Since 2000, several organizations have promulgated and often required reporting of performance measures focusing on venous thromboembolism (VTE) and related conditions. As the science of performance measurement has evolved, there have been substantial changes to VTE measurement which may not be well understood by front-line hospital providers and staff.

This article aims to provide an overview of VTE-related performance measurement activities including:

How the measures were developed and used by various stakeholders

Though a comprehensive patient-centered approach to VTE prevention addresses care across the continuum of care settings (e.g., transitions to ambulatory and post-acute care settings), this article focuses on the inpatient hospital setting where the majority of performance measurement activities have occurred. It expands on the work of Michota et al. by addressing measure changes over time.¹

Much of the measure implementation information comes from The Joint Commission’s experience in VTE measure development and/or published literature. Initial and last reported measure rate information was obtained by downloading the data sets from the CMS Hospitals archived data snapshots website.²

VTE is a common preventable hospital complication, frequently extending the length of stay and increasing hospital mortality.^3–5 Each year in the United States, it is estimated that VTE, including deep vein thrombosis (DVT) and pulmonary embolism (PE), affects as many as 900,000 people, is responsible for up to 300,000 deaths, and is associated with healthcare costs of approximately $10 billion.^6,7 Hospital-associated venous thromboembolism (HA-VTE) is often preventable but VTE prevention strategies, including risk assessment and appropriate prophylaxis, are not applied uniformly or systematically across US hospitals and healthcare systems.⁸

Performance measures indicate the level to which care processes conform to established care standards including clinical guidelines.⁹ When standardized within or across hospitals, performance measures can be used:

To assess organization-level baseline performance and track changes over time, including changes following a quality improvement intervention or to compare performance between specific patient populations or units.¹⁰

Although there are many benefits to using performance measures, publicly reported performance measures can have adverse effects, especially when rewards and penalties are applied.¹² For example, VTE prophylaxis measures may encourage inappropriate anticoagulation in low-risk patients leading to unintended bleeding events when measures do not incorporate risk assessment.¹³ Gaming can occur when staff influence measured performance by intentionally altering factors other than care processes, such as when adding medical record documentation after discharge (e.g., reason for no prophylaxis) intending to ensure passing of a measure.^11,14 Also, the strength of the evidence linking greater hospital-level use of prophylaxis to lower rates of VTE in certain patient populations has been questioned.^15,16 The effectiveness of applying financial incentives and penalties to performance measure data remains controversial.^17–19

Over the years, several major US stakeholders have engaged in VTE measure-related activities, including the National Quality Forum, the Centers for Medicare and Medicaid Services, The Joint Commission, and the Agency for Healthcare Research and Quality.

In 2003, the National Quality Forum (NQF), a not-for-profit membership-based organization, convened an expert steering committee to identify consensus standards and measures for public reporting that would advance VTE prevention and care. Shortly thereafter, NQF released a seminal report which included a framework for the relationship among policies, practices, and performance measures as well as an initial set of policy and practice statements, 17 key characteristics of preferred practices, and two measures of VTE prophylaxis for surgery patients.⁹ The report called for the use of multidisciplinary teams to determine VTE prevention and care activities that address four domains:

Risk assessment: Standardized, externally validated VTE risk assessment models (RAMs) are tools that can promote comparison across different units within hospitals as well as across hospitals.

In approximately 2009, the field of performance measurement transitioned to an electronic data collection approach as a result of the American Recovery and Reinvestment Act (ARRA) and the Health Information Technology for Economic and Clinical Health (HITECH) Act. ARRA included a mandate to incentivize healthcare providers to use an electronic health record. HITECH promoted the adoption and meaningful use of health information technology and encouraged the development of health information systems through the establishment of standards and requirements for the electronic transmission of certain health information.²⁸

The collection of performance measure data electronically started with developing electronic clinical quality measures (eCQMs). CMS describes eCQMs as tools that help measure and track the quality of health care services that eligible hospitals and critical access hospitals (CAHs) provide, as generated by a provider’s electronic health record (EHR).²⁹

In 2010, The Joint Commission began to re-engineer the chart abstracted VTE measures as eCQMs. VTE-1 and VTE-2 were developed as eCQMs (eVTE-1 and eVTE-2) and trialed by hospitals on a voluntary basis in 2015. CMS implemented eVTE-1 and eVTE-2 in 2016 as part of the IQR program. As of January 2023, eCQM data are publicly available from CMS at the hospital level.³⁰ In alignment with CMS, The Joint Commission ORYX performance measurement requirements for 2023 include submitting a minimum of four eCQMs from a list of 16 available eCQMs, which includes VTE eCQMs. As of June 2023, The Joint Commission has not publicly reported any eCQM data on its website.³¹

Beginning in 2020, the concept of electronic measurement evolved to encompass digital quality measures (dQMs).³² dQMs accommodate data from electronic sources beyond the EHR (e.g., registries, case management systems). dQMs use structured, coded data to pull health information that can be electronically transmitted via interoperable systems. An eCQM is one example of a dQM.³³ Eventually, the intent is to evolve from semi-automated to fully automated electronic measures.

In 2007, the CDC’s Paul Coverdell National Acute Stroke Registry partnered with the Joint Commission and the American Heart Association/American Stroke Association to develop and release stroke performance measures.³⁹ The VTE-related measure used in this program relates to the timeliness of VTE prophylaxis as described in Table 1 (STK-1). This same measure is used by The Joint Commission in its Disease-Specific Care (DSC) stroke center certification programs.⁴⁰ Certified primary stroke centers, thrombectomy-capable stroke centers, and comprehensive stroke centers are required to collect and submit data monthly for this measure to achieve and sustain these certification designations.

The American College of Surgeons (ACS) stewards a hospital based, risk and case-mix-adjusted measure related to DVT and PE to monitor clinical outcomes as defined by the ACS National Surgical Quality Improvement Program for adult surgical patients 65 years of age and older. The American Academy of Neurology (AAN) developed a VTE measure related to stroke and stroke rehabilitation, and the American Society of Plastic Surgeons (ASPS) developed a VTE measure related to perioperative care.⁴¹

Recently, CMS contracted with IMPAQ International, LLC research services to develop new eCQMs about harm to hospitalized patients for possible use in CMS programs. A medication-related bleeding measure is under modification following the completion of initial testing.⁴² Another measure targeting surgical patients who experience harm related to a postoperative VTE during hospitalization is under development.⁴³

Despite the large number of measures and their widespread use nationally, there remain several domains of VTE-related care that are not currently being addressed.⁹ Examples of gaps in measuring performance include VTE and bleeding risk assessment at admission, changing risk levels during hospitalization, surveillance for hospital-associated VTE events, appropriateness of treatment, continuity of care at discharge to other settings, and patient medication adherence over time. Regarding risk assessment, though more than 15 externally validated VTE and bleeding RAMs exist, there is currently no consensus on which VTE RAMs are preferred in the United States.⁴⁴ Incorporating validated risk assessment models into clinical decision support systems should facilitate accurate and less burdensome measurement of the appropriateness of prophylaxis and enable linkage to VTE outcomes.^45,46 Another domain for future measures is diagnostic testing. Testing to confirm a VTE diagnosis is a necessary component for future measure development related to therapeutic management.

Beyond topic area gaps, most existing measures reflect processes rather than outcomes. Though process measures may be easier to interpret and may not require risk adjustment, outcome measures are needed to validate the appropriateness of the processes.⁴⁷ However, outcome measures need to be carefully constructed to minimize surveillance bias.⁴⁸

Others are exploring the development of patient-reported outcome-based performance measures (PRO-PM). A PRO-PM “is a way to aggregate the information from patients into a reliable, valid measure of performance at the measured entity level”; it is unclear whether these measures have been developed for VTE-related topics.⁴⁹ Artificial intelligence (AI) presents an opportunity to proactively predict and prevent VTE. Researchers have applied a variety of AI methods to predicting differing types of VTE; however, guidance for implementing AI models within clinical practice remains immature and policymakers are actively considering frameworks for safe and ethical use of AI in health care.^50,51

One key step toward enhancing VTE measures nationally is prioritizing what is most important to measure. This often requires a consensus process among stakeholders. To be considered for national use, performance measures must meet criteria established for measure endorsement including (a) important to measure, (b) scientifically acceptable, (c) feasible to collect, and (d) useable and relevant.⁵² After development, a measure must undergo preimplementation testing.⁵³ Ongoing maintenance and evaluation are essential steps to ensure continued relevance and assess the impact and usefulness of measures over time.

Measure implementation considerations are paramount to minimize the burden of data collection and maximize the effectiveness of using the measures for improvement. CMS recognized the burden and misalignment potential that multiple performance measurement requirements can create and proposed a new approach to aligning measures across its more than 20 quality programs. The Universal Foundation, based on a “building block” approach, will address gaps and increase the use of outcome, patient-reported, and digitally reportable measures.⁵⁴ It is not yet known whether the Universal Foundation effort will address VTE-related measures.

CMS will also leverage the Fast Healthcare Interoperability Resources (FHIR) technology which emphasizes interoperability in data collection for digital quality measures using a standardized terminology, a data model, and a transport mechanism.⁵⁵ FHIR is the modern, internet-based approach selected by the Office of the National Coordinator (ONC) to transport data between various IT applications. As of December 2022, all ONC-certified IT developers must provide their customers with FHIR-based application programming interfaces (APIs).⁵⁶ The next phase of implementation will require healthcare organizations to exchange quality measure information in FHIR. The Joint Commission has since translated the VTE eCQMs to FHIR for this purpose.

In collaboration with CMS, CDC has begun evaluating the feasibility of implementing FHIR-based healthcare-associated VTE surveillance through the CDC’s National Healthcare Safety Network (NHSN), the nation’s largest tracking system for healthcare-associated infections. The NHSN is developing FHIR-based digital patient safety measures that include the topic of VTE.⁵⁷ This effort has the potential to become a national surveillance system for VTE that captures process and outcome performance measure data across the continuum of care settings.

This article describes the evolution of data collection and uses for hospital-level VTE-related performance measures over time. Additionally, it is intended to stimulate thinking regarding future measurement options and activities.

Perhaps the most substantial change in VTE measurement has been the shift from chart-based measures to digital measures. The extent to which shifting to digital quality measures will improve measure accuracy and reduce the burden remains to be determined. The usefulness of current and future measures depends on how interpretable the data are and how sensitive the measures are to changes in care – the most important measure of usefulness for quality improvement. Of course, perceived usefulness will vary depending on who is using the measures and for what purpose; history has shown it is impossible for individual measures to satisfy all possible users.⁵⁸

Though this article has focused on standardized measurement for external entities, it is important to note the value of measurement for internal quality improvement. Measuring performance within hospitals over time is integral to a promising approach for improving performance known as anticoagulation stewardship, which is modeled after major activities involved in antimicrobial and diagnostic stewardship. In 2022, the NQF together with The Anticoagulation Forum released a guidance document listing core elements for anticoagulation stewardship which include securing administrative leadership commitment, establishing accountability with a multidisciplinary approach, measuring performance, and implementing reliable care processes across the continuum.⁵⁹

Though measurement of VTE-related performance has been in the national spotlight for many years, there remain gaps in processes being measured and several opportunities for improvement. Examples of topic areas for future measurement include risk assessment, appropriateness of prophylaxis and treatment, as well as continuity of care across settings. The transition to digital quality measures has the potential to incorporate greater clinical detail with less data collection burden while addressing more topic areas with enhanced specificity. Additional research is needed to demonstrate the accuracy of data collection and the value of using these measures in public reporting. Innovative initiatives are needed to progress toward the goal of improving patient safety and health outcomes for all at risk of VTE.