Clinician reviewing patient records during oncology trial eligibility screening
    Clinical Operations

    When Manual Screening Fails Oncology Trials | Kitsa

    Screen failure in oncology trials can reach 50% in published audits. Learn how clinical trial software reduces manual screening burden and supports enrollment.

    Published by Kitsa Editorial Team·December 18, 2024·~18 min read

    When Manual Screening Fails: How Clinical Trial Software Helps Prevent Screening-Driven Enrollment Failure in Oncology

    The following scenario is illustrative, drawn from patterns documented across published analyses of oncology trial enrollment, including the Tata Memorial Centre audit described below. An oncology study is enrolling on schedule in its first quarter. The coordinator team is experienced. The sites have been selected carefully. The protocol has survived three rounds of review. Then, six weeks into active recruitment, the numbers tell a different story: of 74 patients who have consented and entered formal screening, more than half have screen failed. Every one of those failures has required days of manual chart review to discover.

    The costs are not abstract. Coordinator hours consumed. Patients distressed by a process that ultimately excludes them. Timelines compressed. A sponsor watching its per-day trial expenditure accumulate while enrollment stalls. The study does not collapse, but it comes close, and one possible intervention that helps recover screening momentum is automated patient pre-screening software.

    This scenario is not a cautionary tale from the margins of clinical research. It is a pattern documented in published oncology screening audits from institutions across multiple countries and trial phases. Understanding why manual screening fails in oncology, how the failure cascades through trial operations, and what software-driven approaches actually change about the process is increasingly a prerequisite for anyone building or running an oncology study today.

    Oncology Screening Failure Snapshot

    [1]
    51.0%
    Screen failure rate in Tata Memorial Centre audit of 7,481 screened patients [1]
    [2]
    23%
    Screen failure before treatment in early-phase trials across three French cancer centers [2]
    [3]
    17.5%
    Recent FDA-approved oncology trials enrolling ECOG PS 2 or higher patients [3]
    [4]
    8.1%
    Pooled cancer trial enrollment rate in Unger et al. pathway meta-analysis [4]
    [5]
    22.74%
    U.S. cancer trials terminated early in Zhang and DuBois analysis [5]
    [6]
    31.62%
    Coordinator time spent on patient screening in a time-and-motion study [8]

    Why Oncology Trials Are Unusually Vulnerable to Screening Failure

    Screen failure, in the simplest operational sense, occurs when a patient who has consented to trial participation is subsequently found ineligible and cannot be enrolled. The rate at which this happens in oncology is notable. An audit of 15 randomized studies conducted at Tata Memorial Centre in India found that, of 7,481 patients screened, 3,815 (51.0%) screen failed [1]. The most common reason was failure to meet inclusion criteria, accounting for 54.9% of all failures. A 2025 multicenter retrospective study published in ESMO Open, covering three French cancer centers, found that 23% of patients who consent to early-phase trials are screen failed before treatment starts [2].

    Those numbers reflect a real biological reality: oncology trials are designed around patient populations defined by disease subtype, prior therapy lines, performance status, biomarker status, organ function thresholds, and increasingly, molecular targets. The resulting eligibility criteria are inherently narrow. A 2024 review of FDA-approved oncology trials found that, over the 15-year period from 2009 to 2023, the proportion of studies enrolling patients with an Eastern Cooperative Oncology Group (ECOG) performance status of 2 or higher dropped to just 17.5% in the most recent five-year window, from 43.2% a decade earlier [3]. Trials are, in practice, narrowing their eligible pools even as researchers call for broader inclusion.

    The consequence is predictable. Adult cancer trial participation is commonly cited at 2% to 3% in the literature, while Unger et al.'s pathway meta-analysis found an 8.1% enrollment rate among the studies it included [4]. The same analysis found that among patients for whom no structural barriers existed, approximately 22% were not eligible due to restrictive criteria, and roughly 56% could not enroll because no suitable trial was available at their treatment location [4]. A national analysis of U.S. cancer trials by Zhang and DuBois found that 22.74% were terminated early, with accrual problems cited as one of the most common drivers [5]. Screen failure is not a peripheral operational problem. It sits at the center of one of oncology research's most persistent structural challenges.

    What Manual Screening Actually Requires

    Before attributing these failures to unavoidable disease complexity, it is worth examining what manual eligibility screening actually demands from a research team in operational terms.

    Manual screening for a clinical trial requires a coordinator to pull a candidate patient's full electronic health record, locate the relevant clinical documents (pathology reports, imaging, laboratory results, prior treatment records, performance status assessments), map each element against every inclusion and exclusion criterion in the protocol, document the result, and repeat the process for the next patient. Stanford researchers studying this problem estimated that manual eligibility assessment takes up to one hour per patient [6]. A separate estimate from Penberthy et al., cited in the Veterans Health Administration's MPACT platform study, placed the figure at 6.33 to 8.78 hours per patient enrolled across Phase I through III trials, when all associated activities are included [7].

    A time-and-motion study published in PMC observed clinical research coordinators in a pediatric emergency department and found that patient screening consumed 31.62% of their total working time [8]. That study was conducted in a pediatric ED context rather than an adult oncology setting; the precise proportion varies by site type and trial portfolio, but the general finding that screening dominates coordinator hours is consistent with reported experience across oncology research settings.

    The scope of the manual labor is one problem. The accuracy of manual chart review is another. Research examining patient chart review accuracy for clinical summarization tasks found that 36.6% of reviewed cases required correction in either the pathophysiological process identified or the management decision reached [9]. That figure concerns general chart review rather than trial eligibility determination specifically, and the error distribution in eligibility screening may differ. But the directional implication holds: manual review of complex, multi-field EHR data by individual reviewers is not inherently accurate, and when applied to eligibility gates that carry regulatory weight, even lower error rates than this could mean enrolling excluded patients or missing eligible ones.

    In fast-moving oncology cases, there is a third failure mode that purely labor-driven screening cannot easily manage: disease progression during the screening window. As the Korakis et al. study noted, the time required for screening tests "can impact patients' ability to actually start treatment" and contributes directly to high screen failure rates [2]. A review process that takes days to complete is simply not calibrated to the pace of oncology, where performance status can deteriorate, organ function can shift, and new lesions can appear within the time it takes to complete a manual chart review.

    Manual Screening Failure Cascade

    1
    Narrow oncology eligibility
    Disease subtype, prior therapy, ECOG, biomarkers, organ function, molecular targets
    2
    Manual chart review
    Coordinator reviews pathology, imaging, labs, treatment history, and notes one patient at a time
    3
    Screen failure
    Patients consent, undergo formal screening, and are later found ineligible
    4
    Operational burden
    Coordinator hours, patient distress, site workload, delayed timelines
    5
    Trial-level risk
    Enrollment stalls, costs accumulate, timelines compress, sponsor-site trust weakens

    What the Published Evidence Shows on Automated Screening Tools

    The published evidence on automated eligibility screening tools in oncology has grown substantially since roughly 2015, and the patterns across studies are consistent enough to draw operational conclusions.

    A foundational study by Ni Y, Wright, and colleagues, published in BMC Medical Informatics and Decision Making, evaluated an automated eligibility screening algorithm developed for pediatric oncology patients. The algorithm was validated against 169 historical trial-patient enrollment decisions. Without automation, an oncologist needed to review an average of 163 patients per trial to replicate historical enrollment. With automated screening, that workload fell by 85% to 24 patients [10]. The recall was 100% at the patient level, meaning eligible patients were not missed. The same team found that, from the patient side, automated screening reduced the average number of trials a patient profile needed to be checked against from 42 to 4, a 90% reduction in workload. The pediatric oncology context limits direct generalization to adult solid tumor trials, but the efficiency findings have since been replicated in adult settings, as described below.

    A study published in JCO Clinical Cancer Informatics evaluated an AI clinical trial matching system (WCTM) across four breast cancer trials at a community oncology practice. The tool excluded ineligible patients with 76% to 99% specificity and identified eligible patients with greater than 90% sensitivity in three of the four trials [11]. Breast cancer trials were specifically chosen as a test case because, as the authors noted, they carry some of the most complex inclusion and exclusion criteria in oncology due to molecular subtyping and the growing number of targeted therapies available for treatment.

    At Memorial Sloan Kettering Cancer Center, Rosenthal and colleagues developed MSK-MATCH, a multi-agent AI system for automated eligibility screening from clinical text. Published as a preprint and tested retrospectively across 731 patients spanning six breast cancer trials, the system automatically resolved 61.9% of cases without requiring human review, triaging the remaining 38.1% for coordinator attention [12]. The architecture used a large language model integrated with a curated oncology trial knowledge base, producing audit trails for every determination.

    TrialMatchAI, a system evaluated using expert validation across more than 1,000 patient-criterion pairs including molecularly driven cases, reported an average accuracy exceeding 90% in eligibility assessments, with 92% of oncology patients having at least one relevant trial retrieved within the top 20 recommendations [13].

    Across all of these studies, the pattern is consistent: well-designed automated screening substantially reduces the volume of patients coordinators need to manually review while maintaining or improving the recall of truly eligible patients. The evidence base spans pediatric and adult populations, early-phase and later-phase trials, prospective and retrospective designs, and preprint and peer-reviewed publications. Most systems represented here were evaluated retrospectively; prospective implementation validation in live trial workflows remains an active area of development. Readers should weigh those distinctions when applying findings to a specific trial context. Automated screening does not replace the protocol, the coordinator's clinical judgment, or the final eligibility determination; it narrows the field of candidates who require intensive manual review.

    What Happens to a Trial When Screen Failure Runs High

    The operational consequences of sustained high screen failure rates extend well beyond the immediate recruitment shortfall. They accumulate across the trial timeline in ways that are difficult to recover from without direct operational intervention.

    The Tufts Center for the Study of Drug Development (Tufts CSDD) conducted an empirical analysis published in 2024 in Therapeutic Innovation and Regulatory Science, estimating that a single day of delay in drug development costs approximately $500,000 in unrealized prescription drug or biologic sales, based on an analysis of 645 drugs launched since 2000 [14]. With direct trial costs running roughly $40,000 per day [14], even a one-month recruitment delay driven by excessive screen failure represents a material operational cost, exclusive of the compounding effect on post-approval revenue.

    That figure reflects drug-development economics broadly, not screen-failure costs specifically. But the connection is direct: every week a trial runs below enrollment targets because eligible patients are not being identified in time is a week added to the study timeline, and by extension to the commercialization horizon. The exact multiplier depends on the study, the indication, and the competitive timing, but the direction of impact is operationally plausible and supported by delay-cost evidence.

    The impact on sites is also significant and often underappreciated by sponsors. Sites expend the same resources on a screen failure as on a successful enrollment through most of the screening window. When screen failure rates are high, site coordinators become progressively more burdened, morale deteriorates, and the site's willingness to take on future studies from the same sponsor is diminished. In oncology research networks where site relationships are built over years and depend on realistic workload expectations, this is not a theoretical risk.

    Regulatory Context: FDA's Push to Address Eligibility at the Source

    Part of the screen failure problem in oncology originates in the protocols themselves. The FDA has been explicit about this for several years. The November 2020 FDA guidance on Enhancing the Diversity of Clinical Trial Populations recommended that sponsors include a plan for broadening eligibility criteria no later than the end of the Phase II meeting [15]. This guidance is non-binding, meaning it represents the FDA's current thinking rather than a mandatory regulatory requirement, but it has meaningfully influenced protocol design discussions.

    In April 2024, the FDA issued draft guidance documents specifically addressing cancer clinical trial eligibility criteria, covering performance status [16] and laboratory values [17]. The performance status guidance recommends expanding eligibility to include patients with a wider range of ECOG scores unless there is a clear scientific or safety justification for restriction. The laboratory values guidance recommends that lab-based criteria be calibrated to the investigational drug's mechanism of action rather than applied as categorical exclusions based on conventional clinical norms.

    Draft guidance does not carry binding regulatory weight. The FDA notes in its guidance documents that these recommendations do not establish legally enforceable responsibilities [17]. However, sponsors who design protocols without engaging with this guidance face increasing questions from FDA reviewers and ASCO-Friends of Cancer Research working groups about whether their eligibility criteria are appropriately justified. An ASCO-Friends joint statement from 2021 affirmed directly that eligibility criteria should be broadened to expand patient access and enroll cohorts more representative of the general population [18].

    The practical consequence for screening operations is that protocols with overly restrictive, historically motivated criteria will continue to drive high screen failure rates until those criteria are revised. Software-based screening tools do not fix a flawed protocol, but they surface the failure pattern much faster, at far lower cost, than weeks of manual chart review would.

    FHIR Interoperability and Its Role in Pre-Screening

    The technical foundation underlying effective automated pre-screening in oncology is interoperability between the clinical trial protocol criteria and the patient's existing health record data. The HL7 Fast Healthcare Interoperability Resources (FHIR) standard Release 4 is required for certified health IT API functionality under the ONC Interoperability and Information Blocking Final Rule, issued in May 2020 under the 21st Century Cures Act [20]. That certification requirement established standardized, API-level patient data access across certified EHR systems, creating the technical substrate through which clinical trial pre-screening platforms can query and evaluate patient eligibility without requiring manual data extraction.

    A proof-of-concept study published in PMC, focused on molecularly guided cancer trial matching using FHIR and HL7 Clinical Quality Language (CQL), demonstrated that FHIR-based standards including FHIR Genomics and FHIR mCode (Minimal Common Oncology Data Elements) are maturing as a technical substrate for EHR-integrated eligibility decision support [19]. The study highlighted that physicians often fail to refer patients to trials not because of patient ineligibility but because they are simply unaware of which trials are open and what their criteria require. Software-based screening embedded in or connected to the EHR workflow addresses this referral gap directly.

    This matters operationally because the data required for oncology eligibility assessment is distributed across the EHR in ways that manual reviewers find laborious to aggregate. Pathology reports establishing cancer subtype, radiology establishing measurable disease, laboratory reports establishing organ function, genomic sequencing establishing biomarker status, and medication records establishing prior therapy lines all reside in different data domains. FHIR-connected screening platforms can access and parse these domains concurrently, applying eligibility criteria at a speed and consistency that no manual process matches.

    Automated Oncology Pre-Screening Workflow

    1
    Protocol criteria structuring
    Eligibility criteria converted into structured logic
    2
    FHIR-connected EHR search
    Labs, pathology, imaging, prior therapy, biomarkers, and medications queried where available
    3
    Candidate shortlist
    Likely eligible patients surfaced earlier in the recruitment cycle
    4
    Coordinator review
    Human team confirms eligibility context and resolves ambiguous criteria
    5
    Formal screening
    Only qualified candidates proceed into consent, formal screening, and investigator review

    Where Pre-Screening Infrastructure Fits

    Kitsa's KScreener is designed to address the manual pre-screening bottleneck through FHIR-based EHR integration, allowing research teams to surface likely-eligible patient candidates before the formal eligibility determination process begins. Kitsa describes KScreener as a patient pre-screening and eligibility tool built for HIPAA-compliant clinical research environments, designed to reduce the volume of patients requiring full manual chart review by identifying candidates earlier in the recruitment cycle. Pre-screening of this kind identifies who warrants closer review; it does not substitute for the coordinator's formal eligibility confirmation or the investigator's clinical judgment.

    This type of infrastructure matters most in oncology trials, where eligibility criteria are complex, disease progression is rapid, and the cost of each wasted screening event compounds across a patient population that is already small relative to the trial's accrual target.

    There are problems automated pre-screening cannot fix. A protocol with scientifically unjustified exclusion criteria will still produce high screen failure regardless of how efficiently candidates are identified. Missing biomarker testing infrastructure at a site, poor trial availability in a patient's geography, and insufficient sponsor or site capacity are structural problems that technology alone does not resolve. What pre-screening tools do is remove the inefficiency that obscures these problems and allows teams to see them, and address them, sooner.

    KScreener · KScout

    Manual eligibility review carries most of the screening burden in oncology trials. KScreener connects FHIR-based EHR data to protocol eligibility criteria, surfaces likely-eligible candidates earlier, and reduces the volume of records requiring full coordinator review. Used alongside KScout for site feasibility intelligence, it helps sponsors, CROs, and oncology sites move from one-record-at-a-time screening toward structured, scalable patient matching.

    Key Takeaways

    • In published oncology trial audits, screen failure rates range from 23% in early-phase trials to above 50% in some randomized study settings, with failure to meet inclusion criteria accounting for the majority of failures [1][2].
    • Manual eligibility screening requires up to one hour per patient and as many as 6.33 to 8.78 hours per enrolled patient across all associated activities, consuming more than 30% of research coordinator time at some sites [6][7][8].
    • Research on patient chart review accuracy found error rates in clinical summarization tasks that highlight the inherent limits of manual review processes; the implications for eligibility determination, where incorrect inclusions carry regulatory consequences, are significant [9].
    • Automated screening systems have demonstrated 85% to 90% reductions in the number of patients requiring individual manual review while maintaining 100% recall of eligible patients in pediatric and adult oncology validation studies [10][11].
    • The Tufts CSDD estimates each day of clinical trial delay costs approximately $500,000 in lost drug sales, making sustained screening inefficiency a material financial risk to the sponsor [14].
    • The FDA's 2024 draft guidance on cancer clinical trial eligibility recommends expanding performance status criteria [16] and calibrating laboratory value thresholds to the drug's mechanism of action rather than applying blanket exclusions [17].
    • FHIR-based EHR integration is the technical substrate enabling scalable automated pre-screening in oncology, allowing structured and unstructured patient data to be evaluated concurrently against protocol eligibility criteria [19].

    FAQ

    What is the typical screen failure rate in oncology clinical trials?
    Published data varies by trial phase and disease type, but rates above 20% are common in early-phase oncology trials, and rates above 50% have been documented in some randomized study audits. A 2025 multicenter study across three French cancer centers found a screen failure rate of 23% among patients who consented to early-phase trials, while an audit of 15 randomized oncology studies at Tata Memorial Centre found a rate of 51%. Disease complexity, narrow eligibility criteria, and manual screening inefficiency all contribute.
    Why does manual eligibility screening lead to high screen failure rates?
    Manual screening is slow, error-prone, and operates on retrospective chart data that may not reflect the patient's current clinical status. Researchers estimate that manual screening takes up to one hour per patient for eligibility assessment alone. During that time, disease can progress, patients can initiate competing therapies, or organ function can drift outside acceptable ranges. Research on general chart review accuracy has found high correction rates for manual clinical summarization tasks, and eligibility determination faces similar risks when individual reviewers must cross-reference complex, multi-domain criteria against voluminous EHR data under time pressure.
    What does the FDA say about overly restrictive eligibility criteria in oncology?
    The FDA has issued multiple guidance documents and draft guidance documents on this topic. Its November 2020 guidance, Enhancing the Diversity of Clinical Trial Populations, recommended that sponsors plan for broader eligibility no later than the end-of-Phase-II meeting. In April 2024, the FDA issued draft guidance documents specifically covering performance status and laboratory values in cancer trials. These documents represent the FDA's current thinking and are non-binding in nature, but they reflect a sustained regulatory position that many oncology trial eligibility criteria are more restrictive than science requires.
    How much does a screen failure actually cost a clinical trial?
    The direct cost of a screen failure includes coordinator time for the chart review and eligibility determination, which can run several hours per patient, plus site administrative costs. Indirect costs include delayed enrollment, extended trial timelines, and the opportunity cost of delayed drug approval. The Tufts CSDD estimated in 2024 that each day of trial delay costs approximately $500,000 in lost prescription drug sales, and direct daily trial costs average around $40,000. That analysis concerns drug-development delay economics broadly rather than screen-failure costs specifically, but the connection is direct: screening delays extend timelines, and extended timelines carry compounding financial cost across the development program.
    Can automated screening tools miss eligible patients?
    Well-validated automated screening systems are specifically designed to prioritize recall of eligible patients over absolute precision. The pediatric oncology study by Ni Y et al. found 100% recall at the patient level in its validation dataset, meaning no eligible patients were missed. However, errors do occur, particularly for criteria that depend on interpretable clinical definitions, nested logical conditions, or data not explicitly recorded in structured EHR fields. Human review of AI-generated eligibility assessments remains standard practice, and formal eligibility confirmation remains the responsibility of the investigator and site team, with automated systems reducing the volume of cases requiring that review rather than eliminating it.
    What is FHIR, and why does it matter for clinical trial screening?
    FHIR, the HL7 Fast Healthcare Interoperability Resources standard, provides a structured format through which clinical information from EHRs can be accessed and shared in a machine-readable form via standardized APIs. The ONC Interoperability and Information Blocking Final Rule (2020) required certified health IT developers to implement FHIR Release 4-based APIs, supporting standardized patient data access across certified EHR platforms. In the context of pre-screening, FHIR-connected platforms use these APIs to query structured patient data including lab values, imaging reports, medication records, and genomic results, then evaluate them against protocol eligibility criteria to identify likely-eligible candidates before the formal eligibility confirmation process begins. FHIR Genomics and FHIR mCode are specific extensions relevant to oncology trial matching, enabling decision support tools to work with molecular and biomarker data in a standardized format.

    References

    1. [1]Parekh D, Patil VM, Nawale K, Noronha V, et al. "Audit of screen failure in 15 randomised studies from a low and middle-income country." ecancer. 2022;16:1476. Link
    2. [2]Korakis I, et al. "Addressing screening failures in early-phase clinical trials in oncology: impact on patient outcomes and strategies for improvement." ESMO Open. 2025;10(8):105331. Link
    3. [3]Iannantuono GM, Floudas CS, Filetti M, et al. "Performance status eligibility requirements and enrollment characteristics in cancer clinical trials leading to US Food and Drug Administration drugs approval (2009-2023)." European Journal of Cancer. 2025;225:115589. Link
    4. [4]Unger JM, Vaidya R, Hershman DL, Minasian LM, Fleury ME. "Systematic Review and Meta-Analysis of the Magnitude of Structural, Clinical, and Physician and Patient Barriers to Cancer Clinical Trial Participation." JNCI. 2019;111(3):245-255. Link
    5. [5]Zhang E, DuBois SG. "Early Termination of Oncology Clinical Trials in the United States." Cancer Medicine. 2023;12(5):5517-5525. Link
    6. [6]Wornow M, Lozano A, et al. "Zero-Shot Clinical Trial Patient Matching with LLMs." Stanford University. arXiv preprint. 2024. Link
    7. [7]Divita G, et al. "Matching Patients to Accelerate Clinical Trials (MPACT): Enabling Technology for Oncology Clinical Trial Workflow." AMIA Summits on Translational Science Proceedings. 2024. Link
    8. [8]Dexheimer JW, Tang H, Kachelmeyer A, Hounchell M, Kennebeck S, Solti I, Ni Y. "A Time-and-Motion Study of Clinical Trial Eligibility Screening in a Pediatric Emergency Department." Pediatric Emergency Care. 2019;35(12):868-873. Link
    9. [9]Lee J, et al. "Assessing the role of clinical summarization and patient chart review within communications, medical management, and diagnostics." arXiv preprint. 2024. Link
    10. [10]Ni Y, Wright J, Perentesis J, et al. "Increasing the efficiency of trial-patient matching: automated clinical trial eligibility pre-screening for pediatric oncology patients." BMC Medical Informatics and Decision Making. 2015;15:28. Link
    11. [11]Beck JT, Rammage M, Jackson GP, et al. "Artificial Intelligence Tool for Optimizing Eligibility Screening for Clinical Trials in a Large Community Cancer Center." JCO Clinical Cancer Informatics. 2020;4:50-59. Link
    12. [12]Rosenthal J, et al. "AI-assisted workflow enables rapid, high-fidelity breast cancer clinical trial eligibility prescreening." arXiv preprint. 2025. Link
    13. [13]Abdallah M, Nakken S, Georges M, et al. "TrialMatchAI: an end-to-end AI-powered clinical trial recommendation system to streamline patient-to-trial matching." Nature Communications. 2026. Link
    14. [14]Smith Z, DiMasi J, Getz K. "New Estimates on the Cost of a Delay Day in Drug Development." Therapeutic Innovation and Regulatory Science. 2024. Link
    15. [15]U.S. Food and Drug Administration. "Enhancing the Diversity of Clinical Trial Populations: Eligibility Criteria, Enrollment Practices, and Trial Designs. Guidance for Industry." November 2020. Link
    16. [16]U.S. Food and Drug Administration. "Cancer Clinical Trial Eligibility Criteria: Performance Status. Draft Guidance for Industry, IRBs, and Clinical Investigators." April 2024. Link
    17. [17]U.S. Food and Drug Administration. "Cancer Clinical Trial Eligibility Criteria: Laboratory Values. Draft Guidance for Industry, IRBs, and Clinical Investigators." April 2024. Link
    18. [18]Kim ES, Uldrick TS, Schenkel C, et al. "Continuing to Broaden Eligibility Criteria to Make Clinical Trials More Representative and Inclusive: ASCO-Friends of Cancer Research Joint Research Statement." Clinical Cancer Research. 2021;27(9):2394-2399. Link
    19. [19]Dolin RH, et al. "Molecularly-Guided Cancer Clinical Trial Matching using FHIR and HL7 Clinical Quality Language: A Proof of Concept." AMIA Summits on Translational Science Proceedings. 2025. Link
    20. [20]Office of the National Coordinator for Health Information Technology (ONC). "21st Century Cures Act: Interoperability, Information Blocking, and the ONC Health IT Certification Program." Federal Register. May 2020. Link

    Related posts