIPS has over 30 measures related to surgery, with seven directly related to TKA and THA:Total Knee Replacement: Identification of Implanted Prosthesis in Operative ReportTotal Knee Replacement: Preoperative Antibiotic Infusion with Proximal TourniquetTotal Knee Replacement: Shared Decision-Making: Trial of Conservative (Non-surgical) TherapyTotal Knee Replacement: Venous Thromboembolic and Cardiovascular Risk EvaluationFunctional Status Assessment for Total Hip ReplacementFunctional Status Assessment for Total Knee ReplacementAverage Change in Functional Status Following Total Knee Replacement SurgeryThe measure Total Knee Replacement: Venous Thromboembolic and Cardiovascular Risk Evaluation deals with complications associated with TKA, but does not deal with all complications, nor does it address THA related complications. There are currently 23 measures addressing the Meaningful Measurement Area - Preventable Healthcare Harm, a critical quality objective for CMS to address, but no measures related to TKA and THA.
  
  
• Is the measure evidence-based and either strongly linked to outcomes or an outcome measure? Yes. This measure is an outcome measure that assesses the rates of complications associated with TKA and THA.The developer noted that clinicians, particularly the surgeon performing the procedure, can influence the outcome of surgery for better or worse, both through their technical skill and through their influence on the care team and hospital safety culture. Therefore, many of the strategies and best practices used by hospitals to reduce the risk of complications can also be adopted by individual clinicians and groups of clinicians to improve patient outcomes. Further evidence of surgeons’ influence are data indicating that increasing surgeon volume is associated with reductions in adverse surgical outcomes.
  
  
• Does the measure address a quality challenge? Yes. In 2010, there were 168,000 THAs and 385,000 TKAs performed on Medicare beneficiaries 65 years and older and were increasingly common. Complications associated with TKA and THA are potentially life-threatening. The developer notes that the median risk standardized complication rate (RSCR) is 2.7% at the clinician level and 2.8% at the clinician group level. The 10th and 90th percentile RSCRs (2.2 and 3.7, respectively, for clinicians; 2.2 and 3.5, respectively, for clinician groups) represent meaningful deviations from the median: a clinician performing at the 10th percentile is performing 18.5% better than an average performer (21.4% for clinician group), while a clinician performing at the 90th percentile is performing nearly 37.0% worse than an average performer (25% for clinician groups). Best performing clinicians (1.2%) are performing 55.6% better than an average performer, while worst performing clinicians (7.2%) are performing 166.7% worse than an average performer. The best performing clinician groups (1.4%) are performing 50% better than an average performer, while worst performing clinicians (5.7%) are performing 103.6% worse than an average performer.
  
  
• Does the measure contribute to efficient use of measurement resources and/or support alignment of measurement across programs? Yes. The hospital version of this measure (NQF 1550) is currently being used in IQR.
  
  
• Can the measure can be feasibly reported? Yes. This claims-based measure draws on data generated through the routine provision of care.
  
  
• Is the measure applicable to and appropriately specified for the program's intended care setting(s), level(s) of analysis, and population(s)? Yes. Developer submitted appropriate specifications for this measure to be used within MIPS. Measure is currently endorsed as NQF 3493 at the clinician: individual and group levels.
  
  
• Measure development status: Fully Developed
  
  
• If the measure is in current use, have negative unintended issues to the patient been identified? No. Developer did not note any unintended consequences in the 2017 maintenance endorsement submission for NQF 1550, nor in the submission for NQF 3493.
  
  
• MAP Rural Workgroup Finding:
  • Relative priority/utility: Hip and knee replacements are high-volume procedures for rural residents.
  • Data collection issues: None.
  • Calculation issues: Unclear if the measure will be limited to clinicians/clinician groups with at least 25 patients. If not, low case-volume likely to be an issue for rural providers.
  • Unintended consequences: Access to supportive services prior to surgery will be even more critical when these procedures are done in the outpatient setting; however, access to such services may be more limited in rural areas.
  • Votes: Range is 1 – 5, where higher reflects more agreement regarding suitability for the program
  • Average agreement= 3.2
  • Vote Counts: (1 – 1 vote1; 2 – 1 vote; 3 – 4 votes; 4 – 5 votes; 5 – 0 votes)
  • Program gap areas: None identified.
  
  
  

Rationale for measure provided by HHS
HHS has provided additional documentation to support this measure under consideration. There is evidence that over time, hospital Total Hip Arthroplasty and/or Total Knee Arthroplasty (THA/TKA) volumes have increased, while hospital THA/TKA risk-standardized complication rates (RSCRs) have decreased. This evidence supports the fact that improving complication rates is possible and feasible. There is evidence that specific practices can reduce the chances of complications [1-2]. By attributing the outcome to clinicians who care for inpatient THA/TKA patients, the Merit Based Incentive Payment System (MIPS) THA/TKA complication measure will incentivize those clinicians to promote practices known to reduce post-operative complications and identify new interventions at the clinician level that may also do so. Studies have demonstrated that hospital-based interventions targeting critical aspects of care can reduce the risk of complications such as strategies to reduce blood loss, reduce length of stay, and routine wound care [3-4]. References:1.Kocher MS, Frank JS, Nasreddine AY, et al. Intra-abdominal fluid extravasation during hip arthroscopy: a survey of the MAHORN group. Arthroscopy : the journal of arthroscopic & related surgery : official publication of the Arthroscopy Association of North America and the International Arthroscopy Association. 2012;28(11):1654-1660.e1652.2.Ponzio DY, Poultsides LA, Salvatore A, Lee YY, Memtsoudis SG, Alexiades MM. In-Hospital Morbidity and Postoperative Revisions After Direct Anterior vs Posterior Total Hip Arthroplasty. J Arthroplasty. 2017.3.Chen AF, Heyl AE, Xu PZ, Rao N, Klatt BA. Preoperative Decolonization Effective at Reducing Staphylococcal Colonization in Total Joint Arthroplasty Patients. The Journal of Arthroplasty. 2013;28(8, Supplement):18-20.4.Rao N, Cannella BA, Crossett LS, Yates AJ, McGough RL, Hamilton CW. Preoperative Screening/Decolonization for Staphylococcus aureus to Prevent Orthopedic Surgical Site Infection: Prospective Cohort Study With 2-Year Follow-Up. The Journal of Arthroplasty. 2011;26(8):1501-1507.

Summary of NQF Endorsement Review

• Year of Most Recent Endorsement Review: 2019
  
  
• Project for Most Recent Endorsement Review: Surgery
  
  
• Review for Importance: 1a. Evidence: Pass-17; No Pass-0; 1b. Performance Gap: H-2; M-15; L-0; I-0Rationale:•Because this is a re-specified measure (1550), the Committee questioned whether the distinctions between 1550 and 3493 were great enough to justify having both hospital and provider versions. The Committee discussed that complication rates can be brought down by standardization of process and agreed that there was great value to having provider-level outcomes data available.•The Committee agreed that there is significant variation in complication rates, as evidenced by the hospital measure (1550), which demonstrates first and tenth decile rates of 1.9% and 4.3%). Additionally, for the clinician level, the risk-standardized measure scores had a mean (SD) of 2.83% (0.65%) and for the clinician group level, the risk-standardized measure scores had a mean (SD) of 2.81% (0.51%), demonstrating a performance gap that the Committee deemed moderate.
  
  
• Review for Scientific Acceptability: 2a. Reliability: H-2; M-14; L-1; I-0; 2b. Validity: H-2; M-13; L-2; I-0Rationale:•This measure was reviewed by the Scientific Methods Panel (SMP) and passed both reliability and validity.•Reliability testing was conducted at the measure score level and used Adams method to estimate entity-level reliability. The entity-level reliability testing indicated that for entities with 25 procedures or more, the median signal-to-noise ratio reliability was 0.793 [IQR 0.695-0.878] for clinicians, and 0.790 [IQR 0.647-0.907] for clinician groups. The median reliability scores reflected the reliability of the hospital-level measure score (NQF 1550).•Validity testing was demonstrated through empirical validity testing and by systematic assessment of the measure’s face validity by a technical expert panel (TEP) of national experts and stakeholder organizations. For empirical validity testing, the developer examined the relationship between volume and the measure score for clinicians and clinician groups. Correlations between volume and measure score were calculated for each provider type, and the measure score for each decile of volume was summarized. There was a moderate, yet meaningful, inverse relationship between volume and measure outcome for both clinicians (correlation coefficient of -0.2379; p<0.0001) and clinician groups (correlation coefficient of -0.19026; p<0.0001). Furthermore, the TEP supported the final measure with high agreement.•The Committee addressed a comment made by the American Medical Association (AMA) that questioned whether the case minimum of 25 cases was acceptable, given the low reliability results (0.582 to 0.988 and 0.463 to 0.996 for clinicians and clinician groups, respectively). The developer reported that the 25 cases provided was acceptable reliability while capturing lower volume providers.•The Committee noted that the database is an administrative database for CMS that is based on submitted diagnosis codes for billing, which is less valid than registry data.•The Committee discussed a comment made by the AMA that stated additional testing is needed to evaluate clinical factors in conjunction with social risk factors, as opposed to prioritizing clinical factors. The Committee agreed that including volume as a risk adjuster would not identify important modifiable risk factors that the measure should identify.•The Committee raised questions regarding the inclusion criteria for the measure: continuous 12-month enrollment in Medicare Part A. The developer clarified that this criterion is in place to ensure that all co-morbidities are captured adequately for risk adjustment, as well as for the duration that they are evaluating for complications. Additionally, it was pointed out that this measure is specified for fee-for-service (FFS) beneficiaries only and therefore does not capture Medicare Advantage patients.•It was recommended that it would be worth researching whether there is a substantial difference between FFS beneficiaries and those enrolled in Medicare Advantage. Since Medicare Advantage is more cost-effective for parts of the country, leaving out this population may blunt an important socioeconomic risk adjustment.
  
  
• Review for Feasibility: 3. Feasibility: H-6; M-11; L-0; I-0(3a. Clinical data generated during care delivery; 3b. Electronic sources; 3c. Susceptibility to inaccuracies/ unintended consequences identified 3d. Data collection strategy can be implemented)Rationale:•The data elements can be found in defined fields in electronic claims, and administrative data are routinely collected as part of the billing process. The measure was designed to capture data that are already present in administrative data collection. There are no fees, licensing, or other requirements reported to use any aspect of the measure.•NQF measure 1550 was deemed feasible when it was originally evaluated and endorsed. The Committee agreed that feasibility is moderate to high for this measure.
  
  
• Review for Usability: 4. Use and Usability4a. Use; 4a1. Accountability and transparency; 4a2. Feedback on the measure by those being measured and others; 4b. Usability; 4b1. Improvement; 4b2. The benefits to patients outweigh evidence of unintended negative consequences to patients)4a. Use: Pass-17; No Pass-0 4b. Usability: H-3; M-13; L-1; I-0Rationale:•Since this is a new measure, there are currently no public reporting targets. However, per the developer, the primary goal of the measure is to provide information necessary to implement focused quality improvement efforts.•The Committee discussed that expanding this measure to all-payer or to a broader population would have great usability.
  
  
• Review for Related and Competing Measures: This measure is harmonized with measure 1550 regarding cohort definition, outcome, and risk adjustment approach. The only discrepancy is the attribution approach, but instead of assigning each index admission to a hospital (1550), it assigns it to a clinician or a clinician group.
  
  
• Endorsement Committee Recommendation: Standing Committee Recommendation for Endorsement: Y-17; N-0
  
  

Measure Specifications

• NQF Number (if applicable): 0
  
  
• Description: Annual risk-standardized rate of acute, unplanned hospital admissions among Medicare Fee-for-Service (FFS) patients aged 65 years and older with multiple chronic conditions (MCCs).
  
  
• Numerator: The outcome for this measure is the number of acute, unplanned hospital admissions per 100 person-years at risk for admission during the measurement period. Time PeriodThe outcome includes inpatient admissions to an acute care hospital during the measurement year. Excluded AdmissionsThis measure does not include the following types of admissions in the outcome because they do not reflect the quality of care provided by ambulatory care clinicians who are managing the care of MCC patients: 1. Planned hospital admissions.2. Admissions that occur directly from a skilled nursing facility (SNF) or acute rehabilitation facility.3. Admissions that occur within a 10-day “buffer period†of time after discharge from a hospital, SNF, or acute rehabilitation facility.4. Admissions that occur after the patient has entered hospice.5. Admissions related to complications from procedures or surgeries.6. Admissions related to accidents or injuries.7. Admissions that occur prior to the first visit with the assigned clinician.To identify planned admissions, the measure adopted an algorithm CORE previously developed for CMS's hospital readmission measures, CMS's Planned Readmission Algorithm Version 4.0. [1,2] In brief, the algorithm uses the procedure codes and principal discharge diagnosis code on each hospital claim to identify planned admissions. A few specific, limited types of care are always considered planned (for example, major organ transplant, rehabilitation, and maintenance chemotherapy). Otherwise, a planned admission is defined as a non-acute admission for a scheduled procedure (for example, total hip replacement or cholecystectomy). Admissions for an acute illness are never considered planned. To identify complications of procedures or surgeries, we use the Agency for Healthcare Research and Quality's (AHRQ's) Clinical Classifications Software (CCS), which clusters diagnoses into clinically meaningful categories using International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) or International Classification of Diseases, Tenth Revision, and Clinical Modification (ICD-10-CM) codes. We exclude the following 23 CCS categories. 1. 145: Intestinal obstruction without hernia2. 237: Complication of device; implant or graft3. 238: Complications of surgical procedures or medical care4. 257: Other aftercareb) Accidents or injuries 5. 2601 E Codes: Cut/pierce6. 2602 E Codes: Drowning/submersion7. 2604 E Codes: Fire/burn8. 2605 E Codes: Firearm9. 2606 E Codes: Machinery10. 2607 E Codes: Motor vehicle traffic (MVT)11. 2608 E Codes: Pedal cyclist; not MVT12. 2609 E Codes: Pedestrian; not MVT13. 2610 E Codes: Transport; not MVT14. 2611 E Codes: Natural/environment15. 2612 E Codes: Overexertion16. 2613 E Codes: Poisoning17. 2614 E Codes: Struck by; against18. 2615 E Codes: Suffocation19. 2616 E Codes: Adverse effects of medical care20. 2618 E Codes: Other specified and classifiable21. 2619 E Codes: Other specified; NEC22. 2620 E Codes: Unspecified23. 2621 E Codes: Place of occurrencePerson-time at riskPersons are considered at risk for hospital admission if they are alive, enrolled in Medicare FFS, and not in the hospital during the measurement period. In addition to time spent in the hospital, we also exclude from at-risk time: 1) time spent in a SNF or acute rehabilitation facility; 2) the time within 10 days following discharge from a hospital, SNF, or acute rehabilitation facility; and 3) time after entering hospice care.Citations1. Yale New Haven Health Services Corporation – Center for Outcomes Research & Evaluation (YNHHSC/CORE). 2018 All-Cause Hospital Wide Measure Updates and Specifications Report - Hospital-Level 30-Day Risk-Standardized Readmission Measure – Version 7.0. Centers for Medicare & Medicaid Services; March 2018.2. Horwitz L, Grady J, Cohen D, et al. Development and validation of an algorithm to identify planned readmissions from claims data. Journal of Hospital Medicine. October 2015;10(10):670-677.
  
  
• Denominator: Patients included in the measure (target patient population)The cohort is comprised of patients whose combinations of chronic conditions put them at high risk of admission and whose admission rates could be lowered through better care. This definition reflects NQF's Multiple Chronic Conditions Measurement Framework, which defines patients with MCCs as people having two or more concurrent chronic conditions that. . .act together to significantly increase the complexity of management, and affect functional roles and health outcomes, compromise life expectancy, or hinder self-management. [1] The specific inclusion criteria are as follows.Patient is alive at the start of the measurement period and has two or more of nine chronic disease groups in the year prior to the measurement period. Chronic conditions, except for diabetes, are defined using CMS's Chronic Conditions Data Warehouse (CCW). For diabetes, we used the diabetes cohort definition from the Accountable Care Organization (ACO) diabetes admission measure developed by CORE (v2018a ACO-36) as opposed to the definition used in CCW; CCW includes diagnoses for secondary and drug-induced diabetic conditions that are not the focus of the MIPS MCC admission measure. 1. Acute myocardial infarction (AMI),2. Alzheimer's disease and related disorders or senile dementia,3. Atrial fibrillation,4. Chronic kidney disease (CKD),5. Chronic obstructive pulmonary disease (COPD) or asthma,6. Depression,7. Diabetes,8. Heart failure, and9. Stroke or transient ischemic attack (TIA).Patient is aged >=65 years at the start of the year prior to the measurement period.Patient is a Medicare FFS beneficiary with continuous enrollment in Medicare Parts A and B during the year prior to the measurement period.Provider types included for measurement Primary care providers (PCPs): CMS designates PCPs as physicians who practice internal medicine, family medicine, general medicine, or geriatric medicine, and non-physician providers, including nurse practitioners, certified clinical nurse specialists, and physician assistants.Relevant specialists: Specialists covered by the measure are limited to those who provide overall coordination of care for patients with MCCs and who manage the chronic diseases that put the MCC patients in the measure at risk of admission. These specialists were chosen with input from our Technical Expert Panel (TEP) and include cardiologists, pulmonologists, nephrologists, neurologists, endocrinologists, and hematologists/oncologists.Outcome attributionWe begin by assigning each patient to the clinician most responsible for the patient’s care, based on the pattern of outpatient visits with PCPs and relevant specialists. The patient can be assigned to a PCP, a relevant specialist, or can be left unassigned. A patient who is eligible for attribution can be assigned to a relevant specialist only if the specialist has been identified as "dominant". A specialist is considered "dominant" if they have two or more visits with the patient, as well as at least two more visits than any primary care provider or other relevant specialist.There are two scenarios where a patient can be assigned to a PCP. First, the patient must have seen at least one PCP. The patient will then be assigned to the PCP with the highest number of visits if there are no relevant specialists who are considered “dominantâ€. Second, if the patient has had more than one visit with a relevant specialist, no "dominant" specialist has been identified, and has two or more visits with a PCP, they will be assigned to that PCP.Finally, the patient will be unassigned if they only saw non-relevant specialists, if the patient has not seen a PCP and no "dominant" specialist can be identified, or if the patient has not had more than one visit with any individual PCP.Patients are then assigned at the Taxpayer Identification Number (TIN) level, which includes solo clinicians and groups of clinicians who have chosen to report their quality under a common TIN. Patients "follow" their clinician to the TIN designated by the clinician (i.e. they are assigned to their clinician's TIN). Patients unassigned at the individual clinician-level, therefore, continue to be unassigned at the TIN level.Citations1. National Quality Forum. Multiple Chronic Conditions Measurement Framework. http://www.qualityforum.org/WorkArea/linkit.aspx?LinkIdentifier=id&ItemID=71227. Accessed February 20, 2019.
  
  
• Exclusions: The cohort excludes the following patients:1) Patients without continuous enrollment in Medicare Part A or B during the measurement period.2) Patients who were in hospice at any time during the year prior to the measurement year or at the start of the measurement year.3) Patients who had no Evaluation & Management (E&M) visits to a MIPS-eligible clinician.
  
  
• HHS NQS Priority: Communication and Care Coordination
  
  
• HHS Data Source: Claims
  
  
• Measure Type: Outcome
  
  
• Steward: Centers for Medicare & Medicaid Services
  
  
• Endorsement Status:
  
  
• Meaningful Measure Area: Management of chronic conditions
  
  
• Is the measure specified as an electronic clinical quality measure? No
  
  

Preliminary Analysis of Measure

• Preliminary analysis result: Conditional Support for Rulemaking
  
  
• Preliminary analysis summary
  • Contribution to program measure set:This measure addresses a critical outcome of hospital admission for patients with MCC. However, the analysis presented by the measure developer does not include any analysis of how this measure performs. This measure would be appropriate for SSP if the developer can present analyses that the re-specified measure is also reliable and valid by resubmitting the measure for endorsement to the NQF All Cause Admission and Readmission Standing Committee.
  • Impact on quality of care for patients:ACOs in SSP will focus on processes and interventions that reduce disease progression and undesirable sequalae that lead to hospital admission for Medicare patients with MCC. With over 80% of adults over the age of 65 having MCCs, this measure has the potential to significantly impact the quality of care for the Medicare beneficiary population.
  
  
  
• Does the measure address a critical quality objective not currently adequately addressed by the measures in the program set? Yes. SSP currently has four measures that address the priority for Care Coordination/Patient Safety in the 2019 Measure Set:ACO-8 Risk-Standardized, All Condition ReadmissionACO-13 Falls: Screening for Future FallsACO-38 Risk-Standardized Acute Admission Rates for Patients with Multiple Chronic ConditionsACO-43 Ambulatory Sensitive Condition Acute Composite This measure is re-specified to replace ACO-38/ NQF 2888 and to align with the same measure submitted for MIPS.
  
  
• Is the measure evidence-based and either strongly linked to outcomes or an outcome measure? Yes. This is an outcome measure with the same title but modified specifications to an existing NQF endorsed measure (NQF 2888), last reviewed for endorsement in 2016. In the submission for NQF 2888, the developer notes that “providers can potentially lower the risk of acute admissions in this high-risk population through better coordinated, timelier, and more effective health care. Hence, efforts to redesign care for patients with MCCs have used admission rates as one outcome to evaluate the success of interventions.” Such interventions could include creating and sharing a centralized care plan for high-risk patients, ensuring transportation to and from appointments, and referrals to specialty care for patients with behavioral health diagnoses.
  
  
• Does the measure address a quality challenge? Yes. Over 40% of all US adults, and over 80% of adults over the age of 65 have MCCs—a figure that has remained steady over time. While the population with 3+ chronic conditions is fewer than 1/3 of the total US population, they constitute more than 2/3 of the total medical spend. Hospitalization from preventable causes represents a costly poor health outcome associated with disease progression for beneficiaries with MCCs. This measure encourages ACOs to focus on coordinating care and avoiding preventable hospitalizations.A gap analysis for the respecified measure for SSP using ACO data was not submitted. In the NQF endorsement submission for NQF 2888, the developer used data from the 2012 Medicare Full Sample with 239,551 patients in 114 ACOs. The mean risk-standardized acute admission rate (RSAAR) among ACOs for year 2012 was 69.3 per 100 person years. 45 ACOs (39.5%) had RSAARs that were ‘no different than the national rate’ and 22 ACOs (19.3%) had RSAAR scores ‘worse than the national rate,’ and 47 ACOs (41.2%) were ‘better than the national rate.’
  
  
• Does the measure contribute to efficient use of measurement resources and/or support alignment of measurement across programs? Yes. The measure is currently in the SSP measure set and is under consideration for inclusion in MIPS.
  
  
• Can the measure can be feasibly reported? Yes. The data source for this measure is Medicare administrative claims and enrollment data and is readily available to CMS. Calculating the measure score imposes no data collection burden for CMS or entities measured.
  
  
• Is the measure applicable to and appropriately specified for the program's intended care setting(s), level(s) of analysis, and population(s)? Yes. The measure as modified could replace NQF 2888. The re-specified measure should be submitted to the NQF All Cause Admission and Readmission Standing Committee for maintenance of endorsement prior to implementation within SSP.
  
  
• Measure development status: Fully Developed
  
  
• If the measure is in current use, have negative unintended issues to the patient been identified? No. No negative consequences reported by developer. Also, none reported in the 2016 endorsement submission for NQF 2888.
  
  
• MAP Rural Workgroup Finding:
  • Relative priority/utility: The chronic conditions included in this measure are prevalent in rural residents. The Workgroup noted that many rural residents are attributed to ACOs that operate in rural areas.
  • Data collection issues: None
  • Calculation issues: The Workgroup believes this measure is appropriate for accountability at the ACO level.
  • Unintended consequences: The relative lack of access to social, human and medical services in rural areas could make this unfair to ACOs, particularly given the lack of adjustment for social risk factors. While telehealth could help, there often is a lack of broadband technology in rural areas.
  • Votes: Range is 1 – 5, where higher reflects more agreement regarding suitability for the program
  • Average agreement= 3.9
  • Vote Counts: (1 – 0 votes; 2 – 0 votes; 3 – 3 votes; 4 – 8 votes; 5 – 2 votes)
  • Program gap areas: None identified.
  
  
  

Rationale for measure provided by HHS
Hospital admission rates are an effective marker of ambulatory care quality. Hospital admissions from the outpatient setting reflect a deterioration in patients’ clinical status and as such reflect an outcome that is meaningful to both patients and providers. Patients receiving optimal, coordinated high-quality care should use fewer inpatient services than patients receiving fragmented, low-quality care. Thus, high population rates of hospitalization may, at least to some extent, signal poor quality of care or inefficiency in health system performance.Patients with MCCs are at high risk for hospital admission, often for potentially preventable causes, such as exacerbation of pulmonary disease. [1] Evidence from several Medicare demonstration projects suggests that care coordination results in decreased hospital admission rates among high-risk patients. [2] In addition, studies have shown that the types of ambulatory care clinicians this measure targets (for example, primary care providers and specialists caring for patients with MCCs) can influence admission rates through primary care clinician supply, continuity of care, and patient-centered medical home interventions such as team-based and patient-oriented care. [3-5]Given evidence that ambulatory care clinicians can influence hospital admission rates through optimal care and coordination, this measure will incentivize quality improvement efforts leading to improved patient outcomes.Citations:1. Abernathy K, Zhang J, Mauldin P, et al. Acute Care Utilization in Patients With Concurrent Mental Health and Complex Chronic Medical Conditions. Journal of primary care & community health. 2016;7(4):226-233. 2. Brown RS, Peikes D, Peterson G, Schore J, Razafindrakoto CM. Six features of Medicare coordinated care demonstration programs that cut hospital admissions of high-risk patients. Health Aff (Millwood). 2012;31(6):1156-1166. 3. van Loenen T, van den Berg MJ, Westert GP, Faber MJ. Organizational aspects of primary care related to avoidable hospitalization: a systematic review. Fam Pract. 2014;31(5):502-516.4. Dale SB, Ghosh A, Peikes DN, et al. Two-Year Costs and Quality in the Comprehensive Primary Care Initiative. N Engl J Med. 2016;374(24):2345-2356.5. Casalino LP, Pesko MF, Ryan AM, et al. Small primary care physician practices have low rates of preventable hospital admissions. Health Aff (Millwood). 2014;33(9):1680-1688.

Measure Specifications

• NQF Number (if applicable): 0
  
  
• Description: Percentage of adult hemodialysis patient-months using a catheter continuously for three months or longer for vascular access attributable to an individual practitioner or group practice.
  
  
• Numerator: The numerator is the number of adult patient-months in the denominator who were on maintenance hemodialysis using a catheter continuously for three months or longer as of the last hemodialysis session of the reporting month.
  
  
• Denominator: All patients at least 18 years old as of the first day of the reporting month who are determined to be maintenance hemodialysis patients (in-center and home HD) for the complete reporting month under the care of the same practitioner or group partner.
  
  
• Exclusions: Exclusions that are implicit in the denominator definition include:Pediatric patients (<18 years old);Patients on Peritoneal Dialysis for any portion of the reporting month; Patient-months where there are more than one MCP provider listed for the month. In addition, patients with a catheter that have limited life expectancy, as defined by the following criteria are excluded: Patients under hospice care in the current reporting month;Patients with metastatic cancer in the past 12 months;Patients with end stage liver disease in the past 12 months;Patients with coma or anoxic brain injury in the past 12 months
  
  
• HHS NQS Priority: Communication and Care Coordination
  
  
• HHS Data Source: Claims;CROWNWeb
  
  
• Measure Type: Intermediate Outcome
  
  
• Steward: Centers for Medicare & Medicaid Services
  
  
• Endorsement Status:
  
  
• Meaningful Measure Area: Management of chronic conditions
  
  
• Is the measure specified as an electronic clinical quality measure? No
  
  

Preliminary Analysis of Measure

• Preliminary analysis result: Conditional Support for Rulemaking
  
  
• Preliminary analysis summary
  • Contribution to program measure set:This measure addresses the critical quality objective of MIPS by seeking to promote effective prevention and treatment of chronic disease while also filling a gap currently evident by only having three catheter and/or hemodialysis measures in the MIPS measure set. By using widely accessible claims and CROWNWeb data this measure would provide value MIPS by adding a hemodialysis measure without increasing burden on clinicians. This measure has not been submitted to NQF for review of reliability and validity testing.
  • Impact on quality of care for patients:This measure has the potential to impact the over 726,000 individuals who were on dialysis or received a kidney transplant as of 2016. In 2016, 9.7% of patient-months on hemodialysis used a long-term catheter. The use of a long-term catheter has been observed with a higher mortality rate than the use of arteriovenous fistula, thus this measure has the potential to provide greater quality of care for patients by reducing their mortality rate.
  
  
  
• Does the measure address a critical quality objective not currently adequately addressed by the measures in the program set? Yes. This measure addresses the critical quality objective of Making Care Safer, by aiming to promote effective prevention and treatment of chronic disease. The measure seeks to reduce deaths due to the use of a long-term catheter for vascular access in hemodialysis patients.Of the 258 MIPS measures, only three address catheters, and while all seek to improve care, none of the current MIPS measures are directed towards reducing catheter-associated mortality.
  
  
• Is the measure evidence-based and either strongly linked to outcomes or an outcome measure? Yes. This measure is an intermediate outcome measure. Observational studies show that long-term catheters use in hemodialysis patient have the highest risk of mortality.
  
  
• Does the measure address a quality challenge? Yes. The developer notes that 9.7% of patients-months on hemodialysis used a long-term catheter.
  
  
• Does the measure contribute to efficient use of measurement resources and/or support alignment of measurement across programs? Yes. This measure is used in the End-Stage Renal Disease Quality Incentive Program (ESRD QIP).
  
  
• Can the measure can be feasibly reported? Yes. This measure uses both widely available claims data and CROWNWeb data. These data points are collected by and used by healthcare professionals during provision of care.
  
  
• Is the measure applicable to and appropriately specified for the program's intended care setting(s), level(s) of analysis, and population(s)? Yes. The developer notes that the measure was tested at the clinician level. Testing results have not been provided.
  
  
• Measure development status: Fully Developed
  
  
• If the measure is in current use, have negative unintended issues to the patient been identified? No. This measure is used in ESRD QIP. No negative unintended consequences have been reported. However, a potential unintended consequence of this measure would be pushing patients towards having arteriovenous fistula when they may not realize a benefit from this type of access due to either comorbidities or a limited life expectancy.
  
  
• MAP Rural Workgroup Finding:
  • Relative priority/utility: Diabetes and kidney disease are prevalent conditions in rural populations.
  • Data collection issues: None.
  • Calculation issues: None.
  • Unintended consequences: The workgroup noted that rural patients on dialysis are older and have more comorbidities, and voiced concern that these patients might be pressed to use a fistula, even when there is little benefit.
  • Votes: Range is 1 – 5, where higher reflects more agreement regarding suitability for the program
  • Average agreement= 3.7
  • Vote Counts: (1 – 0 votes; 1; 2 – 0 votes; 3 – 3 votes; 4 – 7 votes; 5 – 0 votes)
  • Program gap areas: None identified.
  
  
  

Rationale for measure provided by HHS
HHS has provided additional documentation to support this measure under consideration. Several observational studies have demonstrated an association between type of vascular access used for hemodialysis and patient mortality. Long term catheter use is associated with the highest mortality risk while arteriovenous fistula use has the lowest mortality risk. Arteriovenous grafts (AVG) have been found to have a risk of death that is higher than AVF but lower than catheters. The measure focus is the process of assessing long term catheter use at chronic dialysis facilities.This process leads to improvement in mortality as follows:Measure long term catheter rate -> Assess value -> Identify patients who do not have an AV Fistula or AV graft->Evaluation for an AV fistula or graft by a qualified dialysis vascular access provider -> Increase Fistula/Graft Rate -> Lower catheter rate ->Lower patient mortality.

Measure Specifications

• NQF Number (if applicable): 0
  
  
• Description: Annual risk-standardized rate of acute, unplanned hospital admissions among Medicare Fee-for-Service (FFS) patients aged 65 years and older with multiple chronic conditions (MCCs).
  
  
• Numerator: The outcome for this measure is the number of acute, unplanned hospital admissions per 100 person-years at risk for admission during the measurement period. Time PeriodThe outcome includes inpatient admissions to an acute care hospital during the measurement year. Excluded AdmissionsThis measure does not include the following types of admissions in the outcome because they do not reflect the quality of care provided by ambulatory care clinicians who are managing the care of MCC patients: 1. Planned hospital admissions.2. Admissions that occur directly from a skilled nursing facility (SNF) or acute rehabilitation facility.3. Admissions that occur within a 10-day “buffer period†of time after discharge from a hospital, SNF, or acute rehabilitation facility.4. Admissions that occur after the patient has entered hospice.5. Admissions related to complications from procedures or surgeries.6. Admissions related to accidents or injuries.7. Admissions that occur prior to the first visit with the assigned clinician.To identify planned admissions, the measure adopted an algorithm CORE previously developed for CMS's hospital readmission measures, CMS's Planned Readmission Algorithm Version 4.0. [1,2] In brief, the algorithm uses the procedure codes and principal discharge diagnosis code on each hospital claim to identify planned admissions. A few specific, limited types of care are always considered planned (for example, major organ transplant, rehabilitation, and maintenance chemotherapy). Otherwise, a planned admission is defined as a non-acute admission for a scheduled procedure (for example, total hip replacement or cholecystectomy). Admissions for an acute illness are never considered planned. To identify complications of procedures or surgeries, we use the Agency for Healthcare Research and Quality's (AHRQ's) Clinical Classifications Software (CCS), which clusters diagnoses into clinically meaningful categories using International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) or International Classification of Diseases, Tenth Revision, and Clinical Modification (ICD-10-CM) codes. We exclude the following 23 CCS categories. 1. 145: Intestinal obstruction without hernia2. 237: Complication of device; implant or graft3. 238: Complications of surgical procedures or medical care4. 257: Other aftercareb) Accidents or injuries 5. 2601 E Codes: Cut/pierce6. 2602 E Codes: Drowning/submersion7. 2604 E Codes: Fire/burn8. 2605 E Codes: Firearm9. 2606 E Codes: Machinery10. 2607 E Codes: Motor vehicle traffic (MVT)11. 2608 E Codes: Pedal cyclist; not MVT12. 2609 E Codes: Pedestrian; not MVT13. 2610 E Codes: Transport; not MVT14. 2611 E Codes: Natural/environment15. 2612 E Codes: Overexertion16. 2613 E Codes: Poisoning17. 2614 E Codes: Struck by; against18. 2615 E Codes: Suffocation19. 2616 E Codes: Adverse effects of medical care20. 2618 E Codes: Other specified and classifiable21. 2619 E Codes: Other specified; NEC22. 2620 E Codes: Unspecified23. 2621 E Codes: Place of occurrencePerson-time at riskPersons are considered at risk for hospital admission if they are alive, enrolled in Medicare FFS, and not in the hospital during the measurement period. In addition to time spent in the hospital, we also exclude from at-risk time: 1) time spent in a SNF or acute rehabilitation facility; 2) the time within 10 days following discharge from a hospital, SNF, or acute rehabilitation facility; and 3) time after entering hospice care.Citations1. Yale New Haven Health Services Corporation – Center for Outcomes Research & Evaluation (YNHHSC/CORE). 2018 All-Cause Hospital Wide Measure Updates and Specifications Report - Hospital-Level 30-Day Risk-Standardized Readmission Measure – Version 7.0. Centers for Medicare & Medicaid Services; March 2018.2. Horwitz L, Grady J, Cohen D, et al. Development and validation of an algorithm to identify planned readmissions from claims data. Journal of Hospital Medicine. October 2015;10(10):670-677.
  
  
• Denominator: Patients included in the measure (target patient population)The cohort is comprised of patients whose combinations of chronic conditions put them at high risk of admission and whose admission rates could be lowered through better care. This definition reflects NQF's Multiple Chronic Conditions Measurement Framework,which defines patients with MCCs as people having two or more concurrent chronic conditions that. . .act together to significantly increase the complexity of management, and affect functional roles and health outcomes, compromise life expectancy, or hinder self-management. [1] The specific inclusion criteria are as follows.Patient is alive at the start of the measurement period and has two or more of nine chronic disease groups in the year prior to the measurement period. Chronic conditions, except for diabetes, are defined using CMS’s Chronic Conditions Data Warehouse (CCW). For diabetes, we used the diabetes cohort definition from the Accountable Care Organization (ACO) diabetes admission measure developed by CORE (v2018a ACO-36) as opposed to the definition used in CCW; CCW includes diagnoses for secondary and drug-induced diabetic conditions that are not the focus of the MIPS MCC admission measure. 1. Acute myocardial infarction (AMI),2. Alzheimer's disease and related disorders or senile dementia,3. Atrial fibrillation,4. Chronic kidney disease (CKD),5. Chronic obstructive pulmonary disease (COPD) or asthma,6. Depression,7. Diabetes,8. Heart failure, and9. Stroke or transient ischemic attack (TIA).Patient is aged >=65 years at the start of the year prior to the measurement period.Patient is a Medicare FFS beneficiary with continuous enrollment in Medicare Parts A and B during the year prior to the measurement period.Provider types included for measurement Primary care providers (PCPs): CMS designates PCPs as physicians who practice internal medicine, family medicine, general medicine, or geriatric medicine, and non-physician providers, including nurse practitioners, certified clinical nurse specialists, and physician assistants.Relevant specialists: Specialists covered by the measure are limited to those who provide overall coordination of care for patients with MCCs and who manage the chronic diseases that put the MCC patients in the measure at risk of admission. These specialists were chosen with input from our Technical Expert Panel (TEP) and include cardiologists, pulmonologists, nephrologists, neurologists, endocrinologists, and hematologists/oncologists.Outcome attributionWe begin by assigning each patient to the clinician most responsible for the patient’s care, based on the pattern of outpatient visits with PCPs and relevant specialists. The patient can be assigned to a PCP, a relevant specialist, or can be left unassigned. A patient who is eligible for attribution can be assigned to a relevant specialist only if the specialist has been identified as “dominantâ€. A specialist is considered “dominant†if they have two or more visits with the patient, as well as at least two more visits than any primary care provider or other relevant specialist.There are two scenarios where a patient can be assigned to a PCP. First, the patient must have seen at least one PCP. The patient will then be assigned to the PCP with the highest number of visits if there are no relevant specialists who are considered "dominant". Second, if the patient has had more than one visit with a relevant specialist, no "dominant" specialist has been identified, and has two or more visits with a PCP, they will be assigned to that PCP.Finally, the patient will be unassigned if they only saw non-relevant specialists, if the patient has not seen a PCP and no "dominant" specialist can be identified, or if the patient has not had more than one visit with any individual PCP.Patients are then assigned at the Taxpayer Identification Number (TIN) level, which includes solo clinicians and groups of clinicians who have chosen to report their quality under a common TIN. Patients follow their clinician to the TIN designated by the clinician (i.e. they are assigned to their clinician's TIN). Patients unassigned at the individual clinician-level, therefore, continue to be unassigned at the TIN level.Citations1. National Quality Forum. Multiple Chronic Conditions Measurement Framework. http://www.qualityforum.org/WorkArea/linkit.aspx?LinkIdentifier=id&ItemID=71227. Accessed February 20, 2019.
  
  
• Exclusions: The cohort excludes the following patients:1) Patients without continuous enrollment in Medicare Part A or B during the measurement period.2) Patients who were in hospice at any time during the year prior to the measurement year or at the start of the measurement year.3) Patients who had no Evaluation & Management (E&M) visits to a MIPS-eligible clinician.
  
  
• HHS NQS Priority: Communication and Care Coordination
  
  
• HHS Data Source: Claims
  
  
• Measure Type: Outcome
  
  
• Steward: Centers for Medicare & Medicaid Services
  
  
• Endorsement Status:
  
  
• Meaningful Measure Area: Management of chronic conditions
  
  
• Is the measure specified as an electronic clinical quality measure? No
  
  

Preliminary Analysis of Measure

• Preliminary analysis result: Conditional Support for Rulemaking
  
  
• Preliminary analysis summary
  • Contribution to program measure set:This measure addresses a critical outcome of hospital admission for patients with MCC. However, the NQF Scientific Methods Panel should consider whether this measure has adequate reliability at the score level, a critical element to ensuring that clinicians are appropriately graded on their performance.
  • Impact on quality of care for patients:Clinicians that elect to use this measure for MIPS would focus on processes and interventions that reduce disease progression and undesirable sequalae that lead to hospital admission for Medicare patients with MCC. With over 80% of adults over the age of 65 having MCCs, this measure has the potential to significantly impact the quality of care for the Medicare beneficiary population.
  
  
  
• Does the measure address a critical quality objective not currently adequately addressed by the measures in the program set? Yes. MIPS currently has 30 measures in the priority area of Communication and Care Coordination, with three measures related to readmissions in the 2019 measure set. All-cause Hospital ReadmissionUnplanned Hospital Readmission within 30 Days of Principal ProcedureUnplanned Reoperation within the 30 Day Postoperative Period There are no measures for admissions for patients with multiple chronic conditions.
  
  
• Is the measure evidence-based and either strongly linked to outcomes or an outcome measure? Yes. This is an outcome measure with the same title but modified specifications to an existing NQF endorsed measure (NQF 2888), last reviewed for endorsement in 2016. In the submission for NQF 2888, the developer notes that “providers can potentially lower the risk of acute admissions in this high-risk population through better coordinated, more timely, and more effective health care. Hence, efforts to redesign care for patients with MCCs have used admission rates as one outcome to evaluate the success of interventions.”
  
  
• Does the measure address a quality challenge? Yes. Over 40% of all US adults, and over 80% of adults over the age of 65 have MCCs—a figure that has remained steady over time. While the population with 3+ chronic conditions is fewer than 1/3 of the total US population, they constitute more than 2/3 of the total medical spend. Hospitalization from preventable causes represents a costly poor health outcome associated with disease progression for beneficiaries with MCCs.The measure developer reports a performance gap for clinicians on this measure, stating that across the 65,242 TINs who had at least one MCC patient, RSAAR measure scores ranged from 17.3 to 113.5 per 100 person-years, with a median of 42.1 and an IQR of 39.6 to 45.4. Overall, measure results suggest that there is opportunity to reduce the number of admissions for this patient population, decrease the variation in admissions across providers, and that improvement goals are achievable.
  
  
• Does the measure contribute to efficient use of measurement resources and/or support alignment of measurement across programs? Yes. This measure is currently in SSP as NQF 2888, now under consideration for MIPS and SSP with slightly different specifications from NQF 2888. The SSP measure will be updated to align with the specifications of the MIPS measure under consideration.
  
  
• Can the measure can be feasibly reported? Yes. The data source for this measure is Medicare administrative claims and enrollment data and is readily available to CMS. Calculating the measure score imposes no data collection burden for CMS or entities measured.
  
  
• Is the measure applicable to and appropriately specified for the program's intended care setting(s), level(s) of analysis, and population(s)? No. The measure is modified to be appropriate for clinicians at the individual and group level. However, in the submission, the developer states that they “determined the minimum sample size needed to achieve provider-level measure score reliability of >0.5 (an acceptable cutoff for outcome measures)…” Reliability testing with a cutoff of >0.5 may not be sufficient to demonstrate reliability for use at the clinician or clinician group level.
  
  
• Measure development status: Fully Developed
  
  
• If the measure is in current use, have negative unintended issues to the patient been identified? No. No negative consequences reported by developer. Also, none reported in the endorsement submission for NQF 2888.
  
  
• MAP Rural Workgroup Finding:
  • Relative priority/utility: The chronic conditions included in this measure are prevalent in rural residents. The Workgroup noted that many rural residents are attributed to ACOs that operate in rural areas.
  • Data collection issues: None
  • Calculation issues: The Workgroup believes this measure is appropriate for accountability at the ACO level.
  • Unintended consequences: The relative lack of access to social, human and medical services in rural areas could make this unfair to ACOs, particularly given the lack of adjustment for social risk factors. While telehealth could help, there often is a lack of broadband technology in rural areas.
  • Votes: Range is 1 – 5, where higher reflects more agreement regarding suitability for the program
  • Average agreement= 3.9
  • Vote Counts: (1 – 0 votes; 2 – 0 votes; 3 – 3 votes; 4 – 8 votes; 5 – 2 votes)
  • Program gap areas: None identified.
  
  
  

Rationale for measure provided by HHS
HHS has provided additional documentation to support this measure under consideration. Hospital admission rates are an effective marker of ambulatory care quality. Hospital admissions from the outpatient setting reflect a deterioration in patient's clinical status and as such reflect an outcome that is meaningful to both patients and providers. Patients receiving optimal, coordinated high-quality care should use fewer inpatient services than patients receiving fragmented, low-quality care. Thus, high population rates of hospitalization may, at least to some extent, signal poor quality of care or inefficiency in health system performance.Patients with MCCs are at high risk for hospital admission, often for potentially preventable causes, such as exacerbation of pulmonary disease. [1] Evidence from several Medicare demonstration projects suggests that care coordination results in decreased hospital admission rates among high-risk patients. [2] In addition, studies have shown that the types of ambulatory care clinicians this measure targets (for example, primary care providers and specialists caring for patients with MCCs) can influence admission rates through primary care clinician supply, continuity of care, and patient-centered medical home interventions such as team-based and patient-oriented care. [3-5]Given evidence that ambulatory care clinicians can influence hospital admission rates through optimal care and coordination, this measure will incentivize quality improvement efforts leading to improved patient outcomes.Citations:1. Abernathy K, Zhang J, Mauldin P, et al. Acute Care Utilization in Patients With Concurrent Mental Health and Complex Chronic Medical Conditions. Journal of primary care & community health. 2016;7(4):226-233. 2. Brown RS, Peikes D, Peterson G, Schore J, Razafindrakoto CM. Six features of Medicare coordinated care demonstration programs that cut hospital admissions of high-risk patients. Health Aff (Millwood). 2012;31(6):1156-1166. 3. van Loenen T, van den Berg MJ, Westert GP, Faber MJ. Organizational aspects of primary care related to avoidable hospitalization: a systematic review. Fam Pract. 2014;31(5):502-516.4. Dale SB, Ghosh A, Peikes DN, et al. Two-Year Costs and Quality in the Comprehensive Primary Care Initiative. N Engl J Med. 2016;374(24):2345-2356.5. Casalino LP, Pesko MF, Ryan AM, et al. Small primary care physician practices have low rates of preventable hospital admissions. Health Aff (Millwood). 2014;33(9):1680-1688.

Measure Specifications

• NQF Number (if applicable): 0
  
  
• Description: The percent of emergency department visits for Medicare beneficiaries ages 18 and older with multiple high-risk chronic conditions (MCC) who had a follow-up service within 7 days of the ED visit. Multiple high-risk chronic conditions include 2 or more of the following: Alzheimer's disease, atrial fibrillation, chronic kidney disease, COPD, depression, heart failure, cardiovascular disease evidenced by acute myocardial infarction, and stroke or transient ischemic attack. Appropriate follow-up services include but not limited to: an outpatient visit; telephone visit; transitional or complex care management services, outpatient or telehealth behavioral health visit, or observation visit.
  
  
• Numerator: A follow-up service within 7 days after the ED visit (8 total days)
  
  
• Denominator: ED visits for Medicare beneficiaries ages 18 and older with multiple high-risk chronic conditions
  
  
• Exclusions: 1. Medicare beneficiaries in hospice2. ED visits followed by admission to an acute or non-acute inpatient care setting on the date of the ED visit or within 7 days after the ED visit
  
  
• HHS NQS Priority: Communication and Care Coordination
  
  
• HHS Data Source: Encounter Data
  
  
• Measure Type: Process
  
  
• Steward: National Committee for Quality Assurance
  
  
• Endorsement Status:
  
  
• Meaningful Measure Area: Management of chronic conditions
  
  
• Is the measure specified as an electronic clinical quality measure? No
  
  

Preliminary Analysis of Measure

• Preliminary analysis result: Conditional Support for Rulemaking
  
  
• Preliminary analysis summary
  • Contribution to program measure set:Care coordination is the deliberate organization of patient care activities between two or more participants involved in a patient’s care to facilitate the appropriate delivery of health care services. This is an additional process measure to the Medicare Part C & D Star Ratings, but one that lends itself to better care efficiencies for health plans and their beneficiaries.
  • Impact on quality of care for patients:There is an increase of utilization and costs associated with use of EDs for Medicare beneficiaries, particularly those with dual-eligible status and with behavioral health diagnosis, both of which are much higher cost demographics. Coordinating the care of beneficiaries who utilize emergency services is an important component to ensuring that they also are receiving outpatient care and preventative services with the potential to mitigate disease progression that results in further unnecessary use of EDs.Recommend this measure for inclusion in the measure set pending NQF review and endorsement.
  
  
  
• Does the measure address a critical quality objective not currently adequately addressed by the measures in the program set? Yes. CMS has characterized the healthcare priority Promote Effective Communication and Coordination of Care as a critical quality objective. The Part C & D Star Rating measure set currently has six measures under the Communication/Care Coordination healthcare priority:Care for Older Adults – Medication ReviewMedication Reconciliation Post-DischargePlan All-Cause ReadmissionsCare Coordination (CAHPS)MPF Price AccuracyMTM Program Completion Rate for CMRThis measure most closely relates to Plan All-Cause Readmissions, but readmissions are not the only outcome of interest related to poor care coordination. A follow-up visit after an ED visit helps to ensure coordination with primary care.
  
  
• Is the measure evidence-based and either strongly linked to outcomes or an outcome measure? Yes. None of the articles in the cited evidence tested the hypothesis that follow-up post ED discharge resulted in improved outcomes for patients with multiple chronic conditions. However, a comparable measure was reviewed in Fall 2018 by the NQF Patient Experience and Function Standing Committee: Timely Follow-Up After Acute Exacerbations of Chronic Conditions. The evidence for COPD/asthma, coronary artery disease and heart failure achieved a moderate rating; diabetes and hypertension were ranked insufficient with exception. The rationale for the exception was the direction and strength of the evidence for some conditions, and the expert recommendations from current practice guidelines.Generally speaking, evidence does suggest that follow-up to primary care within 7 days results in better outcomes for patients (Takahashi et al., 2012; Hernandez et al., 2010; Misky et al., 2010; Weinberger et al., 1996).
  
  
• Does the measure address a quality challenge? Yes. Care coordination is a national-level quality challenge and a key consideration for Medicare Part C & D plans:Analysis consisted of over 3M Medicare beneficiaries.An average of 52.7 percent of ED visits for high-risk Medicare members had a follow-up service within 7 days.The measure demonstrates a range of performance of 47% percent of plans with follow-up within 7 days in the lowest performance region to 56% percent of plans with follow-up within 7 days in the highest performance region. Generally, the measure demonstrates a lower than optional performance across all plans. Multiple chronic conditions consistitute 75% of medical spend nationwide.Submission states that “in 2016, NCQA worked with a data analytics firm to test this measure using a large dataset of almost 3 million Medicare Advantage members. Testing revealed that more than half of this population who visited the emergency department (ED) did not receive any follow-up service within seven days. … patients who did not have follow-up after visiting the ED show increased rates of hospital admissions and 30-day unplanned readmissions.” This does not characterize the size of the increase or the strength of the association.
  
  
• Does the measure contribute to efficient use of measurement resources and/or support alignment of measurement across programs? Yes. There is not a similar measure in the program. Resources used by plans to improve on this measure would also presumably be directed toward overall coordination of care, resulting in improved efficiencies in care delivery.
  
  
• Can the measure can be feasibly reported? Yes. Measure was initially introduced into HEDIS in 2018. The measure draws on encounter data—a low burden data source that is routinely generated as part of the normal provision of care. Submission states that “first year results and advisory panel feedback support the findings from field testing that calculating the percent of members with multiple high-risk chronic conditions who had a follow-up service within 7 days of the ED visit is feasible and demonstrates an important performance gap among health plans.”
  
  
• Is the measure applicable to and appropriately specified for the program's intended care setting(s), level(s) of analysis, and population(s)? Yes. Measure is specified as a plan-level measure, which is appropriate for this program.
  
  
• Measure development status: Fully Developed
  
  
• If the measure is in current use, have negative unintended issues to the patient been identified? No. None reported in submission.
  
  
• MAP Rural Workgroup Finding:
  • Relative priority/utility: The chronic conditions included in this measure are prevalent in rural residents.
  • Data collection issues: None.
  • Calculation issues: None.
  • Unintended consequences: TLack of access to care in rural areas may make performance on this measure more difficult for plans that cover rural residents.
  • Votes: Range is 1 – 5, where higher reflects more agreement regarding suitability for the program
  • Average agreement= 3.6
  • Vote Counts: (1 – 0 votes; 2 – 1 vote; 3 – 4 votes; 4 – 8 votes; 5 – 1 vote)
  • Program gap areas: None identified.
  
  
  

Rationale for measure provided by HHS
The Medicare population includes a large number of individuals and older adults with high-risk multiple chronic conditions (MCC) who often receive care from multiple providers and settings and, as a result, are more likely to experience fragmented care and adverse healthcare outcomes, including an increased likelihood of ED visits (1,2). Medicare beneficiaries with MCCs require high levels of care coordination, particularly as the transition from the ED to the community. During these transitions, they often face communication lapses between ED and outpatient providers and inadequate patient, caregiver and provider understanding of diagnoses, medication and follow-up needs (3,4,5,6). This poor care coordination results in an increased risk for medication errors, repeat ED visits, hospitalization, nursing home admission and death (7,8). Medicare beneficiaries with MCCs not only experience poorer health outcomes, but also greater health care utilization (e.g., physician use, hospital and ED use, medication use) and costs (e.g., medication, out-of-pocket, total health care) (9). Medicare beneficiaries with MCCs are some of the heaviest users of high-cost, preventable services such as those offered by the ED (10,11). An estimated 75 percent of health care spending is on people with MCCs (12,13).REFERENCES1. AHRQ. 2010. Multiple Chronic Conditions Chartbook. “2010 Medical Expenditure Panel Survey Data.†https://www.ahrq.gov/sites/default/files/wysiwyg/professionals/prevention-chronic-care/decision/mcc/mccchartbook.pdf (Accessed January 11, 2017)2. Agency for Healthcare Quality and Research (AHRQ). 2012. “Coordinating Care for Adults with Complex Care Needs in the Patient-Centered Medical Home: Challenges and Solutions.†https://pcmh.ahrq.gov/sites/default/files/attachments/coordinating-care-for-adults-with-complex-care-needs-white-paper.pdf3. Altman, R., J.S. Shapiro, T. Moore and G.J. Kuperman. 2012. “Notifications of hospital events to outpatient clinicians using health information exchange: a post-implementation survey.†Journal of Innovation in Health Informatics 20(4).4. Coleman, E.A., R.A. Berenson. 2004. “Lost in transition: challenges and opportunities for improving the quality of transitional care.†Annals of Internal Medicine 141(7).5. Dunnion, M.E., and B. Kelly. 2005. “From the emergency department to home.†Journal of Clinical Nursing 14(6), 776–85.6. Rowland, K., A.K. Maitra, D.A. Richardson, K. Hudson and K.W. Woodhouse. 1990. “The discharge of elderly patients from an accident and emergency department: functional changes and risk of readmission.†Age Ageing 19(6), 415–18.7. Hastings, S.N., E.Z. Oddone, G. Fillenbaum, R.J. Sloane and K.E. Schmader. 2008. “Frequency and predictors of adverse health outcomes in older Medicare beneficiaries discharged from the emergency department.†Medical Care 46(8), 771–7.8. Niedzwiecki, M., K. Baicker, M. Wilson, D.M. Cutler and Z. Obermeyer. 2016. “Short-term outcomes for Medicare beneficiaries after low-acuity visits to emergency departments and clinics.†Medical Care 54(5), 498–503.9. Lehnert, T., D. Heider, H. Leicht, S. Heinrich, S. Corrieri, M. Luppa, S. Riedel-Heller and H.H. Konig. 2011. “Review: health care utilization and costs of elderly persons with multiple chronic conditions.†Medical Care Research & Review 68(4), 387–420.10. CMS. 2012. Chronic Conditions among Medicare Beneficiaries, Chartbook, 2012 Edition. Baltimore, MD. https://www.cms.gov/research-statistics-data-and-systems/statistics-trends-and-reports/chronic-conditions/downloads/2012chartbook.pdf (Accessed July 19, 2016)11. Lochner, K.A., and C.S. Cox. 2013. Prevalence of multiple chronic conditions among Medicare beneficiaries, United States, 2010. https://www.cdc.gov/pcd/issues/2013/12_0137.htm (Accessed January 11, 2017)12. CDC. 2009. The power of prevention: Chronic disease…the public health challenge of the 21st century. http://www.cdc.gov/chronicdisease/pdf/2009-power-of-prevention.pdf (Accessed January 24, 2017)13. Care Innovations. 2013. “Cost Control for Chronic Conditions: An Imperative for MA Plans.†The Business Case for Remote Care Management (RCM). https://www.rmhpcommunity.org/sites/default/files/resource/The%20Business%20Case%20for%20RCM.pdf (Accessed January 24, 2017).

Measure Specifications

• NQF Number (if applicable):
  
  
• Description: The intent of the measure is to improve the coordination of care for Medicare Advantage members as they transition between inpatient and outpatient settings. The measure assesses the percentage of discharges for members 18 years of age and older who had each of the following four indicators: notification of inpatient admission; receipt of discharge information; patient engagement after inpatient discharge; and medication reconciliation post-discharge. Plans report separate rates for individuals 18-64 years of age and those 65 years and older, as well as a total rate for each indicator in the measure.
  
  
• Numerator: 1.Notification of inpatient admission: Documentation of receipt of notification of inpatient admission on the day of admission or the following day and2.Receipt of discharge information: Documentation of receipt of discharge information on the day of discharge or the following day and3.Patient engagement after inpatient discharge: Documentation of patient engagement (e.g., office visits, visits to the home, telehealth) provided within 30 days after discharge and4.Medication reconciliation post-discharge: Documentation of medication reconciliation on the date of discharge through 30 days after discharge (31 total days).
  
  
• Denominator: Acute or non-acute inpatient discharges for Medicare beneficiaries 18 years and older. The denominator is based on discharges, not members. Members may appear more than once.For Administrative Specification, the denominator is the eligible population.For Hybrid Specification, the denominator is a systematic sample drawn from the eligible population.
  
  
• Exclusions: Members in Hospice
  
  
• HHS NQS Priority: Communication and Care Coordination
  
  
• HHS Data Source: Claims; Record Review
  
  
• Measure Type: Composite
  
  
• Steward: National Committee for Quality Assurance
  
  
• Endorsement Status:
  
  
• Meaningful Measure Area: Transfer of health information and interoperability
  
  
• Is the measure specified as an electronic clinical quality measure? No
  
  

Preliminary Analysis of Measure

• Preliminary analysis result: Conditional Support for Rulemaking
  
  
• Preliminary analysis summary
  • Contribution to program measure set:CMS has identified Communication and Care Coordination as a high priority Meaningful Measure Area for the Part C & D Star Ratings. Medication reconciliation post discharge is currently in the set, as mentioned in the submission. The set also has a Plan All Cause Readmission measure, and the Care Coordination measure that are in the same quality domain, but would be complimented by this measure of transitions of care. There is not currently a measure that addresses care transitions in the measure set.
  • Impact on quality of care for patients:Medicare beneficiaries are at particular risk during transitions of care because of higher comorbidities, declining cognitive function and increased medication use. There is observed variance in performance among health plans on all four components of the measure. Further, evidence indicates that good care transitions and care coordination reduce health care costs and improve outcomes.
  
  
  
• Does the measure address a critical quality objective not currently adequately addressed by the measures in the program set? Yes. CMS has identified Communication and Care Coordination as a high priority Meaningful Measure Area for the Part C & D Star Ratings. Medication reconciliation post discharge is currently in the set, as mentioned in the submission. The set also has a Plan All Cause Readmission measure, and the Care Coordination measure that are in the same quality domain but would be complimented by this measure of transitions of care. There is not currently a measure that addresses care transitions in the measure set.
  
  
• Is the measure evidence-based and either strongly linked to outcomes or an outcome measure? Yes . The Medicare population includes older adults and individuals with complex health needs who often receive care from multiple providers and settings. This population is at particular risk during transitions of care because of higher comorbidities, declining cognitive function and increased medication use. Evidence indicates that good care transitions and care coordination reduce health care costs and improve outcomes.
  
  
• Does the measure address a quality challenge? Yes. Two components of the measure notification of inpatient admission and receipt of discharge information have low benchmark performance rates. The developer observed in a small subset of plans a much higher rate which indicates room for improvement. The third component patient engagement after inpatient discharge had an overall high rate of performance but the developer observed a variance of approximately 18% which indicates room for improvement. The fourth component medication reconciliation post-discharge had a large observed variance.
  
  
• Does the measure contribute to efficient use of measurement resources and/or support alignment of measurement across programs? Yes. One of the four parts of the composite measure is a stand-alone measure reported in Part C, Medication Reconciliation Post-Discharge. The proposed measure captures a broader range of transition issues. Further, NCQA stated that it will work with CMS so that plans will not have to report both this proposed measure and the standalone medication reconciliation measure.
  
  
• Can the measure can be feasibly reported? Yes. The measure is claims and medical record review. Medical record review may create an increased reporting burden.
  
  
• Is the measure applicable to and appropriately specified for the program's intended care setting(s), level(s) of analysis, and population(s)? Yes. The measure is tested at the health plan level.
  
  
• Measure development status: Fully Developed
  
  
• If the measure is in current use, have negative unintended issues to the patient been identified? N/A. The measure is not in current use and the developers did not indicate any potential unintended consequences.
  
  
• MAP Rural Workgroup Finding:
  • Relative priority/utility: The Workgroup noted the importance of measures to assess transitions of care for rural residents.
  • Data collection issues: Requires chart abstraction, which can be particularly burdensome for small rural providers. The Workgroup noted that a yes/no checkbox measure of medication reconciliation may not drive improvements in care quality.
  • Calculation issues: The Workgroup believes this measure is appropriate for accountability at the ACO level.
  • Unintended consequences: There was some concern with the medication reconciliation component, particularly given the lack of pharmacist/pharmacy tech providers in rural areas.
  • Votes: Range is 1 – 5, where higher reflects more agreement regarding suitability for the program
  • Average agreement= 3.3
  • Vote Counts: (1 – 0 votes; 2 – 0 votes; 3 – 10 votes; 4 – 2 votes; 5 – 1 vote)
  • Program gap areas: None identified.
  
  
  

Rationale for measure provided by HHS
The Medicare population includes older adults and individuals with complex health needs who often receive care from multiple providers and settings, and thus experience highly fragmented care and adverse health care utilization patterns and outcomes. This population is at particular risk during transitions of care because of higher comorbidities, declining cognitive function and increased medication use (1). Transitions from the inpatient setting to home often results in poor care coordination, including communication lapses between inpatient and outpatient providers, intentional and unintentional medication changes, incomplete diagnostic work-ups and inadequate beneficiary, caregiver and provider understanding of diagnoses, medication and follow-up needs (2).Poor hospital transitions are not only associated with poor health outcomes, but also increased health care utilization and cost, including duplicate medical services, medication errors and increased emergency department visits and readmissions (3). In 2010, Medicare beneficiaries 65 years and older accounted for 11.9 million (approximately 34 percent) of all hospital discharges in the United States (4). One study estimated that inadequate care coordination and poor care transitions resulted in $25 billion-$45 billion in unnecessary spending in 2011 (5). Other studies have found that care coordination programs that do not incorporate timely transitional care elements are unlikely to result in reduced hospitalizations and associated Medicare spending (6), and current payment structures do not provide much incentive for the collaboration necessary to implement effective care coordination post-discharge (7). Hospital transitions require clear communication between inpatient and outpatient providers to ensure optimal health outcomes during patient handoffs (8, 9, 10, 11, 12). Effective care coordination efforts must include notifying patients' primary care practitioners (PCP) of admission, PCP receipt of meaningful and timely discharge information (13), patient engagement through follow-up provided post-discharge and medication reconciliation post-discharge.REFERENCES1. Vognar, L., and N. Mujahid. 2015. “Healthcare transitions of older adults: An overview for the general practitioner.†Rhode Island Medical Journal http://www.rimed.org/rimedicaljournal/2015/04/2015-04-15-ltc-vognar.pdf (Accessed July 12, 2016)2. Rennke, S., O.K. Nguyen, M.H. Shoeb, Y. Magan, R.M. Wachter and S.R. Ranji. 2013. “Hospital-initiated transitional care as a patient safety strategy: A systematic review.†Annals of Internal Medicine 158(5, Pt. 2), 433–40.3. Sato, M., T. Shaffer, A.I. Arbaje and I.H. Zuckerman. 2011. “Residential and health care transition patterns among older Medicare beneficiaries over time.†The Gerontologist 51(2), 170–8.4. Centers for Disease Control and Prevention (CDC). 2010. Number, rate, and average length of stay for discharges from short-stay hospitals, by age, region, and sex: United States, 2010. http://www.cdc.gov/nchs/data/nhds/1general/2010gen1_agesexalos.pdf (Accessed June 22, 2016)5. Health Affairs. 2012. Health Policy Brief: Care Transitions. September 13, 2012. http://healthaffairs.org/healthpolicybriefs/brief_pdfs/healthpolicybrief_76.pdf (Accessed July 12, 2016)6. Peikes, D., A. Chen, J. Schore and R. Brown. 2009. “Effects of care coordination on hospitalization, quality of care, and health care expenditures among Medicare beneficiaries.†Journal of the American Medical Association 301(3).7. Coleman, E.A. and R.A. Berenson. 2004. “Lost in transition: Challenges and opportunities for improving the quality of transitional care.†Annals of Internal Medicine 141(7), 533–6.8. Kripalani, S., A.T. Jackson, J.L. Schnipper and E.A. Coleman. 2007. “Promoting effective transitions of care at hospital discharge: A review of key issues for hospitalists.†Journal of Hospital Medicine 2(5).9. Kripalani, S., F. LeFevre, C.O. Phillips, M.V. Williams, P. Basaviah and D.W. Baker. 2007. “Deficits in communication and information transfer between hospital-based and primary care physicians: Implications for patient safety and continuity of care.†Journal of the American Medical Association 297(8), 831–41.10. Peart, K. N. 2015. When used effectively, discharge summaries reduce hospital readmissions. http://news.yale.edu/2015/01/15/when-used-effectively-discharge-summaries-reduce-hospital-readmissions (Accessed May 4, 2015)11. van Walraven, C., R. Seth and A. Laupacis. 2002. “Dissemination of discharge summaries. Not reaching follow-up physicians.†Canadian Family Physician 48, 737–4212. van Walraven, C., R. Seth, P.C. Austin and A. Laupacis, A. 2002. “Effect of discharge summary availability during post-discharge visits on hospital readmission.†Journal of General Internal Medicine 17(3), 186–92.13. Kind, A.J.H., and M.A. Smith. 2008. “Documentation of Mandated Discharge Summary Components in Transitions from Acute to Subacute Care.†In: Henriksen, K., J.B. Battles, M.A. Keyes, and M.L. Grady, editors. Advances in Patient Safety: New Directions and Alternative Approaches (Vol. 2: Culture and Redesign). Rockville, MD: Agency for Healthcare Research and Quality, August.Notification of inpatient admissions14. Commonwealth Fund. 2015. Reducing Care Fragmentation. http://www.improvingchroniccare.org/downloads/reducing_care_fragmentation.pdf (Accessed May 4, 2015)15. Jones, C.D., M.B. Vu, C.M. O’Donnell, M.E. Anderson, S. Patel, H.L. Wald, … and D.A. DeWalt. 2015. “A failure to communicate: A qualitative exploration of care coordination between hospitalists and primary care providers around patient hospitalizations.†Journal of General Internal Medicine 30(4), 417–24. 16. Moran, W.P., K.S. Davis, T.J. Moran, R. Newman and P.D. Mauldin. 2012. “Where are my patients? It is time to automate notification of hospital use to primary care practices.†Southern Medical Journal 105(1), 18–23.17. Oregon Health Quality Corporation. 2011. Transitions in Care Hospital Survey. http://q-corp.org/sites/qcorp/files/Transitions-in-Care-Hospital-Survey.pdf (Accessed May 4, 2015) 18. Pantilat, S.Z., P.K. Lindenauer, P.P. Katz and R.M. Wachter. 2002. “Primary care physician attitudes regarding communication with hospitalists.†DM 8(4), 218–29.19. UT Health Science Center San Antonio. 2015. Clinical Safety and Effectiveness, Session Five. http://uthscsa.edu/cpshp/CSEProject/To%20increase%20the%20notification%20of%20primary%20care%20physicians%20(PCP)%20when%20their%20patients%20are%20admitted%20or%20discharged.pdf (Accessed May 4, 2015) 20. Ventura, T., D. Brown, T. Archibald, A. Goroski and J. Brock. 2010. “Improving care transitions and reducing hospital readmissions: Establishing the evidence for community-based implementation strategies through the care transitions theme.†The Remington Report. http://www.communitysolutions.com/assets/2012_Institute_Presentations/caretransitioninterventions051812.pdf (Accessed July 26, 2016) 21. Bell, C.M., J.L. Schnipper, A.D. Auerbach, P.J. Kaboli, T.B. Wetterneck, D.V. Gonzales, V.M. Arora, J.X. Zhang and D.O. Meltzer. 2009. “Association of communication between hospital-based physicians and primary care providers with patient outcomes.†Journal of General Internal Medicine 24(3).Receipt of discharge information22. Alpers, A. 2001. “Key legal principles for hospitalists.†American Journal of Medicine 111(9B), 5S–9S.23. Goldman, L., S.Z. Pantilat and W.F. Whitcomb. 2001. “Passing the clinical baton: 6 principles to guide the hospitalist.†American Journal of Medicine 111(9B), 36S–39S.24. Jack, B.W., K.C. Veerappa, D. Anthony, J.L. Greenwald, G.M. Sanchez, A.E. Johnson, S.R. Forsythe, J.K., O’Donnell, M.K. Paasche-Orlow, C. Manasseh, S. Martin and L.A. Culpepper. 2009. “Reengineered hospital discharge program to decrease rehospitalization: A randomized trial.†Annals of Internal Medicine 150(3).25. RAND. 2014. “Evaluation and Development of Outcome Measures for Quality Assessment in Medicare Advantage and Special Needs Plans.†Validation Study Final Report. Santa Monica, CA: RAND.Patient engagement after inpatient discharge26. Arbaje, A.I., D.L. Kansagara, A.H. Salanitro, H.L. Englander, S. Kripalani, S.F. Jencks and L.A. Lindquist. 2014. “Regardless of age: Incorporating principles from geriatric medicine to improve care transitions for patients with complex needs.†Journal of General Internal Medicine 29(6), 932–9.27. Berkowitz, R.E., Z. Fang, B.K. Helfand, R.N. Jones, R. Schreiber and M.K. Paasche-Orlow. 2013. “Project ReEngineered Discharge (RED) lowers hospital readmissions of patients discharged from a skilled nursing facility.†Journal of the American Medical Directors Association 14(10) 736–40.28. Bisognano, M., and A. Boutwell. 2009. “Improving transitions to reduce readmissions.†Frontiers of Healthcare Services Management. https://www.ache.org/pdf/secure/gifts/July10-frontiers.pdf (Accessed July 27, 2016)29. Braun, E., A. Baidusi, G. Alroy and Z.S. Azzam. 2009. “Telephone follow-up improves patients satisfaction following hospital discharge.†European Journal of Internal Medicine 20(2), 221–5.30. Coleman, E.A., C. Parry, S. Chalmers, et al. 2006. “The Care Transitions Intervention: Results of a randomized controlled trial.†Archives of Internal Medicine 166(17), 1822–8.31. Forster, A.J., H.J. Murff, J.F. Peterson, T.K. Gandhi and D.W. Bates. 2003. “The incidence and severity of adverse events affecting patients after discharge from the hospital.†Annals of Internal Medicine 138(3).32. Hansen, L.O., J.L. Greenwald, T. Budnitz, E. Howell, L. Halasyamani, G. Maynard, ... and M.V. Williams. 2013. “Project BOOST: Effectiveness of a multihospital effort to reduce rehospitalization.†Journal of Hospital Medicine 8(8), 421–7.33. Harrison, P.L., P.A. Hara, J.E. Pope, M.C. Young and E.Y. Rula. 2011. “The impact of postdischarge telephonic follow-up on hospital readmissions.†Population Health Management 14(1), 27–32.34. Hernandez, A.F., M.A. Greiner, G.C. Fonarow, B.G. Hammill, P.A. Heidenreich, C.W. Yancy, E.D. Peterson and L.H. Curtis. 2010. “Relationship between early physician follow-up and 30-day readmission among Medicare beneficiaries hospitalized for heart failure.†Journal of the American Medical Association 303(17), 1716–22.35. Lin, C.Y., A.E. Barnato and H.B. Degenholtz. 2011. “Physician follow-up visits after acute care hospitalization for elderly Medicare beneficiaries discharged to noninstitutional settings.†Journal of The American Geriatrics Society 59(10), 1947–54.36. Misky, G.J., H.L. Wald and E.A. Coleman. 2011. “Post-hospitalization transitions: Examining the effects of timing of primary care provider follow-up.†Journal of Hospital Medicine 5(7), 392–7.37. Muus, K.J., A. Knudson, M.G. Klug, J. Gokun, M. Sarrazin and P. Kaboli. 2010. “Effect of post-discharge follow-up care on re-admissions among US veterans with congestive heart failure: A rural-urban comparison.†Rural Remote Health 10(2), 1447.38. Naylor, M. D., Brooten, D. A., Campbell, R., et al. 2003. “Comprehensive discharge planning and home follow-up of hospitalized elders.†Journal of the American Medical Association 281, 613–20. 39. Naylor, M.D. 2003. Transitional care of older adults. Annual Review of Nursing Research 20, 127–47.40. The Bridge Model. 2016. The Bridge Model. http://www.transitionalcare.org/the-bridge-model/ (Accessed August 22, 2016)41. Balaban, R.B., J.S. Weissman, P.A. Samuel and S. Woolhandler. 2008. “Redefining and redesigning hospital discharge to enhance patient care: A randomized controlled study.†Journal of General Internal Medicine 23(8), 1228–33.Medication reconciliation post-discharge42. Patterns of medications use in the United States 2006: a report from the Slone Survey. http://www.bu.edu/slone/files/2012/11/SloneSurveyReport2006.pdf (Accessed July 17, 2014)43. Vogeli, C., A.E. Shields, T.A. Lee, et al. 2007. “Multiple Chronic Conditions: Prevalence, Health Consequences, and Implications for Quality, Care Management, and Costs.†J Gen Intern Med 22(suppl 3): 391–5.

Measure Specifications

• NQF Number (if applicable): 2940
  
  
• Description: Percent of beneficiaries receiving opioid prescriptions with an average daily morphine milligram equivalent (MME) greater than or equal to 90 mg over a period of 90 days or longer.
  
  
• Numerator: Number of member-years of Medicare Part D beneficiaries 18 years and older in the denominator with an average MME greater than or equal to 90 mg.
  
  
• Denominator: Number of member-years of enrolled Medicare Part D beneficiaries 18 years and older with at least 2 fills of a prescription opioid on unique dates of service (DOS) and at least 15 total opioid days supply over a period of 90 days or longer during the measurement period.
  
  
• Exclusions: Medicare beneficiaries with a cancer diagnosis or that are enrolled in hospice at any time during the measurement period are excluded from the denominator.
  
  
• HHS NQS Priority: Communication and Care Coordination
  
  
• HHS Data Source: Prescription Drug Event Data Elements
  
  
• Measure Type: Process
  
  
• Steward: Pharmacy Quality Alliance
  
  
• Endorsement Status:
  
  
• Meaningful Measure Area: Prevention and treatment of opioid and substance use disorders
  
  
• Changes to Endorsed Measure Specifications?: The MUC list indicates the measure has been modified from its endorsed version.
  
  
• Is the measure specified as an electronic clinical quality measure? No
  
  

Preliminary Analysis of Measure

• Preliminary analysis result: Conditional Support for Rulemaking
  
  
• Preliminary analysis summary
  • Contribution to program measure set:This measure is used in the SSP’s Opioid Utilization Reports while also being NQF endorsed. This measure would benefit Part C & D beneficiaries receiving opioid prescriptions with an average daily morphine milligram equivalent (MME) greater than or equal to 90 mg over a period of 90 days or longer by providing them with information about plan quality and performance indicators and addressing quality objective gaps currently evident by a lack of opioid measures in the Parts C & D measure set.The measure is related to MUC2019-61 and may be somewhat redundant were both to be added.
  • Impact on quality of care for patients:This measure has the potential to impact approximately 13 million individuals, who are prescribed opioid treatment through Medicare Part D, by reducing the risk of opioid use disorder, overdose, and death.
  
  
  
• Does the measure address a critical quality objective not currently adequately addressed by the measures in the program set? Yes. The measure addresses the Effective Prevention and Treatment of Chronic Disease quality objective in the Medicare Part D program. Lowering the number of Part D beneficiaries receiving opioid prescriptions with an average daily morphine milligram equivalent (MME) greater than or equal to 90 mg over a period of 90 days or longer can ensure patients have access to safer, more effective chronic pain treatment while reducing their risk of opioid use disorder, overdose, and death. There are currently no opioid measures in the Part C and D measure sets.
  
  
• Is the measure evidence-based and either strongly linked to outcomes or an outcome measure? Yes. This process measure meets clinical guidelines by tracking Part C and D beneficiaries receiving opioid prescriptions with an average daily intake greater than or equal to 90 mg/day. There is little evidence that dosing >90 mg/day improves function at these higher doses (CDC, 2016) Among patients on chronic opioid therapy, the risk of an overdose increases with the average daily dose. For example, one study found that compared with doses <20 MME/day, those taking higher doses were at increased risk for overdose (adjusted hazard ratio [HR] 1.44 for 20 to 49 MME/day, 3.73 for 50 to 99 MME/day, and 8.87 for =100 MME/day). However, it is not clear if the correlation between overdose and prescription dosing reflects patient differences or the impact of higher opioid doses (Dowell et al., 2016).Guidelines also focus on the use of opioids for treating chronic pain in non-cancerous, palliative care, and end-of-life care siutations, to improve opioid prescription to provide safer access to chronic pain treatment, and reduce risk of opioid use disorder, for patients. This measure meets clinical guidelines by tracking Part D beneficiaries receiving opioid prescriptions with an average daily intake greater than or equal to 90 mg/day.
  
  
• Does the measure address a quality challenge? Yes. This measure addresses a critical population at increased risk of opioid overdose and death. Risks for serious harm related to opioid therapy increases with higher opioid dosages. Higher opioid dosages are associated with increased risk for opioid use disorder and overdose. According to the Office of Inspector General, nearly 30% (13.4 million) of Medicare Part D beneficiares received opioids in 2018 (OIG, 2019).
  
  
• Does the measure contribute to efficient use of measurement resources and/or support alignment of measurement across programs? Yes. This measure supports efficient use of resources in that it is used in SSP, specifically in SSP’s Opioid Utilization Reports; not in the SSP measure set. The measure is related to MUC2019-61 and may be somewhat duplicative were both to be added.
  
  
• Can the measure can be feasibly reported? Yes. All data elements are in defined fields in electronic claims. Additionally, CMS provides Medicare Part D sponsors monthly reports for monitoring and case management purposes by calculating and reporting the measure using prescription drug event data from electronic prescription claims submitted by pharmacies to health plans.
  
  
• Is the measure applicable to and appropriately specified for the program's intended care setting(s), level(s) of analysis, and population(s)? Yes. This measure was tested at the Medicare Drug Plan level and intended for use for Medicare Part D Beneficiaries through Part D plan sponsors.
  
  
• Measure development status: Fully Developed
  
  
• If the measure is in current use, have negative unintended issues to the patient been identified? No. This measure has been used in SSP Opioid Utilization Reports since 2018, with no unintended consequences reported. However, concerns have been raised that prescribing changes could be associated with the unintended consequence of patients seeking illicitly obtained opioids or heroin.
  
  
• MAP Rural Workgroup Finding:
  • Relative priority/utility: The Workgroup agreed that opioid use is a relevant issue for rural residents. There was some feeling that excluding cancer patients is a “blunt instrument” but that the measure itself reflects a reasonable approach for identifying patients who could benefit from counseling or other interventions.
  • Data collection issues: None
  • Calculation issues: None.
  • Unintended consequences: Without a balancing measure, there is a potential for patient harm due to forced tapering and potential for seeking illicit drugs to treat pain. Rural residents have relatively less access to alternative pain treatment, fewer counselors, etc. Due to types of occupations in rural areas, rural residents have higher rates of physical label and injury, making lack of access to pain management more problematic.
  • Votes: Range is 1 – 5, where higher reflects more agreement regarding suitability for the program
  • Average agreement= 4.2
  • Vote Counts: (1 – 0 votes; 2 – 1 vote; 3 – 2 votes; 4 – 4 votes; 5 – 7 votes)
  • Program gap areas: The Workgroup suggested that a different measure developed by PQA (which considers opioids and benzodiazepines) may be a better alternative than the three proposed measures.
  
  
  

Rationale for measure provided by HHS
CMS adapted three PQA opioid overuse measures related to opioid use, including this OHD measure, to examine the quality of use related to the dose of the medications over time, access to the medications and the combination of both of these criteria. CMS has provided each Part D sponsor monthly reports using these metrics, and will publish these as CMS display measures beginning for 2020. Pending rule-making, we will consider adding one of these into the CMS Part C & D Star Ratings.Claims data from commercially insured patients indicate that approximately 8% of opioidprescriptions for acute pain and 12% for chronic pain specify a daily dosage of 120 MED ormore (1). The proportion of patients being treated at this dosage for more than 90 days has not been described. However, one study of veterans treated with 180 MED/day or more for 90+ days (2) found that this group was characterized by high rates of psychiatric and substance abuse disorders and frequently did not receive care consistent with clinical guidelines. Other studies have suggested the people at high opioid dosage are at greater risk of overdoses and fractures (3, 4, 5).The Washington State Agency Medical Directors Group has suggested 120 MED as a dosagelevel that should not be exceeded without special consideration (6). Prescription drug monitoring programs, which track the use of multiple providers by patients, indicate that such use is typically found among a small proportion of patients, with the proportion declining as the number of providers increases. In Massachusetts in 2006, considering only Schedule II opioids, 0.5% of patients saw 4+ prescribers and 4+ pharmacies (7). A national study found that 13% of patients had overlapping prescriptions from two or more different prescribers during an 18-month period. Of these, 0.5% used 4+ prescribers and 4+ pharmacies (8). People who see multiple prescribers or use multiple pharmacies are more likely to die of drug overdoses (4). Data from the California PDMP indicates that people with higher daily dosages are more likely to see multiple prescribers or go to multiple pharmacies (9). The data above suggest that prevention of opioid overdose deaths should focus on strategies that target (1) high-dose opioid users as well as (2) persons who seek care from multiple doctors and pharmacies. The data suggest that these criteria can be considered separately, as measures related to prescribed opioids for legitimate uses versus diverted uses. Thus, we will consider use of 3 measures, one for each criteria and one that is the intersection of both criteria. For the Part C and D Star Ratings, we would add only one of these measures. REFERENCES1. Ying Liu, PhD; Joseph E. Logan, PhD; Leonard J. Paulozzi, MD, MPH; Kun Zhang, MS; and Christopher M. Jones, PharmD. Potential Misuse and Inappropriate Prescription Practices Involving Opioid Analgesics. Am J Manag Care. 2013;19(8):648-658.2. Clinical characteristics of veterans prescribed high doses of opioid medications for chronic noncancer pain. Benjamin J. Morasco, Jonathan P. Duckart, Thomas P. Carr, Richard A. Deyo, Steven K. Dobscha. PAIN. 151 (2010) 625–632.3. Kate M. Dunn, PhD; Kathleen W. Saunders, JD; Carolyn M. Rutter, PhD; Caleb J. Banta-Green, MSW, MPH, PhD; Joseph O. Merrill, MD, MPH; Mark D. Sullivan, MD, PhD; Constance M. Weisner, DrPH, MSW; Michael J. Silverberg, PhD, MPH; Cynthia I. Campbell, PhD; Bruce M. Psaty, MD, PhD; and Michael Von Korff, ScD. Opioid Prescriptions for Chronic Pain and Overdose - A Cohort Study. Ann Intern Med. 2010;152:85-92.4. Paulozzi, et al. A History of Being Prescribed Controlled Substances and Risk of Drug Overdose Death. Pain Medicine 2011.5. Kathleen W. Saunders, JD, Kate M. Dunn, Ph. D, Joseph O. Merrill, M.D, M.P.H., Mark Sullivan, M.D., Ph.D., Constance Weisner, DrPH, M.S.W, Jennifer Brennan Braden, M.D., M.P.H., Bruce M. Psaty, M.D., Ph.D., and Michael Von Korff, Sc.D. Relationship of Opioid Use and Dosage Levels to Fractures in Older Chronic Pain Patients. Society of General Internal Medicine. 2009. DOI: 10.1007/s11606-009-1218-z.6. Agency Medical Directors Group (AMDG). Interagency Guideline on Opioid Dosing for Chronic Non-cancer Pain: An educational aid to improve care and safety with opioid therapy. 2010 Update. www.cdc.gov/HomeandRecreationalSafety/ Poisoning/brief.htm7. Nathaniel Katz, Lee Panas, MeeLee Kim, Adele D. Audet, Arnold Bilansky, John Eadie, Peter Kreiner, Florence C Paillard, Cindy Thomas and Grant Carrow. Usefulness of prescription monitoring programs for surveillance – analysis of Schedule II opioid prescription data in Massachusetts, 1996–2006y. Pharmacoepidemiology and drug safety 2010; 19: 115–123.8. M. Soledad Cepeda, Daniel Fife, Wing Chow, Gregory Mastrogiovanni and Scott C. Henderson. Assessing Opioid Shopping Behaviour - A Large Cohort Study from a Medication Dispensing Database in the US. Drug Saf 2012.9. Han H, Kass PH, Wilsey BL, Li C-S (2012) Individual and County-Level Factors Associated with Use of Multiple Prescribers and Multiple Pharmacies to Obtain Opioid Prescriptions in California. PLoS ONE 7(9): e46246. doi:10.1371/journal.pone.0046246.

Summary of NQF Endorsement Review

• Year of Most Recent Endorsement Review: 2017
  
  
• Project for Most Recent Endorsement Review: Patient Safety Project 2015-2017
  
  
• Review for Importance: 1a. Evidence: 16-H; 3-M; 0-L; 0-I; 1b. Performance Gap: 13-H; 7-M; 0-L; 0-IRationale:• The developer provided a systematic review of the evidence demonstrating the benefits of highdose opioids for chronic pain are not established and the risks for serious harm related to opioidtherapy increases at higher doses.• Lower dosages of opioids reduce the risk for overdose, but a single dosage threshold for safeopioid use has not been identified.• The measure was tested in three different health plan data sources – the Medicare population(mean rate=39.27 per 1,000), one commercial heath plan (mean rate= 32.003 per 1,000), andthe Medicaid population (mean rate =34.04 per 1,000). The Committee noted that these ratesdemonstrate a significant performance gap.• The Committee noted this is highly important to measure given the current national opioidoveruse problem.
  
  
• Review for Scientific Acceptability: 2a. Reliability: 13-H; 7-M; 0-L; 0-I 2b. Validity: 14-M; 7-L; 0-IRationale:• The developer used several data sets for reliability testing:o For Medicare testing, the analysis included a convenience sample of over 700 MedicarePart D prescription drug plans (comprising a total of 7,067,445 individuals aged 18 andolder)o Testing was also conducted in one Commercial health plan (comprising a total of209,191 individuals age 18 and older)o For Medicaid testing, the analysis included 8 state-based prescription drug planscovering 6 states (comprising a total of 1,437,410 individuals age 18 and older)• The mean reliability score across all plans is 0.9938.• The developer assessed the face validity (only) of the measure using a technical expert panelfrom the Pharmacy Quality Alliance (PQA). 67 percent strongly agreed that the measure resultsreflected quality of care. Five PQA member organizations also tested the measure using theirown data, and all strongly agreed that the measure reflected the quality of care provided fortheir populations.
  
  
• Review for Feasibility: 3. Feasibility: 13-H; 8-M; 0-L; 0-I(3a. Clinical data generated during care delivery; 3b. Electronic sources; 3c.Susceptibility to inaccuracies/unintended consequences identified 3d. Data collection strategy can be implemented)Rationale:• Pilot test sites indicated the measure was feasible and results were able to be reportedefficiently and accurately.• All the data elements are in defined fields in electronic claims
  
  
• Review for Usability: 4. Usability and Use: 11-H; 9-M; 1-L; 0-I(Used and useful to the intended audiences for 4a. Accountability and Transparency; 4b. Improvement;and 4c. Benefits outweigh evidence of unintended consequences)Rationale:• The measure is currently being used in the Medicare Part D Overutilization Monitoring Systemto monitor the utilization of opioids for members with the Medicare drug benefit.• Although no unintended negative consequences to individuals or populations were identifiedduring testing, concerns have been raised that prescribing changes such as dose reduction(without offering or arranging evidence-based treatment for patients with opioid use disorder)might be associated with unintended negative consequences, such as patients seeking heroin orother illicitly obtained opioids (1,2) or interference with appropriate pain treatment.
  
  
• Review for Related and Competing Measures: 5. Related and Competing MeasuresRelated measures:• Measure 2950: Use of Opioids from Multiple Providers in Persons without Cancer- Theproportion (XX out of 1,000) of individuals without cancer receiving prescriptions for opioidsfrom four (4) or more prescribers AND four (4) or more pharmacies.• Measure 2951: Use of Opioids from Multiple Providers and at High Dosage in Persons withoutCancer- The proportion (XX out of 1,000) of individuals without cancer receiving prescriptionsfor opioids with a daily dosage greater than 120mg morphine equivalent dose (MED) for 90consecutive days or longer, AND who received opioid prescriptions from four (4) or moreprescribers AND four (4) or more pharmacies.• These measures are also being considered for endorsement. The Committee determined thatthey are related but not competing.
  
  
• Endorsement Public Comments: 6. Public and Member CommentComments:This measure received 3 comments. The commenters noted that the measure may be too inclusive andthe developer should consider narrowing the measure to specific chronic conditions or diagnoses to bemore meaningful.Developers Response:The recommendations in the 2015 American Geriatrics Society Beers Criteria are based on a systematicevidence review conducted by American Geriatrics Society Beers Criteria Expert Panel. The review isfocused on the evidence for potential harms of medications in older adults. Medications then includedin the Beers Criteria recommendations are those that the panel found evidence indicating that themedications should in general be avoided in all older adults or avoided in older adults with certainconditions or diseases, due to their associated risks for these populations. The Beers Criteria is updatedregularly based on currently available literature. We believe it's important for this quality measure to bebased on the systematic evidence review that is conducted by the Beers Criteria Expert Panel. Thecomplete evidence tables for the systematic review can be accessed on the American Geriatrics Society'swebsite here: http://geriatricscareonline.org/toc/american-geriatrics-society-updated-beers-criteria-forpotentially-inappropriate-medication-use-in-older-adults/CL001NCQA recognizes that some of the medications that are most attributable to adverse drug events inolder adults that result in ED visits and hospitalizations are not included in the Beers Criteria asmedications to be generally avoided (e.g., warfarin, antidiabetics and oral antiplatelets - although someoral antiplatelets are in fact included in the Beers Criteria and this measure: Dipyridamole, Ticlopidine).These other high-risk medications should be addressed in separate quality measures that focus on safeprescribing and appropriate monitoring, rather than this measure which focuses on medications thatshould be generally avoided. We agree with the need for such quality measures to improve safeprescribing of anticoagulants, antidiabetics, and opioids and have current work underway at NCQA toexplore development of measures in these areas. Of note, the Pharmacy Quality Alliance has severalmeasures addressing opioid prescribing that are currently being considered for NQF endorsement aspart of this Patient Safety project. NCQA supports the endorsement of these measures and has plans toadapt them for health plan reporting in the near future.In terms of the way this measure is currently specified to include a number of different medications, webelieve that creating separate quality measures or indicators for all the specific medications in the BeersCriteria, or for each drug-disease interaction, would be burdensome for measurement and reporting byhealth plans. Plans can look at medications on an individual basis to see where improvements andinterventions are needed, however we do not think this level of detail would be desirable for nationalreporting by health plans.As a measure of potentially inappropriate medication use, NCQA does not expect this measure'sperformance to ever reach 0% (i.e., no prescribing of high-risk medications). There will always be caseswhere the benefits of prescribing a high-risk medication may outweigh the risks for certain patients.Clinicians should take into account various factors when considering the risk-benefit ratio of prescribinga high-risk medication to an individual. A companion paper to the Beers Criteria was published by theAmerican Geriatrics Society Workgroup on Improving Use of the Beers Criteria in 2015. The paperspecifically states "the AGS 2015 Beers Criteria are reasonable to use for performance measurementacross large groups of patients and providers but should not be used to judge care for any individual"(Steinman et al., 2015, JAGS). We believe measuring this concept of potentially inappropriatemedication use among elderly at the health plan (i.e., population) level is an important and usefulmedication safety measure that health plans can use to identify high-risk medication prescribing.34Committee Response:The Committee agrees with the developer response and maintains their decision to recommend thismeasure for continued endorsement.
  
  
• Endorsement Committee Recommendation: Steering Committee Recommendation for Endorsement: 21-Y; 0-N
  
  

Measure Specifications

• NQF Number (if applicable): 2950
  
  
• Description: Percent of beneficiaries receiving opioid prescriptions from 4 or more prescribers and 4 or more pharmacies within 180 days or less.
  
  
• Numerator: Number of member-years of Medicare Part D beneficiaries 18 years and older in the denominator who received opioids from 4 or more prescribers and 4 or more pharmacies within 180 days or less during the measurement period.
  
  
• Denominator: Number of member-years of enrolled Medicare Part D beneficiaries 18 years and older with at least 2 fills of a prescription opioid on unique dates of service (DOS) and at least 15 total opioid days supply over a period of 90 days or longer during the measurement period.
  
  
• Exclusions: Medicare beneficiaries with a cancer diagnosis or that are enrolled in hospice at any time during the measurement period are excluded from the denominator.
  
  
• HHS NQS Priority: Communication and Care Coordination
  
  
• HHS Data Source: Prescription Drug Event Data Elements
  
  
• Measure Type: Process
  
  
• Steward: Pharmacy Quality Alliance
  
  
• Endorsement Status:
  
  
• Meaningful Measure Area: Prevention and treatment of opioid and substance use disorders
  
  
• Changes to Endorsed Measure Specifications?: The MUC list indicates the measure has been modified from its endorsed version.
  
  
• Is the measure specified as an electronic clinical quality measure? No
  
  

Preliminary Analysis of Measure

• Preliminary analysis result: Support for Rulemaking
  
  
• Preliminary analysis summary
  • Contribution to program measure set:This measure is used in the SSP’s Opioid Utilization Reports while also being NQF endorsed. This measure would benefit Part C & D beneficiaries receiving opioid prescriptions from 4 or more prescribers and 4 or more pharmacies by providing them with information about plan quality and performance indicators and addressing quality objective gaps currently evident by a lack of opioid measures in the Parts C & D measure set.The measure is related to MUC2019-61, and may be somewhat redundant were both to be added.
  • Impact on quality of care for patients:This measure has the potential to impact the 13% of patients who have concurring prescriptions from two or more different providers, and the 0.5% of those patients who use 4 or more pharmacies (Soledad et al., 2012).
  
  
  
• Does the measure address a critical quality objective not currently adequately addressed by the measures in the program set? Yes. The measure addresses the Effective Prevention and Treatment of Chronic Disease quality objective in the Medicare Part D program. Monitoring the number of prescribers and pharmacies can ensure patients have access to safer, more effective chronic pain treatment while reducing their risk of opioid use disorder, overdose, and death. There are currently no opioid measures in the Part C and D measure sets.
  
  
• Is the measure evidence-based and either strongly linked to outcomes or an outcome measure? Yes. This process measure tracks Part C and D beneficiares receiving receiving opioid prescriptions from 4 or more prescribers and 4 or more pharmacies within 180 days or less. Studies have shown that such beneficiaries are at increased risk for overdose and death, with the risk increasing with the number of pharmacies and prescribers (Yang et al., 2015; Gwira Baumblatt et al., 2014)
  
  
• Does the measure address a quality challenge? Yes. This measure addresses a critical population at increased risk of opioid overdose and death. Risks for serious harm related to opioid therapy increases with prescriptions for opioids from multiple prescribers and pharmacies. According to a national study, 13% of patients patients had concurrent prescriptions from two or more different providers during an 18-month period (Soledad Cepada et al., 2012).
  
  
• Does the measure contribute to efficient use of measurement resources and/or support alignment of measurement across programs? Yes. This measure supports efficient use of resources in that it is used in SSP, specifically in SSP’s Opioid Utilization Reports. The measure is related to MUC2019-61 and may be somewhat duplicative were both to be added.
  
  
• Can the measure can be feasibly reported? Yes. All data elements are in defined fields in electronic claims. Additionally, CMS provides Medicare Part D sponsors monthly reports for monitoring and case management purposes by calculating and reporting the measure using prescription drug event data from electronic prescription claims submitted by pharmacies to health plans.
  
  
• Is the measure applicable to and appropriately specified for the program's intended care setting(s), level(s) of analysis, and population(s)? Yes. This measure was tested at the Medicare Drug Plan level and intended for use for Medicare Part D Beneficiaries through Part D plan sponsors.
  
  
• Measure development status: Fully Developed
  
  
• If the measure is in current use, have negative unintended issues to the patient been identified? No. This measure has been used in SSP Opioid Utilization Reports since 2018, with no unintended consequences reported.
  
  
• MAP Rural Workgroup Finding:
  • Relative priority/utility: The Workgroup agreed that opioid use is a relevant issue for rural residents.
  • Data collection issues: None
  • Calculation issues: None.
  • Unintended consequences: Although this measure could promote use of drug monitoring programs in rural areas, on the whole, it may not be particularly applicable, due to the relatively few pharmacies in rural areas.
  • Votes: Range is 1 – 5, where higher reflects more agreement regarding suitability for the program
  • Average agreement= 2.5
  • Vote Counts: (1 – 1 vote; 2 – 6 votes; 3 – 4 votes; 4 – 2 votes; 5 – 0 votes)
  • Program gap areas: The Workgroup suggested that a different measure developed by PQA (which considers opioids and benzodiazepines) may be a better alternative than the three proposed measures.
  
  
  

Rationale for measure provided by HHS
CMS adapted three PQA opioid overuse measures related to opioid use, including this OMP measure, to examine the quality of use related to the dose of the medications over time, access to the medications and the combination of both of these criteria. CMS has provided each Part D sponsor monthly reports using these metrics, and will publish these as CMS display measures beginning for 2020. Pending rule-making, we will consider adding one of these into the CMS Part C & D Star Ratings.Claims data from commercially insured patients indicate that approximately 8% of opioidprescriptions for acute pain and 12% for chronic pain specify a daily dosage of 120 MED ormore (1). The proportion of patients being treated at this dosage for more than 90 days has not been described. However, one study of veterans treated with 180 MED/day or more for 90+ days (2) found that this group was characterized by high rates of psychiatric and substance abuse disorders and frequently did not receive care consistent with clinical guidelines. Other studies have suggested the people at high opioid dosage are at greater risk of overdoses and fractures (3, 4, 5).The Washington State Agency Medical Directors Group has suggested 120 MED as a dosagelevel that should not be exceeded without special consideration (6). Prescription drug monitoring programs, which track the use of multiple providers by patients, indicate that such use is typically found among a small proportion of patients, with the proportion declining as the number of providers increases. In Massachusetts in 2006, considering only Schedule II opioids, 0.5% of patients saw 4+ prescribers and 4+ pharmacies (7). A national study found that 13% of patients had overlapping prescriptions from two or more different prescribers during an 18-month period. Of these, 0.5% used 4+ prescribers and 4+ pharmacies (8). People who see multiple prescribers or use multiple pharmacies are more likely to die of drug overdoses (4). Data from the California PDMP indicates that people with higher daily dosages are more likely to see multiple prescribers or go to multiple pharmacies (9). The data above suggest that prevention of opioid overdose deaths should focus on strategies that target (1) high-dose opioid users as well as (2) persons who seek care from multiple doctors and pharmacies. The data suggest that these criteria can be considered separately, as measures related to prescribed opioids for legitimate uses versus diverted uses. Thus, we will consider use of 3 measures, one for each criteria and one that is the intersection of both criteria. For the Part C and D Star Ratings, we would add only one of these measures.REFERENCES1. Ying Liu, PhD; Joseph E. Logan, PhD; Leonard J. Paulozzi, MD, MPH; Kun Zhang, MS; and Christopher M. Jones, PharmD. Potential Misuse and Inappropriate Prescription Practices Involving Opioid Analgesics. Am J Manag Care. 2013;19(8):648-658.2. Clinical characteristics of veterans prescribed high doses of opioid medications for chronic noncancer pain. Benjamin J. Morasco, Jonathan P. Duckart, Thomas P. Carr, Richard A. Deyo, Steven K. Dobscha. PAIN. 151 (2010) 625–632.3. Kate M. Dunn, PhD; Kathleen W. Saunders, JD; Carolyn M. Rutter, PhD; Caleb J. Banta-Green, MSW, MPH, PhD; Joseph O. Merrill, MD, MPH; Mark D. Sullivan, MD, PhD; Constance M. Weisner, DrPH, MSW; Michael J. Silverberg, PhD, MPH; Cynthia I. Campbell, PhD; Bruce M. Psaty, MD, PhD; and Michael Von Korff, ScD. Opioid Prescriptions for Chronic Pain and Overdose - A Cohort Study. Ann Intern Med. 2010;152:85-92.4. Paulozzi, et al. A History of Being Prescribed Controlled Substances and Risk of Drug Overdose Death. Pain Medicine 2011.5. Kathleen W. Saunders, JD, Kate M. Dunn, Ph. D, Joseph O. Merrill, M.D, M.P.H., Mark Sullivan, M.D., Ph.D., Constance Weisner, DrPH, M.S.W, Jennifer Brennan Braden, M.D., M.P.H., Bruce M. Psaty, M.D., Ph.D., and Michael Von Korff, Sc.D. Relationship of Opioid Use and Dosage Levels to Fractures in Older Chronic Pain Patients. Society of General Internal Medicine. 2009. DOI: 10.1007/s11606-009-1218-z.6. Agency Medical Directors Group (AMDG). Interagency Guideline on Opioid Dosing for Chronic Non-cancer Pain: An educational aid to improve care and safety with opioid therapy. 2010 Update. www.cdc.gov/HomeandRecreationalSafety/ Poisoning/brief.htm 7. Nathaniel Katz, Lee Panas, MeeLee Kim, Adele D. Audet, Arnold Bilansky, John Eadie, Peter Kreiner, Florence C Paillard, Cindy Thomas and Grant Carrow. Usefulness of prescription monitoring programs for surveillance – analysis of Schedule II opioid prescription data in Massachusetts, 1996–2006y. Pharmacoepidemiology and drug safety 2010; 19: 115–123.8. M. Soledad Cepeda, Daniel Fife, Wing Chow, Gregory Mastrogiovanni and Scott C. Henderson. Assessing Opioid Shopping Behaviour - A Large Cohort Study from a Medication Dispensing Database in the US. Drug Saf 2012.9. Han H, Kass PH, Wilsey BL, Li C-S (2012) Individual and County-Level Factors Associated with Use of Multiple Prescribers and Multiple Pharmacies to Obtain Opioid Prescriptions in California. PLoS ONE 7(9): e46246. doi:10.1371/journal.pone.0046246.

Summary of NQF Endorsement Review

• Year of Most Recent Endorsement Review: 2017
  
  
• Project for Most Recent Endorsement Review: Patient Safety Project 2015-2017
  
  
• Review for Importance: 1a. Evidence: 0-H; 20-M; 0-L; 0-I 1b. Performance Gap: 13-H; 7-M; 0-L; 0-IRationale:• The evidence suggests that prescriptions for opioids from multiple prescribers and pharmaciescorrelate with undesired health outcomes. The use of multiple prescribers and pharmacies are 35associated with increased risks for opioid overdose. The Committee noted this is highlyimportant to measure given the current national opioid overuse problem.• The measure was tested in three different health plan data sources – the Medicare population(mean was 23.31 per 1,000 and the median was 26.12 per 1,000), one commercial heath plan(rate for this plan was 20.57 per 1,000), and the Medicaid population (mean was 72.28 per 1,000and the median was 69.93 per 1,000). The Committee noted that these rates demonstrate asignificant performance gap.
  
  
• Review for Scientific Acceptability: 2a. Reliability: 9-H; 11-M; 0-L;0-I 2b. Validity: 19-M; 0-L; 1-IRationale:• The developer tested the measure at the score level using several data sets for reliability testing:o For Medicare testing, the analysis included a convenience sample of over 700 MedicarePart D prescription drug plans (comprising a total of 7,067,445 individuals aged 18 andolder)o Testing was also conducted in one Commercial health plan (comprising a total of209,191 individuals age 18 and older)o For Medicaid testing, the analysis included 8 state-based prescription drug planscovering 6 states (comprising a total of 1,437,410 individuals age 18 and older)• To demonstrate reliability, the developer conducted a signal-to-noise analysis of the computedmeasure score using a beta-binomial model.• The mean reliability score across all plans is 0.9355.• The developer assessed the face validity (only) of the measure using a technical expert panelfrom the Pharmacy Quality Alliance (PQA). 67 percent strongly agreed that the measure resultsreflected quality of care. Five PQA member organizations also tested the measure using theirown data, and all strongly agreed that the measure reflected the quality of care provided fortheir populations.
  
  
• Review for Feasibility: 3. Feasibility: 18-H; 2-M; 0-L; 0-I(3a. Clinical data generated during care delivery; 3b. Electronic sources; 3c.Susceptibility to inaccuracies/unintended consequences identified 3d. Data collection strategy can be implemented)Rationale:• All data elements are in defined field in electronic claims.• Pilot test sites indicated the measure was feasible and results were able to be reportedefficiently and accurately.
  
  
• Review for Usability: 4. Usability and Use: 10-H; 9-M; 1-L; 0-I(Used and useful to the intended audiences for 4a. Accountability and Transparency; 4b. Improvement;and 4c. Benefits outweigh evidence of unintended consequences)Rationale:• The measure is currently being used in the Medicare Part D Overutilization Monitoring Systemto monitor the utilization of opioids for members with the Medicare drug benefit.36• Although no unintended negative consequences to individuals or populations were identifiedduring testing, , concerns have been raised that prescribing changes such as dose reduction(without offering or arranging evidence-based treatment for patients with opioid use disorder)might be associated with unintended negative consequences, such as patients seeking heroin orother illicitly obtained opioids (1,2) or interference with appropriate pain treatment
  
  
• Review for Related and Competing Measures: 5. Related and Competing Measures• Measure 2940: Use of Opioids at high Dosage in Persons without Cancer- The proportion (XX outof 1,000) of individuals without cancer receiving prescriptions for opioids with a daily dosagegreater than 120mg morphine equivalent dose (MED) for 90 consecutive days or longer.• Measure 2951: Use of Opioids from Multiple Providers and at High Dosage in Persons withoutCancer- The proportion (XX out of 1,000) of individuals without cancer receiving prescriptionsfor opioids with a daily dosage greater than 120mg morphine equivalent dose (MED) for 90consecutive days or longer, AND who received opioid prescriptions from four (4) or moreprescribers AND four (4) or more pharmacies.• These measures are also being considered for endorsement. The Committee determined thatthey are related but not competing.
  
  
• Endorsement Public Comments: Comment:The measure received 1 comment in support of the measure with a few recommendations for how themeasure could be improved.Developer Response:The recommendations in the 2015 American Geriatrics Society Beers Criteria are based on a systematicevidence review conducted by American Geriatrics Society Beers Criteria Expert Panel. The review isfocused on the evidence for potential harms of medications in older adults. Medications then includedin the Beers Criteria recommendations are those that the panel found evidence indicating that themedications should in general be avoided in all older adults or avoided in older adults with certainconditions or diseases, due to their associated risks for these populations. The Beers Criteria is updatedregularly based on currently available literature. We believe it's important for this quality measure to bebased on the systematic evidence review that is conducted by the Beers Criteria Expert Panel. Thecomplete evidence tables for the systematic review can be accessed on the American Geriatrics Society'swebsite here: http://geriatricscareonline.org/toc/american-geriatrics-society-updated-beers-criteria-forpotentially-inappropriate-medication-use-in-older-adults/CL001NCQA recognizes that some of the medications that are most attributable to adverse drug events inolder adults that result in ED visits and hospitalizations are not included in the Beers Criteria asmedications to be generally avoided (e.g., warfarin, antidiabetics and oral antiplatelets - although someoral antiplatelets are in fact included in the Beers Criteria and this measure: Dipyridamole, Ticlopidine).These other high-risk medications should be addressed in separate quality measures that focus on safeprescribing and appropriate monitoring, rather than this measure which focuses on medications thatshould be generally avoided. We agree with the need for such quality measures to improve safeprescribing of anticoagulants, antidiabetics, and opioids and have current work underway at NCQA to 37explore development of measures in these areas. Of note, the Pharmacy Quality Alliance has severalmeasures addressing opioid prescribing that are currently being considered for NQF endorsement aspart of this Patient Safety project. NCQA supports the endorsement of these measures and has plans toadapt them for health plan reporting in the near future.In terms of the way this measure is currently specified to include a number of different medications, webelieve that creating separate quality measures or indicators for all the specific medications in the BeersCriteria, or for each drug-disease interaction, would be burdensome for measurement and reporting byhealth plans. Plans can look at medications on an individual basis to see where improvements andinterventions are needed, however we do not think this level of detail would be desirable for nationalreporting by health plans.As a measure of potentially inappropriate medication use, NCQA does not expect this measure'sperformance to ever reach 0% (i.e., no prescribing of high-risk medications). There will always be caseswhere the benefits of prescribing a high-risk medication may outweigh the risks for certain patients.Clinicians should take into account various factors when considering the risk-benefit ratio of prescribinga high-risk medication to an individual. A companion paper to the Beers Criteria was published by theAmerican Geriatrics Society Workgroup on Improving Use of the Beers Criteria in 2015. The paperspecifically states "the AGS 2015 Beers Criteria are reasonable to use for performance measurementacross large groups of patients and providers but should not be used to judge care for any individual"(Steinman et al., 2015, JAGS). We believe measuring this concept of potentially inappropriatemedication use among elderly at the health plan (i.e., population) level is an important and usefulmedication safety measure that health plans can use to identify high-risk medication prescribing.Committee Response:The Committee agrees with the developer response and maintains their decision to recommend thismeasure for continued endorsement.
  
  
• Endorsement Committee Recommendation: Steering Committee Recommendation for Endorsement: 20-Y; 0-N
  
  

Measure Specifications

• NQF Number (if applicable): 2951
  
  
• Description: Percent of beneficiaries receiving opioid prescriptions with an average daily morphine milligram equivalent (MME) greater than or equal to 90 mg over a period of 90 days or longer, and opioid prescriptions from 4 or more prescribers and 4 or more pharmacies within 180 days or less.
  
  
• Numerator: Number of member-years of Medicare Part D beneficiaries 18 years and older in the denominator with an average MME greater than or equal to 90 mg during the measurement period and who received opioid prescriptions from 4 or more prescribers and 4 or more pharmacies within 180 days or less during the measurement period.
  
  
• Denominator: Number of member-years of enrolled Medicare Part D beneficiaries 18 years and older with at least 2 fills of a prescription opioid on unique dates of service (DOS) and at least 15 total opioid days supply over a period of 90 days or longer during the measurement period.
  
  
• Exclusions: Medicare beneficiaries with a cancer diagnosis or that are enrolled in hospice at any time during the measurement period are excluded from the denominator.
  
  
• HHS NQS Priority: Communication and Care Coordination
  
  
• HHS Data Source: Prescription Drug Event Data Elements
  
  
• Measure Type: Process
  
  
• Steward: Pharmacy Quality Alliance
  
  
• Endorsement Status:
  
  
• Meaningful Measure Area: Prevention and treatment of opioid and substance use disorders
  
  
• Changes to Endorsed Measure Specifications?: The MUC list indicates the measure has been modified from its endorsed version.
  
  
• Is the measure specified as an electronic clinical quality measure? No
  
  

Preliminary Analysis of Measure

• Preliminary analysis result: Support for Rulemaking
  
  
• Preliminary analysis summary
  • Contribution to program measure set:This measure is used in both the Medicaid Adult Core Set and SSP and is also NQF endorsed. This measure would benefit Part C & D beneficiaries receiving opioid prescriptions with an average daily morphine milligram equivalent (MME) greater than or equal to 90 mg over a period of 90 days or longer, and opioid prescriptions from 4 or more prescribers and 4 or more pharmacies within 180 days or less by providing them with information about plan quality and performance indicators and addressing quality objective gaps currently evident by a lack of opioid measures in the Parts C & D measure set.
  • Impact on quality of care for patients:This measure has the potential to impact approximately 13 million individuals, who are prescribed opioid treatment through Medicare Part D by reducing the risk of opioid use disorder or death.
  
  
  
• Does the measure address a critical quality objective not currently adequately addressed by the measures in the program set? Yes. The measure addresses the Effective Prevention and Treatment of Chronic Disease quality objective in the Medicare Part D program. Improving the way opioids are prescribed among multiple prescribers and lowering the number of Part D beneficiaries receiving opioid prescriptions with an average daily morphine milligram equivalent (MME) greater than or equal to 90 mg over a period of 90 days or longer can ensure patients have access to safer, more effective chronic pain treatment while reducing their risk of opioid use disorder, overdose, and death. There are currently no opioid measures in the Part C and D measure sets.
  
  
• Is the measure evidence-based and either strongly linked to outcomes or an outcome measure? Yes. This process measure meets clinical guidelines by tracking Part C and D beneficiaries receiving opioid prescriptions with an average daily morphine milligram equivalent (MME) greater than or equal to 90 mg over a period of 90 days or longer, and opioid prescriptions from 4 or more prescribers and 4 or more pharmacies within 180 days or less.Guidelines also focus on the use of opioids for treating chronic pain in non-cancerous, palliative care, and end-of-life care siutations, to improve opioid prescription to provide safer access to chronic pain treatment, and reduce risk of opioid use disorder, for patients (CDC, 2016).
  
  
• Does the measure address a quality challenge? Yes. This measure addresses a critical population at increased risk of opioid overdose and death. The 2016 to 2017 National Survey on Drug Use and Health in the United States estimated that among approximately 76 million United States adults prescribed opioid drugs in the prior year, and 12 percent reported prescription opioid misuse (Griesler et al., 2019).
  
  
• Does the measure contribute to efficient use of measurement resources and/or support alignment of measurement across programs? Yes. This measure is used in both Medicaid Adult Core Set and SSP, specifically in SSP’s Opioid Utilization Reports, not in the SSP measure set. This measure contains elements of MUC2019-57 and MUC2019-60 and may be somewhat duplicative should multiple measures be added.
  
  
• Can the measure can be feasibly reported? Yes. All data elements are in defined fields in electronic claims. Additionally, CMS provides Medicare Part D sponsors monthly reports for monitoring and case management purposes by calculating and reporting the measure using prescription drug event data from electronic prescription claims submitted by pharmacies to health plans.
  
  
• Is the measure applicable to and appropriately specified for the program's intended care setting(s), level(s) of analysis, and population(s)? Yes. This measure was tested at the Medicare Drug Plan level and intended for use for Medicare Part D Beneficiaries through Part D plan sponsors.
  
  
• Measure development status: Fully Developed
  
  
• If the measure is in current use, have negative unintended issues to the patient been identified? No. This measure has been used in SSP Opioid Utilization Reports since 2018, with no unintended consequences reported. However, concerns have been raised that prescribing changes could be associated with the unintended consequence of patients seeking illicitly obtained opioids or heroin.
  
  
• MAP Rural Workgroup Finding:
  • Relative priority/utility: The Workgroup agreed that opioid use is a relevant issue for rural residents.
  • Data collection issues: None
  • Calculation issues: None.
  • Unintended consequences: Although this measure could promote use of drug monitoring programs in rural areas, on the whole, it may not be particularly applicable, due to the relatively few pharmacies in rural areas. Of the three proposed opioid measures, the Workgroup agreed this one was the least useful.
  • Votes: Range is 1 – 5, where higher reflects more agreement regarding suitability for the program
  • Average agreement= 2.3
  • Vote Counts: (1 – 2 votes; 2 – 7 votes; 3 – 2 votes; 4 – 2 votes; 5 – 0 vote)
  • Program gap areas: The Workgroup suggested that a different measure developed by PQA (which considers opioids and benzodiazepines) may be a better alternative than the three proposed measures.
  
  
  

Rationale for measure provided by HHS
CMS adapted three PQA opioid overuse measures related to opioid use, including this OHDMP measure, to examine the quality of use related to the dose of the medications over time, access to the medications and the combination of both of these criteria. CMS has provided each Part D sponsor monthly reports using these metrics, and will publish these as CMS display measures beginning for 2020. Pending rule-making, we will consider adding one of these into the CMS Part C & D Star Ratings.Claims data from commercially insured patients indicate that approximately 8% of opioidprescriptions for acute pain and 12% for chronic pain specify a daily dosage of 120 MED ormore (1). The proportion of patients being treated at this dosage for more than 90 days has not been described. However, one study of veterans treated with 180 MED/day or more for 90+ days (2) found that this group was characterized by high rates of psychiatric and substance abuse disorders and frequently did not receive care consistent with clinical guidelines. Other studies have suggested the people at high opioid dosage are at greater risk of overdoses and fractures (3, 4, 5).The Washington State Agency Medical Directors Group has suggested 120 MED as a dosagelevel that should not be exceeded without special consideration (6). Prescription drug monitoring programs, which track the use of multiple providers by patients, indicate that such use is typically found among a small proportion of patients, with the proportion declining as the number of providers increases. In Massachusetts in 2006, considering only Schedule II opioids, 0.5% of patients saw 4+ prescribers and 4+ pharmacies (7). A national study found that 13% of patients had overlapping prescriptions from two or more different prescribers during an 18-month period. Of these, 0.5% used 4+ prescribers and 4+ pharmacies (8). People who see multiple prescribers or use multiple pharmacies are more likely to die of drug overdoses (4). Data from the California PDMP indicates that people with higher daily dosages are more likely to see multiple prescribers or go to multiple pharmacies (9). The data above suggest that prevention of opioid overdose deaths should focus on strategies that target (1) high-dose opioid users as well as (2) persons who seek care from multiple doctors and pharmacies. The data suggest that these criteria can be considered separately, as measures related to prescribed opioids for legitimate uses versus diverted uses. Thus, we will consider use of 3 measures, one for each criteria and one that is the intersection of both criteria. For the Part C and D Star Ratings, we would add only one of these measures.REFERENCES1. Ying Liu, PhD; Joseph E. Logan, PhD; Leonard J. Paulozzi, MD, MPH; Kun Zhang, MS; and Christopher M. Jones, PharmD. Potential Misuse and Inappropriate Prescription Practices Involving Opioid Analgesics. Am J Manag Care. 2013;19(8):648-658.2. Clinical characteristics of veterans prescribed high doses of opioid medications for chronic noncancer pain. Benjamin J. Morasco, Jonathan P. Duckart, Thomas P. Carr, Richard A. Deyo, Steven K. Dobscha. PAIN. 151 (2010) 625–632.3. Kate M. Dunn, PhD; Kathleen W. Saunders, JD; Carolyn M. Rutter, PhD; Caleb J. Banta-Green, MSW, MPH, PhD; Joseph O. Merrill, MD, MPH; Mark D. Sullivan, MD, PhD; Constance M. Weisner, DrPH, MSW; Michael J. Silverberg, PhD, MPH; Cynthia I. Campbell, PhD; Bruce M. Psaty, MD, PhD; and Michael Von Korff, ScD. Opioid Prescriptions for Chronic Pain and Overdose - A Cohort Study. Ann Intern Med. 2010;152:85-92.4. Paulozzi, et al. A History of Being Prescribed Controlled Substances and Risk of Drug Overdose Death. Pain Medicine 2011.5. Kathleen W. Saunders, JD, Kate M. Dunn, Ph. D, Joseph O. Merrill, M.D, M.P.H., Mark Sullivan, M.D., Ph.D., Constance Weisner, DrPH, M.S.W, Jennifer Brennan Braden, M.D., M.P.H., Bruce M. Psaty, M.D., Ph.D., and Michael Von Korff, Sc.D. Relationship of Opioid Use and Dosage Levels to Fractures in Older Chronic Pain Patients. Society of General Internal Medicine. 2009. DOI: 10.1007/s11606-009-1218-z.6. Agency Medical Directors Group (AMDG). Interagency Guideline on Opioid Dosing for Chronic Non-cancer Pain: An educational aid to improve care and safety with opioid therapy. 2010 Update. www.cdc.gov/HomeandRecreationalSafety/ Poisoning/brief.htm 7. Nathaniel Katz, Lee Panas, MeeLee Kim, Adele D. Audet, Arnold Bilansky, John Eadie, Peter Kreiner, Florence C Paillard, Cindy Thomas and Grant Carrow. Usefulness of prescription monitoring programs for surveillance – analysis of Schedule II opioid prescription data in Massachusetts, 1996–2006y. Pharmacoepidemiology and drug safety 2010; 19: 115–123.8. M. Soledad Cepeda, Daniel Fife, Wing Chow, Gregory Mastrogiovanni and Scott C. Henderson. Assessing Opioid Shopping Behaviour - A Large Cohort Study from a Medication Dispensing Database in the US. Drug Saf 2012.9. Han H, Kass PH, Wilsey BL, Li C-S (2012) Individual and County-Level Factors Associated with Use of Multiple Prescribers and Multiple Pharmacies to Obtain Opioid Prescriptions in California. PLoS ONE 7(9): e46246. doi:10.1371/journal.pone.0046246.

Summary of NQF Endorsement Review

• Year of Most Recent Endorsement Review: 2017
  
  
• Project for Most Recent Endorsement Review: Patient Safety Project 2015-2017
  
  
• Review for Importance: 1a. Evidence: 0-H;17-M; 1-L; 0-I; 1b. Performance Gap: 10-H; 6-M; 0-L; 0-IRationale:• The benefits for high dose opioids for chronic pain are not established and the risks for seriousharms related to opioid therapy increase at higher opioid dosage. The use of multipleprescribers and pharmacies are associated with increased risks for opioid overdose. The risk foroverdose increases with the number of prescribers and pharmacies.• The measure’s performance was tested in three different health plan data sources – theMedicare population (mean was 3.03 per 1,000 and the median was 2.89 per 1,000), onecommercial heath plan (mean rate 1.45 per 1,000), and the Medicaid population (mean was2.68 per 1,000 and the median was 2.38 per 1,000).
  
  
• Review for Scientific Acceptability: 2a. Reliability: 11-H; 5-M; 0-L;0-I 2b. Validity: 16-M; 2-L; 0-IRationale:• The measure was tested at the score level. The developer used several data sets for reliabilitytesting:• For Medicare testing, the analysis included a convenience sample of over 700 Medicare Part Dprescription drug plans (comprising a total of 7,067,445 individuals aged 18 and older)• Testing was also conducted in one Commercial health plan (comprising a total of 209,191individuals age 18 and older)39• For Medicaid testing, the analysis included 8 state-based prescription drug plans covering 6states (comprising a total of 1,437,410 individuals age 18 and older)• The mean reliability score across all plans is 0.9208.• The developer assessed the face validity (only) of the measure using a technical expert panelfrom the Pharmacy Quality Alliance (PQA). 83.3 percent strongly agreed that the measureresults reflected quality of care. Five PQA member organizations also tested the measure usingtheir own data, and all strongly agreed that the measure reflected the quality of care providedfor their populations.
  
  
• Review for Feasibility: 3. Feasibility: 15-H; 2-M; 0-L; 0-I(3a. Clinical data generated during care delivery; 3b. Electronic sources; 3c.Susceptibility to inaccuracies/unintended consequences identified 3d. Data collection strategy can be implemented)Rationale:• All data elements are defined in field in electronic claims• Pilot test sites indicated the measure was feasible and results were able to be reportedefficiently and accurately.
  
  
• Review for Usability: 4. Usability and Use: 10-H; 9-M; 1-L; 0-I(Used and useful to the intended audiences for 4a. Accountability and Transparency; 4b. Improvement;and 4c. Benefits outweigh evidence of unintended consequences)Rationale:• The measure is currently being used in the Medicare Part D Overutilization Monitoring Systemto monitor the utilization of opioids for members with the Medicare drug benefit.• Although no unintended negative consequences to individuals or populations were identifiedduring testing, , concerns have been raised that prescribing changes such as dose reduction(without offering or arranging evidence-based treatment for patients with opioid use disorder)might be associated with unintended negative consequences, such as patients seeking heroin orother illicitly obtained opioids (1,2) or interference with appropriate pain treatment.(3) Dataindicate that if access to prescription opioids is limited, some users of opioid analgesics willtransition to heroin or other illicitly obtained opioids, leading to increased overdose deathcoincident with prescribing restrictions
  
  
• Review for Related and Competing Measures: 5. Related and Competing Measures• Measure 2950: Use of Opioids from Multiple Providers in Persons without Cancer- Theproportion (XX out of 1,000) of individuals without cancer receiving prescriptions for opioidsfrom four (4) or more prescribers AND four (4) or more pharmacies.• Measure 2940: Use of Opioids at high Dosage in Persons without Cancer- The proportion (XX outof 1,000) of individuals without cancer receiving prescriptions for opioids with a daily dosagegreater than 120mg morphine equivalent dose (MED) for 90 consecutive days or longer.• These measures are also being considered for endorsement. The Committee determined thatthey are related but not competing.
  
  
• Endorsement Committee Recommendation: Steering Committee Recommendation for Endorsement: 18-Y; 0-N
  
  

Program Index

• Merit-Based Incentive Payment System 
  
  
• Medicare Shared Savings Program 
  
  
• Part C and D Star Ratings 
  
  

Full Program Summaries

Program History and Structure: The Medicare Access and CHIP Reauthorization Act of 2015 (MACRA) ended the Sustainable Growth Rate (SGR) formula, which would have resulted in a significant cut to payment rates for clinicians participating in Medicare. MACRA requires CMS by law to implement an incentive program for clinicians. This program, referred to as the Quality Payment Program, provides two participation pathways for clinicians: (1) The Merit-based Incentive Payment System (MIPS), and (2) Advanced Alternative Payment Models (Advanced APMs). MIPS combines three Medicare “legacy” programs – the Physician Quality Reporting System (PQRS), Value-based Payment Modifier (VM), and the Medicare EHR Incentive Program for Eligible Professionals – into a single program. Under MIPS, there are four connected performance categories that will affect a clinician’s future Medicare payments. Each performance category is scored independently and has a specific weight, indicating its contribution towards the MIPS Final Score. The MIPS performance categories and their 2018 weights towards the final score are: Quality (45%); Advancing Care information (25%); Improvement Activities (15%); and Cost (15%). The final score (100%) will be the basis for the MIPS payment adjustment assessed for MIPS eligible clinicians.

High Priority Domains for Future Measure Consideration:

CMS identified the following five domains as high-priority for future measure consideration:

1.Person and Caregiver-centered Experience and Outcomes: This means that the measure should address the experience of each person and their family; and the extent to which they are engaged as partners in their care. a. CMS wants to specifically focus on patient reported outcome measures (PROMs). Person or family-reported experiences of being engaged as active members of the health care team and in collaborative partnerships with providers and provider organizations.

2. Communication and Care Coordination: This means that the measure must address the promotion of effective communication and coordination of care; and coordination of care and treatment with other providers.

3. Efficiency/Cost Reduction: This means that the measure must address the affordability of health care including unnecessary health services, inefficiencies in health care delivery, high prices, or fraud. Measures should cause change in efficiency and reward value over volume.

4. Patient Safety: This means that the measure must address either an explicit structure or process intended to make care safer, or the outcome of the presence or absence of such a structure or process; and harm caused in the delivery of care. This means that the structure, process or outcome must occur as a part of or as a result of the delivery of care.

5. Appropriate Use: CMS wants to specifically focus on appropriate use measures. This means that the measure must address appropriate use of services, including measures of over use.

Measure Requirements: CMS applies criteria for measures that may be considered for potential inclusion in the MIPS. At a minimum, the following criteria and requirements must be met for selection in the MIPS: CMS is statutorily required to select measures that reflect consensus among affected parties, and to the extent feasible, include measures set forth by one or more national consensus building entities. To the extent practicable, quality measures selected for inclusion on the final list will address at least one of the following quality domains: Effective Prevention and Treatment, Making Care Safer, Communication and Coordination of Care, Best Practices of Healthy Living, Making Care Affordable or Person and Family Engagement. In addition, before including a new measure in MIPS, CMS is required to submit for publication in an applicable specialty-appropriate, peer-reviewed journal the measure and the method for developing the measure, including clinical and other data supporting the measure. Measures implemented in MIPS may be available for public reporting on Physician Compare. Measures must be fully developed, with completed testing results at the clinician level and ready for implementation at the time of submission (CMS’ internal evaluation). Preference will be given to measures that are endorsed by the National Quality Forum (NQF). Measures should not duplicate other measures currently in the MIPS. Duplicative measures are assessed to see which would be the better measure for the MIPS measure set. Measure performance and evidence should identify opportunities for improvement. CMS does not intend to implement measures in which evidence identifies high levels of performance with little variation or opportunity for improvement, e.g., measures that are “topped out.” Claims measures must also be reportable via another data submission mechanism (e.g. registry, eCQM). MIPS is not accepting claims only measures. Section 101(c)(1) of the MACRA requires submission of new measures for publication in applicable specialty-appropriate, peer-reviewed journals prior to implementing in MIPS. The Peer-Review Journal template provided by CMS, must accompany each measures submission. Please see the template for additional information. eCQMs must meet EHR system infrastructure requirements, as defined by MIPS regulation. Beginning with calendar year 2019, eCQMs will use clinical quality language (CQL) as the expression logic used in the Health Quality Measure Format (HQMF). CQL replaces the logic expressions currently defined in the Quality Data Model (QDM). The data collection mechanisms must be able to transmit and receive requirements as identified in MIPS regulation. For example, eCQMs being submitted as Quality Reporting Data Architecture (QRDA) III must meet QRDA – III standards as defined in the CMS QRDA III Implementation Guide. eCQMs must have HQMF output from the Measure Authoring Tool (MAT), using MAT v5.4, or more recent, with implementation of the clinical quality language logic. Additional information on the MAT can be found at https://ecqi.healthit.gov/ecqm-tools/tool-library/measure-authoring-tool Bonnie test cases must accompany each measure submission. Additional information on eCQM Tools and resources can be found at https://ecqi.healthit.gov/ecqm-tools-key-resources. Reliability and validity testing must be conducted for measures. In addition to the above, feasibility testing must be conducted for eCQMs. Testing data must accompany submission. For example, if a measure is being reported as registry and eCQM, testing data for both versions must be submitted.

Current Measures: NQF staff have compiled the program's measures in a spreadsheet organized according to concepts.

Program History and Structure: Section 3022 of the Affordable Care Act (ACA) requires the Centers for Medicare & Medicaid Services (CMS) to establish a Shared Savings Program that promotes accountability for a patient population, coordinates items and services under Medicare Parts A and B, and encourages investment in infrastructure and redesigned care processes for high-quality and efficient service delivery. The Medicare Shared Savings Program (Shared Savings Program) was designed to facilitate coordination and cooperation among providers to improve the quality of care for Medicare Fee-For-Service (FFS) beneficiaries and reduce the rate of growth in health care costs. Eligible providers, hospitals, and suppliers may voluntarily participate in the Shared Savings Program by creating or participating in an Accountable Care Organization (ACO). On December 31, 2018 CMS released the Medicare Shared Savings Program: Accountable Care Organizations – Pathways to Success final rule. “Pathways to Success” refers to a combination of policy changes which include: redesigning the participation options available under the program to encourage ACOs to transition to two-sided models (in which they may share in savings and are accountable for repaying shared losses); providing new tools to support coordination of care across settings and strengthen beneficiary engagement; ensuring rigorous benchmarking; promoting interoperable electronic health record technology among ACO providers/suppliers; and improving information sharing on opioid use to combat opioid addiction.

Measure Requirements: Specific measure requirements include: Outcome measures that address conditions that are high-cost and affect a high volume of Medicare patients. Measures that are targeted to the needs and gaps in care of Medicare fee-for-service patients and their caregivers. Measures that align with CMS quality reporting initiatives, such as the Quality Payment Program. Measures that support improved individual and population health. Measures addressing high-priority healthcare issues, such as opioid use. Measures that align with recommendations from the Core Quality Measures Collaborative.

Current Measures: NQF staff have compiled the program's measures in a spreadsheet organized according to concepts.

Program History and Structure: Program Type: Quality Payment Program & Public Reporting Incentive Structure: Medicare Part C: Public reporting & Quality bonus payments—5% if 4 Stars or higher Medicare Part D: Public reporting Program Goal: Provide information about plan quality and performance indicators be provided to beneficiaries to help them make informed plan choices. Incentivize high performing plans (Part C).

High Priority Domains for Future Measure Consideration: Promote Effective Communication and Coordination of Care. A primary goal is to coordinate care for beneficiaries in the effort to provide quality care. The Medicare population includes a large number of individuals and older adults with high-risk multiple chronic conditions (MCC) who often receive care from multiple providers and settings and, as a result, are more likely to experience fragmented care and adverse healthcare outcomes. Promote Effective Prevention and Treatment of Chronic Disease. Medicare beneficiaries with multiple high-risk chronic conditions are at increased risk for fragmented care and poor health outcomes so attention to effectively preventing and treating chronic disease is important.

Current Measures: NQF staff have compiled the program's measures in a spreadsheet organized according to concepts.

Index of Measures (by Program)

General Comments

• General comments
  
  

• Hospital-Wide, 30-Day, All-Cause Unplanned Readmission (HWR) Rate for the Merit-Based Incentive Payment Program (MIPS) Eligible Clinician Groups (Public comments received:11; MUC ID: MUC2019-27)
  
  
• Risk-standardized complication rate (RSCR) following elective primary total hip arthroplasty (THA) and/or total knee arthroplasty (TKA) for Merit-based Incentive Payment System (MIPS) Eligible Clinicians and Eligible Clinician Groups (Public comments received:5; MUC ID: MUC2019-28)
  
  
• Clinician and Clinician Group Risk-standardized Hospital Admission Rates for Patients with Multiple Chronic Conditions; in the Medicare Shared Savings Program, the score would be at the ACO level. (Public comments received:8; MUC ID: MUC2019-37)
  
  
• Hemodialysis Vascular Access: Practitioner Level Long-term Catheter Rate (Public comments received:3; MUC ID: MUC2019-66)
  
  

• Clinician and Clinician Group Risk-standardized Hospital Admission Rates for Patients with Multiple Chronic Conditions; in the Medicare Shared Savings Program, the score would be at the ACO level. (Public comments received:4; MUC ID: MUC2019-37)
  
  

• Follow-up after Emergency Department (ED) Visit for People with Multiple High-Risk Chronic Conditions (Public comments received:3; MUC ID: MUC2019-14)
  
  
• Transitions of Care between the Inpatient and Outpatient Settings including Notifications of Admissions and Discharges, Patient Engagement and Medication Reconciliation Post-Discharge (Public comments received:2; MUC ID: MUC2019-21)
  
  
• Use of Opioids at High Dosage in Persons without Cancer (OHD) (Public comments received:5; MUC ID: MUC2019-57)
  
  
• Use of Opioids from Multiple Providers in Persons without Cancer (OMP) (Public comments received:5; MUC ID: MUC2019-60)
  
  
• Use of Opioids from Multiple Providers and at a High Dosage in Persons without Cancer (OHDMP) (Public comments received:4; MUC ID: MUC2019-61)
  
  

Full Comments (Listed by Measure)

• In general we support these measures. The measure listing is comprehensive encompassing all of the major areas of concern such as: MUC2019-018 and 019 hospital acquired infections (UTIs and central blood lines), Measure 021 on transitions of care, 022 mental health follow-up, 027 preventable readmissions, and 114 maternal morbidity and mortality are all important key areas. (Submitted by: Family Voices NJ)
  
  

• Bravado Health supports measurement MUC1914 -Follow-up after Emergency Department (ED) Visit for People with Multiple High-Risk Chronic Conditions The utilization of follow-up services for ED visits for people with a high risk chronic condition would serve as proactive extension of the ED visit to ensure the patient has retained educational instruction, is following up on the recommended interventions, and ensuring successful self-management of the respective chronic condition. Further, connecting to such services will allow the ability to glean root causes deterring the patient for appropriate follow-up (Ie. SDoH). Further, the proposed chronic conditions (2 or more of Alzheimer’s disease, atrial fibrillation, chronic kidney disease, COPD, depression, heart failure, cardiovascular disease (AMI/Stroke/TIA) fit the high risk, high volume, problem-prone condition that create challenges for successful self-management of care. While measurement is standardized, the way in which an organization chooses to implement is built with flexibility, giving the organization the choice of how that is most appropriate for their service offerings, culture, resources, and time). (Submitted by: Bravado Health)
  
  
• Support with modification: It is unclear what a follow-up service is and that is not explained in the measure detail provided here. Also, the NCQA website posting for this measure does not provide more details on this either. Finally, there could be confusion as the definition of multiple high-risk chronic conditions for ths measure differs from that used elsewhere to describe the population eligible for MA supplemental benefits. That definition is found in section 20.1.2 of Chapter 16b of the Medicare Managed Care Manual. We do appreciate exclusion of Medicare beneficiaries enrolled in hospice (Submitted by: The Coalition to Transform Advanced Care)
  
  
• The American College of Emergency Physicians (ACEP) appreciates the opportunity to provide comments on this measure. ACEP strongly agrees that it is essential for all patients, especially those with multiple high-risk chronic conditions, to receive follow-up care. While we have no significant concerns with the measure, we believe it is important for CMS to invest in care coordination and help support emergency physicians and other ED clinicians to appropriately follow-up with patients once discharged. ACEP has urged CMS to consider the development of a care coordination service that would be specific to the services required for providing non-face-to-face follow-up with both the patient and community health care providers for a complex Medicare patient discharged from the ED. Further, ACEP developed a proposal for a Medicare alternative payment model (APM) called the Acute Unscheduled Care Model (AUCM), which we strongly believe has the potential to transform the way emergency care is delivered. Structured as a bundled payment model, it allows emergency physicians to accept some financial risk for the decisions they make around discharges for certain episodes of acute unscheduled care. It would enhance the ability of emergency physicians to reduce inpatient admissions and observation stays when appropriate through processes that support care coordination. Emergency physicians would become members of the continuum of care as the model focuses on ensuring follow-up, minimizing redundant post-ED services, and avoiding post-ED discharge safety events that lead to follow-up ED visits or inpatient admissions. We presented the AUCM proposal before the Physician-Focused Payment Model Technical Advisory Committee (PTAC) on September 6, 2018. The PTAC recommended the AUCM to the Secretary of the Department of Health and Human Services (HHS) for full implementation. On September 27, 2019, Secretary Alex Azar responded to the PTAC’s recommendation by stating that he believes that core concepts of the AUCM should be incorporated into APMs that the CMS’ Innovation Center (CMMI) is developing. We look forward to working CMMI to advance emergency patient care through the implementation of this model. (Submitted by: American College of Emergency Physicians)
  
  

• Bravado Health supports MUC1921 Transitions of Care between the Inpatient and Outpatient Settings including Notifications of Admissions and Discharges, Patient Engagement and Medication Reconciliation Post Discharge. In particular, Bravado Health supports the post discharge patient engagement with the ability for an organization to choose the best way to operationalize post discharge patient engagement (ie. Digital technology, telehealth, visits, etc). Measuring all aspects of the stay including engagement post discharge effectively closes the loop of measurement and efficacy of that measurement. Additionally, allows an organization of root causes of identified opportunities to improve process. (Submitted by: Bravado Health)
  
  
• We support this measure as improving care transitions is key to improving care for advanced illness. This measure provides a good composite of following key areas: Notification of inpatient admission and/or receipt of discharge information on the day of or following day: Patient engagement via office/home/telehealth visits provided within 30 days after discharge Medication reconciliation post-discharge on the date of discharge through 30 days after discharge We also appreciate the exclusion of Medicare beneficiaries enrolled in hospice from the denominator. (Submitted by: The Coalition to Transform Advanced Care)
  
  

• Additional information about how the measure has changed is needed. The existing measure could be enhanced through additional risk adjustment for beneficiaries’ relative severity of illness and through accounting for social determinants of health. Inclusion of adjustment for these factors is essential in order to ensure that clinicians serving higher acuity or disadvantaged populations are not unduly penalized. (Submitted by: Premier)
  
  
• ASCRS opposes the inclusion of this, or other, population-based measures in the MIPS program. Population-based measures are troubling for specialists like ophthalmologists, who do not manage the overall health of the patient. Population-health measures are primary care-based, and the attribution methodology potentially holds physicians responsible for care they did not provide. Ophthalmologists have no way to predict what patients will be attributed and have no means to influence their scores, particularly for this measure because they do not provide any inpatient care. While we believe that population-based measures should not be used in the MIPS program at all, at a minimum, they should exclude specialists, such as ophthalmologists, from attribution. (Submitted by: American Society of Cataract and Refractive Surgery)
  
  
• For this measure, the outcome is readmission within 30 days of discharge from an admission; planned readmissions are excluded. This adapted measure is intended for use in MIPS, part of the Quality Payment Program, to assess the performance of eligible clinicians (ECs) or EC groups. There is currently a version of the hospital-level HWR measure in use under MIPS, referred to as the All-Cause Readmission (ACR) measure. Where relevant, this proposed MIPS HWR measure was drawn from the ACR measure; however, the original hospital-level measure was used as the foundation for development work because that version has been most rigorously tested and vetted. Earlier measure development work focused on redefining the attribution approach for an EC-or EC group-level measure. ASCO is concerned that while the original ACR measure was attributed only to groups of more than 15 clinicians who also had 200 or more readmissions (“case minimums”), as proposed earlier the HWR measure would be attributed to smaller groups and individual ECs with a case minimum of only 100. Further reliability and validity testing is needed prior to implementing this measure across MIPS. It is our understanding also that social risk factors were not considered in the risk adjustment model for this measure; we have previously commented that these risk factors should be an important part of any risk model and would urge CMS to delay implementation of this measure until such factors can be incorporated, with an allowance for stakeholder input. ASCO appreciates the exclusion from this measure of admissions to a PPS-exempt cancer hospital and admissions primarily for medical treatment of cancer, as those should not be considered “unplanned readmissions.” Additional General Comments on Measures We have earlier expressed our concerns in relation to other claims-based measures regarding the application of comorbidities in risk adjustment wherein the patient’s conditions from the prior calendar year (e.g. 2018) are used to predict the risk of an adverse event in the performance year (e.g. 2019). In the case of chronic conditions where an adverse event is expected to remain stable or increase over time, the use of such a prospective methodology may work. However, in the setting of a cancer diagnosis, there is usually an acute phase, followed by a chronic phase. The use of the prior year’s comorbidities appropriately adjusts for patients living with cancer in a chronic phase – for example, survivors and patients on long-term maintenance. However, this methodology does not account for newly diagnosed patients presenting to an oncologist in the current year. These patients, in their acute phase, have a higher likelihood of adverse events such as admissions, as well as a higher likelihood of being primarily managed by the oncologist. It appears that in some of the proposed measures, their new disease is not accounted for in the risk adjustment, as it was not present in the prior calendar year. Potential options to fix this issue include consideration of all diagnoses from the prior and current calendar year, up to the second visit of the attributed provider (this framework is used in the Medicare Spending per Beneficiary and other episode-based programs), or the exclusion of patients presenting with a new, select diagnosis in the current calendar year. (Submitted by: American Society for Clinical Oncology)
  
  
• The AAMC does not support inclusion of the Hospital-Wide, 30-Day, All-Cause Unplanned Readmission (HWR) Rate for the Merit-Based Incentive Payment Program (MIPS) Eligible Clinician Groups measure in the MIPS program. The measure description does not meaningfully describe whether there is the appropriate clinical risk adjustment necessary for implementation in any federal program. In addition, readmissions are often connected to the broader community and its access to care, and CMS should consider adding an adjustment or stratification to account for socio-demographic factors, as is done in the Hospital Readmission Reduction Program. Additionally, in regard to patient attribution, the AAMC is concerned that during the endorsement process for this measure, the measure developer was unable to provide support for attribution of the measure to up to three physicians or practices. Finally, the measure has not yet been endorsed by NQF due to issues with the reliability of the measure as specified. NQF endorsement is critical to establishing that the measure has been appropriately tested and is proven valid and reliable. We recommend that the issues related to risk adjustment, sociodemographic factors, attribution, and measure reliability be addressed. In light of these concerns, the AAMC recommends that the MAP recommendation be “Do Not Support.” (Submitted by: Association of American Medical Colleges (AAMC))
  
  
• The American Academy of Neurology does not support inclusion of the Hospital-Wide, 30-Day, All-Cause Unplanned Readmission (HWR) Rate for the Merit-Based Incentive Payment Program (MIPS) Eligible Clinician Groups measure in the MIPS program. We do not believe that continuing to include measures based on administrative claims meets the intended goals of this program and while this measure may be useful at the community or population level, it is not appropriate to attribute this utilization to an individual physician or practices. Further testing is also needed to demonstrate how the measure would perform under the MIPS benchmark methodology and Physician Compare Star Ratings since CMS utilizes two different methodologies for ranking and profiling physicians. This measure has not yet been endorsed by NQF due to concerns over the reliability of the measure as specified. Please see this article as well: https://jamanetwork.com/journals/jamasurgery/article-abstract/1879843 (Submitted by: American Academy of Neurology)
  
  
• The American Medical Association (AMA) does not support inclusion of the Hospital-Wide, 30-Day, All-Cause Unplanned Readmission (HWR) Rate for the Merit-Based Incentive Payment Program (MIPS) Eligible Clinician Groups measure in the MIPS program. We do not believe that continuing to include measures based on administrative claims meets the intended goals of this program and while this measure may be useful at the community or population level, it is not appropriate to attribute this utilization to an individual physician or practices. The AMA expressed several concerns regarding this measure during the recent NQF endorsement review. We believe that insufficient evidence was provided to support attribution of the measure to physicians or practices in the absence of some coordinated program or targeted intervention led by the health system or hospital. Assignment of responsibility of the reduction of readmissions to multiple physicians and practices in MIPS is not appropriate and the developer did not provide sufficient information to support the attribution of this measure to up to three physicians or practices. The measure score reliability results were too low when based on the minimum case number of 25 patients and higher minimum acceptable thresholds such as 0.80 should be required. We also believe that the conceptual basis used to explain which social risk factors were tested was inadequate and additional testing is needed to evaluate clinical factors in conjunction with social risk factors as well as the impact that the inclusion of these factors had on the absolute change of the rates. Further testing is also needed to demonstrate how the measure would perform under the MIPS benchmark methodology and Physician Compare Star Ratings since CMS utilizes two different methodologies for ranking and profiling physicians. This measure has not yet been endorsed by NQF due to concerns over the reliability of the measure as specified. In light of these concerns, the AMA recommends that the highest level of MAP recommendation be “Do Not Support”. (Submitted by: American Medical Association)
  
  
• The Federation of American Hospitals (FAH) strongly advocates that any measure that is proposed for use in payment programs should be evidence-based, appropriate for accountability purposes at the designated level of attribution, and demonstrated to be reliable and valid. The FAH does not support attributing this measure to the three types of clinicians due to the lack of sufficient data and empirical evidence to demonstrate that any of these individuals can meaningfully influence readmission rates. In addition, FAH is troubled by the lack of robust testing of the risk adjustment model for social risk factors, the low reliability threshold produced at the group level, and the limited testing to demonstrate that the attribution methodologies provide valid representations of the care provided by clinicians. Because this measure has not been endorsed by NQF yet and in fact is being reconsidered due to concerns with the reliability of the measure, the FAH requests that the highest level of MAP recommendation be “Do Not Support with Potential for Mitigation”. (Submitted by: Federation of American Hospitals)
  
  
• The Heart Rhythm Society (HRS) does not support inclusion of the Hospital-Wide, 30-Day, All-Cause Unplanned Readmission (HWR) Rate for the Merit-Based Incentive Payment Program (MIPS) Eligible Clinician Groups measure in the MIPS program. While these administrative claims measure may be useful at the community or population level, it is not appropriate to attribute this utilization to an individual physician or practices. HRS expressed several concerns regarding this measure during the recent NQF endorsement review. There is insufficient evidence provided to support attribution of the measure to physicians or practices in the absence of some coordinated program or targeted intervention led by the health system or hospital. Assignment of responsibility of the reduction of readmissions to multiple physicians and practices in MIPS is not appropriate and there is insufficient information to support the attribution of this measure to up to three physicians or practices. Furthermore, the measure score reliability results were too low when based on the minimum case number of 25 patients, and higher minimum acceptable thresholds such as 0.80 should be required. The conceptual basis used to explain which social risk factors were tested was inadequate and additional testing is needed to evaluate clinical factors in conjunction with social risk factors, as well as the impact that the inclusion of these factors had on the absolute change of the rates. Further testing also is needed to demonstrate how the measure would perform under the MIPS benchmark methodology and Physician Compare Star Ratings as CMS utilizes two different methodologies for ranking and profiling physicians. This measure has not yet been endorsed by NQF due to concerns over the reliability of the measure as specified. Considering these concerns, HRS recommends that the highest level of MAP recommendation should be “Do Not Support”. (Submitted by: Heart Rhythm Society)
  
  
• The Infectious Diseases Society of America (IDSA) does not support inclusion of the Hospital-Wide, 30-Day, All-Cause Unplanned Readmission (HWR) Rate for the Merit-Based Incentive Payment Program (MIPS) Eligible Clinician Groups measure in the MIPS program. We do not believe that continuing to include measures based on administrative claims meets the intended goals of this program and while this measure may be useful at the community or population level, it is not appropriate to attribute this utilization to an individual physician or practices. We believe that insufficient evidence was provided to support attribution of the measure to physicians or practices in the absence of some coordinated program or targeted intervention led by the health system or hospital. Assignment of responsibility of the reduction of readmissions to multiple physicians and practices in MIPS is not appropriate and the developer did not provide sufficient information to support the attribution of this measure to up to three physicians or practices. The measure score reliability results were too low when based on the minimum case number of 25 patients and higher minimum acceptable thresholds such as 0.80 should be required. We also believe that the conceptual basis used to explain which social risk factors were tested was inadequate and additional testing is needed to evaluate clinical factors in conjunction with social risk factors as well as the impact that the inclusion of these factors had on the absolute change of the rates. Further testing is also needed to demonstrate how the measure would perform under the MIPS benchmark methodology and Physician Compare Star Ratings since CMS utilizes two different methodologies for ranking and profiling physicians. This measure has not yet been endorsed by NQF due to concerns over the reliability of the measure as specified. In light of these concerns, the IDSA recommends that the highest level of MAP recommendation be “Do Not Support”. (Submitted by: Infectious Diseases Society of America)
  
  
• There is not sufficient evidence to support attribution of this measure to clinician groups without developing a coordinated program or intervention led by health systems or hospitals. Attribution remains a concern as it is difficult to discern how multiple physicians may be assigned responsibility for a reduction in readmissions. Further reliability testing is needed, as well as the impact of social risk factors and how the measure performs under the MIPS benchmark methodology. This measure has not yet been endorsed by NQF due to concerns over the reliability of the measure as specified. Procedure categories should be evaluated carefully. For example, TAVR (or transcatheter aortic valve replacement) is an increasingly large driver of valve-related disorders, for which there is a defined therapy and low readmission rate. Lumping this procedure with “heart valve disorders” would not be appropriate and would make the category become too heterogeneous. We encourage a higher case minimum of patients to ensure higher reliability of the measure. The reliability ratings for clinicians in the cardiorespiratory and cardiac groupings were only moderate. (Submitted by: American College of Cardiology)
  
  
• We have serious concerns about using a 30-day readmission window, generally, and with this measure proposed for the MIPS, we have special concerns as it is attributable to providers. Attribution would be better, more closely linked with clinician behavior, and closer to the “preventability” of readmission, if this measure window were shortened to 7 days. We strongly recommend against implementing the measure at this time. Supporting literature: https://www.healthaffairs.org/doi/full/10.1377/hlthaff.2016.0205 https://annals.org/aim/article-abstract/2723790/preventability-early-versus-late-hospital-readmissions (Submitted by: Society of Hospital Medicine)
  
  
• We support this measure with the following modification: that it recognize that a few outlier beneficiaries could severely affect score so it may make sense to exclude additional patient groups like pre/post organ transplant patients. (Submitted by: The Coalition to Transform Advanced Care)
  
  

• The American College of Surgeons does not support the inclusion of Risk-standardized complication rate (RSCR) following elective primary total hip arthroplasty (THA) and/or total knee arthroplasty (TKA) for Merit-based Incentive Payment System (MIPS) Eligible Clinicians and Eligible Clinician Groups (MUC2019-28) in the MIPS program, and instead recommends using functional Patient Reported Outcomes (PROs) to measure outcomes and post-surgical risk and complication for both THA and TKA. We do not support MUC2019-28 because it does not track outcomes that matter to patients for THA and TKA. Instead, using functional PROs preoperatively such as HOOS Jr. (hip disability and osteoarthritis outcome score) and KOOS Jr. (knee injury and osteoarthritis outcome score), and then following up postoperatively, would more accurately measure the success of the procedure based on outcomes that matter to the patient. PROs give patients the opportunity to determine whether their care goals were met, share their post-surgical experience, and provide meaningful, actionable data for surgeons. PROs tailored to a condition or episode allow clinicians to better understand the elements of care their patients value most and empower patients to work with care teams to communicate goals and engage in shared decision making prior to and during care. Collecting PROs in more frequent, but brief, occurrences throughout episode(s) of care can provide meaningful information to physicians throughout the patient’s care journey and enhance patient-clinician communication—including progress on patient goals, post-surgical recovery, pain management, and rehab and therapy, to name a few. Using PROs further reflect the program’s intent to become more patient-centric, in order to assess outcomes that matter to patients. Furthermore, due to the low event rates for both of these procedures—both at the individual clinician and group level—the current measure will not likely be able to reliably differentiate from one clinician or group from the next. In general, with event rate outcome measures, if it is not possible to achieve discernibility at the individual-clinician level, then the hospital/institution level measurement should be used as a proxy for quality. This must be determined on a measure-by-measure basis. However, to further support the above argument for PROs, the use of PROs has demonstrated reliably on the level of the clinician and therefore is not only more patient-centric but can produce statistically valid measurements. (Submitted by: American College of Surgeons)
  
  
• The AAMC does not support inclusion of the Risk-standardized complication rate (RSCR) following elective primary total hip arthroplasty (THA) and/or total knee arthroplasty (TKA) for Merit-based Incentive Payment System (MIPS) Eligible Clinicians and Eligible Clinician Groups measure in the MIPS program. We are concerned that this measure is not appropriate for attribution to an individual clinician or clinician group. While this measure was endorsed by the NQF, the measure has the similar results on the reliability of the measures when applied to individual physicians and groups. These concerns resulted in NQF asking the Admissions/Readmissions Standing Committee to reexamine the HWR measure during the upcoming cycle. We recommend that the issues related to attribution and reliability be addressed. In light of these concerns, the AAMC recommends that the MAP recommendation be “Do Not Support.” (Submitted by: Association of American Medical Colleges (AAMC))
  
  
• The American Medical Association (AMA) does not support inclusion of the Risk-standardized complication rate (RSCR) following elective primary total hip arthroplasty (THA) and/or total knee arthroplasty (TKA) for Merit-based Incentive Payment System (MIPS) Eligible Clinicians and Eligible Clinician Groups measure in the MIPS program. We do not believe that continuing to include measures based on administrative claims meets the intended goals of this program and while this measure may be useful at the community or population level, it is not appropriate to attribute this utilization to an individual physician or practices. The AMA expressed several concerns regarding this measure during the recent NQF endorsement review. We believe that insufficient evidence was provided to support attribution of the measure to physicians or practices. The measure score reliability results were too low when based on the minimum case number of 25 patients and higher minimum acceptable thresholds such as 0.80 should be required. We also believe that the conceptual basis used to explain which social risk factors were tested was inadequate and additional testing is needed to evaluate clinical factors in conjunction with social risk factors as well as the impact that the inclusion of these factors had on the absolute change of the rates. Further testing is also needed to demonstrate how the measure would perform under the MIPS benchmark methodology and Physician Compare Star Ratings since CMS utilizes two different methodologies for ranking and profiling physicians. While this measure was endorsed by NQF in this last cycle, we note that the measure has the similar results on the reliability of the measures when applied to individual physicians and groups. These concerns resulted in NQF asking the Admissions/Readmissions Standing Committee to reexamine the HWR measure during the upcoming cycle but the same request has not been made of the Surgery Standing Committee for this measure. This inconsistency is troubling and also leads us to question whether NQF endorsement of this measure with similar reliability issues was appropriate. As a result, the AMA recommends that the highest level of MAP recommendation be “Do Not Support”. (Submitted by: American Medical Association)
  
  
• The Federation of American Hospitals (FAH) strongly advocates that any measure that is proposed for use in payment programs should be evidence-based, appropriate for accountability purposes at the designated level of attribution, and demonstrated to be reliable and valid. The FAH is troubled by the lack of robust testing of the risk adjustment model for social risk factors, the low reliability threshold produced at the group level, and the limited testing to demonstrate that the attribution methodologies provide valid representations of the care provided by clinicians. In addition, while this measure was recently endorsed by NQF, the HWR measure that yielded similar testing results is being reconsidered due to concerns with the reliability of the measure, indicating some degree of inconsistency in the reviews. As a result, the FAH requests that the highest level of MAP recommendation be “Do Not Support with Potential for Mitigation” until these issues are resolved. (Submitted by: Federation of American Hospitals)
  
  

• ASCO has commented on this measure in earlier stages of measure development and our concerns remain relevant here. The nine “chronic conditions” covered by this measure include the following: acute myocardial infarction (AMI); Alzheimer’s disease and related disorders or senile dementia; atrial fibrillation; chronic kidney disease (CKD); chronic obstructive pulmonary disease (COPD) or asthma; depression; diabetes; heart failure; and stroke or transient ischemic attack (TIA). Specialists who can be attributed this measure “are limited to those who provide overall coordination of care for patients with MCCs and who manage the chronic diseases that put the MCC patients in the measure at risk of admission. These specialists were chosen with input from our Technical Expert Panel (TEP) and include cardiologists, pulmonologists, nephrologists, neurologists, endocrinologists, and hematologists / oncologists.” As we commented earlier, we disagree with the assertion that hematologists / oncologists “provide overall coordination of care for patients with MCCs and who manage the chronic diseases that put the MCC patients in the measure at risk of admission.” There is little reason to believe that a hematologist/oncologist would manage a patient with two or more of the above-listed chronic conditions, unless such a patient also has cancer or a major blood disorder. First, it is unlikely that—absent a cancer diagnosis—oncologists would manage the non-oncologic chronic diseases on this list; second, unlike the other specialties listed for attribution, there is no clear link between hematology/oncology and the MCCs listed. In other words, it is plausible that a cardiologist might manage atrial fibrillation or heart failure; a pulmonologist might manage COPD or asthma; a nephrologist might manage CKD; a neurologist might manage stroke/TIA or Alzheimer’s disease; and an endocrinologist might manage diabetes. However, none of the nine chronic conditions on this list would normally be managed primarily by an oncologist. Given that cancer is not the focus of this model, inclusion of the hematologist/oncologist seems arbitrary and should be removed from the list of relevant specialties. Additional General Comments on Measures We have earlier expressed our concerns in relation to other claims-based measures regarding the application of comorbidities in risk adjustment wherein the patient’s conditions from the prior calendar year (e.g. 2018) are used to predict the risk of an adverse event in the performance year (e.g. 2019). In the case of chronic conditions where an adverse event is expected to remain stable or increase over time, the use of such a prospective methodology may work. However, in the setting of a cancer diagnosis, there is usually an acute phase, followed by a chronic phase. The use of the prior year’s comorbidities appropriately adjusts for patients living with cancer in a chronic phase – for example, survivors and patients on long-term maintenance. However, this methodology does not account for newly diagnosed patients presenting to an oncologist in the current year. These patients, in their acute phase, have a higher likelihood of adverse events such as admissions, as well as a higher likelihood of being primarily managed by the oncologist. It appears that in some of the proposed measures, their new disease is not accounted for in the risk adjustment, as it was not present in the prior calendar year. Potential options to fix this issue include consideration of all diagnoses from the prior and current calendar year, up to the second visit of the attributed provider (this framework is used in the Medicare Spending per Beneficiary and other episode-based programs), or the exclusion of patients presenting with a new, select diagnosis in the current calendar year. (Submitted by: American Society for Clinical ONcology)
  
  
• Concerns: attribution, risk adjustment (including socio-demographic factors). This measure in particular is pretty complex (it has far more detail than the others) and is unclear whether it has been tested by the developer for validity/reliability for reporting for clinician groups and ACOs The AAMC does not support the inclusion of the Clinician and Clinician Group Risk-standardized Hospital Admission Rates for Patients with Multiple Chronic Conditions measure in the MIPS program and Shared Savings Program for ACOs. The AAMC recognizes the need to improve the care of patients with multiple chronic conditions, but has significant concerns with the use of these measures in these programs. Attribution of this measure to individual clinicians or clinician groups is of great concern. We do not believe that sufficient evidence was provided to support the theory that physicians or practices, in the absence of some coordinated program or payment offset (e.g., care management fee), can implement structures or processes that can lead to improved outcomes for these patients. Furthermore, because the measure uses a retrospective approach, it is unable to provide timely and actionable feedback to the point of care. The characteristics of the attributed Medicare beneficiaries can vary widely by physician group practice. Not accounting for the clinical variation in the underlying population is extremely misleading and disproportionately affects the physicians who care for the most complex patients. These measures should have appropriate clinical risk adjustment prior to implementation in any program. In addition, as admissions and readmissions are often connected to the broader community, CMS should consider adding an adjustment or stratification to account for socio-demographic factors. The use of this measure at the ACO level also would have similar challenges with population size and risk adjustment. An ACO patient population is typically much smaller than 100,000. In addition, an ACO is already accountable for costs and has an incentive to reduce admissions and readmissions. Therefore, use of an additional measure involving admissions and readmissions would be duplicative and inappropriate. Finally, the measure has not been robustly tested for validity to demonstrate attribution of the measure to specific clinicians, groups, and specialties is clinically appropriate. We recommend that the issues related to risk adjustment, sociodemographic factors, and attribution be addressed and that the measure be endorsed by NQF prior to implementation in the MIPS program or the Shared Savings Program. NQF endorsement is critical to establishing that the measure has been appropriately tested, and is proven valid and reliable. In light of these concerns, the AAMC recommends that the MAP recommendation be “Do Not Support.” (Submitted by: Association of American Medical Colleges (AAMC))
  
  
• The American Medical Association (AMA) does not support inclusion of the Clinician and Clinician Group Risk-standardized Hospital Admission Rates for Patients with Multiple Chronic Conditions measure in the MIPS program. We do not believe that continuing to include measures based on administrative claims meets the intended goals of this program and while this measure may be useful at the community or population level, it is not appropriate to attribute this utilization to an individual physician or practices. This concern is due to several factors. Specifically, the lack of evidence to support applying this measure to individual physicians or practices is particularly concerning since what was provided by CMS during public comment of this measure demonstrated that improved care coordination and programs focused on care management can lead to reductions in hospital admissions but required the involvement of multiple partners such as a health system and/or hospital. We do not believe that sufficient evidence was provided to support the theory that physicians or practices, in the absence of some coordinated program or payment offset (e.g., care management fee), can implement structures or processes that can lead to improved outcomes for these patients. The AMA is also extremely concerned with the lack of alignment of attribution models within and across the existing claims-based measures in this program, as well as the lack of information on the specifications of this measure and its alignment with other measures and definitions in this program. The retrospective approach also prevents this measure from providing timely, meaningful and actionable data at the point of care. Additionally, the AMA believes that the measure must demonstrate a high level of reliability (0.80 at a minimum). We believe that the current approach lacks robust testing of the validity of the measure including demonstrating that the assignment of the measure to specific physicians, groups, and specialties is clinically appropriate and tied to their ability to meaningfully influence the outcome as well as empiric validity testing. This measure has not yet been specified and tested at the individual or group level and as a result, information is not available on the appropriateness of the attribution approach nor do we know the reliability and validity results from testing. As a result, the AMA recommends that the highest level of MAP recommendation be “Do Not Support”. (Submitted by: American Medical Association)
  
  
• The FAH does not support inclusion of this measure in the MIPS program as there is insufficient evidence to support attribution to individuals or groups, particularly with the attribution assigned retrospectively, the minimum sample size and reliability threshold remain too low, and additional information on the validity of the measure when applied at these levels is needed. The National Quality Forum should also endorse the measure prior to finalization. As a result, the FAH requests that the highest level of MAP recommendation be “Do Not Support”. (Submitted by: Federation of American Hospitals)
  
  
• The measure should be submitted for NQF endorsement. Endorsement review should consider the need for risk adjustment, including, but not limited to, socioeconomic status and potential shrinkage factor for physicians with small sample sizes. Review should also assess how the measure would be attributed to a specific physicians. Finally, we are concerned that this measure is duplicative of existing condition specific admission measures for ACOs and could result in ACOs being penalized twice for the same patients. (Submitted by: Premier)
  
  
• There are concerns with measure attribution at the individual level for this measure, as many unplanned readmissions are outside of the individual clinician’s control. This measure is primary-care based, and the attribution methodology holds physicians responsible for care they did not provide. Evidence has shown that reductions in admissions can be achieved but often require the involvement of multiple partners, however, clinicians and/or groups may not be in a position to participate in or influence the development of such coordination. This measure has not been tested at the individual or group level of analysis and as a result, information is not available on the appropriateness of the attribution approach. Reliability and validity results remain unknown. We encourage adequate risk adjustment and social risk factors are accounted for appropriately. As CMS is already aware, acute myocardial infarction, is, by definition, acute and therefore not chronic. Consideration for this as a substitute or marker for heart disease should be evaluated further. (Submitted by: American College of Cardiology)
  
  
• We have concerns about this measure and its assertion in the measure rationale that hospital admission rates are an effective marker of ambulatory care quality. We believe admission rates are more of a surrogate for what occurs in the ambulatory setting, which includes quality of ambulatory care in addition to other factors like social determinants of health. If it does not do so already, this measure should have some form of risk-adjustment or other method for accounting for sociodemographic factors that may impact admission rates. (Submitted by: Society of Hospital Medicine)
  
  
• We support the measure as is and appreciate the exclusion of patients in hospice (Submitted by: The Coalition to Transform Advanced Care)
  
  

• Concerns: attribution, risk adjustment (including socio-demographic factors). This measure in particular is pretty complex (it has far more detail than the others) and is unclear whether it has been tested by the developer for validity/reliability for reporting for clinician groups and ACOs The AAMC does not support the inclusion of the Clinician and Clinician Group Risk-standardized Hospital Admission Rates for Patients with Multiple Chronic Conditions measure in the MIPS program and Shared Savings Program for ACOs. The AAMC recognizes the need to improve the care of patients with multiple chronic conditions, but has significant concerns with the use of these measures in these programs. Attribution of this measure to individual clinicians or clinician groups is of great concern. We do not believe that sufficient evidence was provided to support the theory that physicians or practices, in the absence of some coordinated program or payment offset (e.g., care management fee), can implement structures or processes that can lead to improved outcomes for these patients. Furthermore, because the measure uses a retrospective approach, it is unable to provide timely and actionable feedback to the point of care. The characteristics of the attributed Medicare beneficiaries can vary widely by physician group practice. Not accounting for the clinical variation in the underlying population is extremely misleading and disproportionately affects the physicians who care for the most complex patients. These measures should have appropriate clinical risk adjustment prior to implementation in any program. In addition, as admissions and readmissions are often connected to the broader community, CMS should consider adding an adjustment or stratification to account for socio-demographic factors. The use of this measure at the ACO level also would have similar challenges with population size and risk adjustment. An ACO patient population is typically much smaller than 100,000. In addition, an ACO is already accountable for costs and has an incentive to reduce admissions and readmissions. Therefore, use of an additional measure involving admissions and readmissions would be duplicative and inappropriate. Finally, the measure has not been robustly tested for validity to demonstrate attribution of the measure to specific clinicians, groups, and specialties is clinically appropriate. We recommend that the issues related to risk adjustment, sociodemographic factors, and attribution be addressed and that the measure be endorsed by NQF prior to implementation in the MIPS program or the Shared Savings Program. NQF endorsement is critical to establishing that the measure has been appropriately tested, and is proven valid and reliable. In light of these concerns, the AAMC recommends that the MAP recommendation be “Do Not Support.” (Submitted by: Association of American Medical Colleges (AAMC))
  
  
• The American Medical Association (AMA) does not support inclusion of the Clinician and Clinician Group Risk-standardized Hospital Admission Rates for Patients with Multiple Chronic Conditions measure in the MSSP. This measure is based on hospitalizations that would be included in the overall spending assigned to the accountable care organization (ACO). An ACO will be penalized through a reduction in shared savings if it has a high rate of any of these admissions; therefore, it is unnecessary to include this as quality measures. The measures might be appropriate if there were reason to believe that ACOs would avoid addressing these areas but because they represent large patient populations for ACOs with known ways to reduce high rates of admissions and readmissions ACOs are unlikely to ignore these areas if there are opportunities to reduce them. Including them as quality measures could also force the ACO to focus on areas that do not represent the best opportunity to improve patient care and reduce spending. There are also existing condition specific admission measures in the ACO program and having a multiple chronic condition measure is duplicative and unnecessary. ACOs are penalized for the same patients twice due to the duplicative nature of the condition specific measures, which overlaps with the revised multiple chronic condition measures. If CMS insists on admission measures, then it is preferable to have condition specific measures because they are more actionable. In addition, the risk adjustment model for these measures also does not adequately address the ongoing concerns around socioeconomic factors (SES). As noted by the NQF All-Cause Admission and Readmission Committee and the developer (Yale/CMS), the analyses demonstrated that ACOs with higher numbers of low-SES patients performed worse than the national rate. These shifts in performance scores based on SES adjustment indicate that there may be other variables influencing admission rates that are outside of the ACO’s control. As a result, the AMA recommends that the highest level of MAP recommendation be “Do Not Support”. (Submitted by: American Medical Association)
  
  
• The Federation of American Hospitals (FAH) does not support inclusion of this measure in MSSP due to the lack of information on how this measure provides reliable and valid representations of accountable care organizations’ (ACOs) performance. The measure must produce a minimum reliability threshold of sufficient magnitude (e.g. 0.7 or higher) and represent valid assessments of quality at the ACO level. In addition, it is important that the measure yield variation in performance scores that would inform clinicians, practices, ACOs, CMS, and patients on the quality of care provided and demonstrates that it is capable of measuring and driving change toward meaningful improvements in patient care. As a result, the FAH requests that the highest level of MAP recommendation be “Do Not Support with Potential for Mitigation”. (Submitted by: Federation of American Hospitals)
  
  
• Regarding addition of MUC1937 - Clinician and Clinician Group Risk-standardized Hospital Admission Rates for Patients with Multiple Chronic Conditions; in the Medicare Shared Savings Program, the score would be at the ACO level. NAACOS notes duplication with the current measure ACO-38, Risk Standardized Acute Admission Rates for Patients with Multiple Chronic Conditions (NQF 2888). (Submitted by: National Association of Accountable Care Organizations)
  
  

• Given the current national opioids dependency crisis, many providers are under the mistaken presumption that they should avoid prescribing opioids, thus withholding otherwise appropriate therapies from patients. Our concern about measures calling for opioids use reporting is that such measures could drive behavior in an unintended way. We strongly urge that proper risk adjustment and benchmarking be conducted for MUC2019-057 to exclude patients treated legitimately for chronic pain and substance use disorders from the measure. (Submitted by: Cleveland Clinic)
  
  
• Thank you for the opportunity to submit comments on the inclusion of this opioid measure in the Part C & Part D Star Ratings. CAPC appreciates the steps that the measure stewards, the MAP, and other stakeholders are taking to address the serious and growing public health crisis caused by the inappropriate use of opioid analgesics. We understand the need to implement measures that will encourage guideline-adherent prescribing. That said, poorly managed pain in patients with serious illness can contribute to intense suffering, decreased productivity, poorer quality of life, increased health care utilization, and even increased mortality. We want to impress upon the MAP that there is a critical distinction between patients in serious, chronic pain, for whom taking opioids means that they can get out of bed, go to work, and spend time with loved ones vs. patients with substance use disorder. We are concerned that this distinction may be lost in measures that undermine clinicians’ ability to comprehensively assess each patient, weigh the benefits and burdens of opioids, and make appropriate treatment decisions. We recognize that this is an aggregate measure, and it contains exclusions for Medicare beneficiaries with a cancer diagnosis or that are enrolled in hospice. However, a significant number of patients who do not have cancer require (and do well on) high dose opioids. We are concerned that – particularly in the context of other policies intended to limit opioid prescribing – this measure could discourage the development and implementation of individualized care plans. Our experience has shown that it could have a chilling effect on prescribing and drive inappropriate forced tapers. Therefore, until there is a better way to identify patients who legitimately need opioids beyond just those with cancer, we are against the inclusion of this measure in CMS accountability programs. (Submitted by: Center to Advance Palliative Care)
  
  
• The singular focus on the dose and duration of opioid prescriptions with the two measures is counter to the important steps that the administration has taken to address the national epidemic of opioid-related overdose deaths, which the AMA strongly supports. The final report of the Department of Health and Human Services (HHS) Interagency Pain Management Best Practices Task Force, for example, made a compelling case for the need to focus on patients experiencing pain as individuals and to develop treatment plans that meet their individual needs, not employ one-size-fits-all approaches that assume 120 MME for 90 days is an indication of overuse. Likewise, a CDC publication in the New England Journal of Medicine, “No Shortcuts to Safer Opioid Prescribing,” expressed concern that its opioid prescribing guidelines have been misapplied and wrongly used to nonconsensually discontinue or reduce prescriptions for patients with pain, with some actions likely to result in harm to patients. Finally, the CMS Overutilization Monitoring System already employs a thoughtful patient-centered approach to potential opioid overuse, requiring that Medicare Part D plans consult with the individual patient’s prescribing physician(s) to understand and confirm the appropriateness of prescribed medications. CMS has also recently clarified in the 2020 MA Call letter that beneficiaries in hospice care, those receiving palliative or end of life care should be excluded from opioid policies. They also recommend that beneficiaries with sick cell disease be excluded from opioid limits based on the CDC guidelines for prescribing opioid for chronic pain. In light of these concerns, the AMA recommends that the highest level of MAP recommendation be “Do Not Support”. (Submitted by: American Medical Association)
  
  
• We support this measure and appreciate the recognition that people other than those with cancer may be on high opioid doses, such as those with sickle cell anemia, and that such appropriate use should not be discouraged. We also share a recent study of Florida PDMP showed higher dose ranges in general, so more beneficiaries may fall into this category than expected (https://www.pallimed.org/2019/11/mandated-queries-of-florida.html ) Finally, we appreciate the exclusion of those in hospice (Submitted by: The Coalition to Transform Advanced Care)
  
  
• While written comments were not provided, the commenter indicated their support for this measure in this program (Submitted by: American Urological Association)
  
  

• Thank you for the opportunity to submit comments on the inclusion of this opioid measure in the Part C & Part D Star Ratings. CAPC appreciates the steps that the measure stewards, the MAP, and other stakeholders are taking to address the serious and growing public health crisis caused by the inappropriate use of opioid analgesics. We understand the need to implement measures that will encourage guideline-adherent prescribing. That said, poorly managed pain in patients with serious illness can contribute to intense suffering, decreased productivity, poorer quality of life, increased health care utilization, and even increased mortality. We want to impress upon the MAP that there is a critical distinction between patients in serious, chronic pain, for whom taking opioids means that they can get out of bed, go to work, and spend time with loved ones vs. patients with substance use disorder. We are concerned that this distinction may be lost in measures that undermine clinicians’ ability to comprehensively assess each patient, weigh the benefits and burdens of opioids, and make appropriate treatment decisions. We recognize that this is an aggregate measure, and it contains exclusions for Medicare beneficiaries with a cancer diagnosis or that are enrolled in hospice. However, a significant number of patients who do not have cancer require (and do well on) high dose opioids. We are concerned that – particularly in the context of other policies intended to limit opioid prescribing – this measure could discourage the development and implementation of individualized care plans. Our experience has shown that it could have a chilling effect on prescribing and drive inappropriate forced tapers. Therefore, until there is a better way to identify patients who legitimately need opioids beyond just those with cancer, we are against the inclusion of this measure in CMS accountability programs. (Submitted by: Center to Advance Palliative Care)
  
  
• The singular focus on the dose and duration of opioid prescriptions with the two measures is counter to the important steps that the administration has taken to address the national epidemic of opioid-related overdose deaths, which the AMA strongly supports. The final report of the Department of Health and Human Services (HHS) Interagency Pain Management Best Practices Task Force, for example, made a compelling case for the need to focus on patients experiencing pain as individuals and to develop treatment plans that meet their individual needs, not employ one-size-fits-all approaches that assume 120 MME for 90 days is an indication of overuse. Likewise, a CDC publication in the New England Journal of Medicine, “No Shortcuts to Safer Opioid Prescribing,” expressed concern that its opioid prescribing guidelines have been misapplied and wrongly used to nonconsensually discontinue or reduce prescriptions for patients with pain, with some actions likely to result in harm to patients. Finally, the CMS Overutilization Monitoring System already employs a thoughtful patient-centered approach to potential opioid overuse, requiring that Medicare Part D plans consult with the individual patient’s prescribing physician(s) to understand and confirm the appropriateness of prescribed medications. CMS has also recently clarified in the 2020 MA Call letter that beneficiaries in hospice care, those receiving palliative or end of life care should be excluded from opioid policies. They also recommend that beneficiaries with sick cell disease be excluded from opioid limits based on the CDC guidelines for prescribing opioid for chronic pain. In light of these concerns, the AMA recommends that the highest level of MAP recommendation be “Do Not Support”. (Submitted by: American Medical Association)
  
  
• We are aware of a competing HEDIS measure, Use of Opioids from Multiple Providers and request the developer investigate harmonization. The measure is also similar to PQA MUC measure 19-61. (Submitted by: American Medical Association)
  
  
• We support this measure as well but note that same recent study found 76% of people obtaining opioids in Florida did so from more than one prescriber and those with fragmented care had more than this measure’s 4 prescribers (https://www.pallimed.org/2019/11/mandated-queries-of-florida.html ) This measure should not be used to discourage such appropriate use. We further appreciate the exclusion of those in hospice (Submitted by: The Coalition to Transform Advanced Care)
  
  
• While written comments were not provided, the commenter indicated their support for this measure in this program (Submitted by: American Urological Association)
  
  

• Thank you for the opportunity to submit comments on the inclusion of this opioid measure in the Part C & Part D Star Ratings. CAPC appreciates the steps that the measure stewards, the MAP, and other stakeholders are taking to address the serious and growing public health crisis caused by the inappropriate use of opioid analgesics. We understand the need to implement measures that will encourage guideline-adherent prescribing. That said, poorly managed pain in patients with serious illness can contribute to intense suffering, decreased productivity, poorer quality of life, increased health care utilization, and even increased mortality. We want to impress upon the MAP that there is a critical distinction between patients in serious, chronic pain, for whom taking opioids means that they can get out of bed, go to work, and spend time with loved ones vs. patients with substance use disorder. We are concerned that this distinction may be lost in measures that undermine clinicians’ ability to comprehensively assess each patient, weigh the benefits and burdens of opioids, and make appropriate treatment decisions. We recognize that this is an aggregate measure, and it contains exclusions for Medicare beneficiaries with a cancer diagnosis or that are enrolled in hospice. However, a significant number of patients who do not have cancer require (and do well on) high dose opioids. We are concerned that – particularly in the context of other policies intended to limit opioid prescribing – this measure could discourage the development and implementation of individualized care plans. Our experience has shown that it could have a chilling effect on prescribing and drive inappropriate forced tapers. Therefore, until there is a better way to identify patients who legitimately need opioids beyond just those with cancer, we are against the inclusion of this measure in CMS accountability programs. (Submitted by: Center to Advance Palliative Care)
  
  
• We are aware of a competing HEDIS measure, Use of Opioids from Multiple Providers and request the developer investigate harmonization. The measure is also similar to PQA MUC measure 19-60. (Submitted by: American Medical Association)
  
  
• We support this measure and appreciate the recognition that some without cancer may legitimately need opioids, such as in sickle cell anemia. However, we would note that some situations of long use/high doses and multiple prescribers may be appropriate and this measure should not be used to discourage such appropriate use. We appreciate the exclusion of those in hospice. (Submitted by: The Coalition to Transform Advanced Care)
  
  
• While written comments were not provided, the commenter indicated their support for this measure in this program (Submitted by: American Urological Association)
  
  

• The AAMC does not support inclusion of the Hemodialysis Vascular Access: Practitioner Level Long-term Catheter Rate measure in the MIPS program as the measure is currently specified. During public comment on the measure and its testing during NQF endorsement review in 2018, stakeholders identified concerns with the reliability and validity of the measure as specified and the belief that additional testing must be conducted to further improve the measure prior to implementation. We question whether the empiric validity of the measure has been sufficiently demonstrated since it compared the trends of catheter use to mortality but the results do not identify whether those trends were statistically significant. We recommend that the issues related to reliability and validity be addressed. In light of our concerns, the AAMC recommends that the highest level of MAP recommendation be “Do Not Support With Potential For Mitigation.” (Submitted by: Association of American Medical Colleges (AAMC))
  
  
• The American Medical Association (AMA) does not support inclusion of the Hemodialysis Vascular Access: Practitioner Level Long-term Catheter Rate measure in the MIPS program. During public comment on the measure and its testing, the AMA identified concerns with the reliability and validity of the measure as specified and believes that additional testing must be conducted to further improve the measure prior to implementation. Specifically, we note that the reliability results demonstrate that the measure inter-unit reliability was 0.602, which is lower than the facility level results of 0.765 that CMS provided to NQF in 2017. We believe that measures must meet minimum acceptable thresholds of 0.7 for reliability and additional work is needed to determine whether minimum sample sizes must be increased or other methods used to further improve the reliability of the measure score. We also question whether the empiric validity of the measure has been sufficiently demonstrated since the trends of catheter use to mortality was compared but the results do not identify whether those trends were statistically significant. As a result, the AMA recommends that the highest level of MAP recommendation be “Do Not Support With Potential For Mitigation”. (Submitted by: American Medical Association)
  
  
• The Federation of American Hospitals (FAH) believes that additional work is required prior to implementation of this measure in the MIPS program. Specifically, the reliability of the measures should achieve a minimum reliability threshold of sufficient magnitude (e.g. 0.7 or higher) and the testing that was released for public comment earlier this year demonstrated an inter-unit reliability of 0.602. In addition, the empiric validity testing performed did not yield results that were statistically significant when this measure was compared to mortality. As a result, the FAH requests that the highest level of MAP recommendation be “Do Not Support with Potential for Mitigation”. (Submitted by: Federation of American Hospitals)
  
  

Navigating the Discussion Guide

• How do I get back to the section I was just looking at?
  The easiest way is to use the back button on your browser. Other options are using your backspace button (which works for many browsers on laptops), or using the permanent links at the upper right hand corner of the discussion guide. But the back button is the best choice in most situations.
  
  
• Can I print the discussion guide out?
  You can, but we don't recommend it. Besides using a lot of paper (probably a couple hundred pages at least), you'll lose all the links that allow you to move around the document. For instance, if you're scrolling through the agenda and want to see more information about a particular measure, the electronic format will allow you to click a link, read more, and then bo back. If you're on paper, there will be a lot of flipping through paper.
  
  
• If I can't print this out, how can I read it on the plane?
  Although the Discussion Guide opens in a web browser, it does not require an internet connection if you have downloaded and saved the HTML file to your hard drive.
  
  
• How do I know that I'm looking at the most recent version?
  At the top left corner of the discussion guide is a version number. At the beginning of the in person meetings, the NQF staff will ask everyone to load the most recent discussion guide version and will check that everyone has the same version loaded.
  
  
• What electronic devices can I use to view the discussion guide?
  We tried to make this as universal as possible, so it should work on your laptop (PC, Mac, Linux), your tablet (iPad, Android), or your phone (iPhone, Android). It should also work on many types of browsers (IE, Firefox, Chrome, Safari, Opera, Dolphin,....). Please let us know if you have any problems, and we'll troubleshoot with you (and improve the discussion guide for the next go around).
  
  
• Why do I see weird characters in some places?
  Because we're joining data from many different sources, we do find some technical challenges. This generally shows up as strange characters--extra question marks, accented characters, or otherwise unusual items. We've been able to fix many of these problems, but not all. We ask that you bear with us as we improve this over time!
  
  

Content

• What is included in the discussion guide?
  There are four sections within this document:
  • Agenda, with summaries of each measure under consideration
  • Full information about each measure, including its specifications, preliminary analysis of how this measure can advance the program's goals, and the rationale by HHS for being included in the list
  • Summaries for each federal health program being considered
  • Public comments that have been received to date (Note that the discussion guide may be released before the public comment period is finished, in which case there will just be a placeholder for where comments will go)
  
  
• How are the meeting discussions organized?
  The meeting sessions are organized around consent calendars, which are groups of measures being considered for a particular program or groups of measures for a particular condition or topic area. For each measure being discussed, this document will show you the description, the public comments (if any), the summary of the preliminary analysis, and the result of the preliminary analysis algorithm.

Remote Participation Instructions:

• Direct your web browser to: the web link.
• 2. Enter your email address, first name, and last name in the appropriate fields and click “Submit.”

• All participants dial 1-800-768-2983 and enter the passcode 9676803 to access the audio platform.