Clinical incident reporting behaviors and associated factors among health professionals in Dessie comprehensive specialized hospital, Amhara Region, Ethiopia: a mixed method study.

BACKGROUND: Understanding the type and causes of errors are necessary for the prevention of occurrence or reoccurrence. Therefore addressing the behavior of health professionals on reporting clinical incidents is crucial to create spontaneous knowledge from mistakes and enhance patient safety. METHOD: A mixed type institution-based cross-sectional study design was conducted from March 1 - 30, 2020 in Dessie comprehensive specialized hospital among 319 and 18 participants for the quantitative and qualitative study, respectively. The professions and participants with their assigned proportions were selected using a simple random sampling technique. For quantitative and qualitative data, semi structured questionnaires and interviewer-guided questions were used to collect data, respectively. Finally, qualitative findings were used to supplement the quantitative result. RESULT: The finding showed that the proportion of clinical incident reporting behavior among health professionals was 12.4%. Having training (AOR=3.6, 95% CI, 1.15-11.45), incident reporting help to minimize errors (AOR=2.8, 95% CI, 1.29-6.02), fear of legal penalty (AOR= 0.3, 95% CI, 0.13-0.82), and lack of feedback (AOR=0.3, 95% CI, 0.11-0.90) were identified as significant factors for clinical incident reporting behavior of the health professionals. CONCLUSIONS: This study showed that the clinical incident reporting behavior of the health professionals was very low. Therefore health professionals should get training on clinical incident reporting and the hospital should have an incident reporting system and guideline.

Descripción de las características de los incidentes clínicos reportados en el año 2020 al sistema de reporte voluntario de un hospital pediátrico de Costa Rica / Description of the characteristics of the clinical incidents reported in 2020 to the voluntary reporting system of a pediatric hospital in Costa Rica

El objetivo del presente manuscrito fue describir los incidentes clínicos que fueron enviados al sistema de reporte voluntario durante el año 2020, que funciona en el Hospital Nacional de Niños de Costa Rica, perteneciente a la Caja Costarricense de Seguro Social. Se realizó un estudio observacional descriptivo de los datos consolidados que se enviaron durante los meses de enero a diciembre del año 2020. Durante el año 2020 el 1,6% de los pacientes atendidos en el hospital experimentó algún tipo de incidente clínico. El total de egresos disminuyó un 38,4% en comparación con los egresos del año 2019, sin embargo, los incidentes clínicos reportados aumentaron en el año 2020 un 37,6%, especialmente a partir del mes de agosto. No se reportaron eventos centinela en este año. Los servicios que realizaron mayor cantidad de reportes fueron Cuidados Intensivos (14,3%), Cirugía General (12%), Neonatología (9,8%) e Infectología (9%). El día en el cual se reportaron más incidentes fue el miércoles (27,8%), en el primer turno hospitalario se reportaron la mayoría de los casos (48,1%) y estos incidentes ocurrieron predominantemente a individuos masculinos (66%). Respecto de la edad de los pacientes, la mayoría estuvo en el rango de edad de 1 año a menos de 5 años (36,1%), seguido por el rango de edad de mayores de 29 días a menores de 1 año (24,1%). La mayor parte de los casos estuvieron relacionados con el cuidado que se brindaba al paciente (63,9%). El 41,4% de los incidentes requirieron medidas clínicas pero las secuelas fueron transitorias. El 51,1% de los casos ameritó algún tipo de cuidado médico adicional a su esquema terapéutico al ingreso. El 96% de los incidentes clínicos fueron reportados por personal de enfermería. La mayoría de los incidentes clínicos (35,3 %) en este período fueron errores relacionados con notas en el expediente digital.

The objective of this manuscript was to describe the clinical incidents that were sent to the voluntary reporting system during 2020 at the National Children's Hospital of Costa Rica, belonging to the Costa Rican Social Security Fund.A descriptive observational study of the consolidated data that was sent during the months of January to December of the year 2020 was carried out.During 2020, 1.6% of the patients treated in the hospital experienced some type of clinical incident. The total discharges decreased by 38.4% compared to the discharges of the year 2019, however, the reported clinical incidents increased in the year 2020 by 37.6%, especially from the month of August. Sentinel events were not reported this year. The services that made the highest number of reports were Intensive Care (14.3%), General Surgery (12%), Neonatology (9.8%) and Infectiology (9%). The day on which the most incidents were reported was Wednesday (27.8%), in the first hospital shift most of the cases were reported (48.1%) and these incidents occurred predominantly to male individuals (66%). Regarding the age of the patients, the majority were in the age range from 1 year to less than 5 years (36.1%), followed by the age range from over 29 days to under 1 year (24, 1%). Most of the cases were related to the care provided to the patient (63.9%). 41.4% of the incidents required clinical measures but the sequelae were transitory. 51.1% of the cases merited some type of additional medical care to their therapeutic scheme upon admission. 96% of clinical incidents were reported by nursing staff. Most of the clinical incidents (35.3%) in this period were errors related to notes in the digital file.

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.

Bow-tie diagrams for risk management in anaesthesia.

Bow-tie analysis is a risk analysis and management tool that has been readily adopted into routine practice in many high reliability industries such as engineering, aviation and emergency services. However, it has received little exposure so far in healthcare. Nevertheless, its simplicity, versatility, and pictorial display may have benefits for the analysis of a range of healthcare risks, including complex and multiple risks and their interactions. Bow-tie diagrams are a combination of a fault tree and an event tree, which when combined take the shape of a bow tie. Central to bow-tie methodology is the concept of an undesired or 'Top Event', which occurs if a hazard progresses past all prevention controls. Top Events may also occasionally occur idiosyncratically. Irrespective of the cause of a Top Event, mitigation and recovery controls may influence the outcome. Hence the relationship of hazard to outcome can be viewed in one diagram along with possible causal sequences or accident trajectories. Potential uses for bow-tie diagrams in anaesthesia risk management include improved understanding of anaesthesia hazards and risks, pre-emptive identification of absent or inadequate hazard controls, investigation of clinical incidents, teaching anaesthesia risk management, and demonstrating risk management strategies to third parties when required.