Data consistency in a voluntary medical incident reporting system

Information Technology and Value-Based Healthcare Systems: A Strategy and Framework

Value-based healthcare offers a pathway for enhancing patient satisfaction and population health and reducing healthcare costs. In addition, it provides a means to enhance physicians' perception and experience in healthcare delivery. The foundation of the said system is the notion that community wellness can only be benefited when the health effects of many people are also addressed. The provision of healthcare services incurs costs. However, a value-based model addresses this issue by establishing teams that cater to individuals with similar needs. This approach fosters expertise and efficiency, ultimately leading to cost savings without rationing. Furthermore, entrusting decision-making authority regarding healthcare delivery to the clinical team enhances doctors' professionalism and the integrity of clinician-patient interactions, resulting in more effective and relevant treatments. Currently, various information technology (IT)-based solutions are the main focus for accomplishing the desired value-based healthcare system. The establishment of a coordinated framework that can help organizations create value-based healthcare systems is covered in the current article. Additionally listed are many IT-based solutions used to create a value-based healthcare system.

A Machine Learning Approach with Human-AI Collaboration for Automated Classification of Patient Safety Event Reports: Algorithm Development and Validation Study

BACKGROUND

Adverse events refer to incidents with potential or actual harm to patients in hospitals. These events are typically documented through patient safety event (PSE) reports, which consist of detailed narratives providing contextual information on the occurrences. Accurate classification of PSE reports is crucial for patient safety monitoring. However, this process faces challenges due to inconsistencies in classifications and the sheer volume of reports. Recent advancements in text representation, particularly contextual text representation derived from transformer-based language models, offer a promising solution for more precise PSE report classification. Integrating the machine learning (ML) classifier necessitates a balance between human expertise and artificial intelligence (AI). Central to this integration is the concept of explainability, which is crucial for building trust and ensuring effective human-AI collaboration.

OBJECTIVE

This study aims to investigate the efficacy of ML classifiers trained using contextual text representation in automatically classifying PSE reports. Furthermore, the study presents an interface that integrates the ML classifier with the explainability technique to facilitate human-AI collaboration for PSE report classification.

METHODS

This study used a data set of 861 PSE reports from a large academic hospital's maternity units in the Southeastern United States. Various ML classifiers were trained with both static and contextual text representations of PSE reports. The trained ML classifiers were evaluated with multiclass classification metrics and the confusion matrix. The local interpretable model-agnostic explanations (LIME) technique was used to provide the rationale for the ML classifier's predictions. An interface that integrates the ML classifier with the LIME technique was designed for incident reporting systems.

RESULTS

The top-performing classifier using contextual representation was able to obtain an accuracy of 75.4% (95/126) compared to an accuracy of 66.7% (84/126) by the top-performing classifier trained using static text representation. A PSE reporting interface has been designed to facilitate human-AI collaboration in PSE report classification. In this design, the ML classifier recommends the top 2 most probable event types, along with the explanations for the prediction, enabling PSE reporters and patient safety analysts to choose the most suitable one. The LIME technique showed that the classifier occasionally relies on arbitrary words for classification, emphasizing the necessity of human oversight.

CONCLUSIONS

This study demonstrates that training ML classifiers with contextual text representations can significantly enhance the accuracy of PSE report classification. The interface designed in this study lays the foundation for human-AI collaboration in the classification of PSE reports. The insights gained from this research enhance the decision-making process in PSE report classification, enabling hospitals to more efficiently identify potential risks and hazards and enabling patient safety analysts to take timely actions to prevent patient harm.

The effect of refresher training on fact description in medical incident report writing in the Japanese language

To maintain the effectiveness of the training (1st-Training Session: 1st-TS) to accurate describe facts in the medical incident reports (IRs) in Japanese, a refresher TS was designed and its effectiveness was examined. First, textual analysis showed that IRs' accuracy significantly decreased six months after the 1st-TS. Based on this result, the refresher TS was designed and conducted with 64 residents. To verify the refresher TS' effectiveness, IRs after the 1st-TS, six months later, and after the refresher TS were compared via text analysis. The results showed that the refresher TS restored the description rate of patient's background, safety check procedures, original work procedures, information on equipment used, reporter's actions, and post-incident response. The questionnaire was also administered and showed that the refresher TS contributed to residents' motivation to learn about IRs. In conclusion, the refresher TS contributed to sustaining the effect of the 1st-TS on accurately describing IRs.

Change in the Number of Patient Safety Reports Through a 16-Year Patient Safety Initiative: A Retrospective Study Focusing on the Incident Severity and Type in a Japanese Hospital

Purpose

To describe the long-term quantitative change in the number of submissions of patient safety reports after the introduction of a patient safety reporting system, focusing on incident severity and type.

Patients and Methods

This study was performed at a tertiary care hospital in Japan. Patient safety reports from 2006 to 2020 were retrospectively reviewed. Incident severity was classified from level 0 (near miss) to level 5 (fatality). The incident types included those related to medication, patient care, drains and catheters, procedures and interventions, examinations, medical devices, and blood transfusions. The study period was divided into 1. 2004–2007; 2. 2008–2014; and 3. 2015–2020 based on the implementation of hospital patient safety strategies. The number of reports per hospital worker was compared among the study periods and the incident levels and types.

Results

We analyzed 96,332 reports extracted from the patient safety reporting system of the hospital. The total number of reports per hospital worker has increased over time. The numbers of levels 0 and 1 incidents increased throughout the study period. In addition, levels 3a and 3b incidents increased between periods 2 and 3. All incident types, except for procedure and intervention-related incidents, increased between periods 1 and 2 and between periods 1 and 3. The number of procedure and intervention-related incidents increased between periods 2 and 3, although it did not between periods 1 and 2.

Conclusion

We found increases in the number of patient safety reports according to the incident severity and type. This suggests two contextual changes occurring during the cultural maturity process, which reflected the development of organizational patient safety culture in our institution. The first was the establishment of a reporting attitude in the institution. The second was to overcome barriers to patient safety.

How do healthcare practitioners use incident data to improve patient safety in Japan? A qualitative study

BACKGROUND

Patient incident reporting systems have been widely used for ensuring safety and improving quality in care settings in many countries. However, little is known about the way in which incident data are used by frontline clinical staff. Furthermore, while the use of a systems perspective has been reported as an effective way of learning from incident data in a multidisciplinary team, the level of adaptability of this perspective to a different cultural context has not been widely explored. The primary aim of the study, therefore, was to investigate how healthcare practitioners in Japan perceive the reporting systems and utilize a systems perspective in learning from incident data in acute care and mental health settings.

METHODS

A non-experimental, descriptive and exploratory research design was adopted with the following two data-collection methods: 1) Sixty-one semi-structured interviews with frontline staff in two hospitals; and 2) Non-participatory observations of thirty-seven regular incident review meetings. The two hospitals in the Greater Tokyo area which were invited to take part were: 1) a not-for-profit, privately-run, acute care hospital with approximately 500 beds; and 2) a publicly-run mental health hospital with 200 beds.

RESULTS

While the majority of staff acknowledge the positive impacts of the reporting systems on safety, the observation data found that little consideration was given to systems aspects during formal meetings. The meetings were primarily a place for the exchange of practical information, as opposed to in-depth discussions regarding causes of incidents and corrective measures. Learning from incident data was influenced by four factors: professional boundaries; dealing with a psychological burden; leadership and educational approach; and compatibility of patient safety with patient-centered care.

CONCLUSIONS

Healthcare organizations are highly complex, comprising of many professional boundaries and risk perceptions, and various communication styles. In order to establish an optimum method of individual and organizational learning and effective safety management, a fine balance has to be struck between respect for professional expertise in a local team and centralized safety oversight with a strong focus on systems. Further research needs to examine culturally-sensitive organizational and professional dynamics, including leader-follower relationships and the impact of resource constraints.

A Machine Learning Approach to Reclassifying Miscellaneous Patient Safety Event Reports

BACKGROUND AND OBJECTIVES

Medical errors are a leading cause of death in the United States. Despite widespread adoption of patient safety reporting systems to address medical errors, making sense of the reports collected in these systems is challenging in practice. Event classification taxonomies used in many reporting systems can be complex and difficult to understand by frontline reporters, leading reporters to classify reports as "miscellaneous" as opposed to assigning a specific event-type category, which may facilitate analysis.

METHODS

To assist patient safety analysts in their analysis of "miscellaneous" reports, we developed an ensemble machine learning natural language processing model to reclassify these reports. We integrated the model into a clinical workflow dashboard, evaluated user feedback, and compared differences in user thresholds for model performance.

RESULTS AND CONCLUSIONS

Integrating an ensemble model to classify "miscellaneous" event reports with an interactive visualization was helpful to patient safety analysts review "miscellaneous" reports. However, patient safety analysts have different thresholds for model reclassification depending on their role and experience with "miscellaneous" event reports.

Evaluating resampling methods and structured features to improve fall incident report identification by the severity level

OBJECTIVE

This study aims to improve the classification of the fall incident severity level by considering data imbalance issues and structured features through machine learning.

MATERIALS AND METHODS

We present an incident report classification (IRC) framework to classify the in-hospital fall incident severity level by addressing the imbalanced class problem and incorporating structured attributes. After text preprocessing, bag-of-words features, structured text features, and structured clinical features were extracted from the reports. Next, resampling techniques were incorporated into the training process. Machine learning algorithms were used to build classification models. IRC systems were trained, validated, and tested using a repeated and randomly stratified shuffle-split cross-validation method. Finally, we evaluated the system performance using the F1-measure, precision, and recall over 15 stratified test sets.

RESULTS

The experimental results demonstrated that the classification system setting considering both data imbalance issues and structured features outperformed the other system settings (with a mean macro-averaged F1-measure of 0.733). Considering the structured features and resampling techniques, this classification system setting significantly improved the mean F1-measure for the rare class by 30.88% (P value < .001) and the mean macro-averaged F1-measure by 8.26% from the baseline system setting (P value < .001). In general, the classification system employing the random forest algorithm and random oversampling method outperformed the others.

CONCLUSIONS

Structured features provide essential information for categorizing the fall incident severity level. Resampling methods help rebalance the class distribution of the original incident report data, which improves the performance of machine learning models. The IRC framework presented in this study effectively automates the identification of fall incident reports by the severity level.

Using convolutional neural networks to identify patient safety incident reports by type and severity

OBJECTIVE

To evaluate the feasibility of a convolutional neural network (CNN) with word embedding to identify the type and severity of patient safety incident reports.

MATERIALS AND METHODS

A CNN with word embedding was applied to identify 10 incident types and 4 severity levels. Model training and validation used data sets (n_type = 2860, n_severity = 1160) collected from a statewide incident reporting system. Generalizability was evaluated using an independent hospital-level reporting system. CNN architectures were examined by varying layer size and hyperparameters. Performance was evaluated by F score, precision, recall, and compared to binary support vector machine (SVM) ensembles on 3 testing data sets (type/severity: n_benchmark = 286/116, n_original = 444/4837, n_independent = 6000/5950).

RESULTS

A CNN with 6 layers was the most effective architecture, outperforming SVMs with better generalizability to identify incidents by type and severity. The CNN achieved high F scores (> 85%) across all test data sets when identifying common incident types including falls, medications, pressure injury, and aggression. When identifying common severity levels (medium/low), CNN outperformed SVMs, improving F scores by 11.9%-45.1% across all 3 test data sets.

DISCUSSION

Automated identification of incident reports using machine learning is challenging because of a lack of large labelled training data sets and the unbalanced distribution of incident classes. The standard classification strategy is to build multiple binary classifiers and pool their predictions. CNNs can extract hierarchical features and assist in addressing class imbalance, which may explain their success in identifying incident report types.

CONCLUSION

A CNN with word embedding was effective in identifying incidents by type and severity, providing better generalizability than SVMs.

Can Unified Medical Language System-based semantic representation improve automated identification of patient safety incident reports by type and severity?

Objective

The study sought to evaluate the feasibility of using Unified Medical Language System (UMLS) semantic features for automated identification of reports about patient safety incidents by type and severity.

Materials and Methods

Binary support vector machine (SVM) classifier ensembles were trained and validated using balanced datasets of critical incident report texts (n_type = 2860, n_severity = 1160) collected from a state-wide reporting system. Generalizability was evaluated on different and independent hospital-level reporting system. Concepts were extracted from report narratives using the UMLS Metathesaurus, and their relevance and frequency were used as semantic features. Performance was evaluated by F-score, Hamming loss, and exact match score and was compared with SVM ensembles using bag-of-words (BOW) features on 3 testing datasets (type/severity: n_benchmark = 286/116, n_original = 444/4837, n_independent =6000/5950).

Results

SVMs using semantic features met or outperformed those based on BOW features to identify 10 different incident types (F-score [semantics/BOW]: benchmark = 82.6%/69.4%; original = 77.9%/68.8%; independent = 78.0%/67.4%) and extreme-risk events (F-score [semantics/BOW]: benchmark = 87.3%/87.3%; original = 25.5%/19.8%; independent = 49.6%/52.7%). For incident type, the exact match score for semantic classifiers was consistently higher than BOW across all test datasets (exact match [semantics/BOW]: benchmark = 48.9%/39.9%; original = 57.9%/44.4%; independent = 59.5%/34.9%).

Discussion

BOW representations are not ideal for the automated identification of incident reports because they do not account for text semantics. UMLS semantic representations are likely to better capture information in report narratives, and thus may explain their superior performance.

Conclusions

UMLS-based semantic classifiers were effective in identifying incidents by type and extreme-risk events, providing better generalizability than classifiers using BOW.

A mixed methods analysis of lithium-related patient safety incidents in primary care

Background:

Lithium is a drug with a narrow therapeutic range and has been associated with a number of serious adverse effects. This study aimed to characterise primary care lithium-related patient safety incidents submitted to the National Reporting and Learning System (NRLS) database with respect to incident origin, type, contributory factors and outcome. The intention was to identify ways to minimise risk to future patients by examining incidents with a range of harm outcomes.

Methods:

A mixed methods analysis of patient safety incident reports related to lithium was conducted. Data from healthcare organisations in England and Wales were extracted from the NRLS database. An exploratory descriptive analysis was undertaken to characterise the most frequent incident types, the associated chain of events and other contributory factors.

Results:

A total of 174 reports containing the term ‘lithium’ were identified. Of these, 41 were excluded and, from the remaining 133 reports, 138 incidents were identified and coded. Community pharmacies reported 100 incidents (96 dispensing related, two administration, two other), general practitioner (GP) practices filed 22 reports and 16 reports originated from other sources. A total of 99 dispensing-related incidents were recorded, 39 resulted from the wrong medication dispensed, 31 the wrong strength, 8 the wrong quantity and 21 other. A total of 128 contributory factors were identified overall; for dispensing incidents, the most common related to medication storage/packaging (n = 41), and ‘mistakes’ (n = 22), whereas no information regarding contributory factors was provided in 41 reports.

Conclusion:

Despite the established link between medication packaging and the risk of dispensing errors, our study highlighted storage and packaging as the most commonly described contributory factors to dispensing errors. The absence of certain relevant data limited the ability to fully characterise a number of reports. This highlighted the need to include clear and complete information when submitting reports. This, in turn, may help to better inform the further development of interventions designed to reduce the risk of incidents and improve patient safety.

A characterisation of lithium-related patient safety incidents in primary care

Lithium is an effective treatment for certain mental illnesses, but has a number of harmful side effects. Safety incidents related to medicines in the UK are reported to the National Reporting and Learning System database (NRLS), and concerns relating to lithium have previously been highlighted. This study aimed to characterise lithium incidents reported to the NRLS that occurred in a primary care setting. Reports relating to lithium and submitted between 2002 and 2013 were reviewed, and the information coded. A total of 174 reports containing the term ‘lithium’ were identified. Of these, 41 were excluded and, from the remaining 133 reports, 138 incidents were identified and coded with respect to incident origin, type, contributory factors and outcome. A total of 100 incidents were reported by community pharmacies (96 of which related to medicine dispensing), general practitioner (GP) practices filed 22 reports and 16 reports originated from other sources. Of the dispensing-related incidents, 39 resulted from the wrong medication dispensed, 31 the wrong strength, 8 the wrong quantity and 21 other. A total of 128 contributory factors were identified overall; for dispensing incidents, the most common related to medication storage/packaging (n = 41), and ‘mistakes’ (n = 22) whereas no information regarding contributory factors was provided in 41 reports. Despite the established link between medication packaging and the risk of dispensing errors, our study highlighted storage and packaging as the most commonly cited contributory factors to dispensing errors. The absence of certain relevant data limited the ability to fully characterise a number of reports. This highlighted the need to include clear and complete information when submitting reports. This, in turn, may help to better inform the further development of interventions designed to reduce incident numbers and improve patient safety.

A prototype of knowledge-based patient safety event reporting and learning system

Background

Patient falls, the most common safety events resulting in adverse patient outcomes, impose significant costs and have become a great burden to the healthcare community. Current patient fall reporting systems remain in the early stage that is far away from reaching the ultimate goal toward a safer healthcare. According to the Kirkpatrick model, the key challenge in reaction, learning, behavior and results is the realization of learning stage due to the lack of knowledge management, sharing and growing mechanism.

Methods

Based on the key contributing factors defined by AHRQ Common Formats 2.0, a hierarchical list of contributing factors for patient falls was established by expert review and discussion. Using the list as an infrastructure, we designed and developed a novel reporting system, where a strategy to identify contributing factors is intended to provide reporters knowledge support, in the form of similar cases and potential solutions. A survey containing two scenarios was conducted to evaluate the learning effect of our system.

Results

In both scenarios, potential solutions recommended by the system were annotated with correct contributing factors, and presented only when the corresponding factors were identified from the query report or selected by the user. The five experts show substantial consistency (Fleiss’ kappa > 0.6) and high agreement (ranging between fully agree and mostly agree) in the assessment of the three perspectives of the system, which verifies the effectiveness of the proposed knowledge support toward sharing and learning through the novel reporting system.

Conclusions

This study proposed a profile of contributing factors that could measure the similarity of patient safety events. Based on the profile, a knowledge-based reporting and learning system was developed to bridge the gap between surveillance, reporting, and retrospective analysis in the fall management circle. The system holds promise in improving event reporting toward better and safer healthcare.

Enhancing Patient Safety Event Reporting. A Systematic Review of System Design Features

OBJECTIVES

Electronic patient safety event reporting (e-reporting) is an effective mechanism to learn from errors and enhance patient safety. Unfortunately, the value of e-reporting system (a software or web server based platform) in patient safety research is greatly overshadowed by low quality reporting. This paper aims at revealing the current status of system features, detecting potential gaps in system design, and accordingly proposing suggestions for future design and implementation of the system.

METHODS

Three literature databases were searched for publications that contain informative descriptions of e-reporting systems. In addition, both online publicly accessible reporting forms and systems were investigated.

RESULTS

48 systems were identified and reviewed. 11 system design features and their frequencies of occurrence (Top 5: widgets (41), anonymity or confidentiality (29), hierarchy (20), validator (17), review notification (15)) were identified and summarized into a system hierarchical model.

CONCLUSIONS

The model indicated the current e-reporting systems are at an immature stage in their development, and discussed their future development direction toward efficient and effective systems to improve patient safety.

Developing a similarity searching module for patient safety event reporting system using semantic similarity measures

Background

The most important knowledge in the field of patient safety is regarding the prevention and reduction of patient safety events (PSE) during treatment and care. The similarities and patterns among the events may otherwise go unnoticed if they are not properly reported and analyzed. There is an urgent need for developing a PSE reporting system that can dynamically measure the similarities of the events and thus promote event analysis and learning effect.

Methods

In this study, three prevailing algorithms of semantic similarity were implemented to measure the similarities of the 366 PSE annotated by the taxonomy of The Agency for Healthcare Research and Quality (AHRQ). The performance of each algorithm was then evaluated by a group of domain experts based on a 4-point Likert scale. The consistency between the scales of the algorithms and experts was measured and compared with the scales randomly assigned. The similarity algorithms and scores, as a self-learning and self-updating module, were then integrated into the system.

Results

The result shows that the similarity scores reflect a high consistency with the experts’ review than those randomly assigned. Moreover, incorporating the algorithms into our reporting system enables a mechanism to learn and update based upon PSE similarity.

Conclusion

In conclusion, integrating semantic similarity algorithms into a PSE reporting system can help us learn from previous events and provide timely knowledge support to the reporters. With the knowledge base in the PSE domain, the new generation reporting system holds promise in educating healthcare providers and preventing the recurrence and serious consequences of PSE.

Using multiclass classification to automate the identification of patient safety incident reports by type and severity

BACKGROUND

Approximately 10% of admissions to acute-care hospitals are associated with an adverse event. Analysis of incident reports helps to understand how and why incidents occur and can inform policy and practice for safer care. Unfortunately our capacity to monitor and respond to incident reports in a timely manner is limited by the sheer volumes of data collected. In this study, we aim to evaluate the feasibility of using multiclass classification to automate the identification of patient safety incidents in hospitals.

METHODS

Text based classifiers were applied to identify 10 incident types and 4 severity levels. Using the one-versus-one (OvsO) and one-versus-all (OvsA) ensemble strategies, we evaluated regularized logistic regression, linear support vector machine (SVM) and SVM with a radial-basis function (RBF) kernel. Classifiers were trained and tested with "balanced" datasets (n_ _Type  = 2860, n_ _{SeverityLevel}  = 1160) from a state-wide incident reporting system. Testing was also undertaken with imbalanced "stratified" datasets (n_ _Type  = 6000, n_ _{SeverityLevel} =5950) from the state-wide system and an independent hospital reporting system. Classifier performance was evaluated using a confusion matrix, as well as F-score, precision and recall.

RESULTS

The most effective combination was a OvsO ensemble of binary SVM RBF classifiers with binary count feature extraction. For incident type, classifiers performed well on balanced and stratified datasets (F-score: 78.3, 73.9%), but were worse on independent datasets (68.5%). Reports about falls, medications, pressure injury, aggression and blood products were identified with high recall and precision. "Documentation" was the hardest type to identify. For severity level, F-score for severity assessment code (SAC) 1 (extreme risk) was 87.3 and 64% for SAC4 (low risk) on balanced data. With stratified data, high recall was achieved for SAC1 (82.8-84%) but precision was poor (6.8-11.2%). High risk incidents (SAC2) were confused with medium risk incidents (SAC3).

CONCLUSIONS

Binary classifier ensembles appear to be a feasible method for identifying incidents by type and severity level. Automated identification should enable safety problems to be detected and addressed in a more timely manner. Multi-label classifiers may be necessary for reports that relate to more than one incident type.

Methods for Addressing Technology-induced Errors: The Current State

OBJECTIVES

The objectives of this paper are to review and discuss the methods that are being used internationally to report on, mitigate, and eliminate technology-induced errors.

METHODS

The IMIA Working Group for Health Informatics for Patient Safety worked together to review and synthesize some of the main methods and approaches associated with technology- induced error reporting, reduction, and mitigation. The work involved a review of the evidence-based literature as well as guideline publications specific to health informatics.

RESULTS

The paper presents a rich overview of current approaches, issues, and methods associated with: (1) safe HIT design, (2) safe HIT implementation, (3) reporting on technology-induced errors, (4) technology-induced error analysis, and (5) health information technology (HIT) risk management. The work is based on research from around the world.

CONCLUSIONS

Internationally, researchers have been developing methods that can be used to identify, report on, mitigate, and eliminate technology-induced errors. Although there remain issues and challenges associated with the methodologies, they have been shown to improve the quality and safety of HIT. Since the first publications documenting technology-induced errors in healthcare in 2005, we have seen in a short 10 years researchers develop ways of identifying and addressing these types of errors. We have also seen organizations begin to use these approaches. Knowledge has been translated into practice in a short ten years whereas the norm for other research areas is of 20 years.

Leveraging user's performance in reporting patient safety events by utilizing text prediction in narrative data entry

BACKGROUND

Narrative data entry pervades computerized health information systems and serves as a key component in collecting patient-related information in electronic health records and patient safety event reporting systems. The quality and efficiency of clinical data entry are critical in arriving at an optimal diagnosis and treatment. The application of text prediction holds potential for enhancing human performance of data entry in reporting patient safety events.

OBJECTIVE

This study examined two functions of text prediction intended for increasing efficiency and data quality of text data entry reporting patient safety events.

METHODS

The study employed a two-group randomized design with 52 nurses. The nurses were randomly assigned into a treatment group or a control group with a task of reporting five patient fall cases in Chinese using a web-based test system, with or without the prediction functions. T-test, Chi-square and linear regression model were applied to evaluating the outcome differences in free-text data entry between the groups.

RESULTS

While both groups of participants exhibited a good capacity for accomplishing the assigned task of reporting patient falls, the results from the treatment group showed an overall increase of 70.5% in text generation rate, an increase of 34.1% in reporting comprehensiveness score and a reduction of 14.5% in the non-adherence of the comment fields. The treatment group also showed an increasing text generation rate over time, whereas no such an effect was observed in the control group.

CONCLUSION

As an attempt investigating the effectiveness of text prediction functions in reporting patient safety events, the study findings proved an effective strategy for assisting reporters in generating complementary free text when reporting a patient safety event. The application of the strategy may be effective in other clinical areas when free text entries are required.

Knowledge Representation in Patient Safety Reporting: An Ontological Approach

Purpose

The current development of patient safety reporting systems is criticized for loss of information and low data quality due to the lack of a uniformed domain knowledge base and text processing functionality. To improve patient safety reporting, the present paper suggests an ontological representation of patient safety knowledge.

Design/methodology/approach

We propose a framework for constructing an ontological knowledge base of patient safety. The present paper describes our design, implementation, and evaluation of the ontology at its initial stage.

Findings

We describe the design and initial outcomes of the ontology implementation. The evaluation results demonstrate the clinical validity of the ontology by a self-developed survey measurement.

Research limitations

The proposed ontology was developed and evaluated using a small number of information sources. Presently, US data are used, but they are not essential for the ultimate structure of the ontology.

Practical implications

The goal of improving patient safety can be aided through investigating patient safety reports and providing actionable knowledge to clinical practitioners. As such, constructing a domain specific ontology for patient safety reports serves as a cornerstone in information collection and text mining methods.

Originality/value

The use of ontologies provides abstracted representation of semantic information and enables a wealth of applications in a reporting system. Therefore, constructing such a knowledge base is recognized as a high priority in health care.

An analysis of electronic health record-related patient safety incidents

The aim of this study was to analyse electronic health record-related patient safety incidents in the patient safety incident reporting database in fully digital hospitals in Finland. We compare Finnish data to similar international data and discuss their content with regard to the literature. We analysed the types of electronic health record-related patient safety incidents that occurred at 23 hospitals during a 2-year period. A procedure of taxonomy mapping served to allow comparisons. This study represents a rare examination of patient safety risks in a fully digital environment. The proportion of electronic health record-related incidents was markedly higher in our study than in previous studies with similar data. Human-computer interaction problems were the most frequently reported. The results show the possibility of error arising from the complex interaction between clinicians and computers.

Identifying barriers and benefits of patient safety event reporting toward user-centered design

Background

To learn from errors, electronic patient safety event reporting systems (e-reporting systems) have been widely adopted to collect medical incidents from the frontline practitioners in US hospitals. However, two issues of underreporting and low-quality of reports pervade and thus the system effectiveness remains dubious.

Methods

This study employing semi-structured interviews of health professionals in the Texas Medical Center investigated the perceived benefits and barriers from users who have used e-reporting systems.

Results

As a result, the perceived benefits include the enhanced convenience in data processing and the assistant functions leading to patient safety enhancement. The perceived barriers to the acceptance and quality use of the system include the lack of instructions, lack of reporter-friendly classifications, lack of time, and lack of feedback The identified benefits and barriers help design a user-centered e-reporting system where learning and assistant features are discussed during the interviews.

Conclusions

As a response, the learning and assistant features aiming at enhancing benefits and removing barriers of e-reporting systems should be included for facilitating the acceptance and effective use of the systems.

Patient-Centred Coordinated Care in Times of Emerging Diseases and Epidemics. Contribution of the IMIA Working Group on Patient Safety

OBJECTIVES

In this paper the researchers describe how existing health information technologies (HIT) can be repurposed and new technologies can be innovated to provide patient-centered care to individuals affected by new and emerging diseases.

METHODS

The researchers conducted a focused review of the published literature describing how HIT can be used to support safe, patient-centred, coordinated care to patients who are affected by Ebola (an emerging disease).

RESULTS

New and emerging diseases present opportunities for repurposing existing technologies and for stimulating the development of new HIT innovation. Innovative technologies may be developed such as new software used for tracking patients during new or emerging disease outbreaks or by repurposing and extending existing technologies so they can be used to support patients, families and health professionals who may have been exposed to a disease. The paper describes the development of new technologies and the repurposing and extension of existing ones (such as electronic health records) using the most recent outbreak of Ebola as an example.

Text prediction on structured data entry in healthcare: a two-group randomized usability study measuring the prediction impact on user performance

BACKGROUND

Structured data entry pervades computerized patient safety event reporting systems and serves as a key component in collecting patient-related information in electronic health records. Clinicians would spend more time being with patients and arrive at a high probability of proper diagnosis and treatment, if data entry can be completed efficiently and effectively. Historically it has been proven text prediction holds potential for human performance regarding data entry in a variety of research areas.

OBJECTIVE

This study aimed at examining a function of text prediction proposed for increasing efficiency and data quality in structured data entry.

METHODS

We employed a two-group randomized design with fifty-two nurses in this usability study. Each participant was assigned the task of reporting patient falls by answering multiple choice questions either with or without the text prediction function. t-test statistics and linear regression model were applied to analyzing the results of the two groups.

RESULTS

While both groups of participants exhibited a good capacity of accomplishing the assigned task, the results were an overall 13.0% time reduction and 3.9% increase of response accuracy for the group utilizing the prediction function.

CONCLUSION

As a primary attempt investigating the effectiveness of text prediction in healthcare, study findings validated the necessity of text prediction to structured date entry, and laid the ground for further research improving the effectiveness of text prediction in clinical settings.