SHM Sends Letter on Reducing Regulatory Burdens in Current Pay-for-Performance Programs
May 23, 2017
SHM's Policy Efforts
SHM supports legislation that affects hospital medicine and general healthcare, advocating for hospitalists and the patients they serve.
Thomas Price, MD, Secretary
Department of Health and Human Services
200 Independence Avenue, SW
Washington, DC 20201
Centers for Medicare and Medicaid Services
200 Independence Avenue, SW
Washington, DC 20201
Dear Secretary Price and Administrator Verma,
The Society of Hospital Medicine, on behalf of the nation’s nearly 57,000 hospitalists, welcomes the administration’s emphasis on reducing regulatory burdens, and ask this authority be used to reduce burdens for healthcare providers in existing pay-forperformance programs, including the Physician Quality Reporting System (PQRS), Physician Value-based Payment Modifier (VM) and Meaningful Use (MU). We are supportive of the recent letter from the American Medical Association on the challenges associated with successful participation in the programs and encourage you to act on providing relief for the 2018 calendar year.
As CMS acknowledged in their transition year to the Quality Payment Program (QPP), there are several issues with the pay-for-performance agenda that need to be addressed to maximize participation, validity, and buy-in. The Pick-Your-Pace flexibility for 2017 QPP reporting allows providers who are attempting to gain familiarity with a complex program to report minimally and to avoid penalties in 2019. At the same time, CMS made scoring for the Cost category in the first year of the Merit-based Incentive Payment System (MIPS) worth zero (0%) in response to provider concerns about attribution and comparisons. These same concerns have been consistently present in PQRS and the VM, but were not addressed in rulemaking for those programs. Taken together, we believe this underscores the need for relief in existing programs that will sunset as the QPP begins.
Hospitalists are front-line healthcare providers in America’s hospitals for millions of hospitalized patients each year, many of whom are Medicare and Medicaid beneficiaries. They manage the inpatient clinical care of their patients, while working to enhance the performance of their hospitals and health systems. Although most hospitalists are Board Certified in Internal Medicine or Family Medicine, hospitalists practice exclusively in facility settings and do not have outpatient office practices.
The unique position of hospitalists in the healthcare system affords a distinctive role in facilitating both the individual physician-level and systems or hospital-level performance agendas. Unfortunately, this positioning also creates many challenges and barriers to successful participation in physician pay for performance programs that are more suited to outpatient providers. This results in millions of dollars’ worth of unfair and unsubstantiated penalties being levied against hospitalists and hospitalist groups. Below, we outline the issues hospitalists face in these programs to illustrate the need for broad-based relief from existing programs and their penalties.
PQRS Reporting and Quality Measures
Hospitalists face a dearth of relevant quality measures and significant hardships to successful reporting of the few measures that are specified for their practice. Thus, it is common for hospitalists who are attempting in good faith to report on quality measures, to fail PQRS reporting and therefore be subject to both PQRS and VM penalties. Quality data is often contained in hospital-owned EHRs, making it difficult for providers to collect data and report on measures. Because of this, hospitalists have historically used claims-based reporting as their primary methodology for reporting quality measures. Claims-based reporting is done at the individual level and relies on attaching quality codes to claims as they are submitted.
Quality measures for PQRS are almost universally designed for performance attributed to an individual clinician and are not structured for shared responsibility and performance. In addition, measures are generally developed using clinical and practice logic appropriate for an outpatient setting, where an episode of care is not characterized by multiple providers. A good example of this dynamic in action is Measure 47 (Advance Care Plan). The way the Advance Care Plan measure is structured, a provider must ensure the medical record reflects that advance care planning was discussed and documented for every provider who sees an applicable patient. Hospitalists practice in a team-based model where there are many handoffs over the course of a patient’s hospital stay. Per the measure, Advance Care Plan reported by claims would need to be performed and reported for each billable instance with a new provider, not once for the entire patient stay. If taken literally, this would necessitate repetitive conversations about care plans and end-of-life care by every hospitalist seeing that patient over the course of one hospitalization. This would be damaging to the physician-patient relationship and a waste of provider time and resources. Conversely, Advance Care Plan performed in an outpatient setting would make sense as a single annual conversation, or during major life events, between a patient and their primary provider (either a primary care physician or specialist).
Many hospitalists are currently failing PQRS and experiencing penalties for performing and reporting Advanced Care Plan in a way that makes clinical sense – reporting the measure for the entire care team for the course of a patient’s hospital stay. Unfortunately, when reported this way, only one hospitalist who sees the patient can get credit for the measure while the other hospitalists have denominatoreligible instances with no numerator performance on the measure. These instances count against those providers.
In addition to problems with the measures themselves, hospitalists have reported:
- Their own number of cases in numerators and denominators of measures being vastly different from those reported by CMS;
- Receiving conflicting information from CMS and its contractors on how to properly report on a quality measure; and
- Failing performance on a measure because it is designed to evaluate patient clinical conditions in outpatient settings rather than the inpatient setting, which would necessitate different benchmarks.
Even when hospitalists report on all available quality measures, they fall short of the 9 required measures because there are not 9 measures applicable to hospitalists.
When a provider reports on fewer than 9 measures for PQRS, it triggers a Measure Applicability Validation (MAV) process, which uses clinical relation clusters to determine if that provider could have reported on additional measures. In practice, the MAV functions as a punitive mechanism for providers who are attempting to sort through hundreds of measures and then report on those measures they find relevant to their practice. In the past, we have worked with CMS to try to adjust the MAV process for hospitalists so that it recognizes the limited number of relevant measures available to them and does not mechanically trigger penalties for failure to report measures that either are not available for hospitalists for reporting or do not make clinical sense for hospitalist practice. Despite these efforts, hospitalists who report measures in good faith are still receiving PQRS penalties after failing the MAV, even when there are no other reasonable measures for them to report.
Finally, when hospitalists fail PQRS despite making a substantial investment in time and resources to comply with reporting requirements, the process of seeking an informal review and the decisions of those reviews are not transparent. Informal review decisions are frequently made instantaneously or via form letter, preempting any conversation, fact finding, or discussion with CMS about why providers failed and what they can do to avoid such failure in the future. Consequently, the appeal of a PQRS failure decision by CMS is frequently met with denial and providers are not only left with unfair and unexplained financial penalties, but they are also increasingly frustrated and unsure of how to successfully proceed in future reporting periods.
We encourage relief that enables providers to be protected after attempting to report quality measures in good faith. It is difficult and costly to set up systems to accurately capture the data for reporting and, as outlined above, hospitalists may still fail PQRS reporting because of measure logic that is not reflective of their practice or because of the sheer complexity of the program and resulting confusion around measure reporting and validation.
Inappropriate Cost Measure Comparisons and Calculations
The VM uses PQRS quality measures and cost measures to make assessments of “value,” assigning payment adjustments by performance. Due to the nature of VM cost measures, even the most efficient hospitalists are set up for failure and, at best, may be able to achieve average cost under the CMS Value Modifier quality tiering scoring. The cost measures being used make it impossible for CMS to create accurate and fair assessments under the VM. This warrants relief from the penalties many hospitalists now face due to the inaccurate comparisons being made under current cost measure framework.
Attribution for these administrative claims-derived cost measures is a significant part of the problem. For example, the Total Per Capita Costs measure uses a two-step primary care attribution process, similar to that of Accountable Care Organizations (ACOs), to attribute patients to providers. Hospitalists are increasingly practicing in Skilled Nursing Facilities (SNFs) to more closely follow their patients across settings and to provide more seamless care. Those hospitalists who practice in SNFs have patients attributed to them under this measure, even though SNF stays are not and should not be considered a primary care setting.
By having SNF patients attributed, the costs for these hospitalists appears to be higher than their outpatient primary care colleagues because they are seeing patients who have had expensive hospitalizations and other resource-intensive facility care. In policies for the MIPS and in ACO patient attribution, this process was rightly changed to exclude SNF patients from attribution based on plurality of primary care services. Although this problem within the VM Total Per Capita Costs measure was identified early enough and raised with CMS, the measures were not changed for the VM. This leaves hospitalists and other providers who practice in SNFs subject to poor performance on the cost assessment for the remainder of the VM.
Additionally, several of the measures in the VM use specialty comparison as part of the cost analysis for the measures. Hospitalists predominantly have Internal Medicine and Family Medicine as their specialty and when compared against their outpatient counterparts of the same specialty, they look enormously expensive. One hospitalist group relayed that their performance was, on average, 3 standard deviations worse on cost measures when compared against outpatient primary care doctors.
Higher costs for hospitalists are expected given that they practice in a high-cost setting (hospital vs outpatient primary care office). To alleviate this problem with inappropriate comparisons among broad specialties, CMS recommended that hospitalists apply for a distinct Medicare specialty code. SHM responded and applied for a Hospitalist specialty code in May of 2012, which was finally awarded in February of 2016, but usage of the code did not become effective until April of 2017. Due to this timing, the new hospitalist specialty code does not provide any relief for hospitalists in the VM as they have continued to be compared to outpatient primary care physicians on cost.
CMS has acknowledged these problems and made changes that may improve equity in cost measures for the MIPS. However, absent relief, hospitalists face unfair comparisons and inappropriate attribution for cost measures under the VM. They are currently being penalized due to their practice location rather than being objectively high cost.
Exceptions and Exclusions from Meaningful Use
The structure of the Meaningful Use (MU) program was bifurcated along the lines of the Medicare payment systems. Therefore, there are two separate programs – one for hospitals and one for individual providers. Hospitalists, who practice in facilities, are not able to carry their own Electronic Health Records (EHRs) and use hospital EHR systems. It would be duplicative and wasteful to encourage providers like hospitalists to meet the criteria for Eligible Providers (EP) under MU. For this reason, the HITECH Act and the subsequent Continuing Extension Act of 2010, created a hospital-based exemption using a threshold of 90% of billing in Place of Service 21 (hospital inpatient) and 23 (emergency department). Hospitalists provide a significant amount of observation care, post-acute care, and perioperative care, all of which have Place of Service codes associated with their billing that do not fall within the hospital based, POS 21 and 23, exemption. Consequently, hospitalists would frequently fall outside of the exemption and experience penalties under MU despite their use of hospital EHRs and their inability to comply with the program as individual providers.
CMS acknowledged the narrowness of the definition of “hospital-based” in developing MACRA policies and created a wider exemption threshold from the Advancing Care Information category in the QPP (75% of services in Place of Service 21, 22, and 23). In prior years of the MU program, SHM worked with CMS to develop an automatic “hospitalist exception” (90% of services in Place of Service 21, 23 and observation services under 22). This automatic exception from MU is critical protection for hospitalists and acknowledges their positioning in the healthcare system.
It is unclear whether this “hospitalist exception” from MU will continue in 2018. We want to ensure protection for hospitalists remains in place for the final MU penalty year (2018) and would welcome this protection as part of broader relief around MU for providers.
Taken together, the issues with PQRS, VM and MU programs represent undue administrative burdens and are resulting in penalties being inappropriately assessed against significant numbers of providers who are attempting to comply with requirements. The problems hospitalists face in these programs are widespread, and may not be limited to just our specialty. We also note that penalties in 2018 would be concurrent with providers ramping up their efforts for the second year of reporting under the QPP, which not only increases the expense of compliance, but also risks creating increased confusion. We therefore urge you to act and provide administrative relief from all penalties under PQRS, VM and MU programs for 2018.
Thank you in advance for your prompt attention to this matter.
Ron Greeno, MD, FCCP, MHM
President, Society of Hospital Medicine
cc: Patrick Conway, MD
Kate Goodrich, MD