SHM Submits Comments to CMS on Measures Under Consideration (MUC) for Reporting Programs
January 31, 2013
SHM's Policy Efforts
SHM supports legislation that affects hospital medicine and general healthcare, advocating for hospitalists and the patients they serve.
Marilyn Tavenner, Administrator
Centers for Medicare & Medicaid Services
7500 Security Boulevard
Baltimore, MD 21244
The Society of Hospital Medicine (SHM) welcomes the invitation to comment on the “List of Measures Under Consideration December 2013.” SHM represents the more than 40,000 hospitalists currently practicing in the US. Hospitalists provide care to more hospitalized patients, including Medicare beneficiaries, than any other specialty. Hospitalists have a distinctive role in facilitating both the individual physician-level and the hospital-level performance agendas. SHM has been active in educating and encouraging our members to participate in many of the 20 applicable programs for which these performance measures are being considered, as appropriate for hospital medicine practice.
SHM has a goal to broaden the performance measures used for performance improvement or accountability in the Medicare programs including the Hospital-Acquired Condition (HAC) Reduction, Hospital Inpatient Quality Reporting (IQR), Hospital Readmission Reduction, Hospital Value-Based Purchasing, Medicare Shared Savings Program, Medicare Physician Quality Reporting System (PQRS), Physician Compare, Physician Feedback/Quality and Resource Utilization Reports (QRUR), and the Physician Value-Based Modifier Programs. This will allow more robust participation by hospitalists, as the current measures do not always adequately represent the scope of our work.
Hospital-Acquired Condition (HAC) Reduction Program
SHM has specific comments regarding the following measures in the Hospital Acquired Condition Payment Reduction Program:
- MUC ID XDDLA, PSI 10: Postoperative Physiologic and Metabolic Derangement Rate and MUC ID E0533, PSI 11: Post-Operative Respiratory Failure refer to elective surgeries. As such, ‘elective’ should be included in the measure title.
- MUC ID E0349: PSI 16: Transfusion Reaction effectiveness rationale should be changed to, “This measure is intended to reduce the number of transfusion reactions on medical and surgical discharges,” instead of “…transfusion reactions after surgery.”
Hospital Inpatient Quality Reporting, Medicare Shared Savings, Medicare and Medicaid EHR Incentive Program for Hospitals and CAHs, Physician Compare, Physician Feedback/QRUR, Physician Value-Based Payment Modifier, Medicare Physician Quality Reporting System
- MUC ID XDBGA, Adverse Drug Events- Hypoglycemia should include an exclusion for patients with insulinoma in the denominator.
- MUC ID XDEEL, Hospital 30-day Risk-standardized Acute Myocardial Infarction (AMI) Mortality eMeasure should include an exclusion criterion in the denominator for patients on hospice or palliative care, unless the risk adjustment methodology already accounts for patients with noncardiac terminal illnesses such as advanced cancer.
Hospital Readmissions Reduction Program, Physician Feedback/QRUR, Physician Value-Based Payment Modifier
SHM appreciates the exclusion of planned readmissions from the MUC ID E1789, Hospital-Wide All-Cause Unplanned Readmission Measure (HWR). Exclusion criteria should be added to the denominator such that a planned hospital transfer is not included as a ‘readmission,’ or alternatively the definition of a planned admission should include intrafacility transfers.
Hospital Value-Based Purchasing Program
SHM supports the following stroke measures as they would be considered best practices in the care of stroke patients including, MUC ID E0434 ,STK-1 Venous Thromboembolism (VTE) Prophylaxis, MUC ID E0441, and MUC ID E0435, STK-2 Antithrombotic Therapy for Ischemic Stroke.
Further, SHM supports the following HVBP measures with specific comments including:
- MUC ID E0439, STK-6Discharged on Statin Medication. As written, it is unclear as to whether patients with sickle cell disease or other conditions that would predispose younger patients to stroke, who develop ischemic stroke, would equally benefit from a statin, and perhaps should be excluded.
- MUC ID D0440, STK-8 Stroke Education. SHM supports this important care transitions measure. However, if the numerator includes all patients discharged home, then the exclusion criteria should also include patients transferred to another facility or discharged to a SNF, as they might not receive all of these details in their discharge summary.
- MUC ID D0376 VTE-6: Incidence of Potentially Preventable VTE. SHM supports this patient safety measure so that institutions continue to emphasize the critical importance of identifying and initiating VTE prophylaxis in appropriate medical and surgical patients.
SHM does not support the following measures for inclusion in the Hospital Value-Based Purchasing Program:
- MUC ID E0371, VTE-1: Venous Thromboembolism Prophylaxis. Historically, SHM has supported this as an important process measure; however, the momentum behind this movement has slowed down quite a bit. New data suggests that perhaps not everyone benefits from VTE prophylaxis, and in fact, many “low-risk” patients may be harmed. SHM does not endorse this measure in its current form, until more definitive data about appropriate patient selection is available, or it can be denoted that “high risk” patients are in the denominator (rather than “all”). There should also be clarification about whether VTE prophylaxis refers specifically to pharmacologic VTE prophylaxis or is also inclusive of mechanical compression devices (although the evidence for this on medical patients is not as well-established as it is on surgical patients). This measure is currently under category “E”—endorsed by NQF, but it might be considered for the “D” category, if this is strictly about pharmacologic VTE prophylaxis.
- MUC ID E0373 VTE-3: VTE Patients with Anticoagulation Overlap Therapy. This measure has become somewhat outdated due to the prevalence of newer anticoagulants, many of which have an onset of action in just 24 hours and do not require the 5 days overlap. While Warfarin is still most commonly used, the environment is changing rapidly and separating out which patients are on Warfarin versus a newer agent would be problematic. The specifications of this measure should be changed to reflect that it is applicable to Warfarin only.
- MUC ID D0374 VTE-4: Patients Receiving Un-Fractionated Heparin with Doses/Labs Monitored by Protocol. Much like the measure above, this would have been a good measure a decade or so ago. Very few patients with DVT/PE get managed with IV Heparin. In fact, most hospitals have gone to pharmacist protocol monitoring, so it would be collecting data on a very few outliers. Also, tracking the performance would be difficult as IV Heparin is primarily used as bridge therapy now, which may be used only for 6-24 hrs. and in that short time frame the protocol monitoring is less important.
Medicare and Medicaid EHR Incentive Program for Hospitals and CAHs
SHM supports MUC ID E0500: Severe Sepsis/Septic Shock: Management Bundle, and would recommend that this measure would also be appropriate for other incentive programs such as Hospital Inpatient Quality Reporting and Hospital Value-Based Purchasing.
Medicare Shared Savings Program
SHM supports the following measures for inclusion in the Medicare Shared Savings Program:
- MUC ID E0543: Adherence to Statin Therapy for Individuals with Coronary Artery Disease. SHM supports this measure to affect the consistent use of statin therapy in appropriate patients with coronary artery disease.
- MUC ID E0576: Follow-up after Hospitalization for a Mental Illness. SHM supports the timely mental health follow-up within 30 days after an acute hospitalization with a principal mental health diagnosis.
- MUC ID E0053: Osteoporosis Management in Women who had a Fracture. This measure highlights the evidence-based need to screen and treat women with acute fractures for osteoporosis. Of note, the exclusion criteria should include patients who have been unable to tolerate or have other contraindications to the FDA-approved medications: hormone replacement (risk of DVT, estrogen-sensitive cancer history), bisphosphonates (esophagitis, renal failure).
SHM has the following comments on measures being proposed for the Medicare Shared Savings, Physician Compare, Physician Feedback/QRUR, Physician Value-Based Payment Modifier, and Medicare Physician Quality Reporting System Programs:
- MUC ID XDFDA, Appropriate In Vitro Susceptibility Testing. Exclusion criteria should include patients who leave the ED or hospital against medical advice, prior to results of susceptibility testing.
- MUC ID XDFHL, Appropriate Treatment of MSSA. Exclusion criteria should include patients who left the hospital against medical advice, before results of susceptibility testing were available. Similarly, exclusion criteria should include patients discharged with a plan to follow up susceptibility results in the outpatient setting who do not return for the follow up visit.
- MUC ID XDFCM, Minimum Antimicrobial Therapy for Staph A. Exclusion criteria should include patients who leave the hospital against medical advice, as well as patients who refuse appropriate therapy.
- MUC ID E0465, Perioperative Anti-Platelet Therapy for Patients Undergoing Carotid Endarterectomy. SHM supports this measure. Exclusion criteria should include patient refusal.
- MUC ID XDFCC (not yet submitted to NQF for endorsement), Use of Pre-Medication before Contrast-Induced Imaging Studies Inpatient with Documented Contrast Allergy. SHM supports this measure, as it is a patient safety concern and therefore awareness of a patient's allergies is appropriate.
Physician Feedback/QRUR, Physician Value-Based Payment Modifier
The following three measures are draft measures comparing the resource use for similar clinical situations in order to drive towards achieving best outcomes at lowest cost. CMS has not yet solidified the length of time of tracking, although they indicate that “…most acute conditions will likely have an episode length of 30 days, and most chronic conditions will likely have a length of a year…”
- MUC ID XDECL: Diabetes Condition Episode for CMS Episode grouper
- MUC ID XDEBL: Heart Failure Condition Episode for CMS Episode Grouper
- MUC ID XDEDA: Ischemic Cerebral Artery Disease Condition Episode for CMS Episode Grouper
Hospitalists will not have direct control over all the costs and resource use for these patients, since subspecialists will influence resource use (perhaps to a greater degree). At the same time, it is not unreasonable to ask that hospitalists be engaged in evaluating appropriate resource use for this group of patients while achieving good outcomes. We appreciate that CMS is trying to decrease variability in care for specific diagnoses; tracking resource use and comparing groups may be one way to achieve this. However, SHM has concerns that if CMS is tracking total dollars spent, this would not accurately capture whether one group achieved lower costs because of less actual resources used (e.g. less inappropriate imaging used) versus achieved lower costs because ancillary resources are less expensive in a particular part of the country or region (e.g. lower cost MRI's of head, for example). Thus, we are not certain that CMS will achieve the goal of appropriate resource use and better outcomes for patients.
For draft measures, MUC ID XDDMG/ Draft: Ischemic Heart Disease Condition Episode for CMS Episode Grouper, MUC ID XDECD/ Draft: Pneumonia Condition Episode for CMS Episode Grouper, and MUC ID XDECE/ Draft: Respiratory Failure Condition Episode for CMS Episode Grouper, the language remains vague, and SHM would request the opportunity to provide more specific input, including selecting the data inputs that would allow proper provider comparison, as these measures are operationalized. This may be an opportunity to collaborate with other specialty organizations. One obvious question is whether ICD-9 or ICD-10 usage would affect the program design.
In general, SHM concurs that it would need more information on how these draft measures would be structured in order to provide more specific feedback as relates to hospital medicine.
SHM supports the MUC ID 2158, Medicare Spending Per Beneficiary measure for hospital level performance monitoring, but we feel the measure needs further development before it can reasonably be applied to individual providers or even groups of physicians. As this measure is slated to be included in the 2016 Physician Value-Based Payment Modifier, SHM believes that these concerns need to be addressed now. Hospitalists coordinate care for medically complex patients in the inpatient setting; however it is not clear how costs would be attributed to the various providers on the care team who drive them. Patients consult with varying subspecialists throughout the hospital stay, all of whom impact decisions about resource utilization. While hospitalists are in a unique position to manage appropriate use of such resources, as of yet there is not a legitimate way to distinguish hospitalists from outpatient providers within the same specialty (e.g. internal medicine, family medicine, or pediatrics). This will interfere with accurate performance measurement and confuse the data. We do feel that with further development, this measure may serve as a unique means of measuring the efficiency of inpatient care.
Performance Reporting Alignment between Clinicians and Facilities
In the 2014 Physician Fee Schedule Proposed Rule, CMS examined options for individual clinician and group reporting of hospital-based measures from the IQR program. This proposed flexibility would represent a significant harmonization within the healthcare system, while allowing hospitalists to report measures that better fit their practice patterns and structures. The proposed rule included two different options: retooling measures from the IQR to be reportable by individual physicians and direct performance alignment with that of an associated hospital. SHM would like to reiterate its support for including both of these options as meaningful additions to physician quality reporting programs. Each alternative seems to address, in different ways, the questions of measure attribution and alignment between physicians and facilities, while broadening the number and type of measures that may be reported.
SHM greatly appreciates the opportunity to provide comment and feedback on the HHS List of Measures Under Consideration for future reporting years. We believe that hospitalists have a unique and important perspective on both the physician-level and hospital-level performance agendas. We welcome the opportunity to work with NQF and CMS around the important performance measure issues of attribution and measure specifications.
Eric Howell, MD