1
|
Pradhan R, Dayama N, Morris M, Elliott K, Felix H. Enhancing nursing home quality through electronic health record implementation. HEALTH INF MANAG J 2024:18333583241274010. [PMID: 39183673 DOI: 10.1177/18333583241274010] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 08/27/2024]
Abstract
Background: The quality of care in nursing homes (NHs) in the United States has long been a matter of policy concern. Although electronic health records (EHRs) are argued to improve quality, implementation has lagged due to various factors such as financial constraints and limited research on their impact on NH quality. Objective: This study examined the relationship between EHR implementation and NH quality using Donabedian's structure-process-outcome model. Method: Data on EHR implementation were collected via a 2018 survey of all Federally certified Arkansas NHs (n = 223). Of the 63 responding NHs, 48 reported EHR implementation. Survey data were merged with secondary sources such as Certification and Survey Provider Enhanced Reporting. A total of 744 NH-years for the period 2008-2020 were included in the final sample. A pre-post negative binomial panel data regression was used to examine the relationship between EHR implementation (dichotomous variable) and NH deficiencies (dependent count variable) with facility/community-level control variables. Results were reported as incidence rate ratios (IRR). Results: NHs that had implemented EHR experienced an 18% reduction in the rate of deficiencies compared to those without EHR systems (IRR = 0.82, 95% CI [0.70, 0.99], p = 0.035). Conclusion: EHR implementation had a favourable impact on NH quality. Implications: Past research suggests that higher NH quality may be associated with improved financial performance. Therefore, EHR implementation has the potential to address two critical challenges: enhancing care quality and improving financial outcomes. However, government financial incentives may be necessary to address the high-cost of implementing EHR systems.
Collapse
Affiliation(s)
| | | | | | | | - Holly Felix
- University of Arkansas for Medical Sciences, USA
| |
Collapse
|
2
|
Wang X, Rihari-Thomas J, Bail K, Bala N, Traynor V. Care quality and safety in long-term aged care settings: A systematic review and narrative analysis of missed care measurements. J Adv Nurs 2024. [PMID: 39092879 DOI: 10.1111/jan.16358] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/27/2024] [Revised: 07/08/2024] [Accepted: 07/19/2024] [Indexed: 08/04/2024]
Abstract
AIM To critically evaluate missed care measurement approaches and their application in long-term aged care (LTAC) settings. DESIGN Systematic review using Tawfik's guideline. DATA SOURCES PubMed, Scopus, Web of Science, CINAHL and ProQuest were searched. Supplemental searching was from reference lists of retrieved records, first authors' ORCID homepages and Google advanced search for grey literature. Search limitations were English language, published between 1 January 2001 and 31 December 2022. REVIEW METHOD COVIDENCE was utilized for screening, data extraction and quality appraisal. JBI Critical Appraisal Tools and COSMIN Risk of Bias Tool were used for quality appraisal. Data were summarized and synthesized using narrative analysis. RESULTS Twenty-four publications across 11 regions were included, with two principal methods of missed care measurement: modified standard scales and tailored specific approaches. They were applied inconsistently and generated diverse measurement outcomes. There were challenges even with the most commonly used tool, the BERNCA-NH, including absence of high-quality verification through comparative analysis against an established 'gold standard', reliance on self-administration, incomplete assessment of constructs and inadequate exploration of psychometric properties. CONCLUSION Globally, there are deficiencies in the effectiveness and comprehensiveness of the instruments measuring missed care in LTAC settings. Further research on theoretical and practical perspectives is required. IMPLICATIONS Findings highlighted a critical need to establish a standardized, validated approach to measure missed care in LTAC settings. This review calls for collaborative efforts by researchers, clinical staff and policymakers to develop and implement evidence-based practices as a way of safeguarding the well-being of older clients living in LTAC settings. IMPACT Measurements of missed care in LTAC settings rely on adapting acute care tools. There is a critical gap in measuring missed care in LTAC settings. Developing a new tool could improve care quality and safety in LTAC settings globally. REPORTING METHOD Adhered to PRISMA guideline. PATIENT OR PUBLIC CONTRIBUTION No patient or public contribution.
Collapse
Affiliation(s)
- Xinxia Wang
- School of Nursing, Faculty of Health, Medicine and Health, University of Wollongong, Wollongong, Australian Capital Territory, Australia
| | - John Rihari-Thomas
- Sydney Nursing School, Faculty of Medicine and Health, University of Sydney, Sydney, Australia
| | - Kasia Bail
- Centre for Ageing Research and Translation, University of Canberra and Synergy Nursing and Midwifery Research Centre ACT Health Directorate, Canberra, Australian Capital Territory, Australia
| | - Nina Bala
- School of Nursing, Faculty of Health, Medicine and Health, University of Wollongong, Wollongong, Australian Capital Territory, Australia
| | - Victoria Traynor
- School of Nursing, Faculty of Health, Medicine and Health, University of Wollongong, Wollongong, Australian Capital Territory, Australia
- Aged and Dementia Health Education and Research (ADHERe) Centre, Wollongong, Australian Capital Territory, Australia
| |
Collapse
|
3
|
Saladino C, Favez L, Serdaly C, Ausserhofer D, De Geest S, Zúñiga F. Characteristics and Activities of Nurses in Expanded Roles Employed in Swiss Nursing Homes: A Cross-Sectional Study. Int J Older People Nurs 2024; 19:e12631. [PMID: 38989647 DOI: 10.1111/opn.12631] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/22/2023] [Revised: 04/29/2024] [Accepted: 06/22/2024] [Indexed: 07/12/2024]
Abstract
BACKGROUND As the number of nursing home residents with multiple healthcare needs grows, the demand for nursing expertise increases. The implementation of new care models involving nurses with expanded roles is crucial for ensuring quality care in nursing homes. OBJECTIVES To investigate the characteristics and activities of nurses employed in nursing homes in expanded roles and the factors associated with variation in the activities performed. METHODS This multicentre cross-sectional survey in Switzerland collected data from a convenience sample of 118 nursing homes between September 2018 and October 2019. From a subsample of 62 nursing homes, we analysed the characteristics and activities of 104 nurses in expanded roles. Associations between the activities performed and the educational background of the nurses in expanded roles, their direct supervisors' positions and the presence of physicians in the nursing homes were examined. RESULTS Most Registered Nurses in expanded roles were diploma educated (48%), with fewer having a bachelor's (35%) or master's degree (17%). Overall, direct clinical practice and guidance and coaching activities were conducted monthly to weekly; consultation, evidence-based practice, collaboration and ethical decision-making activities were conducted monthly. We saw variations where a higher educational background was associated with more frequent evidence-based practice activities (z = 3.47, p < 0.001), and if direct supervisors were ward managers, nurses in expanded roles worked more frequently below their scope of practice (z = 4.10, p < 0.001). CONCLUSION This is the first study to use Hamric's integrative Advanced Practice Nursing model to examine the activities of nurses in expanded roles in nursing homes. We found considerable variation in their activities, where nursing homes seem to adapt their roles to their educational background and the local context. IMPLICATIONS FOR PRACTICE Our findings show the importance of clarifying role expectations for Registered Nurses in expanded roles, allowing them to practice at the top of the licence to meet residents' complex healthcare needs.
Collapse
Affiliation(s)
- Christian Saladino
- Department of Public Health, Institute of Nursing Science, University of Basel, Basel, Switzerland
- Alterszentrum am Bachgraben, Allschwil, Switzerland
| | - Lauriane Favez
- Department of Public Health, Institute of Nursing Science, University of Basel, Basel, Switzerland
- School of Engineering and Management Vaud, HES-SO University of Applied Sciences and Arts Western Switerland, Delémont, Switzerland
| | | | - Dietmar Ausserhofer
- Department of Public Health, Institute of Nursing Science, University of Basel, Basel, Switzerland
- Claudiana Research, College of Health Care-Professions Claudiana, Bozen, Italy
| | - Sabina De Geest
- Department of Public Health, Institute of Nursing Science, University of Basel, Basel, Switzerland
- Department of Public Health and Primary Care, Academic Center for Nursing and Midwifery, KU Leuven, Leuven, Belgium
| | - Franziska Zúñiga
- Department of Public Health, Institute of Nursing Science, University of Basel, Basel, Switzerland
| |
Collapse
|
4
|
Cocchieri A, Pezzullo AM, Cesare M, De Rinaldis M, Cristofori E, D'Agostino F. Association between health literacy and nursing care in hospital: A retrospective study. J Clin Nurs 2024; 33:642-652. [PMID: 37807642 DOI: 10.1111/jocn.16899] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/15/2023] [Revised: 08/31/2023] [Accepted: 09/27/2023] [Indexed: 10/10/2023]
Abstract
AIMS To describe the health literacy (HL) levels of hospitalised patients and their relationship with nursing diagnoses (NDs), nursing interventions and nursing measures for clinical risks. DESIGN Retrospective study. METHODS The study was conducted from December 2020 to December 2021 in an Italian university hospital. From 146 wards, 1067 electronic nursing records were randomly selected. The Single-Item Literacy Screener was used to measure HL. Measures for clinical risks were systematically assessed by nurses using Conley Index score, the Blaylock Risk Assessment Screening Score, Braden score, and the Barthel Index. A univariable linear regression model was used to assess the associations of HL with NDs. RESULTS Patients with low HL reported a higher number of NDs, interventions and higher clinical risks. HL can be considered a predictor of complexity of care. CONCLUSIONS The inclusion of standardised terms in nursing records can describe the complexity of care and facilitate the predictive ability on hospital outcomes. IMPLICATIONS FOR THE PROFESSION AND/OR PATIENT CARE HL evaluation during the first 24 h. From hospital admission could help to intercept patients at risk of higher complexity of care. These results can guide the development of interventions to minimise needs after discharge. PATIENT OR PUBLIC CONTRIBUTION No patient or public contribution was required to design or undertake this research. Patients contributed only to the data collection.
Collapse
Affiliation(s)
- Antonello Cocchieri
- Section of Hygiene, Woman and Child Health and Public Health, Fondazione Policlinico Universitario A. Gemelli, IRCCS, Rome, Italy
| | - Angelo Maria Pezzullo
- Section of Hygiene, Department of Life Sciences and Public Health, Università Cattolica del Sacro Cuore, Rome, Italy
| | - Manuele Cesare
- Department of Biomedicine and Prevention, University of Tor Vergata, Rome, Italy
| | - Miriam De Rinaldis
- Section of Hygiene, Woman and Child Health and Public Health, Fondazione Policlinico Universitario A. Gemelli, IRCCS, Rome, Italy
| | - Elena Cristofori
- Department of Biomedicine and Prevention, University of Tor Vergata, Rome, Italy
| | - Fabio D'Agostino
- Saint Camillus International University of Health Sciences, Rome, Italy
| |
Collapse
|
5
|
Ausserhofer D, Tappeiner W, Wieser H, Serdaly C, Simon M, Zúñiga F, Favez L. Administrative burden in Swiss nursing homes and its association with care workers' outcomes-a multicenter cross-sectional study. BMC Geriatr 2023; 23:347. [PMID: 37268879 DOI: 10.1186/s12877-023-04022-w] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/30/2022] [Accepted: 05/06/2023] [Indexed: 06/04/2023] Open
Abstract
BACKGROUND Care workers in nursing homes often perform tasks that are rather related to organizational or management activities than 'direct patient care'. 'Indirect care activities', such as documentation or other administrative tasks are often considered by care workers as a burden, as they increase overall workload and keep them away from caring for residents. So far, there is little investigation into what kind of administrative tasks are being performed in nursing homes, by which type of care workers, and to which extent, nor how administrative burden is associated with care workers' outcomes. PURPOSE The objective of this study was to describe care workers' administrative burden in Swiss nursing homes and to explore the association with four care worker outcomes (i.e., job dissatisfaction, emotional exhaustion, intention to leave the current job and the profession). METHODS This multicenter cross-sectional study used survey data from the Swiss Nursing Homes Human Resources Project 2018. It included a convenience sample of 118 nursing homes and 2'207 care workers (i.e., registered nurses, licensed practical nurses) from Switzerland's German- and French-speaking regions. Care workers completed questionnaires assessing the administrative tasks and burden, staffing and resource adequacy, leadership ability, implicit rationing of nursing care and care worker characteristics and outcomes. For the analysis, we applied generalized linear mixed models, including individual-level nurse survey data and data on unit and facility characteristics. RESULTS Overall, 73.9% (n = 1'561) of care workers felt strongly or rather strongly burdened, with one third (36.6%, n = 787) reporting to spend 2 h or more during a "normal" day performing administrative tasks. Ratings for administrative burden ranged from 42.6% (n = 884; ordering supplies and managing stocks) to 75.3% (n = 1'621; filling out the resident's health record). One out of four care workers (25.5%, n = 561) intended to leave the profession, whereby care workers reporting higher administrative task burden (OR = 1.24; 95%CI: 1.02-1.50) were more likely to intend to leave the profession. CONCLUSION This study provides first insights on care workers' administrative burden in nursing homes. By limiting care workers' burdensome administrative tasks and/or shifting such tasks from higher to lower educated care workers or administrative personnel when appropriate, nursing home managers could reduce care workers' workload and improve their job satisfaction and retention in the profession.
Collapse
Affiliation(s)
- Dietmar Ausserhofer
- Claudiana Research, College of Health Care-Professions, Bolzano-Bozen, Italy
- Institute of Nursing Science, Department of Public Health, University of Basel, Bernoullistr. 28, 4056, Basel, Switzerland
| | - Waltraud Tappeiner
- Claudiana Research, College of Health Care-Professions, Bolzano-Bozen, Italy
| | - Heike Wieser
- Claudiana Research, College of Health Care-Professions, Bolzano-Bozen, Italy
| | - Christine Serdaly
- Serdaly&Ankers Snc, 210 Route de Florissant, 1231, Conches, Switzerland
| | - Michael Simon
- Institute of Nursing Science, Department of Public Health, University of Basel, Bernoullistr. 28, 4056, Basel, Switzerland
| | - Franziska Zúñiga
- Institute of Nursing Science, Department of Public Health, University of Basel, Bernoullistr. 28, 4056, Basel, Switzerland
| | - Lauriane Favez
- Institute of Nursing Science, Department of Public Health, University of Basel, Bernoullistr. 28, 4056, Basel, Switzerland.
- School of Engineering and Management Vaud, HES-SO University of Applied Sciences and Arts Western Switzerland, Yverdon-les-Bains, Switzerland.
| |
Collapse
|
6
|
Li HQ, Xie P, Huang X, Luo SX. The experience of nurses to reduce implicit rationing of nursing care: a phenomenological study. BMC Nurs 2023; 22:174. [PMID: 37208756 DOI: 10.1186/s12912-023-01334-5] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/18/2022] [Accepted: 05/10/2023] [Indexed: 05/21/2023] Open
Abstract
BACKGROUND Implicit rationing of nursing care can adversely affect patient safety and the quality of care, and increase nurses' burnout and turnover tendency. Implicit rationing care occurs at the nurse-to-patient level (micro-level), and nurses are direct participants. Therefore, the strategies based on experience of nurses to reduce implicit rationing care have more reference value and promotion significance. The aim of the study is to explore the experience of nurses to reduce implicit rationing care, thereby to provide references for conducting randomized controlled trials to reduce implicit rationing care. METHODS This is a descriptive phenomenological study. Purpose sampling was conducted nationwide. There are 17 nurses were selected and semi-structured in-depth interviews were conducted. The interviews were recorded, transcribed verbatim and analyzed via thematic analysis. RESULTS Our study found that nurses' reported experience of coping with implicit rationing of nursing care contained three aspects: personal, resource, and managerial. Three themes were extracted from the results of the study: (1) improving personal literacy; (2) supplying and optimizing resources and (3) standardizing management mode. The improvement of nurses' own qualities are the prerequisites, the supply and optimization of resources is an effective strategy, and clear scope of work has attracted the attention of nurses. CONCLUSION The experience of dealing with implicit nursing rationing includes many aspects. Nursing managers should be grounded in nurses' perspectives when developing strategies to reduce implicit rationing of nursing care. Promoting the improvement of nurses' skills, improving staffing level and optimizing scheduling mode are promising measures to reduce hidden nursing rationing.
Collapse
Affiliation(s)
- Hui Qin Li
- Mental Health Center, West China School of Nursing, West China Hospital, Sichuan University, No. 37 Guoxue Road, Chengdu, Sichuan Province, 610041, P.R. China
| | - Peng Xie
- Surgical Anesthesia Center, West China School of Nursing, West China Hospital, Sichuan University, No. 28 Telecom South Street, Chengdu, Sichuan Province, 610041, P.R. China
| | - Xia Huang
- Mental Health Center, West China School of Nursing, West China Hospital, Sichuan University, No. 37 Guoxue Road, Chengdu, Sichuan Province, 610041, P.R. China.
| | - Shan Xia Luo
- Mental Health Center, West China School of Nursing, West China Hospital, Sichuan University, No. 37 Guoxue Road, Chengdu, Sichuan Province, 610041, P.R. China.
| |
Collapse
|
7
|
Zhao Y, Rokhani FZ, Sazlina SG, Devaraj NK, Su J, Chew BH. Defining the concepts of a smart nursing home and its potential technology utilities that integrate medical services and are acceptable to stakeholders: a scoping review. BMC Geriatr 2022; 22:787. [PMID: 36207705 PMCID: PMC9540152 DOI: 10.1186/s12877-022-03424-6] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/17/2021] [Accepted: 08/29/2022] [Indexed: 11/30/2022] Open
Abstract
BACKGROUND AND OBJECTIVES Smart technology in nursing home settings has the potential to elevate an operation that manages more significant number of older residents. However, the concepts, definitions, and types of smart technology, integrated medical services, and stakeholders' acceptability of smart nursing homes are less clear. This scoping review aims to define a smart nursing home and examine the qualitative evidence on technological feasibility, integration of medical services, and acceptability of the stakeholders. METHODS Comprehensive searches were conducted on stakeholders' websites (Phase 1) and 11 electronic databases (Phase 2), for existing concepts of smart nursing home, on what and how technologies and medical services were implemented in nursing home settings, and acceptability assessment by the stakeholders. The publication year was inclusive from January 1999 to September 2021. The language was limited to English and Chinese. Included articles must report nursing home settings related to older adults ≥ 60 years old with or without medical demands but not bed-bound. Technology Readiness Levels were used to measure the readiness of new technologies and system designs. The analysis was guided by the Framework Method and the smart technology adoption behaviours of elder consumers theoretical model. The results were reported according to the PRISMA-ScR. RESULTS A total of 177 literature (13 website documents and 164 journal articles) were selected. Smart nursing homes are technology-assisted nursing homes that allow the life enjoyment of their residents. They used IoT, computing technologies, cloud computing, big data and AI, information management systems, and digital health to integrate medical services in monitoring abnormal events, assisting daily living, conducting teleconsultation, managing health information, and improving the interaction between providers and residents. Fifty-five percent of the new technologies were ready for use in nursing homes (levels 6-7), and the remaining were proven the technical feasibility (levels 1-5). Healthcare professionals with higher education, better tech-savviness, fewer years at work, and older adults with more severe illnesses were more acceptable to smart technologies. CONCLUSIONS Smart nursing homes with integrated medical services have great potential to improve the quality of care and ensure older residents' quality of life.
Collapse
Affiliation(s)
- Yuanyuan Zhao
- Department of Family Medicine, Faculty of Medicine & Health Sciences, Universiti Putra Malaysia, 43400 Serdang, Selangor Malaysia
- Global Century Science Group, Hong Kong, China
| | - Fakhrul Zaman Rokhani
- Faculty of Engineering, Universiti Putra Malaysia, Serdang, Malaysia
- Malaysian Research Institute on Ageing (MyAgeingTM), Universiti Putra Malaysia, Serdang, Malaysia
| | - Shariff-Ghazali Sazlina
- Department of Family Medicine, Faculty of Medicine & Health Sciences, Universiti Putra Malaysia, 43400 Serdang, Selangor Malaysia
- Malaysian Research Institute on Ageing (MyAgeingTM), Universiti Putra Malaysia, Serdang, Malaysia
| | - Navin Kumar Devaraj
- Department of Family Medicine, Faculty of Medicine & Health Sciences, Universiti Putra Malaysia, 43400 Serdang, Selangor Malaysia
- Malaysian Research Institute on Ageing (MyAgeingTM), Universiti Putra Malaysia, Serdang, Malaysia
| | - Jing Su
- College of Public Health, Hainan Medical University, Haikou, China
| | - Boon-How Chew
- Department of Family Medicine, Faculty of Medicine & Health Sciences, Universiti Putra Malaysia, 43400 Serdang, Selangor Malaysia
- Clinical Research Unit, Hospital Pengajar Universiti Putra Malaysia (HPUPM Teaching Hospital), Serdang, Malaysia
| |
Collapse
|
8
|
Prevalence, type, and reasons for missed nursing care in municipality health care in Sweden – A cross sectional study. BMC Nurs 2022; 21:95. [PMID: 35462537 PMCID: PMC9035238 DOI: 10.1186/s12912-022-00874-6] [Citation(s) in RCA: 7] [Impact Index Per Article: 3.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/18/2021] [Accepted: 04/14/2022] [Indexed: 11/22/2022] Open
Abstract
Background With an ageing population, there is an increasing need for care, both as home care and in nursing homes. However, some needed care is not carried out for different reasons, which can affect patient safety. The aim of the study was to describe prevalence, type, and reasons for missed nursing care in home care and nursing homes, from nurses’ perspective. Methods A cross sectional design with quantitative and qualitative approach. A Swedish version of Basel Extent of Rationing of Nursing Care for nursing homes and 15 study specific questions were answered by 624 registered nurses, enrolled nurses, or nurse assistants. Both descriptive and analytical, independent-samples t-test, analyses were used. Qualitative content analysis was used for the open-ended question. Results The care activity most often missed in home care was: ‘set up or update care plans’ (41.8%), and in nursing homes: ‘scheduled group activity’ (22.8%). Reasons for missed nursing care were lack of preparedness for unexpected situations, obstacles in a deficient work environment, unsatisfactory planning in the organisation, and/or shortcomings related to the individual. Conclusion Not all care activities needed are performed, due to reasons such as lack of time or organisational issues. Missed nursing care can lead to adverse events and affect patient safety. It is important to be aware of missed nursing care and the reasons for it, which gives a possibility to initiate quality improvement work to ensure patient safety.
Collapse
|
9
|
Zanotto BS, Beck da Silva Etges AP, Dal Bosco A, Cortes EG, Ruschel R, De Souza AC, Andrade CMV, Viegas F, Canuto S, Luiz W, Ouriques Martins S, Vieira R, Polanczyk C, André Gonçalves M. Stroke Outcome Measurements From Electronic Medical Records: Cross-sectional Study on the Effectiveness of Neural and Nonneural Classifiers. JMIR Med Inform 2021; 9:e29120. [PMID: 34723829 PMCID: PMC8593798 DOI: 10.2196/29120] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/29/2021] [Revised: 06/27/2021] [Accepted: 08/05/2021] [Indexed: 01/20/2023] Open
Abstract
BACKGROUND With the rapid adoption of electronic medical records (EMRs), there is an ever-increasing opportunity to collect data and extract knowledge from EMRs to support patient-centered stroke management. OBJECTIVE This study aims to compare the effectiveness of state-of-the-art automatic text classification methods in classifying data to support the prediction of clinical patient outcomes and the extraction of patient characteristics from EMRs. METHODS Our study addressed the computational problems of information extraction and automatic text classification. We identified essential tasks to be considered in an ischemic stroke value-based program. The 30 selected tasks were classified (manually labeled by specialists) according to the following value agenda: tier 1 (achieved health care status), tier 2 (recovery process), care related (clinical management and risk scores), and baseline characteristics. The analyzed data set was retrospectively extracted from the EMRs of patients with stroke from a private Brazilian hospital between 2018 and 2019. A total of 44,206 sentences from free-text medical records in Portuguese were used to train and develop 10 supervised computational machine learning methods, including state-of-the-art neural and nonneural methods, along with ontological rules. As an experimental protocol, we used a 5-fold cross-validation procedure repeated 6 times, along with subject-wise sampling. A heatmap was used to display comparative result analyses according to the best algorithmic effectiveness (F1 score), supported by statistical significance tests. A feature importance analysis was conducted to provide insights into the results. RESULTS The top-performing models were support vector machines trained with lexical and semantic textual features, showing the importance of dealing with noise in EMR textual representations. The support vector machine models produced statistically superior results in 71% (17/24) of tasks, with an F1 score >80% regarding care-related tasks (patient treatment location, fall risk, thrombolytic therapy, and pressure ulcer risk), the process of recovery (ability to feed orally or ambulate and communicate), health care status achieved (mortality), and baseline characteristics (diabetes, obesity, dyslipidemia, and smoking status). Neural methods were largely outperformed by more traditional nonneural methods, given the characteristics of the data set. Ontological rules were also effective in tasks such as baseline characteristics (alcoholism, atrial fibrillation, and coronary artery disease) and the Rankin scale. The complementarity in effectiveness among models suggests that a combination of models could enhance the results and cover more tasks in the future. CONCLUSIONS Advances in information technology capacity are essential for scalability and agility in measuring health status outcomes. This study allowed us to measure effectiveness and identify opportunities for automating the classification of outcomes of specific tasks related to clinical conditions of stroke victims, and thus ultimately assess the possibility of proactively using these machine learning techniques in real-world situations.
Collapse
Affiliation(s)
- Bruna Stella Zanotto
- National Institute of Health Technology Assessment - INCT/IATS (CNPQ 465518/2014-1), Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil.,Graduate Program in Epidemiology, Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil
| | - Ana Paula Beck da Silva Etges
- National Institute of Health Technology Assessment - INCT/IATS (CNPQ 465518/2014-1), Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil.,School of Technology, Pontifícia Universidade Católica do Rio Grande do Sul, Porto Alegre, Brazil
| | - Avner Dal Bosco
- School of Technology, Pontifícia Universidade Católica do Rio Grande do Sul, Porto Alegre, Brazil
| | - Eduardo Gabriel Cortes
- Graduate Program of Computer Science, Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil
| | - Renata Ruschel
- National Institute of Health Technology Assessment - INCT/IATS (CNPQ 465518/2014-1), Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil
| | | | - Claudio M V Andrade
- Computer Science Department, Universidade Federal de Minas Gerais, Belo Horizonte, Brazil
| | - Felipe Viegas
- Computer Science Department, Universidade Federal de Minas Gerais, Belo Horizonte, Brazil
| | - Sergio Canuto
- Computer Science Department, Universidade Federal de Minas Gerais, Belo Horizonte, Brazil
| | - Washington Luiz
- Computer Science Department, Universidade Federal de Minas Gerais, Belo Horizonte, Brazil
| | | | - Renata Vieira
- Centro Interdisciplinar de História, Culturas e Sociedades (CIDEHUS), Universidade de Évora, Évora, Portugal
| | - Carisi Polanczyk
- National Institute of Health Technology Assessment - INCT/IATS (CNPQ 465518/2014-1), Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil.,Graduate Program in Epidemiology, Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil
| | - Marcos André Gonçalves
- Computer Science Department, Universidade Federal de Minas Gerais, Belo Horizonte, Brazil
| |
Collapse
|