BMC Medical Informatics and Decision Making

Intensive care information systems (ICIS) are continuously evolving to meet the ever changing information needs of intensive care units (ICUs), providing the backbone for a safe, intelligent and efficient patient care environment. Although beneficial for the international advancement in building smart environments to transform ICU services, knowledge about the contemporary development of ICIS worldwide, their usage and impacts is limited. This study aimed to fill this knowledge gap by researching the development and implementation of an ICIS in a Chinese hospital, nurses’ use of the system, and the impact of system use on critical care nursing processes and outcomes. This descriptive case study was conducted in a 14-bed Respiratory ICU in a tertiary hospital in Beijing. Participative design was the method used for ICU nurses, hospital IT department and a software company to collaboratively research and develop the ICIS. Focus group discussions were conducted to understand the subjective perceptions of the nurses toward the ICIS. Nursing documentation time and quality were compared before and after system implementation. ICU nursing performance was extracted from the annual nursing performance data collected by the hospital. A participative design process was followed by the nurses in the ICU, the hospital IT staff and the software engineers in the company to develop and implement a highly useful ICIS. Nursing documentation was fully digitized and was significantly improved in quality and efficiency. The wrong data, missing data items and calculation errors were significantly reduced. Nurses spent more time on direct patient care after the introduction of the ICIS. The accuracy and efficiency of medication administration was also improved. The outcome was improvement in ward nursing performance as measured by ward management, routine nursing practices, disinfection and isolation, infection rate and mortality rate. Nurses in this ICU unit in China actively participated in the ICIS development and fully used the system to document care. Introduction of the ICIS led to significant improvement in quality and efficiency in nursing documentation, medication order transcription and administration. It allowed nurses to spend more time with patients to improve quality of care. These led to improvement in overall nursing performance. Further study should investigate how the ICIS system contributes to the improvement in decision making of ICU nurses and intensivists.

The present paper is a first evaluation of the use of "CEAwatch", a clinical support software system for surgeons for the follow-up of colorectal cancer (CRC) patients. This system gathers Carcino-Embryonic Antigen (CEA) values and automatically returns a recommendation based on the latest values. Consecutive patients receiving follow-up care for CRC fulfilling our in- and exclusion criteria were identified to participate in this study. From August 2008, when the software was introduced, patients were asked to undergo the software-supported follow-up. Safety of the follow-up, experiences of working with the software, and technical issues were analyzed. 245 patients were identified. The software-supported group contained 184 patients; the control group contained 61 patients. The software was safe in finding the same amount of recurrent disease with fewer outpatient visits, and revealed few technical problems. Clinicians experienced a decrease in follow-up workload of up to 50% with high adherence to the follow-up scheme. CEAwatch is an efficient software tool helping clinicians working with large numbers of follow-up patients. The number of outpatient visits can safely be reduced, thus significantly decreasing workload for clinicians.

Dữ liệu phi cấu trúc từ hồ sơ sức khỏe điện tử đại diện cho một kho thông tin phong phú. Doc’EDS là một công cụ sàng lọc dựa trên phân tích ngữ nghĩa và văn bản. Hệ thống Doc’EDS cung cấp một giao diện người dùng đồ họa để tìm kiếm tài liệu bằng tiếng Pháp. Mục tiêu của nghiên cứu này là trình bày công cụ Doc’EDS và cung cấp một đánh giá chính thức về các tính năng ngữ nghĩa của nó. Doc’EDS là một công cụ tìm kiếm được xây dựng trên kho dữ liệu lâm sàng phát triển tại Bệnh viện Đại học Rouen. Công cụ này là một công cụ tìm kiếm đa cấp, kết hợp dữ liệu có cấu trúc và phi cấu trúc. Nó cũng cung cấp các tính năng phân tích cơ bản và các tiện ích ngữ nghĩa. Một đánh giá chính thức đã được tiến hành để đo lường tác động của các thuật toán Xử lý Ngôn ngữ Tự nhiên. Khoảng 18,1 triệu tài liệu tường thuật được lưu trữ trong Doc’EDS. Đánh giá chính thức được thực hiện trên 5000 khái niệm lâm sàng đã được thu thập thủ công. Các chỉ số F của các khái niệm tiêu cực và khái niệm giả thuyết lần lượt là 0,89 và 0,57. Trong đánh giá chính thức này, chúng tôi đã chỉ ra rằng Doc’EDS có khả năng xử lý sự tinh tế của ngôn ngữ để nâng cao tìm kiếm toàn văn tiên tiến trong các tài liệu y tế tiếng Pháp. Công cụ Doc’EDS hiện đang được sử dụng hàng ngày để giúp các nhà nghiên cứu xác định các nhóm bệnh nhân nhờ vào dữ liệu phi cấu trúc.

Hand hygiene is one of the most effective attempts to control nosocomial infections, and it is an important measure to avoid the transmission of pathogens. However, the compliance of healthcare workers (HCWs) with hand washing is still poor worldwide. Herein, we aimed to determine the best hand hygiene preference of the infectious diseases and clinical microbiology (IDCM) specialists to prevent transmission of microorganisms from one patient to another. Expert opinions regarding the criteria that influence the best hand hygiene preference were collected through a questionnaire via face-to-face interviews. Afterwards, these opinions were examined with two widely used multi-criteria decision analysis (MCDA) methods, the Multi-Attribute Utility Theory (MAUT) and the Analytic Hierarchy Process (AHP). A total of 15 IDCM specialist opinions were collected from diverse private and public hospitals located in İzmir, Turkey. The mean age of the participants was 49.73 ± 8.46, and the mean experience year of the participants in their fields was 17.67 ± 11.98. The findings that we obtained through two distinct decision making methods, the MAUT and the AHP, suggest that alcohol-based antiseptic solution (ABAS) has the highest utility (0.86) and priority (0.69) among the experts’ choices. In conclusion, the MAUT and the AHP, decision models developed here indicate that rubbing the hands with ABAS is the most favorable choice for IDCM specialists to prevent nosocomial infection.

In Western languages the period character is highly ambiguous, due to its double role as sentence delimiter and abbreviation marker. This is particularly relevant in clinical free-texts characterized by numerous anomalies in spelling, punctuation, vocabulary and with a high frequency of short forms. The problem is addressed by two binary classifiers for abbreviation and sentence detection. A support vector machine exploiting a linear kernel is trained on different combinations of feature sets for each classification task. Feature relevance ranking is applied to investigate which features are important for the particular task. The methods are applied to German language texts from a medical record system, authored by specialized physicians. Two collections of 3,024 text snippets were annotated regarding the role of period characters for training and testing. Cohen's kappa resulted in 0.98. For abbreviation and sentence boundary detection we can report an unweighted micro-averaged F-measure using a 10-fold cross validation of 0.97 for the training set. For test set based evaluation we obtained an unweighted micro-averaged F-measure of 0.95 for abbreviation detection and 0.94 for sentence delineation. Language-dependent resources and rules were found to have less impact on abbreviation detection than on sentence delineation. Sentence detection is an important task, which should be performed at the beginning of a text processing pipeline. For the text genre under scrutiny we showed that support vector machines exploiting a linear kernel produce state of the art results for sentence boundary detection. The results are comparable with other sentence boundary detection methods applied to English clinical texts. We identified abbreviation detection as a supportive task for sentence delineation.

For researchers and public health agencies, the complexity of high–dimensional spatio–temporal data in surveillance for large reporting networks presents numerous challenges, which include low signal–to–noise ratios, spatial and temporal dependencies, and the need to characterize uncertainties. Central to the problem in the context of disease outbreaks is a decision structure that requires trading off false positives for delayed detections. In this paper we apply a previously developed Bayesian hierarchical model to a data set from the Indiana Public Health Emergency Surveillance System (PHESS) containing three years of emergency department visits for influenza–like illness and respiratory illness. Among issues requiring attention were selection of the underlying network (Too few nodes attenuate important structure, while too many nodes impose barriers to both modeling and computation.); ensuring that confidentiality protections in the data do not impede important modeling day of week effects; and evaluating the performance of the model. Our results show that the model captures salient spatio–temporal dynamics that are present in public health surveillance data sets, and that it appears to detect both “annual” and “atypical” outbreaks in a timely, accurate manner. We present maps that help make model output accessible and comprehensible to public health authorities. We use an illustrative family of decision rules to show how output from the model can be used to inform false positive–delayed detection tradeoffs. The advantages of our methodology for addressing the complicated issues of real world surveillance data applications are three–fold. We can easily incorporate additional covariate information and spatio–temporal dynamics in the data. Second, we furnish a unified framework to provide uncertainties associated with each parameter. Third, we are able to handle multiplicity issues by using a Bayesian approach. The urgent need to quickly and effectively monitor the health of the public makes our methodology a potentially plausible and useful surveillance approach for health professionals.

Higher levels of functional health in older adults leads to higher quality of life and improves the ability to age-in-place. Tracking functional health objectively could help clinicians to make decisions for interventions in case of health deterioration. Even though several geriatric assessments capture several aspects of functional health, there is limited research in longitudinally tracking personalized functional health of older adults using a combination of these assessments. We used geriatric assessment data collected from 150 older adults to develop and validate a functional health prediction model based on risks associated with falls, hospitalizations, emergency visits, and death. We used mixed effects logistic regression to construct the model. The geriatric assessments included were Activities of Daily Living (ADL), Instrumental Activities of Daily Living (IADL), Mini-Mental State Examination (MMSE), Geriatric Depression Scale (GDS), and Short Form 12 (SF12). Construct validators such as fall risks associated with model predictions, and case studies with functional health trajectories were used to validate the model. The model is shown to separate samples with and without adverse health event outcomes with an area under the receiver operating characteristic curve (AUC) of > 0.85. The model could predict emergency visit or hospitalization with an AUC of 0.72 (95% CI 0.65–0.79), fall with an AUC of 0.86 (95% CI 0.83–0.89), fall with hospitalization with an AUC of 0.89 (95% CI 0.85–0.92), and mortality with an AUC of 0.93 (95% CI 0.88–0.97). Multiple comparisons of means using Turkey HSD test show that model prediction means for samples with no adverse health events versus samples with fall, hospitalization, and death were statistically significant (p < 0.001). Case studies for individual residents using predicted functional health trajectories show that changes in model predictions over time correspond to critical health changes in older adults. The personalized functional health tracking may provide clinicians with a longitudinal view of overall functional health in older adults to help address the early detection of deterioration trends and decide appropriate interventions. It can also help older adults and family members take proactive steps to improve functional health.

Record linkage is the process of identifying and combining records about the same individual from two or more different datasets. While there are many open source and commercial data linkage tools, the volume and complexity of currently available datasets for linkage pose a huge challenge; hence, designing an efficient linkage tool with reasonable accuracy and scalability is required. We developed CIDACS-RL (Centre for Data and Knowledge Integration for Health – Record Linkage), a novel iterative deterministic record linkage algorithm based on a combination of indexing search and scoring algorithms (provided by Apache Lucene). We described how the algorithm works and compared its performance with four open source linkage tools (AtyImo, Febrl, FRIL and RecLink) in terms of sensitivity and positive predictive value using gold standard dataset. We also evaluated its accuracy and scalability using a case-study and its scalability and execution time using a simulated cohort in serial (single core) and multi-core (eight core) computation settings. Overall, CIDACS-RL algorithm had a superior performance: positive predictive value (99.93% versus AtyImo 99.30%, RecLink 99.5%, Febrl 98.86%, and FRIL 96.17%) and sensitivity (99.87% versus AtyImo 98.91%, RecLink 73.75%, Febrl 90.58%, and FRIL 74.66%). In the case study, using a ROC curve to choose the most appropriate cut-off value (0.896), the obtained metrics were: sensitivity = 92.5% (95% CI 92.07–92.99), specificity = 93.5% (95% CI 93.08–93.8) and area under the curve (AUC) = 97% (95% CI 96.97–97.35). The multi-core computation was about four times faster (150 seconds) than the serial setting (550 seconds) when using a dataset of 20 million records. CIDACS-RL algorithm is an innovative linkage tool for huge datasets, with higher accuracy, improved scalability, and substantially shorter execution time compared to other existing linkage tools. In addition, CIDACS-RL can be deployed on standard computers without the need for high-speed processors and distributed infrastructures.

A growing body of research has shown that machine learning (ML) can be a useful tool to predict how different variable combinations affect out-of-hospital cardiac arrest (OHCA) survival outcomes. However, there remain significant research gaps on the utilization of ML models for decision-making and their impact on survival outcomes. The purpose of this study was to develop ML models that effectively predict hospital’s practice to perform coronary angiography (CA) in adult patients after OHCA and subsequent neurologic outcomes. We utilized all (N = 2398) patients treated by the Chicago Fire Department Emergency Medical Services included in the Cardiac Arrest Registry to Enhance Survival (CARES) between 2013 and 2018 who survived to hospital admission to develop, test, and analyze ML models for decisions after return of spontaneous circulation (ROSC) and patient survival. ML classification models, including the Embedded Fully Convolutional Network (EFCN) model, were compared based on their ability to predict post-ROSC decisions and survival. The EFCN classification model achieved the best results across tested ML algorithms. The area under the receiver operating characteristic curve (AUROC) for CA and Survival were 0.908 and 0.896 respectively. Through cohort analyses, our model predicts that 18.3% (CI 16.4–20.2) of patients should receive a CA that did not originally, and 30.1% (CI 28.5–31.7) of these would experience improved survival outcomes. ML modeling effectively predicted hospital decisions and neurologic outcomes. ML modeling may serve as a quality improvement tool to inform system level OHCA policies and treatment protocols.