Skip to content

Welcome to Society of Hospital Medicine

Your Browser Is No Longer Supported. Please upgrade your browser from Internet Explorer 10 to Internet Explorer 11 or higher

SHM Submits Comments to CMS on the 2020 Physician Fee Schedule Proposed Rule

September 20, 2019

SHM's Policy Efforts

SHM supports legislation that affects hospital medicine and general healthcare, advocating for hospitalists and the patients they serve.

Learn More

Download Letter

Centers for Medicare and Medicaid Services

Department of Health and Human Services

Attn: CMS-1715-P

P.O. Box 8016

Baltimore, MD 21244-8016


Dear Administrator Verma,

The Society of Hospital Medicine (SHM), on behalf of the nation’s hospitalists, is pleased to offer comments on the proposed rule entitled: CY 2020 Revisions to Payment Policies under the Physician Fee Schedule and Other Changes to Part B Payment Policies; Medicare Shared Savings Program Requirements; Medicaid Promoting Interoperability Program Requirements for Eligible Professionals; Establishment of an Ambulance Data Collection System; Updates to the Quality Payment Program; Medicare Enrollment of Opioid Treatment Programs and Enhancements to Provider Enrollment Regulations Concerning Improper Prescribing and Patient Harm; and Amendments to Physician Self-Referral Law Advisory Opinion Regulations (CMS-1715-P).

Hospitalists are clinicians whose professional focus is the general medical care of hospitalized patients. Hospitalists are front-line healthcare providers in America’s hospitals for millions of patients each year, many of whom are Medicare and Medicaid beneficiaries. They manage inpatient clinical care of their patients, while working to enhance the performance of their hospitals and health systems.

Most hospitalists are Board Certified in Internal Medicine or Family Medicine but practice exclusively in the hospital setting. The position of hospitalists within the healthcare system affords them a distinctive role in facilitating both the individual physician-level and hospital-level performance agendas. The rule was reviewed by SHM’s Performance Measurement and Reporting Committee, a group consisting of practicing hospitalists and hospitalist leaders who are experts in measurement and assessment, and the Public Policy Committee, a group of hospitalist experts in healthcare policy. This diversity of perspectives informs our comments on this proposed rule.

We offer the following comments on the proposals: 

Physician Supervision for Physician Assistant (PA) Services

We appreciate CMS’ continued efforts to streamline reporting and reduce unnecessary administrative burdens. CMS proposes to revise its requirements for physician supervision of physician assistant (PA) services to be met when a PA furnishes care in accordance with their state’s laws and scope of practice rules. In the absence of state laws, the physician supervision requirement would be evidenced by documentation in the medical record. The proposal aligns PA requirements with previously existing Medicare regulations for nurse practitioner (NP) and clinical nurse specialist (CNS) services. PAs are integral members of the hospital medicine team and having consistent supervision requirements streamlines the management of team-based care in the hospital. We welcome this proposal and encourage CMS to finalize it. 

Review and Verification of Medical Record Documentation

 CMS proposes to allow the billing physician, PA or APRN to review and verify, rather than re-document, information in the medical record written by physicians, residents, nurses, students and other members of the medical team. The supervising physician will no longer need to redocument notes made by other members of the medical team; instead they will only need to sign and date these notes in the record. This change will reduce unnecessary record duplication and reduce documentation burden. We support finalizing these changes to the regulations guiding documentation.

Comment Solicitation on Opportunities for Bundled Payments under the PFS

Bundled payments offer the opportunity to move farther away from the fee-for-service (FFS) payment system by easing the transition towards APMs. Hospital medicine is one of the few specialties that has heavily engaged with the Bundled Payments for Care Improvement (BPCI) initiative and the BPCI-Advanced Alternative Payment Model (APM) through the Center for Medicare and Medicaid Innovation. Our members have seen firsthand how bundling payments can result in care improvement and reductions in healthcare spending. Appropriately structured bundled payments may better incentivize cost effective, high quality care.

As CMS considers how and whether to proceed, we encourage it to keep in mind:

  • Opportunities for aligning the bundles or payment system with quality measures and reporting programs.
  • Continuing the “Patients Over Paperwork” initiative by reducing or eliminating reporting and documentation requirements in exchange for accepting bundled payments.
  • Finding novel ways for bundled payments to account for care previously un- or under-compensated in the PFS—such as cognitive medicine, care coordination and transitions management.
  • The challenges in scoping and attributing episode-based cost measures, particularly for conditions that have extremely heterogeneous patient populations and highly variable costs.

SHM welcomes more detailed discussions about bundling payments under the PFS. We stand ready to work with CMS on updating the payment system to address the current and future needs of the country.

CY 2020 Updates to the Quality Payment Program

Transforming MIPS: MIPS Value Pathways (MVPs) Request for Information

We share CMS’ desire to streamline and reduce overly burdensome reporting within the Merit-based Incentive Payment System (MIPS) program and are excited to collaborate alongside CMS to improve the program. We also want increased participation in Advanced Alternative Payment Models (APMs). However, the MIPS Value Pathways (MVPs) concept is a significant change from the existing program structure. Considering the scope of this proposed rule, we believe this far-reaching proposal should be presented as a separate Request for Information (RFI) to afford us and other stakeholders time to adequately analyze and respond to these changes. We offer preliminary comments regarding the MVPs below.

We do not support mandatory participation in MVPs in the 2021 Performance Year. This proposal outlines major changes to the MIPS. We are concerned that rushing forward with a requirement for mandatory participation will cause widespread confusion and hinder the overall success of the program. Instead, we suggest voluntary participation to ease the transition to the MVPs and to enable providers to select what makes sense for their practice. We also believe that CMS will need additional rulemaking cycles to develop the concept and work with specialties to identify and create appropriate measures for their MVPs.

MVPs should be flexible. Hospitalists have an incredibly heterogeneous practice, treating and managing patients with a range of diseases, injuries and comorbidities. We believe it will be difficult for CMS to develop MVPs for each specialty, even with the input from specialty societies, that will be relevant and meaningful to those providers. As contemplated in the rule, CMS could automatically enroll clinicians in the most appropriate MVP based on claims. At a minimum, providers should be able to voluntarily change their MVP if they feel they have been assigned to inappropriate MVP. Maximally, providers should be able to choose the MVP that best fits their practice. This helps account for the differences in the day-to-day care delivered in a highly urban hospital versus a critical access hospital, for example. Flexibility will provide clinicians with an increased sense of autonomy, which will likely increase satisfaction within and buy-in for program.

We are concerned that the MVPs, as currently imagined, will create a situation in which many providers will not have enough meaningful measures to report. We agree with CMS’ desire to eliminate unnecessary, overly burdensome, and irrelevant measures; however, we are concerned that eliminating too many measures will prevent many specialties from meaningfully participating in the program. Hospitalists already face a dearth of reportable measures in the MIPS; many are restricted to just three or four measures they can reliably report each year. While we are excited about the potential for partnering with CMS to create relevant and meaningful measures, launching the MVPs now, without new measures in place, or even in the pipeline would be premature. MVPs will not be successful without enough relevant measures for each specialty.

SHM supports increased attention to population health as we believe this is the next generation of measurement and healthcare improvement. Population health-level measures are typically high-level and measure the effects of systems and providers of many specialties on patient and community health outcomes. We encourage CMS to consider how to balance incorporating these important measures with the realities of a clinician-level pay-for-performance program like MIPS that affects individual providers’ payments. Larger structural changes to the MIPS, outside of the MVP concept, may be necessary to realize the true value of population health-focused measures.

We also want to ensure that hospital-based exclusions from the Promoting Interoperability category are included within the MVP framework, as the failure to do so would be incongruous with current rules and the realities of hospital-based practice. CMS has excluded hospital-based and non-patient facing groups from the Promoting Interoperability category due to these providers using their hospital’s EHR systems which are not set up for eligible provider (EP) Promoting Interoperability measures. Within the MVPs, hospital-based providers must continue to be exempt from the Promoting Interoperability category.

We are generally supportive of CMS’ intention to work with specialty societies to develop MVPs relevant to their specialties. However, we are concerned that MVPs, as currently conceived, continue the tendency to silo specialties into separate and distinct assessable pools. This contributes to the fractured nature of the healthcare system and undermines team-based approaches to care. Many hospitalists practice in multispecialty groups that serve a wide range of patients’ needs throughout different settings. By focusing on developing specialty-specific MVPs, CMS is missing an opportunity to promote team-based and cross-specialty care, which are critical elements to improving the quality and efficiency of the healthcare system.

We are supportive of improving and streamlining the MIPS, and we are excited to work alongside CMS to improve the MIPS and reduce clinician burden. However, the MVPs proposal contemplates extensive changes from the existing program, which is just now entering full operation. While we recognize the desire to move forward with this proposal to improve the MIPS, we ultimately believe creating a separate RFI will allow for more concrete and specific feedback. While creating a separate RFI may delay the implementation of the program, allowing for more concrete feedback will hopefully lead to a better, more-widely accepted reporting pathway. We do not support moving forward with this proposal before issuing a separate RFI and without further detailed work on the concept.

Quality Performance Category

Data Completeness Criteria

CMS proposes to increase the data completeness criteria for reporting quality measures in the Quality category from 60% to 70%. While we agree with CMS’ assertion that higher data completeness may yield a more accurate assessment of clinician performance in quality measures, it is not appropriate to increase the data completeness threshold at this time.

In Table 35 in the rule, CMS cites the data completeness rates for individuals, groups and small practices in the CY 2017 reporting period; the rates were 76.14%, 85.27% and 74.76%, respectively. This data comes from the first year of the program which had special “Pick Your Pace” scoring as part of an effort to encourage broad MIPS participation across a wide array of medical practices. Pick Your Pace required groups to submit at a minimum data on a single patient for a single measure in any MIPS category to avoid a penalty in the 2019 Payment Year. Because of this flexibility, providers were more likely to report on a limited number of quality measures and measures with which they are familiar and comfortable. Therefore, the data presented in Table 35 does not necessarily reflect the true data completeness rates for quality measure reporting at full participation. Because of its unique scoring structure, data from Pick Your Pace should not be used in isolation to guide policy changes. Without more detailed analysis, we believe it would be premature to increase the data completeness threshold.

We also have concerns about the ability of groups to meet data completeness criteria when reporting on TINs that practice in multiple facilities. It is common for hospitalist groups to provide coverage in multiple hospitals, meaning data submitted by these groups must come from multiple sources. As a result, the hospitalists submitting data may be working with different processes, IT infrastructure systems, and hospital administrations in order to capture their data. Without additional detail, the analysis of 2017 reporting rates supplied in the proposed rule do not provide the level of granularity to support increasing the data completeness threshold.

Topped Out Measures

CMS seeks comment on how to manage measure removal for specialties with limited options in quality measures. They ask whether they should increase the data completeness threshold for extremely topped out quality measures in instances where there are limited quality measures available for a specialty. We support this approach to balance continual quality improvement, ensuring specialties have relevant measures to report, and create space for new measure development.

Hospital medicine has a very limited supply of relevant MIPS measures. CMS’ proposed hospitalist specialty set, if finalized, would have only 5 measures for hospitalists to report. Our regular review of available quality measures indicates that these five measures are generally the only measures available for hospitalists and that several of these measures are at or near topped-out status. By continuing to remove topped out measures, hospitalists will go from having very few measures to report to having no reportable measures. Therefore, we support increasing the data completeness threshold on extremely topped out measures as a strategy to keep measures and specialties in the program.

We remind CMS that quality measure selection can be determinative and reflective of resource prioritization in practices, particularly in resource-scarce environments. Certain metrics may be programmatically topped out, but clinically still require resources and attention to ensure patient care is improving. By removing these topped out measures, CMS may be inadvertently deprioritizing these clinical areas, which may cascade into resource redeployment into other clinical quality areas. We encourage CMS to identify high priority clinical areas and, when available relevant measures become topped out, prioritize measure development in that clinical area.

Because topped-out measures will continue to be an issue for many specialties, CMS should also consider developing policies to address scoring the scarcity of non-topped-out quality measures for specialties. Providers and groups should be able to report on fewer than six measures in the Quality category and still be able to achieve maximum points. Some specialties, including hospital medicine, have had fewer than the six required measures in the Quality category for the duration of the MIPS program. This reality in the program is unlikely to change soon and CMS should adjust its policies accordingly.

Cost Performance Category

Weight in the Final Score

CMS proposes to use their statutory flexibility authorized in the Bipartisan Budget Act of 2018 to slow the increase of the Cost category weight through 2024. For the 2022 Payment Year, CMS proposes to increase the weight of the Cost category from 15% to 20% of the overall MIPS score. We support the gradual and predictable increase of the Cost category, as it will allow for providers to account for and adjust to the expansion of cost measures accordingly. We understand that CMS is statutorily required to increase the cost category to 30% by 2024, and we believe this gradual transition will allow providers to prepare for the increased weighting of the category.

Episode-Based Measures for the 2020 and Future Performance Periods

MACRA requires CMS to develop and implement episode-based cost measures for inclusion in the Cost category of the MIPS. For the 2020 performance year, CMS proposes to add 10 new episode-based cost measures. Broadly, we support the development and implementation of episode-based cost measures and have been active participants in the Waves I, II, and III clinical subcommittees to develop these measures. Because they are targeted to specific clinical conditions or procedures, episode-based cost measures assess a narrower set of overall provider costs. As such, they may yield more actionable and meaningful information for providers. We also believe that, as providers who influence patient care to some degree in nearly all these measures, cost measures have the potential to encourage better communication and alignment between attributed clinicians and those involved in caring for the patient, but not attributed to the measure.

Two of the proposed measures (Lower Gastrointestinal Hemorrhage and Inpatient Chronic Obstructive Pulmonary Disease (COPD) Exacerbation) are directly relevant and attributable to hospitalists. We encourage CMS to monitor for unintended consequences or issues associated with the attribution methodology and episode windows on these measures. CMS should continue to seek feedback from frontline providers on these episode-based cost measures and address any problems that may arise as these measures are implemented.


CMS uses a reliability threshold of 0.4 for determining how to implement cost measures in the program. Indeed, CMS made proposals to restrict certain episode-based cost measures like Lower Gastrointestinal Hemorrhage to group reporting only as the individual level did not meet this 0.4 reliability threshold. We are concerned about such a low reliability threshold for all measures and ask CMS for more clarification about how and why they use such a low threshold. We would like to see cost measures implemented only when they meet higher reliability thresholds. Providers must be confident in the measures that are being used to adjust their payments; measures with low reliability thresholds call into question the validity of the attribution and the measure specifications. 

Request for Comments on Future Potential Episode-Based Measure for Mental Health

CMS asks for additional feedback on a previously developed episode-based cost measure for psychoses/related conditions. AS CMS notes, the National Quality Forum (NQF) Measures Application Partnership (MAP) voted “Do not support for rulemaking” for several reasons. We appreciate that CMS followed this recommendation and did not implement the Psychoses/Related Conditions cost measure without further consideration.

Although hospitalists would not typically be the primary providers for patients assigned the Psychosis MS-DRG, they are involved in the care of patients with psychosis and related conditions. In some hospitals, however, there may not be psychiatric personnel, leaving hospitalists as the most likely attributable clinicians. We believe the measure should control for these situations by limiting the measure to inpatient psychiatric facilities and certain specialty codes, focusing on those that are key players in the treatment plan such as 13 (neurology), 26 (psychiatry), 27 (geriatric psychiatry), 50 (nurse practitioner), 68 (clinical psychologist), 79 (addiction medicine), 80 (licensed clinical social worker), 86 (neuropsychiatry),  and 97 (physician assistant). Absent these changes, we believe the 90-day post trigger window, while appropriate for attributed clinicians who see patients in both the inpatient and outpatient settings, would not be appropriate for attributed hospitalists and should be adjusted. 

Improvement Activities Performance Category

Group Reporting

CMS proposes to increase the group reporting threshold beginning with the 2020 performance year from at least one clinician attesting to activities to at least 50 percent of the group and that at least 50 percent of a group’s NPIs must perform the same Improvement Activity (IA) for the same continuous 90-day period. SHM is opposed to these proposals.

Both proposals taken together would place serious burdens on group reporting of IAs, particularly for multispecialty groups. A single IA may not be relevant to providers in the group with different clinical functions, whether it be different specialties, different practice roles to care for their patients or different practice locations altogether. Therefore, the requirement for 50 percent of NPIs to report the same activity for the same 90-day performance period would be onerous to achieve. We also note that it is not uncommon for hospitalist groups to assign providers to specific functions within the practice, even within single-specialty groups. For example, a group may have a dedicated clinician(s) for discharge or admissions. These individual functions or roles may be most ripe for IAs that impact the quality of care and transitions of care for all the group’s patients, despite being an activity performed by a minority of providers in the group. Furthermore, by requiring 50 percent of a group to report IAs, CMS may essentially be mandating the group to report on a significantly higher number of IAs than required in the category, increasing the burden of the category for these groups. While we support CMS’ conceptual aim of increasing the impact of IAs on patient care, the proposals in this rule would add significant burden to group reporting in the IA category. 

Promoting Interoperability: Hospital-Based MIPS Eligible Clinicians in Groups

CMS proposes to redefine under § 414.1305 hospital-based groups and virtual groups as those with more than 75% of individual NPIs classified as hospital-based MIPS eligible clinicians. The prior definition of hospital-based group required 100% of individual NPIs to be classified as hospital-based for the group to be exempt from Promoting Interoperability. This change will ensure that hospital-based reporting groups, which utilize their hospitals’ certified electronic health records (EHRs), are excluded from the Promoting Interoperability category.

Hospitalist groups, despite most of their MIPS eligible clinicians qualifying as hospital-based, lost points in the first two years of the MIPS as a result of not achieving the prior 100% threshold for exemption from Promoting Interoperability. In some cases, a single provider in the group, such as a locum tenens provider, resulted in the group unwittingly being held accountable for Promoting Interoperability. The proposed policy aligns with hospital-based practice realities and creates space for the flexibility in staffing that is common in hospital medicine. We note that this change also aligns the hospital-based group designation with the facility-based reporting and non-patient-facing group exemption pathways, minimizing confusion and unfair penalties.

SHM, joined by the Infectious Diseases Society of America, the American College of Emergency Physicians, the American Society of Anesthesiologists and the Emergency Department Practice Management Association, wrote about the issues faced by hospital-based groups under the previous strict definition in a letter to CMS on January 30, 2019. For all our specialties, CMS policy was out of alignment with the unique structure of hospital-based practice and yielded a structural disadvantage for groups reporting as a group in the MIPS. We greatly appreciate CMS’ engagement with us on this issue and strongly support finalization of the proposed new policy.

MIPS Payment Adjustments

Establishing the Performance Threshold

CMS proposes to use flexibility authorized by Congress in the Bipartisan Budget Act of 2018 to continue phasing in the MIP performance threshold. In MIPS Payment Years 1, 2, and 3, CMS set the performance threshold lower than the mean or median of overall MIPS performance to create an on-ramp to MIPS reporting; the performance thresholds were set at 3 points, 15 points, and 30 points, respectively. For Payment Year 4 (2022), CMS proposes to increase the performance threshold to 45 points. CMS indicated that, according to their analysis of Performance Year 1 data (2017), the mean final score for the program is 74.01 points. As such, they hope to continue building the performance threshold up to the mean or median by 2024, as required by the statute.

We support the proposed stepwise increase of the performance threshold, as this approach allows providers to adapt to program changes. A significant and sudden increase of the performance will likely disincentive small group practitioners and individual clinicians from adhering to the MIPS, as program adherence would be difficult and costly. Sudden increases to the program would disproportionately benefit large entities that have the funds and resources to adhere to the MIPS, whereas a stepwise increase allows smaller groups and individual providers to prepare for the program changes.

We cautiously support using the mean for the performance threshold beginning in 2024. We note that the data informing the performance threshold may be skewed towards larger and better resourced practices. Early adopters of full MIPS reporting were those that had the resources to respond nimbly to the new regulations, driving up the mean and median final scores in the first year of the program. We believe CMS should continue to explore what types of practices are successful and not successful in the MIPS to determine whether the scoring policy inadvertently disadvantages smaller (but not small) groups and individual clinicians. If evidence of disparities in scoring exist, we encourage CMS to consider how to address these differences either in the performance threshold or other scoring methodologies.

Public Reporting on Physician Compare

CMS proposes to post aggregate MIPS data on Physician Compare, beginning with Year 2 (CY 2018) data as soon as technically feasible. We broadly support public reporting to give patients and families more information when making decisions about their care, but caution that this approach may not be relevant for certain specialties like hospital medicine. Patients do not generally choose their hospitalist and/or choose which hospital to go to, based on the presence of a single hospitalist.

We support CMS’ proposal to have an indicator for facility-based clinicians and a link to the relevant Hospital Compare page(s). CMS must also include ample explanatory information about what facility-based means. Facility-level measures assess the care provided by the entire institution and associated providers, meaning a single provider does not have full control over performance on those measures.

CMS asks for feedback on adding patient narratives and a “value indicator” to the Physician Compare website in the future to increase the usability and relevance for patients and caregivers using the site to make decisions about their healthcare needs. We recognize the importance of providing patients with transparent, easy to understand information about their providers, and recognize that patient testimonials and narratives can be a useful tool to increase patient satisfaction. Additionally, published narratives and an easy to understand value indicator may help increase a patients’ sense of agency. While we see the value in providing patients with narrative testimonials, we are concerned that attributing a single narrative to a MIPS eligible clinician can provide an incomplete, and potentially inaccurate, depiction of care provided by a single provider or a group. Additionally, attributing these narratives, particularly to a single clinician, will be difficult in instances when multiple clinicians care for a single patient throughout the course of hospitalization.

We also require more information about how CMS plans to establish a single value indicator for the QPP. As informed by issues with the Medicare hospital Star Rating system, we believe CMS must approach developing a value indicator for individual clinicians cautiously. If patients will be making provider and caregiver determinations based off a single value indicator, we will need more information about the design of the metric and encourage CMS to engage with stakeholders more extensively on the concept.


SHM appreciates the opportunity to provide comments on the 2020 Physician Fee Schedule proposed rule. If you have any questions or need more information, please contact Josh Boswell, Director of Government Relations, at or 267-702-2632.


Chris Frost, MD, SFHM

President, Society of Hospital Medicine