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 on CY 2015 Physician Fee Schedule Proposed Rule

September 02, 2014

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


Marilyn Tavenner, Administrator
Centers for Medicare & Medicaid Services
Department of Health and Human Services
P.O. Box 8013 Baltimore, MD 21244-8013

Attention: CMS-1612-P Medicare Program; Revisions to Payment Policies under the Physician Fee Schedule, Clinical Laboratory Fee Schedule, Access to Identifiable Data for the Center for Medicare and Medicaid Innovation Models & Other Revisions to Part B for CY 2015

Dear Administrator Tavenner:

The Society of Hospital Medicine (SHM), representing the nation’s nearly 44,000 hospitalists, is pleased to submit the following comments on the CY 2015 Physician Fee Schedule proposed rule. Hospitalists are experts in primary care for hospitalized patients. In that role, they provide a significant amount of care to Medicare and Medicaid beneficiaries, ensuring safe and efficient delivery of that care during hospital stays and transitions into and out of the hospital. The unique position of hospitalists within the healthcare system affords a distinctive role in facilitating both individual physician-level and systems/hospital-level performance agendas.

This year’s proposed updates to the Physician Quality Reporting System (PQRS) and the physician value-based payment modifier (VBPM) are particularly pertinent for hospitalists. Many of the proposed changes add complexity to these programs, further illuminating the difficulty that hospitalists face in participating. We appreciate that CMS has acknowledged this dynamic in its discussion of the treatment of hospital-based physicians.

SHM offers the following comments on the proposals contained in the CY 2015 Physician Fee Schedule proposed rule.

Reports of Payments or Other Transfers of Value to Covered Recipients

SHM strongly disagrees with the CMS proposal to delete in its entirety the “Continuing Education Exclusion” found in 42 CFR 403.904(g). CMS suggests that the deletion of Section 403.904(g) would remove an unintended redundancy from the final rule and expand the range of educational events that are appropriately exempt from reporting. We believe the deletion of the section would do neither and risks creating additional confusion at a time when industry and physicians are still adjusting to the new open payments requirements.

CMS proposes to delete 42 CFR 403.904(g), in part, because the agency considers it redundant with the exclusion in 42 CFR 403.904(i)(1). Although there may be overlap between the two sections, they are not the same. Section 403.904(i)(1) excludes “indirect payments” or other transfers of value where the applicable manufacturer is “unaware” of the covered recipient’s identity during the reporting year and for two quarters thereafter. While combining the two exemptions into a single exemption may eliminate some redundancy when it comes to the indirect payment aspect, the “unaware” requirement will render the exclusion unworkable for purposes of CME.

CME programs are often planned and promoted months and sometimes years in advance, which makes the “unaware” requirement almost insurmountable. Additionally, many CME programs are planned and promoted before commercial support grants are even determined. As faculty are selected and identified during the activity planning process, it is common to use their names as part of the promotional materials for the activity. Specific faculty members are frequently part of what generates interest in the program; physicians hope to learn from known experts. It is neither realistic, nor would it be perceived as transparent, if faculty names were hidden until the day of the program. Indeed, SHM is deeply concerned physicians would not attend programs without knowing who will be serving as faculty. As a result, even if the applicable manufacturer financially supporting CME providers does not request faculty names, they are likely to learn the names of the faculty, usually before the program and certainly within two quarters after the program, through promotion of the program itself. Therefore, establishing a policy whereby an arbitrary determination of the presence of a relationship is made based on the timing of learning of the faculty names is unworkable – the names of faculty at CME programing cannot and should not be hidden.

The same can be said for attendees. While attendees might not be identified in advance of a CME program, they are certainly identifiable both during and after the program. However, CMS has always recognized that attendees have no relationship with companies that provide grants of commercial support to CME providers for accredited and certified CME. Therefore, it is not necessary to establish an arbitrary timing proxy for attendees. Attending accredited and certified CME does not establish a reportable relationship with any supporting companies.

To avoid confusion and unintended consequences, SHM urges that CMS retain the CME exemption in its current form. However, if CMS feels it cannot retain this exemption, section 42 CFR 403.904(i)(1) must be restructured to properly account for the realities of CME. This could be done by simply removing the “unaware” element when CME is the subject of the exclusion by adding the bolded text below. Section 42 CFR 403.904(i)(1) could then read as follows:

  1. Indirect payments or other transfers of value (as defined in § 403.902), where the applicable manufacturer is unaware of the identity of the covered recipient or when an applicable manufacturer or applicable GPO provides funding to a continuing education provider, but does not either select or pay the covered recipient speaker directly, or provide the continuing education provider with a distinct, identifiable set of covered recipients to be considered as speakers for the continuing education program. An applicable manufacturer is unaware of the identity of a covered recipient if the applicable manufacturer does not know (as defined in § 403.902) the identity of the covered recipient during the reporting year or by the end of the second quarter of the following reporting year.

Physician Compare Website

The proposals for the Physician Compare Website focus on expanding the depth and breadth of
available information and ensuring its utility and relevance for consumers. SHM appreciates that CMS wants to provide actionable information for consumers as they make decisions about their healthcare providers. We continue to support the concept of transparency though public reporting of quality and performance results. At the same time, we continue to have significant concerns about misinterpretation and misunderstanding of reported data due to questions about the accuracy and validity of publicly reported measures as indicators of quality, particularly for hospitalists.

Although SHM concurs that giving patients more information may encourage active participation in healthcare decisions, the proposed expansion of publicly reportable measures is concerning. By making all of the 2015 PQRS group practice reporting option (GPRO) measure sets and individually-reported measures available, this greatly increases the amount of potential information for consumers to digest. At a time when many of the measures in these pay for performance programs are still questionably indicative of quality care, adding additional information may not be appropriate.

Physician Compare treats all physicians the same, regardless of their practice realities and the relevance of their measures as quality indicators. Unfortunately, the measures hospitalists are able to report will likely not be useful information for consumers. Patients generally do not have a choice in their hospitalist, akin to what they experience with physicians in an emergency room. Situation and health condition permitting, patients are more likely to select a hospital or healthcare system and therefore its associated hospitalists, but to do this they would rely on hospital-level data or some other criteria to make these decisions.

Similarly, the proposal to publicly report CG-CAHPS and expand that reporting has severe limitations for hospitalists and other hospital-based providers. These surveys are not appropriate for hospitalist care; many of the domains are either beyond a hospitalist’s control or completely inapplicable to hospital medicine. SHM is deeply concerned with CMS’ reliance on this and related proposals for the CG-CAHPS as a one-size-fits-all tool for evaluating the patient experience with physicians. Much like the questionable utility of many of the publicly reported quality measures for hospitalists, public reporting on CG-CAHPS will not provide useful or even accurate information about individual hospitalists.

To begin to address the above concerns, SHM recommends that CMS create an identifier for hospitalists to contextualize the nature of their practice. It is estimated that there are 44,000 hospitalists currently practicing in the US. Although hospital medicine does not yet have a unique specialty billing code through CMS, the practice patterns and location of hospitalists make them unique from general outpatient internal medicine or family practice. The consumer community would be well-served by website content that defines the unique and important role of hospitalists in providing patient care, managing costs and improving safety in the hospital environment. Without it, results from Physician Compare may be confusing and counterintuitive. SHM is more than willing to work with CMS on further developing this concept and in drafting specific language for the hospitalist description.

Physician Payment, Efficiency, and Quality Improvements—Physician Quality Reporting System

The PQRS continues to be an integral part of quality reporting and pay for performance for physicians. As in previous years, CMS continues to refine and develop requirements for successful participation.

Unfortunately, many of these refinements may have the opposite effect, rendering reporting irrelevant for hospitalists and further highlighting their mismatch with PQRS. Specifically, SHM has concerns with the following proposals and their impact on hospitalists’ ability to successfully participate in PQRS:

  1. The elimination of a large number of measures from PQRS, including 8 measures that are applicable to hospitalists;
  2. Requirements for reporting CG-CAHPS for groups, to be expanded to other physicians in future years; and
  3. The creation of cross-cutting measure sets and requirements to report on cross-cutting measures.

Proposed Elimination of Measures from PQRS

Seventy-three measures are scheduled for removal from PQRS. This represents a major reduction in available measures and will impede many providers’ ability to successfully report in PQRS. SHM is deeply concerned about the removal of so many measures from the program at a time when most providers nationwide have only participated in the program for one year or have yet to participate at all. We also caution against the trend of CMS rapidly changing fundamental aspects of the program. Drastically changing the availability of measures risks confusing providers and hinders their ability to understand the mechanics of PQRS and feel confident in reporting.

SHM strongly recommends that CMS consider a methodology that will warn providers of upcoming changes, particularly when measures are slated for removal. Because of the nature of the rulemaking cycle, proposed changes to the measures available for reporting take effect just a few months after CMS finalizes each year’s PFS rule. Functionally, this means that measures providers are in the midst of reporting may also be in the process of being removed for the very next cycle. This does not produce confidence that information being reported is actually related to quality or practice improvement. A short turnaround on measure removal also does not give practices time to adjust and find alternatives for the next reporting cycle. A warning methodology could mirror the average time for measure development and submission to CMS by instituting three years between when a measure is identified for removal and when it is actually removed. This would give providers time to adjust and find additional measures, while also ensuring that CMS is able to curate the measures within PQRS.

For hospitalists, the measures proposed for removal are especially troubling as their elimination would virtually halve the number of measures applicable to their practice. Specifically, SHM offers comments on the following measures slated for removal:

  • PQRS # 0031 Stroke and Stroke Rehabilitation: Venous Thromboembolism (VTE) Prophylaxis for Ischemic Stroke or Intracranial Hemorrhage
  • PQRS # 0032 Stroke and Stroke Rehabilitation: Discharged on Antithrombotic Therapy
  • PQRS # 0033 Stroke and Stroke Rehabilitation: Anticoagulant Therapy Prescribed for AtrialFibrillation (AF) at Discharge
  • PQRS # 0035 Stroke and Stroke Rehabilitation: Screening for Dysphagia
  • PQRS # 0036 Stroke and Stroke Rehabilitation: Rehabilitation Services Ordered


SHM notes that the CMS’ explanation for the proposed removal of the Stroke and Stroke Rehabilitation measures (PQRS # 0031, 0032, 0033, 0035, and 0036) is due to “represent[ing] clinical concepts that [are] currently included within inpatient standards of care.” SHM strenuously disagrees with this reasoning and objects to removal of these measures at this time. The slated removal of measures relating to “inpatient” concepts of care does not make clinical or practical sense. If physicians are to be required to report on quality measures that are applicable to their practices, and their practices are rooted in inpatient or other facilities, it would follow that their measures should mirror concepts required for reporting in other inpatient or facility-based quality reporting programs. SHM finds the proposal to remove these measures to be contrary to the broader goal of finding measures and methodologies to make PQRS and the VBPM meaningful and actionable for hospitalists and other hospital-based providers.

  • PQRS # 056 Emergency Medicine: Community-Acquired Bacterial Pneumonia (CAP): Vital Signs
  • PQRS # 059 Emergency Medicine: Community-Acquired Bacterial Pneumonia (CAP): Empiric Antibiotic


The Community-Acquired Bacterial Pneumonia measures (PQRS # 056 and 059) are slated for removal due to those measures being “topped out” or having performance rates close to 100%. SHM worked closely with the AMA-PCPI and CMS measure contractors to expand the measures’ denominators to render the measures reportable by hospitalists starting in 2014. This expands the number of providers who may be able to report on these measures, and as such SHM recommends that CMS retain the measures. Since the changes went into effect only this year and with numerous new providers reporting on them, the measures cannot be considered “topped out” and should therefore remain in the program.

Requirements for Reporting CG-CAHPS

CMS is seeking to institute new requirements for groups to report CG-CAHPS as part of a larger commitment to assessing patient experience of care in pay-for-reporting and pay-for-performance programs. Specifically, CMS proposes to make reporting on CG-CAHPS mandatory for groups of 100 or more eligible professionals (EPs) in 2015 for the PQRS payment adjustment in 2017. CMS has also indicated future requirements to make CG-CAHPS reporting mandatory for groups of 25 or more EPs in 2016 for application of the 2018 PQRS payment adjustment. SHM agrees with CMS’ assertion that patient experience is an important dimension of evaluating the quality of care afforded through Medicare. However, a universal requirement for CG-CAHPS reporting will not appropriately meet this aim.

SHM strongly advises against finalizing mandatory reporting of CG-CAHPS for successful PQRS participation and by extension for VBPM performance. The measures contained in CG-CAHPS are best suited for outpatient physician practices, and as such are not applicable to all physician practices. This is pointedly true for hospitalists. For example:

  • Patients cannot call to get an appointment with a hospitalist; hospitalists treat patients who require care in acute care hospitals and other facilities.
  • Hospitalists do not have “offices” for patients to schedule appointments and routine visits.
  • Hospitalist groups generally do not have ongoing relationships with their patients, unless those patients are frequently seen in the hospital.
  • Hospitalists cannot directly control the wait times in an emergency room or for hospital admission; there are many factors that influence those times.


As illustrated, the CG-CAHPS would not capture a patient’s experience with a hospitalist. It would in effect be a double penalty for hospitalists and hospitalist groups to report CG-CAHPS. First, the group would be required to pay for the administration of the survey, which requires partnering with a CMS-certified survey vendor and a significant outlay of resources. Second, and because the surveys are inapplicable to hospitalist care, hospitalists will likely fail any performance standards set for the measure, thus unfairly jeopardizing their performance in the VBPM. This is contrary to the published Governing Principles for the VBPM program, which states that “measures for the VM should consistently reflect differences in performance among groups or solo practitioners, [and] reflect the diversity of services furnished.”1

Creation of Cross-Cutting Measures Sets and Requirements to Report Two Cross-Cutting Measures

CMS proposes to create a new measure set, reclassifying selected measures as “cross-cutting” in an effort to identify core clinical competencies across specialties. SHM agrees there is inherent value in identifying and utilizing cross-cutting measures. If done well, this could help ensure greater consistency across the healthcare system and move forward particular clinical improvement goals.

Within the proposed measure set, there are two measures that are specified to be reportable by
hospitalists: PQRS # 047 Advance Care Plan and PQRS # 130 Documentation of Current Medications in the Medical Record. SHM concurs that these measures may be appropriate cross-cutting measures and should have broad applicability within the healthcare system.

However, SHM is concerned about the proposed requirement to report on two cross cutting measures. A more measured implementation of this requirement should be considered until there is a more robust cross-cutting measure set that incorporates evidence-based, accurate and actionable measures. We recommend starting with one required cross cutting measure. This will help ensure more providers are able to meet this requirement, while emphasizing the importance of identifying and developing measures that are more broadly applicable.

SHM would also like CMS to clarify that any measures reported in the requirement for cross-cutting measures would also be counted towards the requirements to report measures across three National Quality Strategy (NQS) domains. If they are not, hospitalists will face extreme difficulties in meeting the NQS requirements; nearly all of their other potential measures for reporting are from one domain: Effective Clinical Care.

Comments on the MAV (Measure-Applicability Validation) Process

SHM appreciates that CMS is making the MAV process more broadly available throughout the
requirements for PQRS reporting. Doing so will help practices identify measures that are possible for their group to report from year to year, particularly if they are unable to self-identify enough measures to meet the reporting requirements.

However, SHM believes it would be more productive to allow groups access to the MAV process before they are penalized. For example, a list of potential measures could be automatically populated for each group or individual EP upon their registration for PQRS based on prior year’s claims data. This could help practices better understand the PQRS and VBPM program, while ensuring that fewer physicians and groups are penalized simply for not being able to navigate the PQRS program and its daunting measures list. We acknowledge the QualityNet Helpdesk may be used as a resource, but SHM has heard feedback from members indicating significant gaps in the ability of the Helpdesk to provide adequate assistance for this program.

Proposed Change to the Registration Period for the Group Practice Reporting Option

CMS is proposing to move the deadline for registering as a GPRO from September 30 to June 30 of each reporting year. This would mean that groups have until June 30, 2015 to register for the 2015 reporting year. While we acknowledge that an earlier deadline may allow CMS to provide more timely feedback to groups, SHM is concerned this cutoff will have the consequence of requiring any TINs formed in the third quarter of the year to report as individuals. It would effectively mandate individual-level reporting for all TINs formed in the second half of the year. This would add greatly to administrative complexity and burden for groups participating in PQRS, particularly for those that are comprised of multiple TINs and practice across regions or nationwide. SHM recommends that CMS explore yearly trends in TIN formation to see the scope of impact before implementing any proposed changes. If CMS finalizes this proposal, SHM strongly recommends the development of an additional process for groups that form after June 30 that would enable them to either report as groups or be held harmless from PQRS penalties and VBPM quality tiering. Forcing these groups to develop the infrastructure to report as individuals for one year only is counterintuitive and unnecessary.

Proposed Changes to the Requirements for the QCDR

SHM fully supports the QCDR concept and appreciates that CMS will continue to adjust QCDR requirements as a way to make it a more robust reporting tool. However, we note that continuously increasing start up requirements will have the unintended effect of restricting the ability for new QCDRs to enter the market. Functionally, this means that specialties or other interested parties who have not yet started a QCDR may find the learning curve and requisite resources too steep and extensive to warrant the undertaking. SHM recommends that CMS consider a step-wise process of unrolling expanded requirements for QCDRs that enables first-year QCDRs to have the same beginning requirements that currently operating QCDRs had, whether it is this year or three years from now. Each subsequent year could then progressively add requirements, similar to the way Meaningful Use was designed. This would ensure the pathway to creating a new QCDR remains open and available today and in the future.

Comments on Previously-Contemplated Inclusion of IQR Measures in PQRS

SHM notes CMS did not make any recommendations or proposals regarding the inclusion of IQR measures for reporting by EPs in PQRS. The 2014 Physician Fee Schedule proposed rule (CMS 1600-P) contained a discussion section exploring options for reporting and aligning hospital-based measures in PQRS. Although we are heartened by CMS’ discussion this year of a more wholesale alignment with Hospital Value-Based Purchasing in the section “Discussion Regarding Treatment of Hospital-Based Physicians,” we hoped to see proposals or more information for discussion regarding the retooling of hospital-level measures for individual reporting. Indeed, SHM finds the proposal to remove PQRS measures because they contain clinical concepts meant for inpatient settings to be contrary to the goal of finding measures and methodologies to make PQRS and the VBPM more meaningful and actionable for hospitalists and other hospital-based providers.

Value-Based Payment Modifier and Physician Feedback Program

SHM recognizes that 2015 is an important year for the development of the VBPM, with proposals to expand the program to all physicians and non-physician providers. We appreciate that CMS seems ready to work with providers to make the VBPM more relevant and contain actionable data that would be able to drive quality improvement efforts. We also laud the continued effort to align requirements between PQRS and the VBPM and encourage CMS to continue to look for ways to administratively combine the two programs to reduce further complexity and confusion over the two separate programs.

Group Size

SHM understands the statutory requirements placed on CMS to expand the VBPM to all physicians by 2017, which means that under the current structure of the program, all physicians must report on quality measures in 2015 to avoid the application of the full negative payment adjustment. We appreciate that CMS has continued its step-wise approach in adding new participants to the VBPM by proposing to hold harmless from any negative adjustment groups and individuals who do not have enough beneficiaries or are new to the VBPM in 2015.

Application of the VM to Non-physician EPs

SHM agrees with CMS’ contention that shared accountability among all the members of the healthcare team is one of the goals of VBPM. Applying the VBPM to all providers would seem to be an appropriate step to further this goal. Hospitalists and hospital medicine have been long-standing champions of teambased care and support efforts to harmonize reporting requirements and responsibilities with the entire team.

Approach to Setting the VM Adjustment Based on PQRS Participation

SHM appreciates that CMS’ proposals mirror those from previous years, where EPs who are newly integrated into to the program will be held harmless from negative payment adjustments in the VBPM. We believe this gives EPs and their groups a chance to become more familiar with the VBPM and to better understand the connection between their reporting of quality measures in PQRS and a potential pay-for-performance adjustment under the VBPM. It is reasonable and fair to give providers who are new to the VBPM the same starting footing in the program as has been afforded to larger groups in previous years.

Payment Adjustment Amount

CMS proposes to raise the payment adjustment to a potential -4.0% for the full downward adjustment and to +4.0x for the full upward adjustment. SHM does agree that significant payment adjustments may be needed to drive behavior; prior experiences with PQRS bears this out. However, PQRS had previously been a voluntary program whereas the current VBPM is an entirely new program that uses novel methodologies to create performance-based payment adjustments. Given these facts, SHM recommends that increases in adjustment rates should be phased in more gradually and that minimally any increase should occur after all providers have had at least one year with the VBPM applied to their payments. This would help avoid alienating providers as they ramp up their efforts to comply and participate fully.

Potential Methods to Address NQF Concerns Regarding the Total per Capita Cost Measures

SHM agrees with CMS’ proposal to adjust the primary care attribution methodology to include nonphysician providers within the first step of attributing beneficiaries to a group. Physician assistants, nurse practitioners and other non-physician providers deliver critical aspects of patients’ care that should be appropriately captured in a cost assessment. This also seems to align the Total per Capita Cost measures with CMS’ intent to include all non-physician providers in the2017 VBPM. However, this adjustment does not make these measures more applicable to hospitalists.

Although CMS defers making any changes to the cost measures relating to adjustments for
socioeconomic and other sociodemographic factors, SHM reiterates its support for making these adjustments. There is significant data about the social determinants of health and how they influence care, utilization, and outcomes. SHM has been consistently in support of adjusting measures, including those for readmissions, to account for situations beyond the providers’ scope of control. By having these adjustments, measures will better capture meaningful differences in care between providers and drive quality improvement efforts within healthcare systems.

Discussion Regarding Treatment of Hospital-Based Physicians

SHM greatly appreciates that CMS is continuing to consider how PQRS and the VBPM be structured to better reflect the practice patterns of hospital-based physicians. CMS discusses how a structure for hospital-based physicians to incorporate or attribute their hospital’s or hospitals’ performance in Hospital Value-Based Purchasing (HVBP) into the physician-level reporting requirements for the VBPM. This includes a consideration of the mechanics of the option, including who would be eligible and how the HVBP score could be either completely adopted or broken down to be applied to the VBPM. This is an opportunity for some hospitalists who would find the alignment contemplated in the proposed rule to be clinically meaningful and representative of their relationship with their hospital.

We note that any distinct program for hospital-based physicians must be voluntary in nature. A group or individual must not be compelled to align performance in the VBPM with that of an associated institution. There are countless practice structures and employment models that make this alternative appealing for some hospitalists and wholly inappropriate for others. SHM appreciates that CMS seems to indicate this would be a voluntary policy, but we want to ensure this is maintained throughout any future proposals.

SHM also recommends that, although this was not explicitly mentioned in the discussion, any policy implemented for hospital-based physicians must fulfill PQRS reporting requirements as well as satisfying performance requirements for the VBPM.


CMS contemplates two options for determining which providers would be eligible for this option. SHM is strongly in support of the first alternative, self-nomination at the TIN level, which allows a group to attest that it is comprised primarily of hospital-based physicians. This is consistent with our prior comments and would ensure that only those providers who wish to have this level of alignment would be included in the program. It would not be in the best interest of any group or specialty that does not have significant relationships and influence within the hospital to elect this option, pre-empting any concerns about potential misuse of this option.

The second alternative uses set criteria that a TIN will need to satisfy to have the HVBP performance rates automatically included or to open the option to elect the inclusion of hospital data in the VBPM. CMS uses the example of 90 percent or more services performed in a hospital setting, which is similar to the Meaningful Use threshold for hospital-based providers. SHM notes there have been significant issues with the Meaningful Use threshold of 90 percent of “hospital” services because they do not include any care that is considered outpatient, including observation status and SNF care. Due to wide practice variation, SHM does not believe a set threshold would be an appropriate gatekeeper for this option. A simple cross-check could be performed to ensure self-nominating groups are actually working in the hospital(s), which could be established through claims analysis or some other identifying criteria. However if CMS chooses to pursue set criteria for eligibility, we caution that it should be inclusive of all of the services normally considered part of the practice of hospital medicine.

Attribution of Hospital or Hospitals:

CMS contemplates two options for the attribution of a hospital or hospitals to a given TIN that would attribute either a single hospital based on plurality of services provided by the TIN or multiple hospitals based on a threshold of hospital services. SHM believes a methodology that enables the inclusion of multiple hospitals would be necessary. We know that many hospitalist groups practice in multiple locations, and as such it may not be appropriate to assign a single hospital’s performance to the TIN. We caution that a threshold should be high enough to ensure the amount of care delivered is not insignificant and that a hospitalist or group could be expected to have a reasonable amount of influence in the hospital.

CMS should also consider expanding the self-nomination process to enable providers to select which hospital(s) with which they would like to be aligned. As CMS indicates in the proposal, groups would need to register for PQRS reporting before the hospital’s final HVBP performance, meaning they would need to look at historical HVBP data to make decisions. This would be akin to the process that EPs go through in selecting quality measures in PQRS based on prior performance and future expectations.

Scoring Methodology:

CMS also considers how a hospital’s or hospitals’ Total Performance Score (TPS) or subset therein could be incorporated into the VBPM. We agree that it is appropriate for this option to be consistent with the two year performance to adjustment time lag in the PQRS/VBPM.

SHM believes that either the second or third options explored in the discussion could be acceptable, but recommends CMS strongly consider the third option. The third option, which identifies a subset of measures from HVBP to incorporate into the VBPM, would be preferable to ensure measures are sufficiently reflective of hospitalists’ care and influence. This would allow for the selection of measures within the scope of influence of hospitalists and other hospital-based physicians and would be the easiest to relate back to PQRS/VBPM reporting requirements. The second option, which includes the Efficiency and Cost Reduction domain score in the VBPM cost composite and all or some subset of the other domains in the VBPM quality composite, may engender coordination between hospitals and physicians, but would contain some measures that may not be appropriate for physician-level reporting and evaluation. SHM cautions that the first option, total TPS score alignment, is not appropriate in a physician pay-for-performance program at this time.

In sum, hospital-performance alignment must contain the following principles:

  • Voluntary election
  • Flexibility to capture unique and varied practice structures
  • Appropriate and reasonable measures that are within the influence of hospital-based physicians
  • Will fulfill both PQRS and the VBPM requirements

SHM looks forward to working with CMS to identify an eligibility pathway and scoring methodology that would be most appropriate for hospitalists and would encourage more alignment between physicians and their hospitals.

General Comments about the VBPM Cost Comparisons

SHM has consistently advised CMS that the comparison pools in the VBPM are inadequate and not reflective of intraspecialty differences. Hospitalists, by virtue of practicing in hospitals, see a higher proportion of sicker and more expensive patients than their outpatient counterparts. For its part, SHM has applied for a specialty billing code so that hospitalists are not inappropriately compared and unduly penalized in the VBPM. Another potential option for ameliorating concern surrounding comparison pools is to explore using a site-specific breakdown of the specialties. This would directly acknowledge that a provider who practices primarily in an institution will have different costs and patterns than one practicing in an outpatient office. The specialty options in the Provider Enrollment, Chain and Ownership System (PECOS) do not account for these differences, so CMS could use the place of service (POS) codes as part of the consideration in the cost measures. This would help ensure that a hospitalist would only be compared against similarly situated Internal Medicine or Family Practice providers. Not only would this be a preferable way to have fair comparison pools, but this would also help stem some of the hyperspecialization the VBPM seems to be incentivizing.


SHM appreciates the opportunity to provide comments on the proposed 2015 Physician Fee Schedule and the development of physician quality reporting in particular. 

Burke T. Kealey, MD, SFHM
President, Society of Hospital Medicine

 1 Medicare Program; Revisions to Payment Policies under the Physician Fee Schedule, DME Face-to-Face Encounters, Elimination of the Requirements for Termination of Non-Random Prepayment Complex Medical Review and Other Revisions to Part B for CY 2013. 77 Fed. Reg. 69, 307.