Quality measures are a cornerstone in measuring physicians' performance within the Centers for Medicare & Medicaid Services' Quality Payment Program (QPP). Clinicians' performance on quality measures and other categories within the QPP determines Medicare part B payment adjustments. Driven by evidence-based clinical practice guidelines, quality measures should focus on high-priority facets of health care, support a desired patient outcome, and address an area with evidence of a gap or variation in provider performance.
To meet the goals of the QPP, a broad array of quality measures must be developed that allows pathologists the flexibility to choose activities and measures most meaningful to their practice and patient population while also trying to mitigate the challenges of implementation and data collection.
In this second manuscript of the series, we present the development of additional College of American Pathologists–developed quality payment measures for use in the QPP. We also discuss the relationship of quality measure reporting with reimbursement and the challenges with capturing data for quality reporting.
The College of American Pathologists identified 23 new measures for quality performance reporting that reflect rigorous clinical evidence and address areas in need of performance improvement.
Development of quality measures is a necessary and ongoing effort within the College of American Pathologists. Increased awareness about pathology-specific issues in measure development and reporting is essential to ensuring pathology's ability to demonstrate value and meaningfully participate in the QPP.
The Medicare Access and CHIP Reauthorization Act of 2015 led to the creation of a new Quality Payment Program for the Centers for Medicare & Medicaid Services (CMS), which has 2 payment pathways: the Merit-Based Incentive Payment System (MIPS) and Advanced Alternative Payment Models. The default pathway within the Quality Payment Program is MIPS, in which physicians' performance in different categories results in Medicare Part B payment adjustments from CMS. A review1 of the Quality Payment Program was previously published by the College of American Pathologists (CAP) and an in-depth update focusing on year 4 of the program was released along with this series of articles. For pathology, the most heavily weighted MIPS performance category is Quality, in which clinicians' scores represent their overall performance on various clinical quality measures.
Within the CAP, the Measures and Performance Assessment Subcommittee has been tasked with creating an array of pathology-specific quality payment measures for use in MIPS. This process of measure development was reviewed in part I (see article from Bocsi et al in this special section). This article aims to briefly present the relationship of quality payment measure reporting to reimbursement within MIPS and the challenges related to capturing the data needed for reporting on quality measures. This manuscript also reviews 6 of the CAP's newer Qualified Clinical Data Registry quality payment measures for use in the Quality Payment Program.2
QUALITY MEASURES AND REIMBURSEMENT
Eligible providers and/or groups participating in MIPS will earn an overall final score (FS) based on their performance within 4 categories: Quality, Improvement Activities, Promoting Interoperability, and Cost. However, the current requirements and metrics for the Promoting Interoperability and Cost categories will likely not apply to a pathologist and therefore will not be scored. Consequently, the vast majority of MIPS-eligible pathologists will likely need to report only on Quality and Improvement Activities, which represent 85% and 15% of the FS, respectively. An eligible provider or group's FS is compared with that of all other MIPS providers and with standards set by CMS to determine the resulting positive, neutral, or negative payment adjustment. Therefore, a pathologist's score in the quality category has an exaggerated impact on his or her overall MIPS FS compared with other specialties. MIPS performance in 2019 will impact 2021 Medicare part B physician fee schedule reimbursements by up to ±7%1 ; 2020 MIPS performance will impact 2022 Medicare part B physician fee reschedule reimbursements by up to ±9%.
In 2020, eligible providers need to report on 6 quality measures, or all that apply if fewer than 6 are available, to avoid a penalty from CMS. Additionally, clinicians must report on a minimum of 20 cases/patients and at least 70% of eligible patients for each measure in order to fulfill CMS's data completeness criteria. Lastly, one of the 6 measures needs to be either a high-priority or outcome measure. Any additionally reported high-priority or outcome measures are eligible for 1 or 2 bonus points, respectively. Scoring on quality measures ranges from 1 to 10. If 6 measures are applicable to an eligible provider or group, the maximum number of quality points is 60. The performance score for the top 6 reported measures plus any bonus points is summed, divided by 60 (the total possible quality points) and then multiplied by 85% to determine the total number of points that the quality category contributes to the MIPS FS. This scoring methodology further illustrates the importance of quality measures and reporting.3
CHALLENGES WITH MEASURE REPORTING
Historically, collecting and entering accurate quality payment measures data has been a highly time-consuming and manual process. Despite these challenges, CMS continues to increase reporting requirements, such as the number of cases needed to meet data completeness requirements (increased from 60% to 70% for 2020), the need to report on all patients regardless of payer type for non–claims-based reporting, and the gradual removal of claims-based reporting. The need for extensive record review for chart-abstracted measures is becoming an insurmountable task. Furthermore, CMS is requesting increased complexity for new quality measures, emphasizing measures that span the patient care continuum and cross multiple medical specialties. These facts necessitate the need for more readily accessible laboratory information systems (LISs), electronic health records (EHRs), and/or other electronic clinical systems data in order to provide access to a more robust set of clinical information. The CMS considers use of the data routinely collected through EHRs an essential tool for reducing reporting burden. As such, the electronic clinical quality measure (eCQM) has become a critical component of the quality payment reporting framework.
An eCQM is a measure that is expressed and formatted to use data from an electronic information system (eg, LIS or EHR) and capture data in a structured format during the normal clinical workflow. Electronic clinical quality measures, formerly known as eMeasures, can promote greater consistency and improved uniformity in defining clinical concepts and measure logic across measures and increase comparability of performance results. For the eCQM to be reported from a LIS, a standard language is used for electronically documenting the measure specifications. Standardizing a measure's structure, metadata, definitions, and logic promotes quality measure consistency and unambiguous interpretation. In addition to significantly reducing measurement errors due to manual abstraction, eCQMs drastically reduce the burden on providers to report performance and increase accessibility of benchmarking to individual clinicians.
The eCQMs are based on the following:
The types of clinical data that are typically encoded using standardized terminology in EHR systems (ie, Current Procedural Terminology, International Statistical Classification of Diseases and Related Health Problems, Logical Observation Identifiers Names and Codes, and Systematized Nomenclature of Medicine), considering the impact on workflow and data fidelity for organizations that will need to map local codes to standard terminologies used in an eCQM.
Standard data models and languages that provide the ability to represent measures in a way that is human readable yet structured enough for processing a query electronically. Widespread use of the same models and languages will increase interoperability among measures.
The Measure Authoring Tool, a Web-based tool that enables measure developers to author eCQMs using health care industry standard terminologies. The representation of an eCQM is simplified and standardized when measure developers author their eCQMs in the Measure Authoring Tool.
Many of the data for pathology quality measures reside in the free text of reports and the associated orders. Although data in this semistructured format are easier for a human abstractor to find and report on, most LISs do not store discrete data elements within pathology reports. This makes electronic implementation of sophisticated quality measures difficult. Because of the diversity of LIS designs, as well as variability in the configuration of LIS at individual organizations, implementing quality measure reporting may require significant resources. Unstructured data fields, such as pathology narratives or notes, require the use of natural language processing software to extract important clinical concepts from these notes into a reference terminology such as Systematized Nomenclature of Medicine or Logical Observation Identifiers Names and Codes for analysis. Though natural language processing shows promise in some data-extraction cases, this approach is not yet sophisticated enough for interpreting a comprehensive amount of clinical documentation types. Nonetheless, even if these hurdles are overcome, the issue of data ownership also plays a role in obtaining access to the electronic data.
Rebuilding EHRs and LISs to capture data in a structured format at ordering and reporting can provide the unambiguous information needed to track processes to optimize pathology clinical practice. With clinical concepts stored in a uniform way in discrete fields, extraction of eCQM information is greatly simplified compared with natural language processing–guided keyword searches of extensive notes fields. Developing a more structured LIS interface can allow information capture required to measure and report; however, the cost of this approach is an impediment to adoption by pathologists and referring physicians. This approach warrants careful consideration, as it is another trade-off in allocation of health care dollars. Improvements in natural language processing techniques may also provide a pathway for accurately extracting meaningful data from pathology reports as well as their associated orders without the need for human abstractors. The new CAP-approved Qualified Clinical Data Registry measures were designed and specified with these challenges in mind in hopes of minimizing them.
DEVELOPMENT OF QUALITY MEASURES
The CAP had 21 Qualified Clinical Data Registry measures approved by CMS for use in 2019 and 23 measures approved for use in 2020, including 20 of the original 21 measures. Herein we describe a subset of the individual Qualified Clinical Data Registry measures relating to the pathologist's role in the care of patients with hemolysis rates of blood draws, breast cancer, gastroesophageal adenocarcinoma, and oropharyngeal squamous cell carcinoma.
HEMOLYSIS RATE MEASURE
CAP10: Blood Laboratory Samples for Potassium Determination With Hemolysis Drawn in the Emergency Department: Percentage of blood laboratory samples for potassium determination drawn in the emergency department (ED) with hemolysis.
Hemolysis is the rupture of red blood cells resulting in the release of hemoglobin and other intracellular content into plasma. These components can interfere with accuracy and performance of multiple laboratory tests, including determination of potassium concentration. Although hemolyzed specimens may reflect the presence of hemolytic anemia, most cases are due to preanalytical errors related to suboptimal phlebotomy procedures or failure to follow procedures for collection, handling, and storage of the samples. Hemolyzed samples account for the majority of rejected laboratory samples, resulting in delays with test results and patient care decisions, further patient discomfort, and increased costs in the episode of care.4,5 The American Society of Clinical Pathology considers a hemolysis rate below 2% as best practice.5 One study6 found that factors that were statistically significant (P < .05) and associated with the highest rate of hemolysis included right-hand/forearm puncture site, 22-gauge intravenous catheters, difficult draws with multiple attempts, and small tubes (1.8 mL). Other associated factors included blue tubes, 6.0-mL tubes, resistance when aspirating blood using a syringe, and respiratory discharge diagnoses.6 Specimens originating from the ED also account for a large proportion of a hospital's laboratory-rejected specimens for hemolysis.5 Emergency departments are challenged by the need for rapid blood draws and laboratory results, high patient flow, and numerous involved providers, which increase the chance of a hemolyzed blood draw. Rejected laboratory specimens can lead to a stressful ED and clinical laboratory relationship, organizational economic waste, clinical inefficiencies, and patient dissatisfaction, making it essential to develop effective processes for systematically identifying unsuitable specimens, troubleshooting the potential causes, and maintaining good relations.
Gap Analysis and Related Guidelines
The intent of this measure is to lower hemolysis rates in samples collected from the ED (Table 1). A recent study by Dugan et al7 found an overall hemolysis rate of 12.8% and ultimate laboratory rejection rate of 3.7%. A report8 from the Cleveland Clinic states that the hemolysis rate from the ED averaged 13% and could be as high as 18.5% in 1 week. This was in stark contrast to specimens from other sites within the medical campus, whose hemolysis rate averaged 2.3%. Of note, following improvement in blood collection methodology, the hemolysis rate in Cleveland Clinic's ED dropped from 13% to 2%.8 However, this success story is not common. In a study done using the CAP's Survey/Q-Probes Program, Howanitz et al9 revealed that 55% (n = 465) of participants reported hemolysis rates of higher than 1%, with 20% reporting higher than 3%. Out of that same cohort of survey participants, 49% (n = 415) acknowledge taking corrective action during the last year in trying to reduce overall hemolysis rates at their institution. Unfortunately, when asked about the success of their action(s), 70% noted slow to no progress, and 2% had given up on the initiative.9
The majority of hemolyzed and subsequently rejected laboratory specimens are avoidable. This measure aims to improve blood draw practices, which could dramatically improve an organization's hemolysis rate. Reducing this preanalytical error can improve patient care by reducing the need for repeat draws and testing, improve overall turnaround time for results and subsequent management decisions, save in overall costs, and improve patient satisfaction. Accordingly, the intent of this measure aligns with CMS's “Patient Experience with Care” Meaningful Measure Area and with the National Quality Strategy to focus on effective clinical care.
BREAST CANCER MEASURES
CAP11: Accurate Human Epidermal Growth Factor Receptor 2 (HER2) Tumor Evaluation and Repeat Evaluation in Patients With Breast Carcinoma: Percentage of surgical pathology reports for breast carcinoma with appropriate human epidermal growth factor receptor 2 (HER2) breast tumor evaluation.
Breast cancer is one of the most common cancers, with an annual incidence of close to 270 000 in the United States.10,11 Patients with regional lymph node involvement or evidence of distant disease have lower 5-year survival rates compared with those patients with localized disease. Breast cancers with human epidermal growth factor receptor 2 (HER2) gene amplification are also associated with a more aggressive clinical course, with more rapid growth and spread of the disease. Determining HER2 status is essential for appropriate clinical management because these patients are much more likely to benefit from therapies that target the HER2 protein.12,13 Testing for HER2 gene amplification can be assessed either by in situ hybridization or HER2 protein overexpression as assessed with immunohistochemistry (IHC). A positive test result, via either methodology, remains the primary predictor of responsiveness to HER2-targeted therapies in breast cancer.14 For patients with breast cancer, an absent or inaccurate test interpretation of HER2 status can alter their treatment course and clinical outcomes while potentially increasing health care costs. HER2 gene amplification assessed by in situ hybridization or protein overexpression assessed by IHC remains the primary predictor of responsiveness to HER2-targeted therapies in breast cancer.14 HER2 gene amplification and/or protein overexpression occurs in up to 20% of breast cancers.15 The most recent American Society of Clinical Oncology (ASCO)/CAP guideline related to HER2 testing in breast cancer recommends testing be performed on all cases of breast cancer. It also refines the criteria for how to interpret the tests, report the results, and determine when repeat or alternate methodology should be performed.16
Gap Analysis and Related Guidelines
Data confirm less than optimal adherence to existing guidelines. A review of registry data12 found that only 86% of patients had detailed information regarding HER2 testing documented. These data suggest that, based on current breast cancer incidence, approximately 38 000 breast cancer patients may not have proper testing and/or documentation of HER2 status to inform clinical management, specifically chemotherapeutic options. Additionally, the recent ASCO/CAP guidelines updated the triggers for when additional testing should be performed in order to ensure proper test results and ultimately determine the most appropriate patient management. The intent of this measure is to ensure compliance with guideline recommendations for test interpretation and potential additional confirmatory testing (Table 2).
This breast cancer quality measure is intended to improve diagnostic accuracy and thereby appropriate patient management decisions because of reliable reporting of HER2 expression status. The measure will assess the compliance rate with the related guideline recommendations and in turn further promote the adoption of policies and procedures within pathology practices that ensure appropriate biomarker testing and reporting. Patient treatment for breast cancer depends on expression of HER2 status for targeted therapy, along with prognostic connotations of expression or lack of expression of the biomarker.
These approved measures have tremendous clinical relevance and improve delivery of patient care. Specifically, the HER2 measure furthers bolsters the National Quality Strategy focusing on “communication and care coordination.”
GASTROESOPHAGEAL ADENOCARCINOMA MEASURE
CAP12: Accurate Human Epidermal Growth Factor Receptor 2 (HER2) Tumor Evaluation and Repeat Evaluation in Patients With Gastroesophageal Adenocarcinoma (GEA): Percentage of patients diagnosed with GEA cancer (primary or metastatic) for which biopsies, resection, or metastatic specimens that have HER2 evaluation conducted using the current ASCO/CAP recommended manual system or computer-assisted system consistent with the optimal algorithm.
In 2019 the estimated number of new cases of esophageal cancer was 17 650 (1% of all new cancers) and the estimated number of related deaths was 16 080 (2.6% of all cancer deaths) in the United States.17 The estimated number of new cases of gastric cancer in the United States in 2019 was 27 510 (1.6% of all new cancers), with 11 140 deaths (1.8% of all cancer deaths).18 Clinically, GEA is often diagnosed at an advanced stage, resulting in a poor overall prognosis. However, when localized (stages II and III), GEAs are best treated with multimodal therapy, which currently results in a 5-year survival of approximately 40%. Once the GEA is considered advanced (defined as unresectable, clinically recurrent disease, or with evidence of metastasis), therapies are limited and palliative. A cure is extremely rare. More recently, a subset of GEA cases have been shown to possess HER2 gene amplification ranging from 7% to 38% in various studies.19,20 This fact presents an additional second- or third-line targeted therapeutic option when the genetic alteration is present. Accordingly, the National Comprehensive Cancer Network guidelines and the recent ASCO/CAP guidelines recommend a set of reporting criteria and testing algorithm that is different from what is used for breast carcinoma.21 All patients who have documented advanced GEA and who are considered good candidates for combination chemotherapy plus trastuzumab therapy should have their tumor tissue tested for HER2 overexpression and/or amplification. In patients with HER2-positive GEA, the addition of trastuzumab can increase the response rate and prolong progression-free and overall survival.
Gap Analysis and Related Guidelines
In 2012, ASCO and CAP convened an update committee to revise the guideline recommendations. In comparison with breast carcinomas, immunoreactivity for HER2 was reported to be more heterogeneous in GEA. HER2 expression in GEA is often basolateral and infrequently complete, in contrast to breast carcinomas, where completeness of membrane staining is a central feature of positive staining results.22,23 Because there are important distinct differences in HER2 expression, scoring, and outcomes in GEA relative to breast carcinoma, the update committee recognized the need for HER2 guidelines (that include critical clinical and laboratory considerations).20 The committee developed new algorithms for testing and recommended quality assurance monitoring that would make HER2 testing less variable and ensure more analytic consistency among laboratories. This quality measure aims to assess the degree of compliance with the most updated ASCO/CAP guidelines (Table 3).
This quality measure is intended to improve diagnostic accuracy and appropriate use of HER2 testing in GEA. These actions will positively impact patient management decisions related to properly reporting HER2 expression status. The measure will assess the compliance rate with the related guideline recommendations and in turn further promote the adoption of policies and procedures within pathology practices that ensure quality reporting. The measure has tremendous clinical relevance and specifically supports the National Quality Strategy of improving communication and care coordination and indirectly CMS's “Appropriate Use of Healthcare” meaningful measure focus.
OROPHARYNGEAL SQUAMOUS CELL CARCINOMA MEASURES
CAP20: High Risk HPV Testing and p16 Scoring in Surgical Specimens for Patients With Oropharyngeal Squamous Cell Carcinoma (OPSCC): Percentage of surgical pathology reports for invasive OPSCC with high-risk (HR) human papillomavirus (HPV) testing by surrogate marker p16 performed AND that include quantitative p16 IHC results.
CAP21: High-Risk HPV Testing in Cytopathology Specimens for Patients With Oropharyngeal Squamous Cell Carcinoma (OPSCC): Percentage of cytopathology reports from samples of known or suspected OPSCC or metastatic squamous cell carcinoma (SCC) of unknown primary that include HR-HPV testing status.
Human papillomavirus is a major cause of OPSCC and has contributed to its increased incidence.24 The current literature suggests that between 25% and 60% of head and neck cancers are associated with HR-HPV infection, with an increasing incidence in recent years.25 Human papillomavirus–positive OPSCC differs from HPV-negative OPSCC in that the latter is related to other risk factors, including alcohol and tobacco use. Additionally, HPV-positive OPSCC has an improved response to treatment and overall better prognosis.26 Therefore, it is crucial to determine the HPV status of SCCs of the oropharynx or metastatic SCCs suspected of having an origin in the oropharynx, as treating clinicians use this information when developing a treatment plan for patients, which may include less-aggressive treatment modalities. In the clinical setting, p16 IHC is an approach used to reliably diagnose HPV-induced OPSCC; it is felt to best stratify patient survival outcomes while also being practical and inexpensive.25 Furthermore, data suggest that the correlation between HPV positivity and p16 overexpression is highest when a threshold of 70% or higher staining for p16 in the tumor cells is applied.27 However, if p16 testing is not available or appropriate, determination of HPV status is of sufficient import to merit testing by any method deemed appropriate on known or suspected OPSCC samples.
Gap Analysis and Related Guidelines
Recently published guidelines recommend that HR-HPV testing be performed on all patients with newly diagnosed OPSCC, including all histologic subtypes. This testing may be performed on the primary tumor or on a regional lymph node metastasis when the clinical findings are consistent with an oropharyngeal primary.25 For oropharyngeal tissue specimens (ie, noncytology), pathologists should perform high-risk HPV testing via a surrogate marker—p16 IHC. Additional HPV-specific testing may be done at the discretion of the pathologist and/or treating clinician, or in the context of a clinical trial. Tumor staining of at least 70% nuclear and cytoplasmic p16 expression with at least moderate to strong intensity is a positive result and indicates the presence of HR-HPV25 (Table 4).
It is further recommended that pathologists routinely perform HR-HPV testing on patients with metastatic SCC of unknown origin when identified in a cervical upper or mid jugular chain lymph node. If positive, an explanatory note on the significance of the result is recommended.25 With head and neck fine-needle aspiration SCC samples, HR-HPV testing should also be performed on patients with known OPSCC or metastatic SCC of unknown primary but not previously tested. Unlike the specific recommendation for using p16 IHC on noncytology specimens, any testing methodology for HR-HPV testing is allowable for fine-needle aspiration (eg, IHC, in situ hybridization, or polymerase chain reaction). If the result on the fine-needle aspiration sample is negative, testing should be performed on tissue if it becomes available. If pathologists use cytology samples for p16 IHC testing, it is recommended that they validate the criteria (ie, cutoff) for a positive result25 (Table 5).
Patients with HPV-related OPSCC respond better to first-line therapies and have a better prognosis when compared with their HPV-negative counterparts. Therefore, these patients likely benefit from less-aggressive interventions. Based on these facts, recently published guidelines recommend that all patients with OPSCC be tested for HPV. These measures aim to assess compliance with the guideline recommendations and indirectly improve diagnostic accuracy and patient management. In fulfilling the requirements of this measure, pathologists will demonstrate how they further the National Quality Strategy related to communication and care coordination. This action further aligns with CMS's meaningful measures area focusing on Transfer of Health Information and Interoperability.
Pathologists play critical roles on the front lines of diagnosis to guide patient care in a timely and accurate fashion. It is therefore essential that pathologists perform the most appropriate test and document the results in a clear, concise manner in the pathology report. The quality measures described here illustrate the efforts of the CAP to promote adherence to evidence-based guidelines that drive effective and efficient diagnostic processes. Furthermore, these measures align with established meaningful measure areas and national quality priorities as set out by CMS in the MIPS program. Therefore, these metrics represent the highest standard in measurement to drive quality improvement in value-based payment programs.
The authors have no relevant financial interest in the products or companies described in this article.