Nursing process, derived from the clinical care classification system components, as an earlier indicator of nursing care during a pandemic
Introduction
During the surge of admissions of patients with symptoms of or a confirmed diagnosis of COVID, hospital-based clinical operations shifted to daily responsiveness. The immediate clinical operations goal was to support direct care providers as urgently and as efficiently as possible to ensure resources needed for treatment of patients were available. In order to meet the demands of these resources, daily operational reports and daily online/dial in meetings were scheduled. This method of communication allowed direct unit level insights into the critical needs of patient care: medication supply, equipment needs, oxygen utilization, use of personal protective equipment [PPE]. The daily reports, accompanied by the daily phone call/video conferencing software, verbal reporting, allowed greater agility for reallocating resources to where they were needed, and, therefore, meeting the needs of patient care (see Fig. 1).
However, some of the operational reports relied on manually recording of new data and took time to collect. During the pandemic, the question surfaced around more timely, less manually entered data. In other words, there was a question if nursing documentation could address more clinically tied situational awareness directly from patient care [e.g., EHR documentation].
Three key aspects were needed to enable this study: an information model that could link the patient’s condition, interventions associated with that condition, and resources allocated to meet the needs of care. To conduct the study analysis, a codified data string of associated data elements would be required, in order to overcome the aspect of discordant time, as a feature of the data set; and an ability to link resource demands to nursing interventions, even if rudimentary, to account for work intensity for the interventions and the supplies/resources being utilized. Since nursing documentation at HCA Healthcare was based on the CCC Information Model, the data sets and data modeling of the string associations among the CCC model elements were employed for extracting two data sets and analyses of the study. For a complete overview of the study, and the descriptions provided here, please refer to Fig. 2 in the Appendix.
The CCC Information Model was developed by Virginia K Saba and first reported in 1991. Saba’s model was born from the distress of early-days removal of nurses’ notes from the paper-based medical record, as not part of the record. It was Saba’s early contention that the story of the patient, written within the nursing assessments, interventions, and outcomes, was essential to understand patient care, care responses, activities of nursing, and the further ability to conduct inquiries supportive of nursing science. Specifically, the CCC Information Model maps nursing practice and processes. This mapping is based upon the sequential steps taken by nurses for assessment, nursing diagnoses, interventions, goals and evaluations/outcomes of the care provided. The current version, CCC System Version 2.5, four healthcare patterns, 21 care component types,176 nursing diagnoses, 804 nursing interventions/actions, and 528 nursing outcomes [2].
In the early work of Virginia K Saba, two distinct, yet symbiotic measures were considered for the, then titled, Home Health Care Classification. Saba’s goals were to “conduct a descriptive analysis of patient care” and “develop a nursing classification that could predict resource requirements” [3], [4], this last goal as a contemporary gap in the CCC body of literature. As such, nurse informaticists pursued similar taxonomies and nomenclatures for structuring the language, work, and outcomes of nursing. Another critical aspect for the formation of languages was the development, alignment, and incorporation of nursing practice standards or “standards of care”. Ultimately, the computer age, and specifically electronic health record [EHR] development, accelerated the need for structured language for constructing documentation methods for nursing [5].
At HCA Healthcare, the standard for nursing documentation is described by Evidenced-based Clinical Documentation [EBCD][6]. Within the MediTech EHR documentation system, both the structure of nursing practice/processes of care and the nursing standards of care are incorporated. The guiding taxonomy for the design, development, and concurrent governance is the CCC System. As a result, the 21 Care Component/Concepts of CCC should be useful for data modeling across 179 acute care facilities. This study sought to capitalize on the EBCD nursing documentation, inclusive of the CCC Care Component/Concept codes and the documentation of the processes of nursing care, via the Nursing Diagnosis and CCC Intervention codes. Doing so, a test was undertaken to rapidly model nursing care data. This would include modeling the data with the Care Component codes, Nursing Diagnoses, and the associated CCC Intervention codes prior to the pandemic, during and after. Operational reports used during the pandemic time frame would also be used for assessment of demands of critical resources. A goal of the study was to test the ability to link the interventions of nurses to the resources used during COVID care and subsequently satisfy a need for more data informed clinical operations. If possible, then this may close the gap from Saba’s original goal of assessing nursing activities and resource utilization. Refer to Fig. 2 Appendix for a more detailed overview of the study.
Three questions guided this work: 1. Do the Care Component/Concept codes change as the demographic/diagnoses of patients changes on the in-patient units; 2. Are there changes in the CCC Nursing Interventions from baseline to during COVID care; and 3. Can the CCC Nursing Interventions signal changes in nursing work intensity and, therefore, changes in resource needs during COVID care?
Section snippets
Material and methods
As of September 2022, HCA Healthcare is a network of 181 hospitals. These hospital range in size from 50 beds up to 1000 beds. Healthcare providers document in four nationally known EHR systems, with the MediTech EHR being the majority. Two facilities do not use MediTech, bringing the total study facility count to 179. The date range for analysis was from January 2017 to March 2021. From this large data set, two data sets were created for comparison. The first data set, the “2017 Data Set” was
Results
From the study sample of 179 facilities using the MediTech Magic EHR system, twenty-two billion CCC Care Component codes were available for analysis, after loading the data model. Of importance, all CCC Care Component codes were used in this data sample. Most of the code counts were in the care component categories of: K-Physical Regulation, L-Respiratory, F-Fluid Volume, D-Cognitive/Neuro, R-Skin Integrity, and N-Safety [see Fig. 3 for distributions of all care component categories].
For COVID
Discussion
As a retrospective analysis, CCC Care Component/Concept codes are not granular enough to achieve situational awareness for clinical operations. The concept code did not discern demographic patient-level changes for the two unit-types. When incorporated as a data string from CCC Care Component/ Concept codes + Nursing Diagnosis + CCC Interventions + daily Operational Reports did provide early insight into shifts in care components, interventions, and utilization of resources specific to the care
Limitations
By design, this study was a retrospective exploration to determine the utility of data modeling of the CCC Care Component/Concept coding schema and the nursing interventions provided [see Table 2, Appendix]. Dykes et al [7] demonstrated the full structure of the CCC Information Model, in previous work. This study focused on the care components and the interventions, as a rapid data modeling activity for signal detection. This less complete approach limits any analysis to determine the stronger
Conclusion
Virginia K Saba had determination and intent. The authors feel it was important to anchor her pioneering work to contemporary times. The CCC Information Model is a tested and a true organizing architecture for underpinning systems by and for nursing practice. The Information Model and the structure of CCC System are essential to translate between the science of nursing and the computational sciences: serving as an intermediary between domains so that data models, employing the given structure
Authors’ contributions
Fine, S: standardization and structure of EBCD, manuscript development, analysis.
Chaudhri, A: data extraction, data management, and analysis.
Englebright, J: standardization and structure of EBCD, manuscript development.
Roberts, Wm D: study design, data modeling, manuscript development, analysis.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgement
Fine, S: As a nurse of twenty-five years, I have always been drawn to high intensity, not easily understood/abstract clinical reasoning, and complexities of nursing care. My formative years were in critical care, while my passion is in structuring information for advancing the practice of bedside nursing. In 2013, I began to transition and garner new focus in the area of nursing informatics. The pivotal moment was meeting, learning, and supporting the work of Virginia K Saba. Firsthand, I
References (8)
- P.C. Dykes, D. Wantland, L. Whittenburg, S. Lipsitz, V.K. Saba, A pilot study to explore the feasibility of using the...
- V.K. Saba, Clinical Care Classification (CCC) System, (2012), version 2.5....
- V.K. Saba, A. O'Hare, A.E. Zuckerman, J. Boondas, E. Levine, D.M. Oatway, A nursing intervention taxonomy for home...
Final Report: Home Care Classification Project: Develop and demonstrate a method for classifying Home Health Patients to predict resource requirements and to measure outcomes
(1991)
Cited by (1)
Future advancement of health care through standardized nursing terminologies: reflections from a Friends of the National Library of Medicine workshop honoring Virginia K. Saba
2023, Journal of the American Medical Informatics Association