HealthManagement, Volume 24/25 - Issue 6, 2025
Digital Clinical Quality Measures (dCQMs), powered by standards like HL7 FHIR and USCDI, will revolutionise chronic disease management. By integrating real-time clinical and claims data, dCQMs can identify care gaps at scale, supporting preventive care and improved outcomes. This approach requires a robust ecosystem of regulatory frameworks (e.g., TEFCA), advanced data standards and automated quality measures, enabling a shift from retrospective benchmarking to proactive patient care in value-based models.
Key Points
- dCQMs leverage real-time FHIR data to detect care gaps and improve chronic disease outcomes.
- Regulatory frameworks like TEFCA enable seamless clinical and claims data exchange and aggregation.
- USCDI sets high-level standards, aligning data exchange with HL7 FHIR for better interoperability.
- Automated CQL-based dCQMs scale chronic disease management by codifying best practice guidelines.
- Integrated ecosystems transform data into actionable insights for proactive value-based care.
Digital Clinical Quality Measures (dCQMs) have the potential to transform chronic disease management within a well-integrated healthcare ecosystem. By standardising data from across the care continuum in Fast Healthcare Interoperability Resources (FHIR) and combining it with rules derived from clinical quality measures like CQL (Clinical Quality Language), healthcare providers can identify and address care gaps in chronic disease management. While this sounds like a bold claim, this approach is critical for maintaining the affordability of healthcare systems in ageing societies. Over the past decade, the United States has made significant strides in building a disruptive ecosystem of technologies and regulations to enable this transformative change towards preventive care.
Part 1 – The Problem
According to the U.S. Centers for Disease Control and Prevention (CDC 2024), 90% of healthcare costs in the United States, totalling €4.3 trillion ($4.5 trillion), are directed towards chronic and mental health conditions. It represents a staggering €3.8 trillion ($4.05 trillion) annually, equating to roughly 16% of the U.S. GDP.
The top chronic diseases, according to the CDC, are heart disease and stroke, cancer, diabetes and obesity. Common to all four is that they are well researched, medications are readily available and effective if risk factors are identified early in the disease progression, and drugs are prescribed and taken regularly. As an example, statins are highly effective in preventing cardiovascular disease (CVD) in adults between 40 and 75 years old with at least one risk factor (U.S. Preventive Services Task Force et al. 2022). Knowing this, we could scan all 40–75-year-olds for CVD risk factors and, if they are not currently on a statin, flag them for consideration of a regular statin regime. It is way cheaper to detect these risk factors early, prescribe effective drugs such as statins and prevent health escalations such as heart attacks than to deal with the consequences of untreated chronic diseases such as strokes, terminal cancer and the effects of uncontrolled diabetes and obesity. The Centers for Medicare and Medicaid Services (CMS), one of the largest government health insurers in the world, has therefore declared it a priority to shift from fee-for-service models to value-based care by 2030 (Berger 2024). The transition to value-based care marks a significant shift towards prevention and primary care by incentivising providers to detect and manage chronic diseases early rather than focusing on treating costly complications. An important tool in this model is clinical quality metrics (CQMs), which offer insights into how effectively a provider's organisation manages its assigned population against a wide array of quality goals. For example, in the MIPS (Merit-Based Incentive Payment System) Star rating, providers are benchmarked based on their performance relative to top performers in their category and receive higher reimbursement if they are doing well (U.S. Centers for Medicare and Medicaid Services).
At this point, let’s postulate two hypotheses. First, the underlying problems described here, such as the high cost of chronic disease management versus the need to detect risk factors and prevent complications, are not unique to the United States but are common across most advanced societies with ageing populations. Therefore, similar trends could be observed in the UK, the Middle East, Australia and most EU countries. However, since health systems are funded differently worldwide, the shift towards value-based care is at varying stages of development. While the United States is not a pioneer, it has made significant progress supported by a variety of laws and regulations such as the Affordable Care Act of 2009, the 21st Century Cures Act of 2016 (U.S. Food & Drug Administration 2024) and subsequent regulations by CMS and the Office of the National Coordinator of Health IT (ONC) which merit close examination in this context. Second, since many countries share the problem, they could also share the solution. However, since healthcare data regulations vary significantly between nations —and even between states in countries like Germany —implementing regulatory and data ecosystems will require careful adaptation to each region's legislative framework.
Part 2 – Solution: Regulatory and Data Ecosystems
Step 1 – Constructing a longitudinal patient record
The most important ingredient for successful data analytics is the availability of high-quality and specific data. In the context of chronic disease management, this necessitates collecting data from across the care continuum. Depending on the specific chronic disease and potential comorbidities, a variety of care providers could be involved in patient care. To detect care gaps, it is important to gather data from all these providers, including some non-traditional ones that influence social determinants of health (AMA Ed Hub 2022).
The data required falls into two main categories: clinical data, collected during care delivery, and claims data, submitted to payers for reimbursement. Globally, healthcare systems vary significantly–single-payer systems that also manage care delivery, systems with single payers and private providers and those with multiple payers and providers. The United States, with its complex network of multiple payers (government, for-profit and non-profit) and diverse private providers (both for-profit and non-profit), has, over the last decade, developed a robust infrastructure to collect and integrate both clinical and claims data. This system is supported by a carefully designed regulatory framework aimed at improving care coordination:
- Trusted Exchange Framework, Common Agreement (TEFCA) is a regulatory framework designed to facilitate health data exchange. It enables different “onramps” or QHINs (Qualified Health Information Networks) to allow providers to exchange clinical data for treatment purposes. QHINs can include networks of Electronic Medical Records (EMRs) from vendors like EPIC or Cerner, as well as regional or national Health Information Exchanges (HIEs) such as the eHealth Exchange. Initially, TEFCA implementations utilised XML-based Consolidated Clinical Document Architecture (C-CDA) documents to exchange clinical summaries. However, the framework will switch to FHIR-based data exchange by 2025.
- Beneficial Claims Data API (BCDA) is an interface provided by CMS that allows providers to download claims data for CMS members in their care, already formatted in FHIR.
- ONC also made it mandatory for certified EMRs to provide Bulk-FHIR access since 2022, which is very helpful for population health data purposes. Bulk-FHIR allows the export of discrete data for a cohort of patients.
In summary, healthcare providers aiming to identify care gaps and improve chronic disease management can take several steps to leverage data effectively. They can download bulk FHIR-formatted data from their own EMR systems, combine it with claims data received via BCDA and utilise the TEFCA framework to request additional data from other providers. Providers within an Accountable Care Organisation (ACO)—groups that contract to deliver comprehensive care for a population—often agree to share data amongst themselves, utilising the same data exchange technologies such as C-CDA, FHIR or Bulk-FHIR.
While most of the data exchanged through TEFCA is in CDA format, CDA documents can also be parsed and transformed into FHIR, allowing data from different sources and formats to be integrated into a unified FHIR Server. With TEFCA’s planned transition to FHIR payloads, the need to exchange and then parse CDA documents will eventually be eliminated.
Once the data is normalised, cleansed and transformed, it can form the basis for creating a longitudinal patient record. This record, containing both clinical and claims data elements, provides a comprehensive, up-to-date view of the patient’s health. However, this is merely the foundation needed to mine for care gaps.
Step 2 – Creating a high-level data content standard with USCDI and FHIR
One of the drawbacks of clinical data aggregation in the past has been the inconsistency in the formatting and coding of HL7 v2 data. HL7 FHIR has addressed this challenge by creating a stricter standard with narrowly defined resources, enabling the exchange of specific data elements on demand and as required. This improvement also enhances the technology’s suitability for data aggregation.
In order to maximise the utility of such data for population health management, ONC also introduced the U.S. Core Data for Interoperability (USCDI) (HL7 International 2024). It defines the type of data that every certified EMR must record and share as discrete resources. These can be accessed through a FHIR API for individual patients, a Bulk FHIR API for groups or cohorts of patients, or as C-CDA XML documents, with “USCDI defining high-level data requirements and FHIR US Core providing detailed FHIR-based profiles for meeting those requirements.” (Health Level Seven International 2024) This means the ecosystem ensures the ability of its participants to share patient data, as well as adherence of the data to a stricter content standard, in this case, USCDI.
Over different generations, currently at version four, USCDI has incorporated more and more data to cover wider aspects of health, such as social determinants of health. For example, USCDI v4 introduces important elements for chronic disease risk assessments, including alcohol use, substance use, physical activity and average blood pressure (in addition to point-in-time measurements). As USCDI evolves, HL7 FHIR has adapted alongside it to ensure that all defined data elements can be exchanged between providers and, when necessary, between providers and payers. Other countries or regions can define their own core data standards for interoperability and likewise align their regional FHIR implementation accordingly.
Step 3 – Digital Clinical Quality Measures
The final component of the proposed chronic disease management solution is digital Clinical Quality Measures (dCQM). Clinical Quality Measures (CQMs) have a long history that can be traced back to the work of Florence Nightingale, Ernest Codman and Avedis Donabedian (Chun et al. 2014). More recently, in the 1980s, the National Care Quality Association (NCQA) developed a set of HEDIS™ measures. These were initially developed for payers to measure provider effectiveness based on claims data, formatted in the U.S. according to the ANSI EDI standard X-12 and required by HIPAA for provider-to-payer claims submission (Department of Health and Human Services 2009).
Adopting Peter Drucker’s famous principle, “If you can’t measure it, you can’t improve it”, CMS made HEDIS measures mandatory for Health Management Organisations (HMOs) in 1991. Later, this requirement was extended to Accountable Care Organisations (ACOs) participating in Medicare shared savings programs.
While these metrics codify best practices in chronic disease management and incentivise providers by awarding higher scores when a greater percentage of their population adheres to care guidelines, their original claims-based design posed limitations.
Claims data-based CQMs, often months old by the time they are processed, are better suited for performance measurement and reporting, such as determining Merit-Based Incentive Payment System (MIPS) rankings. However, this approach is less effective for actively managing individual patient care. The retrospective nature of claims data means it cannot support real-time adjustments to care plans, making it insufficient to directly manage patients against the very guidelines these measures promote.
This dynamic started to change very recently due to two major technology advancements. The first is the development of Clinical Quality Language (CQL), an ANSI standard championed by HL7 in 2020 (HL7 International 2020). CQL enables the standardised expression of clinical quality measure rules and the parsing of clinical data based on these rules. Tasks that previously required manual effort—such as determining whether a patient meets the inclusion or exclusion criteria for a measure and verifying if clinical requirements are fulfilled—can now be automated at scale.
The second advancement is FHIR data aggregation. Clinical data, when captured as part of a longitudinal patient record, can now be updated almost in real time, offering unprecedented data currency. Meanwhile, the claims data component of the longitudinal record can provide historical depth. For instance, determining whether a patient meeting risk criteria (age, gender, other risk factors) had a colon cancer screening within the past five years could be confirmed through a combination of recent FHIR clinical data and claims history.
This means that now hundreds of clinical quality measures can be digitalised with CQL and executed against an aggregated longitudinal patient record, thereby identifying care gaps at scale.
In 2023, the NCQA published a subset of their HEDIS™ measures in CQL for the first time. Committed to transitioning their entire measure set to CQL, the NCQA is also working on converting additional measure sets into CQL code (National Committee for Quality Assurance 2024). The measure sets align with the content requirements of the USCDI and are designed to parse FHIR resources as input.
With these advancements, we now possess all the necessary components to build a data ecosystem capable of aggregating clinical and claims data and constantly measuring this data against chronic disease management best practices expressed through CQM-based dCQMs. These core elements include:
- A framework and infrastructure for clinical and claims data aggregation. Tools like TEFCA, HIPAA and data use agreements within ACOs enable data exchange across the care continuum, whether for individual patients (using FHIR APIs or C-CDA documents) or patient cohorts (Bulk-FHIR). EMRs are mandated to support this interoperability.
- A high-level data content standard (USCDI) aligned with a data transfer and aggregation standard (HL7 FHIR).
- Codified chronic disease management standards. CQL enables the automation of chronic disease management standards, processing aggregated clinical and claims data (in FHIR format) at scale to identify care gaps for individual patients.
The combination of these technologies in a coordinated ecosystem allows us for the first time to identify care gaps in population cohorts at scale and in real time. This capability is instrumental in advancing preventive care goals, improving patient outcomes and accelerating the transition towards value-based care.
Conclusion
dCQMs, when applied to real-time clinical data, hold transformative potential for chronic disease management. By uncovering care gaps for individual patients within larger populations, they help providers to close these gaps, significantly improving health outcomes. However, achieving this requires the establishment of two types of ecosystems:
- A technical and regulatory framework for data exchange and aggregation. This framework must enable healthcare providers and payers across the care continuum to share and aggregate data. Data aggregation could occur within a group of providers (such as an Accountable Care Organisation), within integrated provider-payer networks (e.g. Kaiser Permanente, Clalit in Israel or the NHS in the UK) or at the payer level.
- Integration of robust data standards and aligned quality measures. High-level healthcare data requirements, such as those defined by USCDI, must align with modern standards for data exchange and aggregation, such as HL7 FHIR. Additionally, a comprehensive set of dCQMs should correspond to the data resources provided by these standards.
If such an ecosystem doesn’t exist or its essential components are missing, as have been in the past, technologies like HL7 FHIR alone will not be able to provide the data foundation of the required quality. Conversely, CQMs measured solely against claims data and/or limited to samples serve merely as tools for reporting and benchmarking but are not sufficient to guide providers in the detection of care gaps for individuals within the population. The integration of these two ecosystems is thus indispensable for attaining the full potential of dCQMs in chronic disease management.
Conflict of Interest
The author is involved in developing a turn-key solution for care gap detection based on FHIR data and dCQMs.
References:
AMA Ed Hub (2022) What Are Social Determinants of Health? [Accessed on 3 December 2024] Available from: edhub.ama-assn.org/health-systems-science/interactive/17498806?resultClick=1&bypassSolrId=M_17498806
Berger R (2024) Shift to Value-Based Care Model Is CMS' Top Priority. ALM Globest.com [Accessed on 3 December 2024] Available from: globest.com/2024/04/17/shift-to-value-based-care-model-is-cms-top-priority/?slreturn=20241114190531
Chun J, Bafford AC (2014). History and background of quality measurement. Clin Colon Rectal Surg., 27(1): 5-9.
Department of Health and Human Services (2009) Health Insurance Reform; Modifications to the Health Insurance Portability and Accountability Act (HIPAA) Electronic Transaction Standards. Federal Register, 74(11).
Health Level Seven International (2020) Clinical Quality Language (CQL) [Accessed on 3 December 2024] Available from: cql.hl7.org
Health Level Seven International (2024) US Core Implementation Guide [Accessed on 3 December 2024] Available from: hl7.org/fhir/us/core/uscdi.html#:~:text=ONC%E2%80%99s%20USCDI%20and%20FHIR%20US%20Core%20are%20complementary,in%20healthcare%20data%20exchange%20within%20the%20United%20States
National Committee for Quality Assurance (2024) HEDIS Digital Quality Measures [Accessed on 3 December 2024] Available from: ncqa.org/hedis/the-future-of-hedis/digital-measures/
U.S. Centers for Disease Control and Prevention (2024) Fast Facts: Health and Economic Costs of Chronic Conditions [Accessed on 3 December 2024] Available from: cdc.gov/chronic-disease/data-research/facts-stats/index.html
U.S. Centers for Medicare and Medicaid Services. Performance information for doctors & clinicians. [Accessed on 3 December 2024] Available from: medicare.gov/care-compare/resources/physician/performance-scores
U.S. Food & Drug Administration (2024) 21st Century Cures Act Deliverables [Accessed on 3 December 2024] Available from: fda.gov/regulatory-information/21st-century-cures-act/21st-century-cures-act-deliverables
U.S. Preventive Services Task Force, Mangione CM, Barry MJ et al. (2022) Statin Use for the Primary Prevention of Cardiovascular Disease in Adults: US Preventive Services Task Force Recommendation Statement. JAMA, 328(8): 746–753.