Journal of evaluation in clinical practice
-
Clinical abbreviations pose a challenge for clinical decision support systems due to their ambiguity. Additionally, clinical datasets often suffer from class imbalance, hindering the classification of such data. This imbalance leads to classifiers with low accuracy and high error rates. Traditional feature-engineered models struggle with this task, and class imbalance is a known factor that reduces the performance of neural network techniques. ⋯ Deep neural network methods, particularly Bi-LSTM, offer promising alternatives to traditional feature-engineered models for clinical abbreviation disambiguation. By employing data generation techniques, we can address the challenges posed by limited-resource and imbalanced clinical datasets. This approach leads to a significant improvement in model accuracy for clinical abbreviation disambiguation tasks.
-
This paper explores how frontline nurses experienced the onset of the coronavirus disease (COVID-19) pandemic to provide appropriate care during a global health crisis. ⋯ Understanding the challenges faced by frontline nurses during the onset of the COVID-19 pandemic may help healthcare practitioners and policy makers to implement targeted interventions, support mechanisms and resource allocation strategies that enhance the well-being of frontline nurses and optimise patient care delivery during health crises.
-
The National Health Service (NHS) Long Term Plan was published in January 2019. One of its objectives was restructuring outpatient services, as part of an Outpatient Transformation initiative. Monitoring of trusts' adherence to the objectives of the Long Term Plan is therefore required to benchmark progress against national objectives. ⋯ There are deficiencies in current outpatient establishments that may hinder the achievement of objectives set in the NHS Long Term Plan. Changes at all levels of healthcare are required, with increased reliance on technologies and investment in support for transformation management.
-
Despite the widespread recommendation to engage in therapeutic exercise for the treatment of low back pain (LBP), there is conflicting evidence regarding clinical outcomes and effectiveness. Poor methodological quality may be to blame for reducing the overall strength of evidence for this intervention, yet little is known about the difficulties researchers encounter when designing and implementing their study methods. ⋯ Statistical power, study length and/or follow-up, and inclusion criteria were the three most commonly reported categories of SALs in exercise trials for LBP. Lack of long-term follow-up, inadequate sample size and inclusion of specific populations were the most common subcategories. Research protocols recognizing and avoiding these limitations will enhance the overall quality of evidence of exercise therapy trials for LBP.
-
Low-value radiological imaging threatens patient safety and efficient use of limited health resources. It is important to evaluate measures for reducing low-value utilisation, to learn and to improve. Accordingly, the objective of this study was to qualitatively evaluate a pilot intervention for reducing low-value imaging in Norway. ⋯ The pilot intervention was deemed acceptable, feasible, engaging and relevant. Specific training in the use of the new procedure was suggested to improve the intervention. The simple design, as well as the positive acceptance demonstrated and the few resources needed, make the pilot intervention and methodology highly relevant for other settings or when aiming to reduce the number of other low-value radiology examinations.