Exploring common data model coverage of nursing flowsheet data: a pilot study using SNOMED CT and LOINC mapping Authors: Robin Austin, Malin Britt Lalich, Katy Stewart, Jonna Zarbano, Matthew Byrne, Melissa D. Pinto, Elizabeth E. Umberfield Published: 2025 ABSTRACT AbstractObjectivesThe primary objective of this research is to assess the content coverage of nursing data within a publicly available common data model (CDM), focusing on how nursing data, documented in flowsheets, are represented within the model.Materials and MethodsThis mapping study was informed by previous evaluation studies and serves as a framework for evaluating information resources, including to guide development and implementation. The overall research process consists of 4 steps: (1) identify a CDM; (2) define evaluation criteria; (3) map nursing flowsheet data; and (4) apply evaluation criteria.ResultsOverall, 65.5% (n = 1170) of the flowsheet concepts were mapped to Systematized Nomenclature of Medicine—Clinical Terms (SNOMED CT) and Logical Observation Identifiers Names and Codes (LOINC) target codes and 56.0% (n = 1831) of the flowsheet values were mapped to SNOMED CT and LOINC target codes. The flowsheet concepts had a higher average mapping time per concept/reviewer (1.19 min) as compared to the average mapping time per value/reviewer (0.64 min).DiscussionThis mapping study demonstrated the progress and ongoing challenges of mapping nursing data to a national common data model. However, the ability to use nursing data at scale in a national CDM remains limited until more comprehensive mapping is completed.ConclusionThis mapping study identifies a significant gap in integrating nursing data into a national common data model, highlighting an opportunity to enhance patient care through improved real-time insights and evidence-based nursing practices. Addressing this gap can help shape policies that prioritize the inclusion of nursing data. Additionally, aligning nursing data at scale can advance research, increase efficiency, and optimize nurse-sensitive patient outcomes. Introduction Nursing is the largest health-care profession in the United States, with nearly 4.3 million registered nurses nationwide.1,2 Despite well-established links between nursing care delivery and patient outcomes,3–5 research examining these relationships has relied on small or institution-specific samples. Recent national initiatives have focused on harmonizing large volumes of electronic health record (EHR) data to support real-world evidence studies, pragmatic clinical trials, and population health research.6,7 However, these initiatives primarily center on physician-generated data, with little attention paid to nursing data. As a result, clinically meaningful information documented by nurses remains underutilized in national research infrastructures. Nurses are among the most frequent users of health information technology and are responsible for some of the most detailed documentation in acute care settings. A common format for nursing documentation is the EHR flowsheet—a structured, template-based tool used to record patient care and clinical observations. Inpatient nurses manually document 631-875 data points in flowsheets per 12-h shift typically capturing both observations and measurements, as well as interventions.8 Patient outcomes influenced by nursing care, such as falls, infection rates, and pressure ulcers, are nurse-sensitive data,9 whereas nurse-specific data reflect the actual nursing process (assessments, interventions, nurse identity, documentation details).10 While flowsheet data are highly structured and potentially rich for secondary use, they are often not standardized within or across health-care systems, limiting their accessibility for nurse-sensitive and nurse-specific research or quality improvement efforts. Because nurse-sensitive and nurse-specific data are drawn directly from nursing documentation, ensuring data quality—along with interoperability—is essential for enabling evidence-based practice, advancing outcomes research, and improving care quality.11,12 Standardizing nursing data using nationally adopted terminologies and incorporating them into common data models (CDMs) can facilitate large-scale analytics, enable cross-institutional comparisons, and support meta-analyses. Such standardization is also a prerequisite for leveraging advanced technologies like artificial intelligence (AI) and machine learning (ML) to identify data-driven insights that inform nursing practice.11,13–16 Including nursing data in CDMs enhances organizational capacity to measure quality, benchmark performance, and ultimately recognize the full value of nursing care—particularly in improving outcomes and addressing health equity gaps.17,18 Currently, a major barrier to integrating nursing care data into national research efforts is the lack of standardized representation of nursing activities. This impedes the ability to study the relationship between nursing interventions and patient outcomes at scale, evaluate high-value nursing practices, and support reimbursement models such as value-based care. Efforts to standardize nursing data are foundational to advancing nursing science, improving care delivery, and ensuring that nursing contributions are visible and measurable within national health data ecosystems.19–21 This work aligns with national priorities, including the National Academy of Medicine’s Future of Nursing Report, which calls for integrating nursing expertise into the design, analysis, and application of health data, particularly in areas related to social determinants of health and health equity. To meet this call, nurse scientists and informaticists must have access to standardized, high-quality nursing data that can be used to drive clinical decision-making and policy reform. Additionally, appropriately representing nursing data in CDMs can reveal the economic value of nursing—such as cost-effectiveness and resource utilization—and reinforce nursing’s critical role in achieving high-quality, efficient, and equitable care.12 Common data models are designed to standardize the structure, format, and content of health data to support harmonization and reuse across institutions. Data elements are standardized through value sets from established terminologies including International Classification of Diseases, Systematized Nomenclature of Medicine—Clinical Terms (SNOMED CT), Current Procedural Terminology, Healthcare Common Procedure Coding System, and Logical Observation Identifiers Names and Codes (LOINC). Leveraging these standards supports interoperability and adaptability to evolving data standards. Integrating nursing data into national CDMs is essential to enabling large-scale studies of nursing care and generating evidence related to nurse-sensitive patient outcomes. Objective This study aimed to evaluate the representation of nursing data documented in flowsheets within a publicly available CDM. The specific objectives were to: (1) identify a CDM and define evaluation criteria encompassing content coverage and technical quality metrics such as standards compatibility; (2) map nursing flowsheet data from a large Midwestern academic medical center to the selected CDM; and (3) apply the evaluation criteria (eg, Integration, Completeness, Flexibility, and Implementability).22,23 Methods The overall research process consisted of 3 steps: (1) identify a CDM and define evaluation criteria; (2) map flowsheet data; (3) apply evaluation criteria (Figure 1). Overall 3-step research process. Step 1: Identify a CDM and define evaluation criteria The mapping evaluation study was informed by Friedman et al., Anatomy of All Evaluation Studies, which serves as a framework for evaluating information resources, including to guide development and implementation.24 This evaluation approach fulfills 2 purposes including both pragmatic and scholarly. We adapted the Anatomy of All Evaluation Studies model based on current literature and similar evaluation studies.24 Drawing on Garza et al.’s20 methodology for CDM evaluation, the initial step was to identify a CDM. Two widely used national CDMs: the Observational Medical Outcomes Partnership (OMOP)22 and the National Patient-Centered Clinical Research Network (PCORnet) were considered.23 Observational Medical Outcomes Partnership enables systematic analysis of disparate observational databases and is actively maintained by the Observational Health Data Sciences and Informatics (OHDSI, pronounced “Odyssey”) collaborative, a global, interdisciplinary, open-science initiative.25 Patient-Centered Clinical Research Network is a research partnership network comprising clinical data research networks and patient-powered research networks and offers a standardized model for organizing data from EHRs, insurance claims, and patient-reported outcomes. The OMOP CDM was selected for this study due to its widespread use, research relevance, incorporation of multiple standardized terminologies, and availability of well-documented and open-source software, and alignment with national health-care priorities. The evaluation criteria selected for this study emphasize the ability of the CDM to effectively represent nursing data, prioritizing expressiveness, usability, and practical value within this context (see Table 1). Table 1.Evaluation criteria and definitions; revised from Garza et al.20CriterionConceptual definitionIntegration: ConceptsThe extent to which flowsheet concepts are mappable to the standardized terminologies which are supported by the common data modelsIntegration: ValuesThe extent to which flowsheet values are mappable to the standardized terminologies which are supported by the common data modelsCompleteness: ConceptsAll integrated flowsheet concepts (ie, those which are mapped to a standardized terminology) are accommodated by the CDMFlexibilityThe ease with which the CDM can adapt to changesImplementability: Field experienceThe number and diversity of uses of the data model (ie, how many years the model has been in use)Implementability: StabilityThe total number of changes to the data modelImplementability: AdoptionThe size of the user community using and supporting the data modelImplementability: CostThe costs of training and mapping to the CDMAbbreviation: CDM, common data model. Evaluation criteria and definitions; revised from Garza et al.20 Abbreviation: CDM, common data model. Step 2: Map flowsheet data Step 2 consisted of premapping and mapping processes. Premapping process The team used OHDSI Usagi (Version 1.4.3), an open access software (Figure 2). The premapping process included the following steps: (1) retrieve terminology files from Athena (OHDSI) June 4, 2024, and limit the terminologies to SNOMED CT (Version 20240101) and LOINC (Version 2.77) and (2) extract flowsheet metadata in CSV format from a Midwestern Academic Medical Center EHR. The team selected flowsheets most representative data of general inpatient nursing care and included the following inpatient flowsheets: (1) Vital Signs; (2) Head-to-Toe Assessment; (3) Intake and Output; (4) Lines, Drains, and Airways; and (4) Daily Cares. Table 2 includes flowsheet inclusion and exclusion criteria. Our intent in selecting these flowsheets was to capture data elements most consistently documented across general inpatient settings by nurses. We acknowledge these flowsheets primarily focus on team-based care tasks often ordered or influenced by the medical team, however recognizing that they represent the foundational layer of nurse-generated data within the EHR. Screenshot of OHDSI Usagi open access mapping software. Abbreviation: OHDSI, Observational Health Data Sciences and Informatics. Table 2.Flowsheet inclusion and exclusion criteria.Flowsheet inclusion/exclusion Inclusion: Sheets most commonly used for nursing assessment documentation of the inpatient population, including: Complex vital signs Head-to-toe assessment Intake and output Lines, drains, airways, wounds, and incisions Daily cares and safety Exclusion: Concepts for documenting care of small subsets of the overall patient population, including: Pediatrics Neonatal Obstetrics/Gynecology Medication administration details not related to nursing assessment Group recordsRetired records All other sheets Flowsheet inclusion and exclusion criteria. Complex vital signs Complex vital signs Head-to-toe assessment Head-to-toe assessment Intake and output Intake and output Lines, drains, airways, wounds, and incisions Lines, drains, airways, wounds, and incisions Daily cares and safety Daily cares and safety Pediatrics Pediatrics Neonatal Neonatal Obstetrics/Gynecology Obstetrics/Gynecology Medication administration details not related to nursing assessment Medication administration details not related to nursing assessment Group records Group records Retired records Retired records Each flowsheet documentation element was stored in the metadata CSV as a discrete row, representing a single observation linked to a specific patient and timestamp. For the purposes of this analysis, each unique documentation element from the nursing flowsheets is referred to as a concept, consistent with terminology used in common data models such as OMOP. Concept rows within each flowsheet CSV contain metadata such as the display name, value type (eg, custom list, numeric, rate, dose, free-text also called, etc.) and may have a custom pick list of values, which correspond to the documented observation. For example, in the Daily Cares flowsheet row for oral hygiene, the concept display name is Oral Care, the value type is Custom List and the associated values are Brush teeth, Denture care, Mouth wash, Swab, and Refused (Comment [allowing for string, or free-text, input]). These concepts and values were used to map flowsheet documentation elements to standardized terminologies, SNOMED CT and LOINC, for evaluation within the context of the OMOP CDM. Flowsheet concepts and corresponding values were first extracted and reviewed to identify and remove duplicates, ensuring each unique clinical concept and value was represented only once in the mapping process. This deduplication was necessary due to how flowsheets are structured within the EHR. For example, the concept “Pain Score” was included in the Vital Signs, Head-to-Toe Assessment, and Lines, Drains, and Airways flowsheets. The flowsheet value “Scant” appeared multiple times across the selected flowsheets in custom lists for concepts such as “Nasal Drainage,” “Subglottic Secretions,” “Drainage Amount,” and “Output Description.” During the deduplication process, repeated concepts and values were consolidated to reduce mapping burden. As this was a pilot study with no immediate plan to implement a full Extract, Transform, and Load (ETL) process into the OMOP CDM, we made a methodological decision to deduplicate and disambiguate flowsheet concepts from their associated values. This approach allowed for a more streamlined evaluation of content coverage by focusing on the semantic representation of nursing flowsheet documentation elements, rather than the full data pipeline. Prioritizing this separation expedited the mapping process and enabled early insights into how well existing standardized terminologies and OMOP data tables support the structure and granularity of nursing flowsheet documentation. Mapping process The process included mapping flowsheet source concepts and values to SNOMED CT or LOINC target codes using Usagi. The Office of the National Coordinator (ONC) for Health IT 2024 Interoperability Standards Advisory Reference Edition for nursing data interoperability guided the mapping process.26 Source concepts for observations were mapped to LOINC target codes; source values for observations were mapped to SNOMED CT except for validated scales, which were mapped to LOINC; source concepts for patient problems and nursing interventions were mapped to SNOMED CT. All flowsheet source concepts and values were mapped to target codes by 2 independent team members (M.B.L., K.S.). We defined source concepts and values as the original flowsheet documentation elements nurses use to record clinical observations and interventions within EHR flowsheets. We systematically mapped these source concepts and values to standardized terminologies (SNOMED CT and LOINC target codes) and categorized each mapping according to established semantic alignment guidelines.27 Each mapping was classified as one of the following: Unmatched: No corresponding standard target code in SNOMED CT or LOINC captured the meaning of the flowsheet source concept or value. Unmatched: No corresponding standard target code in SNOMED CT or LOINC captured the meaning of the flowsheet source concept or value. Inexact: The target code overlapped partially with the source concept or value, but both covered additional or different conceptual meanings. Inexact: The target code overlapped partially with the source concept or value, but both covered additional or different conceptual meanings. Equivalent: The source concept or value and target code conveyed the same clinical concept but differed in syntax or used synonyms. Equivalent: The source concept or value and target code conveyed the same clinical concept but differed in syntax or used synonyms. Equal: The source concept or value and target code matched exactly in both meaning and syntax. Equal: The source concept or value and target code matched exactly in both meaning and syntax. There were specific rules for syntax, which in combination with having the same meaning, the source code (concept or value) and target code followed these syntax exceptions: (1) 2 codes are equal if one code fully spells out the word and the other code is the abbreviation; (2) 2 codes are equal if one code is plural and the other code is singular; (3) 2 codes are equal if one code has smaller words such as “a,” “and,” “the,” “of”; and (4) 2 codes are equal if the codes consist of the same words, but in a different order. When multiple target codes were equivalent, they were mapped to the flowsheet source code. There was an exception to this rule when the multiple target codes consisted of both LOINC and SNOMED target codes: when mapping flowsheet concepts, mapping to SNOMED target codes was prioritized, while when mapping flowsheet values, mapping to LOINC target codes was prioritized per established guidelines.28 Two independent team members met to resolve discrepancies through deliberation (M.B.L., K.S.) and included a third reviewer (R.A. or E.E.U.), as needed. The team did not perform interrater reliability testing, given the exploratory nature of this pilot study and the large number of concepts and values mapped; this is acknowledged as a limitation in assessing the consistency of the mapping process. Lastly, after mapping flowsheet source concepts and values to standardized terminology target codes (SNOMED CT and LOINC), the resulting mappings were exported from Usagi. These files include the original source concepts or values, their corresponding standardized target codes, and the associated Domain ID. The Domain ID identifies the clinical domain of each concept (eg, procedure, observation, measurement, etc.), which facilitates proper categorization within the OMOP CDM data tables. We then assessed content coverage within OMOP data tables. Coverage was defined as the presence of a mapped standard code within the CDM’s vocabulary structure, indicating its potential for representation and reuse in secondary analyses. This approach enabled evaluation of how well the OMOP CDM could accommodate nursing data captured through structured flowsheet documentation. Step 3: Apply evaluation criteria After all flowsheet concepts and values were mapped, we applied the evaluation criteria (Table 1). Two trained reviewers independently applied the Integration evaluation criterion to each mapped flowsheet source concept or value (M.B.L., K.S.). A calibration exercise using a representative sample was completed prior to full review to ensure consistency in interpretation and application of the semantic alignment categories. Discrepancies were discussed and resolved (M.B.L., K.S., E.E.U., J.Z., R.A.) through consensus. This process ensured reliability in classification and minimized subjectivity in assigning mappings as unmatched, inexact, equivalent, or equal. The entire research team participated in iterative discussions related to the Integration, Flexibility, Completeness, and Implementability evaluation criteria. Results Integration and Completeness After deduplication, the total number of included flowsheet concepts was 1786, and the total number of flowsheet values was 3260. A total of 1170 flowsheet concepts (65.5%) were mapped to SNOMED CT and LOINC. Similarly, 1831 flowsheet values (56.0%) were mapped to SNOMED CT and LOINC target codes (Table 3). A total of 203 equal and equivalent flowsheet concepts (11.4% of the total) were mapped to the MEASURMENT table in the OMOP CDM, more than any other data table. This was followed by the OBSERVATION data table with 7.7% (137 concepts). Other tables—PROCEDURE_OCCURRENCE (1.7%), CONDITION_OCCURRENCE (1.4%), and DEVICE_EXPOSURE (0.8%)—contained a relatively small proportion of mapped concepts. Table 3.Integration and Completeness mapping evaluation metrics.CriterionOperational definitionMetricsIntegrationMappedNot mappedConceptsThe frequency and relative frequency of flowsheet concepts which are mapped and not mapped to SNOMED CT or LOINC1170 (65.5%)616 (34.5%)ValuesThe frequency and relative frequency of flowsheet values which are mapped and not mapped to SNOMED CT or LOINC1831 (56.0%)1438 (44.0%)CompletenessOMOP data tableMappedConceptsFrequency and relative frequency of integrated flowsheet equal and equivalent concepts that map to data tables in the CDMsMEASUREMENT203 (11.4%)OBSERVATION137 (7.7%)PROCEDURE_OCCURENCE31 (1.7%)CONDITION_OCCURENCE25 (1.4%)DEVICE_EXPOSURE15 (0.8%)Abbreviations: CDM, common data model; LOINC, Logical Observation Identifiers Names and Codes; OMOP, Observational Medical Outcomes Partnership; SNOMED CT, Systematized Nomenclature of Medicine—Clinical Terms. Integration and Completeness mapping evaluation metrics. Abbreviations: CDM, common data model; LOINC, Logical Observation Identifiers Names and Codes; OMOP, Observational Medical Outcomes Partnership; SNOMED CT, Systematized Nomenclature of Medicine—Clinical Terms. Flexibility and Implementability The Flexibility evaluation indicated that new data elements can be added without changing referential integrity, provided those data elements are mapped to standardized terminologies included in the OMOP CDM and custom concepts can be created for local use. The extensive use and worldwide of OMOP supports the Field Experience and Adoption Implementability evaluation criteria (Table 4). The OMOP CDM has been in use since 2009 and appears to be stable with the latest updates occurring in 2024 (OMOP V5.4). The OMOP CDM has many collaborators (∼4294), including international partners, representing >970 million unique person records. The Cost evaluation criterion showed the flowsheet concepts had a higher average mapping time per concept/reviewer (1.19 min) as compared to the average mapping time per value/reviewer (0.64 min). Table 4.Flexibility and Implementability mapping evaluation metrics.FlexibilityOperational definitionOMOPWhether a new data element can be added to the CDM without changing referential integrity (yes/no, narrative description)Yes. Large corpus of integrated standardized terminologies supports the flexible incorporation of a wide variety of documentation elements. When a local concept is not defined in one of the standardized terminologies represented in OMOP, users have the option to create custom concepts. Custom concepts are defined locally and cannot be used for network research. OHDSI outlines conventions for creating and working with custom conceptsImplementabilityOperational definitionOMOPField experienceThe number of years the model has been in use16 years (2009)StabilityNumber of CDM updates in the last 5 years (differentiate major and minor changes)Total updates: 4 (1 prerelease, 1 major, 2 minor)2021: prerelease v6.0.1 major changes2021: v5.3.2 minor changes2021: v5.4 major changes, 3 new tables2023: v5.4.1 minor changes2024: v5.4.2 minor changesAdoptionNumber of adopters for each CDM (retrieved from host websites)∼4294 collaborators across 83 countries and 6 continents representing >974 million unique person recordsCostMapping time estimate (measured as hours per intern mapping per project; 15-min increments) Concepts: 4789 total concepts mapped; 5720 min (95.33 h); average time per concept/reviewer: 1.19 min Values: 7005 total values mapped; 4485 min (74.75 h); average time per value/reviewer: 0.64 min Abbreviations: CDM, common data model; OHDSI, Observational Health Data Sciences and Informatics; OMOP, Observational Medical Outcomes Partnership. Flexibility and Implementability mapping evaluation metrics. Concepts: 4789 total concepts mapped; 5720 min (95.33 h); average time per concept/reviewer: 1.19 min Concepts: 4789 total concepts mapped; 5720 min (95.33 h); average time per concept/reviewer: 1.19 min Values: 7005 total values mapped; 4485 min (74.75 h); average time per value/reviewer: 0.64 min Values: 7005 total values mapped; 4485 min (74.75 h); average time per value/reviewer: 0.64 min Abbreviations: CDM, common data model; OHDSI, Observational Health Data Sciences and Informatics; OMOP, Observational Medical Outcomes Partnership. Discussion This pilot mapping study demonstrated the ability to map nursing flowsheet concepts and values to standardized terminologies, such as SNOMED CT and LOINC, to assess their representation within the OMOP CDM. The finding that 65.5% of flowsheet concepts were successfully mapped to SNOMED CT and LOINC highlights the current content coverage with standardized terminologies. This research aligns with prior literature and reinforces the persistent gap in the standardized representation of nursing data within widely adopted clinical terminologies.29,30 Furthermore, the observed discrepancy between the mapping rates of flowsheet concepts (65.5%) and their corresponding values (56.0%) echoes previous studies that have highlighted similar limitations.18,20 Our finding that the MEASUREMENT and OBSERVATION data tables contained the majority of equal and equivalent mapped terms aligns with prior work.31 This finding suggests that the OMOP CDM = captures quantitative and observational nursing data (eg, vital signs, validated assessment scales, and clinical observations). These findings underscore a critical challenge, while a substantial portion of nursing documentation can be aligned with standardized terminologies, incomplete or partial content coverage limits the usability of these data for secondary purposes such as research, quality measurement, and clinical decision support. This partial coverage creates barriers to fully integrating nursing data into CDMs, which are foundational for large-scale analytics and cross-institutional research.32,33 There were clear patterns in how nursing flowsheet documentation elements were successfully mapped to standardized terminologies. Many common nursing assessments are performed using validated scales. Through national documentation standardization efforts, instruments such as the Braden Scale for predicting pressure ulcer risk, the NIH Stroke Scale for measuring stroke severity, and the 1-10 numerical pain scale are already included in LOINC Panels. Importantly, when these validated scales appeared in included flowsheets, mapping both the related concepts and values was simple, highlighting the utility of using the Usagi software in the early ETL process. Mapping the broader diversity and contextual nuance of nursing flowsheet documentation presents challenges. For instance, the term “round” can refer to both pupil shape and wound shape in a flowsheet. In LOINC, “Round” (concept code LA24884-1) is part of the Meas[urment] Value Domain and is an answer in the Pupil assessment panel. In SNOMED CT, “round” is represented as “Round shape” (concept code 42700002), also in the Meas[urment] Value Domain. Similarly, terms like “scant” often appear in nursing flowsheets to describe various types of output—such as sputum, urine, or nasal drainage—but the intended meaning depends heavily on context. These examples show how subtle semantic differences can complicate the process of harmonizing nursing data within standardized terminologies. This underscores the critical need for domain expertise and a deep understanding of ONC interoperability standards to ensure accurate mappings. Emerging research has shown the potential for generative AI to assist with mapping data elements that could ultimately decipher semantic nuances.34,35 While still emerging, future research should consider such technology to enhance the ability to integrate various data sources more readily. Additional challenges arose when applying terminology mapping heuristics (ie, mental shortcuts), primarily due to the naming conventions used in nursing flowsheets. Many assessment concepts in our sample flowsheets were recorded as precoordinated concepts that combine multiple semantic parts—such as body location and clinical findings—into a single term. For example, concepts like “right foot dorsiflexion,” “right lower extremity motor response,” or “left breath sounds” combine both the assessment (dorsiflexion) and the body location (right foot, etc.). While this format is efficient for nursing documentation at the bedside, it complicates mapping to standardized terminologies. Accepted heuristics recommend postcoordination, which treats body location and clinical concept as separate, mappable entities.36,37 As ontologies, standardized terminologies lack direct equivalents for these precoordinated concepts that often require subjective interpretation or custom definitions. Precoordinated flowsheet concepts increase the risk of misclassification, inconsistent mappings, and challenge semantic integrity when applying postcoordination transformations, especially with ambiguous or context-dependent terms. The use of precoordinated flowsheet concepts in our sample contributed to the low proportion of successfully mapped concepts and values. To improve interoperability and terminological consistency, nursing documentation structures might benefit from closer alignment with postcoordinated terminology models that clearly distinguish key elements like body location and clinical concept. The relatively low proportion of original flowsheet source codes that could be mapped to SNOMED CT or LOINC substantiates a gap rooted in the precoordination, variability, and nonstandardization of nursing documentation at the source level. Thus, the primary barrier to inclusion in CDMs may not lie within the CDMs themselves, but rather in the upstream lack of standardized representation of nursing data within local EHRs. This distinction highlights the importance of efforts to standardize nursing documentation prior to CDM integration. This finding warrants further investigation and highlights the importance of carefully selecting a CDM in nursing research. Additionally, this research also highlights there are existing representations of nursing activities; however, these activities have not been widely adopted by vendor EHR systems. The integration of multiple complete standardized terminologies in the OMOP CDM allows for the representation of a broader range of health-care data within the CDM. Nurse scientists considering a CDM for research should carefully examine available data tables within candidate CDMs to identify a model that is best suited to their use case. To effectively compare content coverage across CDMs, future research must consider inherent structural differences between the OMOP and alternate models. The overall mapping process (flowsheet concepts and values), included 2 independent reviewers, is a time intensive process (∼170 h). The labor hours needed to complete this work highlights a potential gap in the advancement of nursing data coverage within CDMs.29 Biomedical entity linking using a variety of techniques including ML and natural language processing (NLP) is one area of related research that may reduce the time required to complete large-scale flowsheet mapping projects. Future research should focus on use of technology, such as large language models, NLP, and AI, to enhance automation of this work to increase representation and inclusion of nursing data in CDMs.38 Limitations This research has several limitations. This study utilized flowsheet data from a single large Midwestern academic medical center, which may limit generalizability to other health-care settings. Community hospitals, non-US institutions, and organizations using different EHR vendors may structure and label nursing documentation differently, leading to variations in flowsheet content, granularity, and standardization. For example, community hospitals may rely on fewer structured documentation fields or use more localized terminology, which could affect the ability to map to standardized terminologies such as SNOMED CT or LOINC. Additionally, international institutions may use region-specific coding systems or document in other languages, introducing further variability. Electronic health record vendor differences can also influence how flowsheet templates are configured and customized at the organizational level. Future research should replicate this mapping process across diverse health-care settings to evaluate the consistency of nursing data representation and to inform broader standardization efforts. Additionally, the process of mapping flowsheet concepts and values is inherently subjective. To minimize this potential bias, we had 2 reviewers independently evaluate each concept and value. Future studies should aim to include a broader range of clinical documentation systems to enhance generalizability. Furthermore, establishing standardized guidelines for mapping could help reduce subjectivity in future research. Implications This research highlights implications for clinical practice, research, and policy. For clinical practice, this study identifies gaps in the inclusion of nursing data within the OMOP CDM, taking a crucial step toward improving patient care. To our knowledge, few researchers have evaluated national CDMs for their content coverage specific to nursing flowsheet data.31,39 Standardizing and promoting the exchange of nursing data enables health-care systems to operate more efficiently while demonstrating the critical value of nursing in patient care.12 By integrating and analyzing nursing data, organizations can gain meaningful insights into nurse-sensitive impacts on patient outcomes, resource utilization, and overall cost of care, ultimately driving more informed decision-making and cost-effective health-care delivery. Real-time analysis of nursing data enhances the accuracy of patient outcomes and supports evidence-based nursing practices. This study identifies a key challenges in nursing research, incomplete representation of nursing data limits the visibility of nursing care within big data environments. Without mapping accurate clinical concepts and values, the ability to evaluate the impact of nursing interventions is compromised and thus limiting the inclusion of nursing-sensitive variables in predictive models. Identifying and mapping flowsheet data, guided by previous minimum dataset efforts, will be essential to advance evidence-based practice and nursing-sensitive outcomes research.21,28,29 This evaluation is a critical step toward ensuring the inclusion of nursing data in standardized health-care models to support evidence-based practice and research. Advancing this research has the potential to strengthen the visibility of nursing care, promote equitable representation in national data infrastructures, and enhance the ability of nurse scientists to contribute meaningfully to data-driven healthcare innovation. This research supports national nursing initiatives, including The Future of Nursing 2020-2030, by demonstrating the value of standardized nursing data to generate new knowledge and inform policy.13 Furthermore, integrating nursing data into clinical workflows, and CMDs, such as OMOP, directly supports the recognition and measurement of nursing’s contribution to health outcomes. This alignment can enable the development of policies and reimbursement models that reflect nursing’s essential role in care delivery.12,40 Conclusion This mapping study reveals a substantial gap in the integration of nursing data into national CDMs. Standardizing and exchanging these data is essential for generating real-time insights and supporting evidence-based nursing care. The findings identify areas of underrepresentation, guide the development of targeted solutions, and enable research that highlights nursing’s unique contributions. Prioritizing nursing data in national frameworks such as OMOP has policy implications, enhancing data scalability, advancing research, and improve health-care efficiency and outcomes.