Blog

Heart Failure and SGLT2 Inhibitors: Access Has Improved, Utilization Has Not.

BenefitsPRO just headlined the recent <a data-mce-href="https://www.benefitspro.com/2026/03/10/bcbs-research-shows-hospitals-ai-systems-inflate-health-care-costs/?kw=BCBS%20research%20shows%20hospitals%27%20AI%20systems%20inflate%20health%20care%20costs&amp;utm_source=email&amp;utm_medium=enl&amp;utm_campaign=newsroomupdate&amp;utm_content=20260311&amp;utm_term=bpro&amp;user_id=4c7c55e6ebba5df2d8e44eaae546775b11b67dc4ba5dada7079f3d7824421666" href="https://www.benefitspro.com/2026/03/10/bcbs-research-shows-hospitals-ai-systems-inflate-health-care-costs/?kw=BCBS%20research%20shows%20hospitals%27%20AI%20systems%20inflate%20health%20care%20costs&amp;utm_source=email&amp;utm_medium=enl&amp;utm_campaign=newsroomupdate&amp;utm_content=20260311&amp;utm_term=bpro&amp;user_id=4c7c55e6ebba5df2d8e44eaae546775b11b67dc4ba5dada7079f3d7824421666">Blue Cross Blue Shield Association analysis</a> revealing that AI-driven hospital coding contributed to a 1.8% increase in claims costs has ignited a critical debate: Are hospitals using AI to inflate bills, or are these systems finally capturing clinical complexity that manual processes systematically missed for decades?The answer is likely both, and the implications point toward a fundamental reckoning with how we structure healthcare payment in the digital age.<h3>The Historic Underdocumentation Problem</h3>Manual clinical coding has long suffered from profound accuracy limitations. Research demonstrates error rates of 40-70%, with approximately 26% of diagnoses completely undercoded in manual processes. Concordance between physicians and coders for the most responsible diagnosis can be as low as 20-35%.But these aren't just statistics about code accuracy. They reflect a fundamentally broken workflow. Physicians provide care and document clinical findings as best they can within crushing time constraints. Days or weeks later, medical coders asynchronously review charts and attempt to translate clinical narratives into billing codes, often submitting clarification queries back to already-overwhelmed clinical teams. This process is fraught with errors, gaps, and behavioral adoption hurdles at every step.The barriers are well-documented: illegible handwriting, variability in documentation quality, time constraints, and the fundamental limitation that coding professionals cannot interpret clinical nuance. They can only code what is explicitly documented. The American College of Physicians noted that traditional E guidelines transformed documentation from "what was done" to "what was documented," creating systems where "much of the documentation includes often irrelevant elements" rather than actual clinical decision-making.<h3>AI's Double-Edged Sword</h3>AI-powered ambient documentation systems demonstrate genuine efficiency gains. Recent meta-analyses show these tools reduce documentation burden with a standardized mean difference of -0.71, representing a moderate effect. Documentation time decreases by 46% for complex cases, with F1-scores of 0.62-0.85 for ICD coding accuracy.More importantly, AI collapses the asynchronous physician-coder workflow by capturing clinical conversations in real-time and generating both clinical notes and billing codes simultaneously. This eliminates the communication gaps, query cycles, and weeks-long delays that plague manual coding and further drag out a rather long revenue cycle.However, the BCBS analysis identified a concerning pattern: at 10% of hospitals, postpartum anemia diagnoses surged from 4% to 12.3% while blood transfusion rates remained constant at 1%. This suggests coding that may not reflect treatment-relevant clinical acuity. When one insurer audited a hospital with such spikes, "less than 20% of the cases met established clinical criteria for postpartum anemia."Yet this doesn't necessarily indicate fraud. It could represent AI systems capturing diagnoses that were clinically present but previously undocumented due to the workflow barriers described above. These are conditions mentioned in conversation but never making it into the formal record because physicians lacked time or coders lacked context.The critical question: are we trading one set of systematic errors (underdocumentation due to workflow friction) for another (overdocumentation of clinically irrelevant findings)? AI removes the human coder as a quality check, but that "check" was itself producing 40-70% error rates.<h3>The Fundamental Limitation: A Broken Workflow Built on Inadequate Codes</h3>Even when physicians document thoroughly and coders work diligently, CPTs, DRG, HCPCS, andICD-10 codes themselves were designed for billing and statistics, not differential diagnosis or comprehensive clinical documentation. They lack the compositional structure and semantic richness needed to distinguish between:<ol><li>Legitimate capture of previously undocumented complexity</li><li>Documentation of clinically present but treatment-irrelevant findings</li><li>Actual inappropriate coding</li></ol><a data-mce-href="https://www.snomed.org/" href="https://www.snomed.org/">SNOMED CT</a> offers a solution. As a compositional system, it can represent complex concepts by combining discrete observations: "acute inflammation" + "perforation" + "appendix structure" = acute perforated appendicitis. It provides "consistent and computable framework" that ICD fundamentally lacks. The whole range of medical terminology is covered by SNOMED CT in ways that <a data-mce-href="https://www.who.int/standards/classifications/international-classification-of-health-interventions" href="https://www.who.int/standards/classifications/international-classification-of-health-interventions">ICD-10 and ICHI</a> cannot achieve.This semantic precision could enable payment systems to distinguish between a diagnosis that was documented because it required treatment versus one that was simply observed during an encounter. Current ICD-based payment models cannot make this distinction, regardless of whether humans or AI generate the codes.<h3>The Path Forward: FHIR, SNOMED, and Payment Evolution</h3>The federal government is already moving in this direction. CMS's Interoperability and Patient Access Final Rule (January 2024) identified FHIR as the chosen standard for interoperability and required payers to establish Patient Access APIs using FHIR. The HTI-2 proposed rule extends this framework to create greater transparency across healthcare delivery, payment, and public health domains.FHIR combined with SNOMED-CT achieves both structural and semantic interoperability, with studies demonstrating 0% data loss in transfers between systems. This granularity could enable payment models that distinguish between documentation of clinical findings and documentation of treatment-relevant complexity, the very distinction that current billing codes cannot make.The transition won't be simple. A 2025 scoping review identified 73 challenges to FHIR adoption across organizational, technical, individual, data management, and legal/ethical domains. Yet 43 facilitators were also identified, and federal investments through CDC's Public Health Infrastructure Grant are funding implementation centers to provide technical assistance.<h3>Where the Burden Falls: Payers, Not Providers</h3>Here's the critical asymmetry in this transition: providers have already done much of the heavy lifting, but perhaps the data assymetry is at the root of problem in payer-provider abrasion. Modern EHR systems already support FHIR, SNOMED, and LOINC standards. The clinical data is being captured in these richer formats today. The infrastructure exists on the provider side.The real transformation must happen on the payer side. Actuaries will need to develop entirely new frameworks for understanding risk and pricing when working with compositional ontologies rather than categorical billing codes. Contracting models built around DRGs, CPT bundles, and HCPCS will require fundamental restructuring. CMS and state Medicaid agencies will need to lead foundational shifts in how they define, measure, and reimburse clinical complexity.This is not a small undertaking. Decades of actuarial science, contract templates, and regulatory frameworks are built on the assumption that clinical reality can be adequately represented by billing codes. Unwinding that assumption while maintaining payment system stability will require careful, phased implementation.But the key insight is that this is primarily a payer-side and regulatory transformation, not another massive provider rebuild. Providers have already invested billions in EHR infrastructure that captures clinical data in modern standards. The question is whether payment systems will evolve to use what's already being captured, or continue forcing that rich clinical data through the narrow funnel of billing codes.<h3>The Approaching Breaking Point</h3>We are indeed approaching a breaking point where coarse-grained administrative datasets and ontologies used for payment have underrepresented clinical data for years. <blockquote>AI hasn't created this problem. It has exposed it by eliminating the workflow friction that previously masked the inadequacy of billing codes for capturing clinical reality.</blockquote>The question isn't whether to embrace AI-assisted documentation, but how to evolve payment systems to handle the clinical granularity that modern technology can now capture. The asynchronous physician-coder workflow was never a feature. It was a workaround for inadequate documentation tools and insufficient physician time. AI removes that workaround, forcing us to confront what we're actually trying to measure and pay for.In my mind, the evolution toward FHIR/SNOMED/LOINC-based payment models represents the next horizon. These standards can provide the semantic precision needed to validate AI-assisted coding accuracy while preserving the efficiency gains that reduce clinician burnout and improve documentation quality.The alternative is continuing to force increasingly sophisticated clinical documentation through billing codes designed in an analog era, whether generated by humans or AI. This will only deepen the disconnect between clinical reality and payment systems. <h3>The workflow was already broken. AI is just making it impossible to ignore.</h3>

New BCBS Analysis on AI-Powered Clinical Documentation: Upcoding Crisis or Long-Overdue Accuracy?

<h3>The Strategic Alignment of Professional Organizations with Payer Value Frameworks</h3>Over the past two posts (<a data-mce-href="https://erikabel.org/blog/quality--the-untapped-potential-in-strategically-aligning-professional-organizations-with-payer-value-frameworks---part-1-of-3" href="https://erikabel.org/blog/quality--the-untapped-potential-in-strategically-aligning-professional-organizations-with-payer-value-frameworks---part-1-of-3">Part 1</a> and <a data-mce-href="https://erikabel.org/blog/quality--the-translation-gap-----and-why-fhir-changes-everything---part-2-of-3" href="https://erikabel.org/blog/quality--the-translation-gap-----and-why-fhir-changes-everything---part-2-of-3">Part 2﻿</a>), I've argued that our national quality measurement infrastructure is structurally broken, 1) built on fragmented governance, claims-based data, and marginal incentives and, 2) that FHIR-based interoperability now makes a fundamentally better approach possible.This final post is about what professional organizations should actually do about it.The stakes are real. Professional societies that lead this transition will define how quality is measured and rewarded in their specialties for the next generation. Those that wait will find their clinical standards continue to be interpreted, approximated, and occasionally misrepresented by payers and measurement bodies operating without their input.<h3>Integrate Economic Value Into Guidelines … Systematically</h3>The most powerful thing a professional organization can do to align with payer value frameworks is to speak in the language payers actually use....cost-effectiveness.This doesn't mean subordinating clinical judgment to economic metrics. It means acknowledging that health systems and payers are making allocation decisions whether or not clinical guidelines inform them, and that specialty societies with the deepest clinical evidence base are uniquely positioned to provide that guidance credibly.The model for this exists and it's already in practice for some. The <a data-mce-href="https://professional.heart.org/-/media/PHD-Files/Guidelines-and-Statements/2025-Guideline-Methodology-Manual-November-2025.pdf" href="https://professional.heart.org/-/media/PHD-Files/Guidelines-and-Statements/2025-Guideline-Methodology-Manual-November-2025.pdf">ACC/AHA 2025 Guideline Methodology Manual (see page 9)</a> explicitly incorporates cost and value analysis as a formal component of clinical practice guideline development. Cost-effectiveness is not an afterthought layered onto finished recommendations, and instead it is embedded in the methodology from the outset, with standardized thresholds, structured value classifications, and a framework for rating the certainty of economic evidence alongside clinical evidence. That is a meaningful structural commitment, and it sets a bar that other specialty societies should study and build toward.The <a data-mce-href="https://www.acponline.org/acp-newsroom/acp-urges-a-unified-approach-and-national-focus-on-important-and-common-clinical-conditions-and" href="https://www.acponline.org/acp-newsroom/acp-urges-a-unified-approach-and-national-focus-on-important-and-common-clinical-conditions-and">American College of Physicians</a> has taken a complementary approach on the measurement side. ACP has been publicly vocal about the need to focus quality measurement on a smaller number of high-impact, evidence-grounded conditions — arguing that not everything that can be measured should be, and that the current proliferation of metrics imposes real costs without proportional clinical benefit. Their call for a unified national focus on the most important and common clinical conditions is exactly the kind of institutional position that can shift the conversation at the federal and payer level.What it requires is methodological commitment to treat economic evidence with the same systematic rigor as clinical evidence, establishing reference standards for future analyses, and being willing to produce transparent value classifications even when the results are inconvenient. This also aligns with CMS approaches around HEOR evidence required of the industry to attain new billing codes (e.g., HCPCS, CPT, etc.)<h3>Develop FHIR-Based Quality Measure Specifications</h3>This is where the technical opportunity of Part 2 becomes organizational strategy.Professional organizations should move toward developing FHIR implementation guides for their specialty to define the data elements required for accurate population identification, authoring measure logic in Clinical Quality Language (CQL) that can be executed directly against FHIR repositories and creating payer-ready documentation that eliminates the interpretation gap.The <a data-mce-href="https://codex.hl7.org/" href="https://codex.hl7.org/">CodeX oncology model</a> is instructive where ASCO and ASTRO partnered with technical stakeholders to produce executable quality specifications that payers could adopt without translation. The result was more accurate measurement, less administrative burden, and clinical organizations maintaining control over how their specialty's quality was defined and assessed.This is not work that most specialty societies have traditionally done. It requires informatics capacity, technical partnership, and a willingness to engage deeply with implementation rather than simply publishing specifications and stepping back. But it is the work that actually closes the translation gap and the organizations willing to do it will hold enormous influence over how clinical quality is operationalized in payer contracts and federal value-based programs.<h3>Couple Economic Value With Recommendation Strength</h3>One of the most transformative moves available to professional organizations is to link economic value directly to recommendation classification so that the strength of a clinical recommendation reflects not just clinical evidence but cost-effectiveness evidence as well. This also creates research anchor points where known-unknowns are cleanly evaluated and are set up for active research and validation.This creates something the current system almost entirely lacks - market incentives embedded in guideline structure. When a Class I recommendation requires both strong clinical evidence and cost-effectiveness at an accepted threshold, it changes what manufacturers, health systems, and payers do. Manufacturers have incentives to price interventions in alignment with the value they generate. Payers have a credible, clinically grounded basis for coverage decisions. Clinicians have guidance that integrates both the science of benefit and the reality of resource allocation.This is transformational, not incremental. It moves professional organizations from publishing guidelines that payers interpret inconsistently, to shaping the economic architecture of care delivery in their specialty.<h3>Focus on the Vital Few</h3>One of the clearest lessons from two decades of quality measurement failure is that proliferation is not a strategy. Analysis of AHRQ's quality indicators found that seven of them account for 93% of measurable population health benefit. Yet measurement programs have continued to expand, imposing burden without proportional value.Professional organizations should resist the temptation to be comprehensive. The goal is not to measure everything, but rather &nbsp;to measure what matters, measure it accurately, and create incentives that drive meaningful change.This means actively retiring low-value process measures that do not link to improved outcomes. It means concentrating development resources on the vital few measures with demonstrated population health impact. It means preferring outcome measures over process proxies. And it means being willing to say (publicly and credibly) that measuring certain things is not worth the administrative cost of doing so.<h3>Build Toward Equity From The Start</h3>Value-based payment has enormous potential as a lever for addressing health equity and as a mechanism that inadvertently entrenches disparities, if equity isn't built into program design from the outset.Professional organizations should embed health equity considerations into both guideline development and quality measure design: incorporating differential effectiveness across populations into cost-effectiveness frameworks, structuring data elements to enable equity analysis, adjusting for social determinants of health in risk models, and including patient priorities beyond just clinical endpoints to define what quality means.This is both the right thing and the strategically smart thing. Payers and CMS are increasingly focused on health equity as a dimension of quality. Professional societies that develop credible, clinically grounded equity frameworks will find receptive audiences in exactly the policy and payer stakeholders they need to influence.<h3>Engage Payers as Technical Partners</h3>The old model of professional organization-payer interaction is largely adversarial or transactional where societies publish guidelines, payers develop coverage policies, gaps emerge, where utilization management and appeals processes fill in the cracks. The new model must be different.FHIR-based quality measurement only succeeds if payers can implement it. That requires professional organizations to do something unfamiliar…..sit with payers in technical implementation conversations, pilot clinical data-based quality calculation in real contracting environments and create feedback loops between implementation experience and specification refinement.This is not about giving payers control over clinical standards. It's about ensuring that the specifications professional organizations develop are actually implementable and that when implementation barriers emerge (and they will), they get resolved through clinical input rather than administrative approximation.<h3>A Three-Phase Strategic Roadmap</h3>Phase 1 — Foundation (Years 1–2): Establish quality measurement infrastructure grounded in clinical data, not claims. Critically assess existing measure portfolios and retire low-value indicators. Begin FHIR profile development by mapping specialty-specific clinical data elements to FHIR resources leveraging and stewarding Value-Set Authority Center (VSAC) frameworks. Establish connections with TEFCA-qualified health information networks.Phase 2 — Integration (Years 2–4): Incorporate cost-effectiveness systematically into guideline development. Author FHIR-based quality measures using CQL. Pilot clinical data-based quality calculation with early-adopter payers. Develop payer-ready implementation guides. Generate economic evidence for priority clinical areas.Phase 3 — Transformation (Years 4–5+): Couple economic value with recommendation strength across the guideline portfolio. Scale FHIR-based measurement across payer contracts. Advocate for CMS adoption of FHIR-based measures in federal value-based programs. Establish continuous improvement feedback loops between clinical data, quality measurement, and guideline updates.<h3>What's Actually At Stake</h3>The quality measurement system did not fail because of bad intentions. It failed because of structural misalignment including measurement authority held by bodies without clinical ownership, data systems inadequate to the specifications being enforced, and incentives too small to drive meaningful change.Professional organizations have the clinical authority, the registry infrastructure, the guideline ownership, and the stakeholder trust to build something fundamentally different. They have every incentive to do so because the alternative is continued delegation of clinical quality definition to organizations and data sources that cannot represent what the specialty actually knows about what constitutes good care.The paradigm shift is clear enough shifting from professional organizations developing quality metrics that payers interpret using claims data, to professional organizations developing FHIR-based quality measure specifications that payers implement directly using clinical data. The technology is ready. The regulatory environment is aligning. The infrastructure is being built.The only question is which professional organizations will lead…. and which will be led.&nbsp;----------------------------------------------------------------A note on who's doing this work: <a data-mce-href="https://kisersolutions.com/" href="https://kisersolutions.com/">Kiser Healthcare Solutions (KHS)</a> is actively operating at the crossroads of digital quality, professional organizations, and value-based care helping professional organizations, payers and providers advance quality without adding administrative burden or questionable ROI. Their focus on FHIR interoperability and data-driven strategy reflects exactly the kind of implementation partner professional organizations need as they make this transition. Worth knowing about: kisersolutions.comSpecial thanks to <a data-mce-href="linkedin.com/in/kristyneroeger/" href="linkedin.com/in/kristyneroeger/">Kristyne Reoger</a> and <a data-mce-href="https://www.linkedin.com/in/jenniferkbitterman/" href="https://www.linkedin.com/in/jenniferkbitterman/">Jennifer Bitterman</a> for their contributions to this piece.----------------------------------------------------------------&nbsp; This is the final installment of a three-part series on professional organizations and payer value alignment. <a data-mce-href="https://erikabel.org/blog/quality--the-untapped-potential-in-strategically-aligning-professional-organizations-with-payer-value-frameworks---part-1-of-3" href="https://erikabel.org/blog/quality--the-untapped-potential-in-strategically-aligning-professional-organizations-with-payer-value-frameworks---part-1-of-3" target="_blank" rel="noopener">Part 1 ﻿</a>examined the structural failures of the current quality measurement system. <a data-mce-href="https://erikabel.org/blog/quality--the-translation-gap-----and-why-fhir-changes-everything---part-2-of-3" href="https://erikabel.org/blog/quality--the-translation-gap-----and-why-fhir-changes-everything---part-2-of-3" target="_blank" rel="noopener">Part 2﻿</a> explored the translation gap between clinical guidelines and payer actuaries, and the emerging FHIR/TEFCA infrastructure that makes a different approach possible.

Quality: A Potential Strategic Roadmap for Professional Organizations Ready to Lead | Part 3 of 3

<h3>The Strategic Alignment of Professional Organizations with Payer Value Frameworks</h3>&nbsp; In <a data-mce-href="https://erikabel.org/blog/quality--the-untapped-potential-in-strategically-aligning-professional-organizations-with-payer-value-frameworks---part-1-of-3" href="https://erikabel.org/blog/quality--the-untapped-potential-in-strategically-aligning-professional-organizations-with-payer-value-frameworks---part-1-of-3">Part 1</a>, I described how our national quality measurement infrastructure, despite significant investment from organizations like NCQA, NQF, and AHRQ, has produced fragmented measures, claims-based approximations, and outcomes research that is at best mixed. The system is not merely underperforming. It is structurally misaligned.This post digs into why that misalignment happens at the technical level, and why emerging infrastructure such as FHIR, TEFCA, and USCDI now makes a fundamentally different approach possible.<h3>The Fundamental Disconnect</h3>When a clinical guideline committee defines a quality measure, they think in clinical language: patient populations identified by lab values, imaging findings, symptom severity scales, and biomarker thresholds. This is also where the lifes sciences industry anchors its R&amp;D. A heart failure quality measure might specify a population using ejection fraction below 40%. An oncology measure might require a confirmed pathologic staging. These are the criteria that distinguish the patients for whom a recommended intervention is appropriate.When a payer actuary goes to implement that measure, they face a different reality. They have claims. Claims carry diagnosis codes, procedure codes, and encounter dates. They do not carry ejection fractions, staging results, or biomarker levels. So the actuary approximates using ICD codes that roughly correspond to the clinical population, making assumptions where clinical context is absent, and building logic that cannot fully represent what the guideline intended.This is the translation gap...and it has profound consequences.When population identification is inaccurate, measure results are unreliable. Practices get penalized or rewarded based on measurement artifacts rather than actual care quality. One study comparing measurement methods across cardiovascular quality measures found that for smoking cessation, medical record abstraction suggested a 94% probability of meeting the performance target. All the while, EHR-generated reports from the same patient population suggested only 11%. Same patients. Same care. Wildly different conclusions depending on how measurement was operationalized.That's not a data problem. That's a structural problem with how quality measures are specified and handed off.<h3>Why Single-Site Data Isn't Enough</h3>The translation gap is compounded by the fragmentation of care delivery itself. Research shows that 79% of patients receive care at more than one facility during a calendar year. A quality measure calculated using only the data available to a single practice or health system is working with an incomplete picture.When health information exchange (HIE) data was incorporated into quality calculations in published studies, 15% of all measure calculations changed, affecting nearly one in five patients. These weren't edge cases, they were systematic errors introduced by incomplete data. Under the current model, neither the clinician nor the payer has visibility into the full picture.<h3>FHIR and TEFCA: Infrastructure That Changes the Equation</h3>The healthcare system now has building blocks that make a different approach not just theoretically possible but practically achievable:<ul><li>USCDI (United States Core Data for Interoperability) standardizes the clinical data elements that electronic health systems must make available — demographics, medications, vital signs, diagnoses, laboratory results. This is the vocabulary layer.</li><li>FHIR (Fast Healthcare Interoperability Resources) provides a flexible, modular standard for organizing and exchanging that data through use-case-specific profiles. This is the grammar layer.</li><li>TEFCA (Trusted Exchange Framework and Common Agreement) is the nationwide network enabling secure health information exchange across hundreds of hospitals, thousands of physician offices, and more than 100,000 clinicians. This is the infrastructure layer.</li></ul>CMS has designated FHIR as the standard for interoperability and now requires payers to establish patient access APIs using it. The Interoperability and Prior Authorization Final Rule creates bidirectional data flow enabling payers to access clinical data for quality measurement while returning claims data to clinicians. The regulatory momentum has arrived.<h3>A Proof of Concept Already Exists</h3>This is not a theoretical future. ASCO, ASTRO, and partners have already demonstrated feasibility through the <a data-mce-href="https://codex.hl7.org/" href="https://codex.hl7.org/">CodeX Quality Measures for Cancer project</a>, using FHIR standards and specialty-specific data profiles (mCODE) to author, test, and execute oncology quality measures. Their findings: measures generated accurate results when executed against FHIR repositories, burden was reduced for all parties in the quality measure lifecycle, and the approach creates "a less burdensome path" for everyone involved.The technology works. The infrastructure is coming. The regulatory mandate is in place. What's missing is specialty-level clinical ownership of quality measure specifications built for this new environment.<h3>A Potential Paradigm Shift</h3>The current paradigm is professional organizations develop quality metrics in narrative form. Payers interpret those narratives using claims data. Interpretation gaps are inevitable, measurement accuracy suffers, and clinicians bear the burden of a system that doesn't reflect their actual practice.The emerging paradigm that could take shape includes professional organizations develop FHIR-based quality measure specifications using executable code that payers can implement directly using clinical data accessed through TEFCA. The measure travels with its clinical logic intact. Population identification uses the same clinical criteria the guideline intended. Calculation is automated and auditable. Interpretation gaps are then eliminated by design.This isn't an incremental improvement. It's a structural shift in who controls the quality specification and how accurately care actually gets measured.<h3>What This Asks of Professional Organizations</h3>The opportunity here is significant, but it requires professional organizations to expand their conception of what quality measure development means. Writing a narrative measure specification is not enough. The new standard is developing FHIR implementation guides, authoring Clinical Quality Language (CQL) specifications, mapping specialty-specific data elements to USCDI, and partnering with payers on implementation through TEFCA-enabled exchange.This is a different kind of work than most specialty societies have done. It requires technical infrastructure, informatics capability, and a genuine commitment to clinical data stewardship...far more than just guideline publication.The organizations that build this capability first will define quality measurement for their specialties for the next generation and defend the posture of the professional specialty they serve. Those that don't will find their guideline-based measures continued to be approximated (and misrepresented) by payers working with the only tools available to them....and abrasion will persist in some form or fashion.----------------------------------------------------------------In Part 3, I'll outline what a transformed professional organization strategy could actually look like from integrating cost-effectiveness into guidelines, to building FHIR-based measure specifications, to driving transformational rather than incremental change.

Quality: The Translation Gap ... And Why FHIR Changes Everything - Part 2 of 3

<h3>Why Quality Measurement Has Not Delivered on Aspirations and Why Professional Organizations May be the Answer </h3>Healthcare has spent decades and billions of dollars trying to measure its way to quality. The results have been underwhelming at best, and quietly destructive at worst.The ambition was sound: if we can quantify quality, we can reward it. If we can reward it, we can accelerate it. But somewhere between the clinical guideline and the payer contract, something fundamental broke down. And unfortunately the organizations built to fix it including&nbsp; NCQA, the NQF, AHRQ and other have not been able to create the level of transformation needed in the system.This is the first in a three-part series exploring what went wrong, why FHIR-based interoperability changes the equation, and what professional medical organizations are uniquely positioned to do about it.<h3>The Fragmentation Problem </h3>The original sin of American healthcare quality measurement is fragmentation. Professional organizations, payers, accrediting bodies, and government agencies have each built their own quality ecosystems, often measuring the same conditions with subtly different specifications. The result is a measurement landscape where the same provider can look high-quality on one scorecard and mediocre on another, not because care changed, but because definitions did.The National Quality Forum (NQF) was established specifically to prevent this proliferation. There are now endorses over 400 endorsed quality measures and some may say the quality problem remains elusive and seemingly compounded.This isn't a criticism of the people involved, but rather a structural critique. Fragmented governance produces fragmented output. And fragmented output produces administrative burden with unclear proportional clinical benefit.<h3>The Claims Data Dead End</h3>Alongside fragmentation sits a second, more insidious problem: the persistent reliance on administrative claims data for clinical quality measurement.Claims data were designed for billing. They capture what was done and what was charged. They do not capture why, how well, or with what clinical context. When quality measure developers use claims as their primary substrate, they are forced to approximate populations using diagnosis codes that lack the clinical granularity of the original guideline. A guideline might define a patient population using ejection fraction, biomarker thresholds, or symptom severity scores. A claims-based measure must approximate that same population using ICD codes. The result is systematic misclassification based on data asymmetry, which may be at the root of payer-provider abrasion in many cases.Research makes this concrete: studies have found that between 27% and 94% of hospitals classified as high-quality using administrative data were reclassified as intermediate- or low-quality when richer clinical data sources were applied. For a system making payment and public reporting decisions on these classifications, that is not a measurement error, yet more of over-indexing on measures of perceived value and with consequence at scale.The "streetlight effect" seems appropriately applied in this case. We've been measuring where the light is, but maybe not where quality actually lives.<h3>Burden Without Benefit</h3>The downstream effect of proliferating measures built on flawed data is a system that imposes enormous costs on providers without clear return. Primary care practices carry the weight of thousands of electronic clinical quality measures (eCQMs) developed by independent groups, many of which are inconsistent with actual clinical guidelines. Clinicians and embeded clinical quality auditors spend time on measure compliance that could be spent on care.Analysis of AHRQ's quality indicators found that just seven of them account for 93% of total measurable population health benefit. The rest&nbsp; (hundreds of measures demanding documentation, abstraction, and reporting) account for 7%. This is not a portfolio problem. It is a misallocation crisis.<h3>Marginal Incentives, Marginal Results</h3>Perhaps most damning is that the programs built on this measurement infrastructure haven't worked. Research on Medicare's Hospital Value-Based Purchasing program found no evidence of improved clinical process or patient experience measures in its first nine months, and no reduction in mortality in its first thirty months. Systematic reviews describe the impact of value-based purchasing broadly as "mixed and far from convincing."The critique is not that value-based payment is wrong in principle. It's that value-based payment layered on top of flawed measurement, inadequate data, and marginal incentives produces marginal results.<h3>Why Professional Organizations Are Different</h3>Professional medical organizations occupy a unique position in the healthcare ecosystem. They hold clinical authority, guideline ownership, specialty registry infrastructure, and the trust of the clinicians they represent. These are exactly the assets that measurement organizations like NCQA and NQF are adjacent to and exactly the assets needed to build quality frameworks that actually work.Where NCQA was formed largely by purchasers seeking accountability, whereas professional societies are formed by and for clinicians. Where NQF has accepted measures from multiple competing entities with different methodological standards in the past, but potential future could be led by Professional Organizations that build integrated frameworks where architected measures could flow directly from evidence-based guidelines.The failures of the current measurement system are not evidence that quality measurement is impossible. They are evidence that quality measurement done without clinical ownership, without being harmonized with clinical practice guidelines, without appropriate data, and without meaningful incentives produces bureaucracy rather than improvement.I look back at the great collaborative efforts among organization in Choosing Wisely where tons of great work aligned at healthcare waste and best practices. I also reflect and ponder, what if they took it one understand how payers would use this, and understand how best to steward translation of these for actuaries and utilization management teams. In my mind, professional organizations have a clear stakeholder position at the table and perhaps a critical opportunity (maybe an obligation) to build something better.---------------------------------------------------------------<h3>Stay tuned</h3>In Part 2 (<a data-mce-href="https://erikabel.org/blog/quality--the-translation-gap-----and-why-fhir-changes-everything" href="https://erikabel.org/blog/quality--the-translation-gap-----and-why-fhir-changes-everything">link</a>), I explore the translation gap between clinical guidelines and payer actuaries, and why FHIR and TEFCA now make it possible to close it.In Part 3, I'll lay out what a transformed role for professional organizations could look like and a potential strategic roadmap or framework to get there.

Quality: The Untapped Potential in Strategically Aligning Professional Organizations with Payer Value Frameworks | Part 1 of 3

In <a data-mce-href="erikabel.org/blog/the-ultrasound-tam-mirage--why-billions-haven-t-unlocked-outpatient-pocus-scale" href="erikabel.org/blog/the-ultrasound-tam-mirage--why-billions-haven-t-unlocked-outpatient-pocus-scale">Part 1</a>, I argued that POCUS has not achieved outpatient scale despite billions in investment, not because the technology failed, but because commercial architecture was never engineered around payment realities. TAM models assumed frictionless reimbursement. The system is not frictionless.The pattern becomes clearer when examining specific applications. Not all clinical utility translates to commercial viability. The difference is payment architecture.<h3>Where AI Investment Has Misaligned</h3>Lung ultrasound (LUS) to support heart failure management represents strong signal detection with weak reimbursement persistence. U.S. cardiology practice remains anchored to volume echocardiography and biomarker-driven workflows. Lung ultrasound is episodically useful for symptomatic evaluation, but payer coverage is not structured for longitudinal monitoring....and very few cardiologists even leverage ultrasound or are trained in LUS for this purpose, thus it is not a common tool of consideration. Medicare covers lung ultrasound (CPT 76604) for symptomatic evaluation but does not recognize it as a chronic disease monitoring service with recurring reimbursement. There is no endorsed screening or chronic management pathway and arguably little HEOR evidence (their currency) to support it as standard of care. The clinical utility is real. The payment architecture for routine deployment does not exist.Gestational age estimation offers high social value in access deserts and improves dating accuracy when early prenatal care is available. But reimbursement is anchored to comprehensive obstetric ultrasounds. Most commercial payers reimburse comprehensive OB ultrasounds (CPT 76805) at $150-300, while isolated biometry has no distinct reimbursement pathway outside of that comprehensive study, therefore it become a nice to have feature but likely with added cost. Gestational age alone is a component finding, not a standalone payable service. Without integration into a bundled OB care model with aligned incentives, the economic value remains theoretical.AI-guided limited echocardiography addresses sonographer scarcity and can greatly expand user bases for image acquisition - critical when internationally sonographers are in great shortage and even in the U.S there are only ~14,000 cardiac trained sonographer. However available products are limited by the lack of Ai-guidance for at least color Doppler. Thus the limited scans often fail to answer the clinical question. Quantitative hemodynamic assessment, valve evaluation, and longitudinal surveillance typically require comprehensive studies. Limited echoes if seemingly "abnormal" frequently convert to full echocardiograms. Color Doppler remains non-negotiable for ruling out prescence or progression of valvular disease, murmurs, and heart failure assessment. While the technology is facinating for reduces initial friction, it risks not reducing the total cost and in some cases may increase downstream utilization.Carotid POCUS for stroke risk screening illustrates what happens when clinical logic exists without payment alignment. Screening asymptomatic individuals for carotid stenosis is generally not reimbursed and is considered investigational by most payers. Unlike AAA screening, there is no USPSTF endorsement for population-level carotid screening. The anatomy is accessible. The technology works. But to-date, evidence of asymptomatic stenosis has shown to lead to potentially unecessary additional testing and interventions that ultimately have not improved outcomes. Thus what may be thought of as great idea, has incomplete HEOR evidence, lacks guideline support as well as&nbsp; coverage pathways leaving no clear path to scale.Venous compression ultrasound for DVT is covered... but only with adequate view quality and documentation. Poor documentation leads to frequent denials if studies are abbreviated or incomplete. This is a case where the payment pathway exists but product design often fails to enforce the alignment to view requirements and documentation standards needed to capture it. The reimbursement is available. The workflow to reliably access it is not.These are not failures of engineering. They are mismatches between clinical capability and payment logic.<h3>The Emerging Exception: AI-Assisted Interpretation</h3>CPT pathways for AI-assisted interpretation are evolving, but progress is occurring. Some payers now cover AI-assisted interpretation when the applicable technology has been used by the provider (e.g., Ultromics, US2ai) . This represents an evolving payment architecture and diligent market access and HEOR evidence required apply for a <a data-mce-href="https://www.ama-assn.org/about/cpt-editorial-panel/cpt-code-process" href="https://www.ama-assn.org/about/cpt-editorial-panel/cpt-code-process">CPT </a>or <a data-mce-href="https://mearis.cms.gov/public/home" href="https://mearis.cms.gov/public/home">HCPCS</a> billing code. If successful subsequent deliberate payer engagement for medical and coverage policy would still be needed..The lesson is not that payment architecture is immutable. It is that changing it requires intentional effort, typically years before commercialization.<h3>Why AAA Screening Is Structurally Different</h3>Abdominal aortic aneurysm (AAA) screening represents the structural inverse.AAA screening is USPSTF Grade B recommended for men aged 65-75 who have ever smoked and of undetermined evidence in women of similar age. The clinical pathway is established. The disease progression is well-characterized where testing is justified for preventable harm. Defined surveillance intervals exist based on aneurysm diameter. The payer logic is straightforward: prevent catastrophic rupture, avoid high-cost emergency surgery, reduce mortality. Medical necessity is unambiguous.Yet screening rates remain suboptimal. Estimates suggest that only one-third of eligible adults have been screened, largely due to sonographer shortages, imaging center capacity constraints, and lack of practical primary care deployment models. The anatomy is relatively simple. The finding is binary - aneurysm present or absent, with diameter measurement as an anchor to guide surveillance or intervention. Interpretation variance is low. The workflow is naturally suited to task-shifting with appropriate supervision and quality oversight. Furthermore, beyond surveillance, even post-operative care often calls for AAA ultrasound assess for endoleaks, stent assessment, and sac diameter, along with flow patterns in the sac and branch vessels with the addition of color Doppler.&nbsp;AAA screening has coverage, guideline support, a defined population, and a broken delivery model. Quite the opportunity that few, if any, are actively targeting. AI-enabled ultrasound does not need to create the market. It needs to operationalize an existing one.<h3>The Litmus Test</h3>The pattern is consistent:If the use case already has coverage, guideline support, a defined population, and a broken delivery model, AI scales.If it relies on clinical enthusiasm without payment logic, it stalls.AAA checks every box. Lung ultrasound, gestational age estimation, carotid screening, and limited echo guidance do not unless tightly capitated delivery models that align incentives across the care continuum as an path to gain real-world HEOR evidence.Just because reimbursement mechanisms for the other don't exist, it does not mean they may not be valued in care....but, it does mean that a Market Access Strategy, HEOR evidence, and Payer engagement plan will need solidified beyond just selling devices and features.<h3>What This Means for Product Strategy</h3>Bolster your MRD Approach an Business CaseThe traditional medical device development sequence is: identify clinical need, build technology, validate clinically, then figure out reimbursement. That sequence is why billions have been invested without achieving outpatient scale.In such path, the Market Requirements Document (MRD) often has less rigor before developing and proceeding with the Product Requirements Document (PRD). Commercial evaluation, validation, and payment architecture analysis should precede and constrain product design decisions.<ul><li>Target Population Size and Prevalence: What is the annual incidence or prevalence of the clinical condition? What percentage meets inclusion criteria for the intended use? This determines addressable market size and payer interest. A large prevalent population with clear diagnostic criteria is fundamentally different from a niche indication requiring complex patient identification.</li><li>Care Setting and Continuum Placement: Is the product intended for primary care, specialty clinics, inpatient, outpatient, or home settings? Is it for early detection, diagnosis, triage, monitoring, or follow-up? Site-of-service dictates the incumbunant revenue cycle constraints affecting reimbursement rates, supervision requirements, and who controls adoption decisions.</li><li>Procedure Utilization Volume:&nbsp;How often is the associated procedure performed per patient per year? Can this solution be deployed at scale across provider types? A one-time screening exam has different economics than longitudinal monitoring. High-frequency utilization supports recurring revenue but may trigger utilization management scrutiny.</li><li>Existing CPT or HCPCS Codes: Are there existing procedure codes that can support billing? Are these codes limited by provider type, setting, imaging completeness or supervision requirements? Building around existing codes accelerates reimbursement feasibility. Requiring new codes can add years and uncertainty.</li><li>Payer Mix of Target Population:&nbsp;What percentage of the clinical population is covered by Medicare, Medicaid, Commercial, or cash-pay? Are they in value-based plans (Medicare Advantage, ACOs, MCOs) or traditional fee-for-service? Payer mix shapes coverage strategy and pricing. Medicare Advantage populations may value cost avoidance differently than fee-for-service Medicare.</li><li>Payer Line of Business Alignment: Is this condition managed heavily in Medicaid MCOs, Medicare Advantage, or Commercial plans? Do payers in these segments have incentives aligned with the innovation? A technology that reduces downstream utilization creates value for risk-bearing entities but may represent revenue loss for fee-for-service providers.</li><li>Clinical Workflow and Delegation Potential: Can image acquisition or data input be delegated to non-licensed staff? How is supervision, documentation, and interpretation structured? Delegation potential affects labor economics and scalability. Products that require physician-performed acquisition face different constraints than those enabling MA or RN acquisition with remote interpretation and all of which may have implications on technical fee coverage and reimbursement, independent of State scope of practice allowances.</li><li>HEOR Evidence Requirements. What clinical outcomes, cost offsets, or access improvements must be demonstrated? What value levers matter to payers: utilization reduction, earlier diagnosis, network adequacy and expansion, medical cost-reduction or avoidance? Understanding payer evidence requirements before clinical trial design prevents building a clinical evidence dossier that does not support coverage decisions.</li></ul>These questions should be answered with data and not assumptions before product requirements are finalized. The answers constrain what should be built, not just how it should be marketed.<h3>Connecting Product to Commercialization Strategy</h3>Start with payment architecture, not clinical capability. Before building, map the CPT landscape, medical policy language, prior authorization requirements, and audit exposure. If the payment pathway does not exist or requires policy evolution, that is a commercialization variable—not an afterthought.Engage payers before commercialization, not after. Medical policy language can evolve, but it requires deliberate engagement. The companies building AI interpretation pathways today started payer conversations years ago. Companies that treat payer relations as a post-launch function will face the same structural barriers that have constrained the industry for a decade.Design documentation workflows around medical policy language (or make an informed decision to pursue a new code). If payer policies require specific views, medical necessity documentation, or defined clinical indications, product design should enforce compliance and not assume clinicians will adapt. DVT ultrasound is covered, but poor documentation leads to denials. The product should make compliant documentation the path of least resistance.Model denial rates and prior authorization exposure as core variables. A 20-25% effective revenue haircut from denials, downcoding, or bundling changes unit economics fundamentally. TAM models that ignore this are theoretical.Build for service delivery, not device sales. Ultrasound outside hospital walls is a payment-governed diagnostic service. The companies that recognize this will design business models around adoption, utilization, outcomes, workflows, and payer alignment....not hardware margins around a feature laden commodity.<h3>The Path Forward</h3>The technology is improving. The clinical evidence is evolving. The question is no longer whether AI-features and POCUS can work in outpatient settings, but rather does the evidence support payment architecture that drive scale.AAA screening is the proof of clear untapped opportunity. It demonstrates that when clinical utility aligns with coverage, guidelines, defined populations, and delivery gaps, the potential exists for AI-enabled POCUS to operationalize markets that traditional imaging infrastructure cannot serve.The companies that recognize this pattern will build differently. They will start with payment architecture. The will understand the business relationship of their providers with payers including the revenue cycle. They will engage payers as partners, not obstacles. They will design products around audit defensibility and documentation compliance. They will model economics honestly.Handheld POCUS has the potential to reshape ambulatory medicine. However,&nbsp;that potential will only be realized when commercial architecture matches clinical capability.The litmus test is simple. The execution is not.------------------------------------------------------------If you missed it: <a data-mce-href="https://erikabel.org/blog/the-ultrasound-tam-mirage--why-billions-haven-t-unlocked-outpatient-pocus-scale" href="https://erikabel.org/blog/the-ultrasound-tam-mirage--why-billions-haven-t-unlocked-outpatient-pocus-scale">Part 1 of 2: The Ultrasound TAM Mirage: Why Billions Haven’t Unlocked Outpatient POCUS Scale</a>

The Outpatient POCUS and AI Litmus Test: Why AAA Screening Can Scale and Others Won't (Part 2)

Erik Abel

Thought Leader Portfolio (clean hero, blog, and about).

Erik Abel is a healthcare strategist and operator bridging clinical, payer, and MedTech ecosystems to scale AI and digital health solutions that improve access, outcomes, and profitability across fee-for-service, hybrid, and value-based models. With over two decades of leadership translating innovation into reimbursable, market-ready solutions—including a nine-figure acquisition, nationwide diagnostic deployment, and measurable payer adoption—he is recognized for advancing healthcare innovation and commercial growth through research, thought leadership, and advisory work that converts complex care, technology, and policy environments into sustainable value across the healthcare continuum.


Insights & Opportunities

Heart Failure and SGLT2 Inhibitors: Access Has Improved, ...

New BCBS Analysis on AI-Powered Clinical Documentation: U...

Quality: A Potential Strategic Roadmap for Professional O...

Quality: The Translation Gap ... And Why FHIR Changes Eve...

Quality: The Untapped Potential in Strategically Aligning...

The Outpatient POCUS and AI Litmus Test: Why AAA Screenin...