Predictive models for pressure ulcers from intensive care unit electronic health records using Bayesian networks.

BACKGROUND: We develop predictive models enabling clinicians to better understand and explore patient clinical data along with risk factors for pressure ulcers in intensive care unit patients from electronic health record data. Identifying accurate risk factors of pressure ulcers is essential to determining appropriate prevention strategies; in this work we examine medication, diagnosis, and traditional Braden pressure ulcer assessment scale measurements as patient features. In order to predict pressure ulcer incidence and better understand the structure of related risk factors, we construct Bayesian networks from patient features. Bayesian network nodes (features) and edges (conditional dependencies) are simplified with statistical network techniques. Upon reviewing a network visualization of our model, our clinician collaborators were able to identify strong relationships between risk factors widely recognized as associated with pressure ulcers. METHODS: We present a three-stage framework for predictive analysis of patient clinical data: 1) Developing electronic health record feature extraction functions with assistance of clinicians, 2) simplifying features, and 3) building Bayesian network predictive models. We evaluate all combinations of Bayesian network models from different search algorithms, scoring functions, prior structure initializations, and sets of features. RESULTS: From the EHRs of 7,717 ICU patients, we construct Bayesian network predictive models from 86 medication, diagnosis, and Braden scale features. Our model not only identifies known and suspected high PU risk factors, but also substantially increases sensitivity of the prediction - nearly three times higher comparing to logistical regression models - without sacrificing the overall accuracy. We visualize a representative model with which our clinician collaborators identify strong relationships between risk factors widely recognized as associated with pressure ulcers. CONCLUSIONS: Given the strong adverse effect of pressure ulcers on patients and the high cost for treating pressure ulcers, our Bayesian network based model provides a novel framework for significantly improving the sensitivity of the prediction model. Thus, when the model is deployed in a clinical setting, the caregivers can suitably respond to conditions likely associated with pressure ulcer incidence.

Intravenous Infiltration Risk by Catheter Dwell Time Among Hospitalized Children.

PURPOSE: This study was aimed to examine the cumulative risk for infiltration over IV catheter dwell time by general or catheterization-specific characteristics of pediatric patients with IV therapy. DESIGN AND METHODS: This secondary data analysis was done with the data of 1596 children who received peripheral IV therapy at least once during their hospital stay between August 1st and October 30th, 2011 and in June, 2013 in an academic medical center, Yangsan, Republic of Korea. The survival functions of infiltration were determined by using the Kaplan-Meier analysis. RESULT: The cumulative risk for infiltration had rapidly increased from 1.5% after 24 hours of catheter dwell time to 17.3% after 96 hours. The survival functions were significantly different in the medical than in the surgical department (p=.005), lower extremities than upper ones (p=.001), and use of 10% dextrose (p=.001), ampicillin/sulbactam (p<.001), vancomycin (p=.024), high-concentration electrolytes (p=.001), and phenytoin (p<.001). CONCLUSION: When catheter dwell times are similar, the cumulative risk for infiltration was higher in cases wherein the patient had a risk factor. The cumulative risk for infiltration has rapidly increased after 24 hours in patients who have 10% dextrose, high-concentration electrolytes, and phenytoin. PRACTICE IMPLICATIONS: The results suggest that nurses are required to assess the IV site every hour after 24 hours of catheter dwell time for the infusion of irritants for a safer practice of IV therapy. However, this monitoring time may be modified by the age of child, previous IV complications, and/or hemodynamic issues which may impact IV integrity.

[CORRIGENDUM: Development of a Diabetic Foot Ulceration Prediction Model and Nomogram].

This corrects the article on p. 280 in vol. 51, PMID: 34215707.

[Development of a Diabetic Foot Ulceration Prediction Model and Nomogram].

PURPOSE: This study aimed to identify the risk factors for diabetic foot ulceration (DFU) to develop and evaluate the performance of a DFU prediction model and nomogram among people with diabetes mellitus (DM). METHODS: This unmatched case-control study was conducted with 379 adult patients (118 patients with DM and 261 controls) from four general hospitals in South Korea. Data were collected through a structured questionnaire, foot examination, and review of patients' electronic health records. Multiple logistic regression analysis was performed to build the DFU prediction model and nomogram. Further, their performance was analyzed using the Lemeshow-Hosmer test, concordance statistic (C-statistic), and sensitivity/specificity analyses in training and test samples. RESULTS: The prediction model was based on risk factors including previous foot ulcer or amputation, peripheral vascular disease, peripheral neuropathy, current smoking, and chronic kidney disease. The calibration of the DFU nomogram was appropriate (χ² = 5.85, p = .321). The C-statistic of the DFU nomogram was .95 (95% confidence interval .93~.97) for both the training and test samples. For clinical usefulness, the sensitivity and specificity obtained were 88.5% and 85.7%, respectively at 110 points in the training sample. The performance of the nomogram was better in male patients or those having DM for more than 10 years. CONCLUSION: The nomogram of the DFU prediction model shows good performance, and is thereby recommended for monitoring the risk of DFU and preventing the occurrence of DFU in people with DM.

Using the nursing interventions classification as a potential measure of nurse workload.

Standardized terminologies, such as the Nursing Interventions Classification (NIC) taxonomy, may be used in multiple ways to represent nursing constructs. This study is the first known to explore the NIC as a framework for the development of a nursing workload measure. While the NIC may not represent the complexity of nurses' work, the classification system may represent uniformly the work of nurses in health information systems to yield reliable data for a nursing workload measure.

Exploration of critical care data by using unsupervised machine learning.

BACKGROUND AND OBJECTIVE: Identification of subgroups may be useful to understand the clinical characteristics of ICU patients. The purposes of this study were to apply an unsupervised machine learning method to ICU patient data to discover subgroups among them; and to examine their clinical characteristics, therapeutic procedures conducted during the ICU stay, and discharge dispositions. METHODS: K-means clustering method was used with 1503 observations and 9 types of laboratory test results as features. RESULTS: Three clusters were identified from this specific population. Blood urea nitrogen, creatinine, potassium, hemoglobin, and red blood cell were distinctive between the clusters. Cluster Three presented the highest blood products transfusion rate (19.8%), followed by Cluster One (15.5%) and cluster Two (9.3%), which was significantly different. Hemodialysis was more frequently provided to Cluster Three while bronchoscopy was done to Cluster One and Two. Cluster Three showed the highest mortality (30.4%), which was more than two-fold compared to Cluster One (14.1%) and Two (12.2%). CONCLUSION: Three subgroups were identified and their clinical characteristics were compared. These findings may be useful to anticipate treatment strategies and probable outcomes of ICU patients. Unsupervised machine learning may enable ICU multi-dimensional data to be organized and to make sense of the data.

Iterative evaluation of the Health Level 7--Logical Observation Identifiers Names and Codes Clinical Document Ontology for representing clinical document names: a case report.

The authors summarize their experience in iteratively testing the adequacy of three versions of the Health Level Seven (HL7) Logical Observation Identifiers Names and Codes (LOINC) Clinical Document Ontology (CDO) to represent document names at Columbia University Medical Center. The percentage of documents fully represented increased from 23.4% (Version 1) to 98.5% (Version 3). The proportion of unique representations increased from 7.9% (Analysis 1) to 39.4% (Analysis 4); the proportion reflects the level of specificity in the document names as well as the completeness and level of granularity of the CDO. The authors shared the findings of each analysis with the Clinical LOINC committee and participated in the decision-making regarding changes to the CDO on the basis of those analyses and those conducted by the Department of Veterans Affairs. The authors encourage other institutions to actively engage in testing healthcare standards and participating in standards development activities to increase the likelihood that the evolving standards will meet institutional needs.

Development and evaluation of nursing user interface screens using multiple methods.

Building upon the foundation of the Structured Narrative Electronic Health Record (EHR) model, we applied theory-based (combined Technology Acceptance Model and Task-Technology Fit Model) and user-centered methods to explore nurses' perceptions of functional requirements for an electronic nursing documentation system, design user interface screens reflective of the nurses' perspectives, and assess nurses' perceptions of the usability of the prototype user interface screens. The methods resulted in user interface screens that were perceived to be easy to use, potentially useful, and well-matched to nursing documentation tasks associated with Nursing Admission Assessment, Blood Administration, and Nursing Discharge Summary. The methods applied in this research may serve as a guide for others wishing to implement user-centered processes to develop or extend EHR systems. In addition, some of the insights obtained in this study may be informative to the development of safe and efficient user interface screens for nursing document templates in EHRs.

Exploring the ability of natural language processing to extract data from nursing narratives.

Natural Language Processing (NLP) offers an approach for capturing data from narratives and creating structured reports for further computer processing. We explored the ability of a NLP system, Medical Language Extraction and Encoding (MedLEE), on nursing narratives. MedLEE extracted 490 concepts from narrative text in a sample of 553 oncology nursing process notes. The most frequently monitored and recorded signs and symptoms were related to chemotherapy care, such as adverse reactions, shortness of breath, nausea, pain, and bleeding. In terms of nursing interventions, chemotherapy, blood culture, medication, and blood transfusion were commonly recorded in free text. NLP may provide a feasible approach to extract data related to patient safety/quality measures and nursing outcomes by capturing nursing concepts that are not recorded through structured data entry. For better NLP performance in the domain of nursing, additional nursing terms and abbreviations must be added to MedLEE's lexicon.

Prediction Model for Hospital-Acquired Pressure Ulcer Development: Retrospective Cohort Study.

BACKGROUND: A pressure ulcer is injury to the skin or underlying tissue, caused by pressure, friction, and moisture. Hospital-acquired pressure ulcers (HAPUs) may not only result in additional length of hospital stay and associated care costs but also lead to undesirable patient outcomes. Intensive care unit (ICU) patients show higher risk for HAPU development than general patients. We hypothesize that the care team's decisions relative to HAPU risk assessment and prevention may be better supported by a data-driven, ICU-specific prediction model. OBJECTIVE: The aim of this study was to determine whether multiple logistic regression with ICU-specific predictor variables was suitable for ICU HAPU prediction and to compare the performance of the model with the Braden scale on this specific population. METHODS: We conducted a retrospective cohort study by using the data retrieved from the enterprise data warehouse of an academic medical center. Bivariate analyses were performed to compare the HAPU and non-HAPU groups. Multiple logistic regression was used to develop a prediction model with significant predictor variables from the bivariate analyses. Sensitivity, specificity, positive predictive values, negative predictive values, area under the receiver operating characteristic curve (AUC), and Youden index were used to compare with the Braden scale. RESULTS: The total number of patient encounters studied was 12,654. The number of patients who developed an HAPU during their ICU stay was 735 (5.81% of the incidence rate). Age, gender, weight, diabetes, vasopressor, isolation, endotracheal tube, ventilator episode, Braden score, and ventilator days were significantly associated with HAPU. The overall accuracy of the model was 91.7%, and the AUC was .737. The sensitivity, specificity, positive predictive value, negative predictive value, and Youden index were .650, .693, .211, 956, and .342, respectively. Male patients were 1.5 times more, patients with diabetes were 1.5 times more, and patients under isolation were 3.1 times more likely to have an HAPU than female patients, patients without diabetes, and patients not under isolation, respectively. CONCLUSIONS: Using an extremely large, electronic health record-derived dataset enabled us to compare characteristics of patients who develop an HAPU during their ICU stay with those who did not, and it also enabled us to develop a prediction model from the empirical data. The model showed acceptable performance compared with the Braden scale. The model may assist with clinicians' decision on risk assessment, in addition to the Braden scale, as it is not difficult to interpret and apply to clinical practice. This approach may support avoidable reductions in HAPU incidence in intensive care.

An electronic health record based on structured narrative.

OBJECTIVE: To develop an electronic health record that facilitates rapid capture of detailed narrative observations from clinicians, with partial structuring of narrative information for integration and reuse. DESIGN: We propose a design in which unstructured text and coded data are fused into a single model called structured narrative. Each major clinical event (e.g., encounter or procedure) is represented as a document that is marked up to identify gross structure (sections, fields, paragraphs, lists) as well as fine structure within sentences (concepts, modifiers, relationships). Marked up items are associated with standardized codes that enable linkage to other events, as well as efficient reuse of information, which can speed up data entry by clinicians. Natural language processing is used to identify fine structure, which can reduce the need for form-based entry. VALIDATION: The model is validated through an example of use by a clinician, with discussion of relevant aspects of the user interface, data structures and processing rules. DISCUSSION: The proposed model represents all patient information as documents with standardized gross structure (templates). Clinicians enter their data as free text, which is coded by natural language processing in real time making it immediately usable for other computation, such as alerts or critiques. In addition, the narrative data annotates and augments structured data with temporal relations, severity and degree modifiers, causal connections, clinical explanations and rationale. CONCLUSION: Structured narrative has potential to facilitate capture of data directly from clinicians by allowing freedom of expression, giving immediate feedback, supporting reuse of clinical information and structuring data for subsequent processing, such as quality assurance and clinical research.

Evidence-based staffing: potential roles for informatics.

Over the last 15 years, evidence has been accumulating relating higher levels of nurse staffing in both quantity and experience to lower rates of adverse patient outcomes. Consequently, to promote quality patient outcomes efficiently, making staffing decisions based in evidence is of increasing importance. However, there is still limited data to help decide how to effectively allocate scarce nurse resources in practice. Existing principles, frameworks, and guidelines provide a foundation for nurse staffing decisions but face poor adoption. To determine optimal nurse staffing practices and provide evidence-based recommendations for policy, and integration into operations, comprehensive data are necessary. Information technology can assist nurse staffing decisions. Four informatics processes that may support evidence-based nurse staffing are described: (a) Data acquisition from multiple data sources, (b) Representation of data in a way it can be re-used for multiple purposes, (c) Sophisticated data processing and mining, and (d) Presentation of data in standardized and user-configurable ways.

Is the Health Level 7/LOINC document ontology adequate for representing nursing documents?

The use of nursing documents from different electronic health record (EHR) systems is challenging due to inconsistency in document naming across systems and institutions. Mapping each local document name to standard document ontology may enable health care professionals to navigate and retrieve documents efficiently for multiple purposes such as quality assurance, outcomes research or public health reporting. The purpose of this study was to evaluate the sufficiency of the Health Level 7 (HL7)/Logical Observation Identifiers, Names, and Codes (LOINC) document ontology for representing nursing document names. We collected 94 nursing document types from the Eclipsys Clinical Information System (CIS) and the Columbia Medical Entities Dictionary (MED) and mapped them to the components of the HL7/LOINC document ontology. Seventy-five (79.8%) nursing document names were completely represented and 19 (20.2%) document names were partially represented. In order for the HL7/LOINC document ontology to be of more use in implementing EHRs that support nursing documentation, Subject Matter Domain and Type of Service axes require extension and clarification.

Markup of temporal information in electronic health records.

Temporal information plays a critical role in the understanding of clinical narrative (i.e., free text). We developed a representation for marking up temporal information in a narrative, consisting of five elements: 1) reference point, 2) direction, 3) number, 4) time unit, and 5) pattern. We identified 254 temporal expressions from 50 discharge summaries and represented them using our scheme. The overall inter-rater reliability among raters applying the representation model was 75 percent agreement. The model can contribute to temporal reasoning in computer systems for decision support, data mining, and process and outcomes analyses by providing structured temporal information.

Usefulness of a personal digital assistant-based advanced practice nursing student clinical log: Faculty stakeholder exemplars.

The number of health sciences educational programs that are integrating personal digital assistants (PDAs) into their curricula is on the rise. In this paper, we report an evaluation of the usefulness of a PDA-based advanced practice nursing (APN) student clinical log through faculty stakeholder exemplars in three areas: pediatric asthma care; procedures of Acute Care Nurse Practitioner (NP) students; and diagnostic and screening procedures of Women's Health NP students. We generated descriptive data through routine queries and through custom SQL queries at the request of a specific faculty member who wished to examine a particular aspect of an educational program. In addition, we discussed the potential implications of the data with the respective faculty members. The exemplars provide evidence that faculty stakeholders found the APN student clinical log to be useful for a variety of purposes including monitoring of student performance, benchmarking, and quality of care assessments.

ISO reference terminology models for nursing: applicability for natural language processing of nursing narratives.

Natural language processing (NLP) systems have demonstrated utility in parsing narrative texts for purposes such as surveillance and decision support. However, there has been little work related to NLP of nursing narratives. The purpose of this study was to compare the semantic categories of a NLP system (Medical Language Extraction and Encoding [MedLEE] system) with the semantic domains, categories, and attributes of the International Standards Organization (ISO) reference terminology models for nursing diagnoses and nursing actions. All but two MedLEE diagnosis and procedure-related semantic categories mapped to ISO models. In some instances, we found exact correspondence between the semantic structures of MedLEE and the ISO models. In other situations (e.g. aspects of Site or Location), the ISO model was not as granular as MedLEE. For clinical procedure and non-invasive examination, two ISO nursing action model components (Action and Target) mapped to a single MedLEE semantic category. The ISO models are applicable to NLP of nursing narratives. However, the ISO models require additional specification of selected semantic categories for the abstract semantic domains in order to achieve the objective of using NLP to parse and encode data from nursing narratives. Our analysis also suggests areas for extension of MedLEE particularly in regard to represent nursing actions.

Predictive Modeling for Pressure Ulcers from Intensive Care Unit Electronic Health Records.

Our goal in this study is to find risk factors associated with Pressure Ulcers (PUs) and to develop predictive models of PU incidence. We focus on Intensive Care Unit (ICU) patients since patients admitted to ICU have shown higher incidence of PUs. The most common PU incidence assessment tool is the Braden scale, which sums up six subscale features. In an ICU setting it's known drawbacks include omission of important risk factors, use of subscale features not significantly associated with PU incidence, and yielding too many false positives. To improve on this, we extract medication and diagnosis features from patient EHRs. Studying Braden, medication, and diagnosis features and combinations thereof, we evaluate six types of predictive models and find that diagnosis features significantly improve the models' predictive power. The best models combine Braden and diagnosis. Finally, we report the top diagnosis features which compared to Braden improve AUC by 10%.

Promoting patient safety through informatics-based nursing education.

The Institute of Medicine (IOM) Committee on Quality of Health Care in America identified the critical role of information technology in designing safe and effective health care. In addition to technical aspects such as regional or national health information infrastructures, to achieve this goal, healthcare professionals must receive the requisite training during basic and advanced educational programs. In this article, we describe a two-pronged strategy to promote patient safety through an informatics-based approach to nursing education at the Columbia University School of Nursing: (1) use of a personal digital assistant (PDA) to document clinical encounters and to retrieve patient safety-related information at the point of care, and (2) enhancement of informatics competencies of students and faculty. These approaches may be useful to others wishing to promote patient safety through using informatics methods and technologies in healthcare curricula.

MobileNurse: hand-held information system for point of nursing care.

Healthcare information travels with patients and clinicians and therefore the need for information to be ubiquitously available is key to reliable patient care and reliable medical systems. We have implemented MobileNurse, a prototype point-of-care system using PDA. MobileNurse has four modules each of which performs: (1) patient information management; (2) medical order check; (3) nursing recording; and (4) nursing care plan. MobileNurse provides easy input interface and various outputs for nursing records. The system consists of PDAs and a mobile support system (MSS) which supports clinical data exchange between PDAs and hospital information system. Two synchronization modules have been developed to keep the patient data consistent between PDAs and MSS. Clinical trials were performed with six volunteered nurses. They tried MobileNurse for 1-day caring-simulated patients. According to the survey after the trials, most of volunteers agreed that MobileNurse is more helpful and convenient than other non-mobile care systems to check medical orders and retrieve the results of recent clinical tests at the bedside. Through the involvement, we found out that ease-to-use interface is the most critical successful factor for mobile patient care systems.