Search

Simple Summary: Oesophago-gastric cancer is one of the commonest cancers worldwide, yet it can be particularly difficult to diagnose given that initial symptoms are often non-specific and routine screening is not available. Cancer risk-assessment tools, which calculate cancer risk based on symptoms and other risk factors present in the primary care record, can aid decisions on referrals for cancer investigations, facilitating earlier diagnosis. Diagnosing common cancers earlier could help improve survival rates. Using UK primary care electronic health record data, we compared five different machine learning techniques for probabilistic classification of cancer patients against a current widely used UK primary care cancer risk-assessment tool. The machine learning algorithms outperformed the current risk-assessment tool, with a higher overall accuracy and an ability to reasonably identify 11–25% more cancer patients. We conclude that machine-learning-based risk-assessment tools could help better identify suitable patients for further investigation and support earlier diagnosis. Oesophago-gastric cancer is difficult to diagnose in the early stages given its typical non-specific initial manifestation. We hypothesise that machine learning can improve upon the diagnostic performance of current primary care risk-assessment tools by using advanced analytical techniques to exploit the wealth of evidence available in the electronic health record. We used a primary care electronic health record dataset derived from the UK General Practice Research Database (7471 cases; 32,877 controls) and developed five probabilistic machine learning classifiers: Support Vector Machine, Random Forest, Logistic Regression, Naïve Bayes, and Extreme Gradient Boosted Decision Trees. Features included basic demographics, symptoms, and lab test results. The Logistic Regression, Support Vector Machine, and Extreme Gradient Boosted Decision Tree models achieved the highest performance in terms of accuracy and AUROC (0.89 accuracy, 0.87 AUROC), outperforming a current UK oesophago-gastric cancer risk-assessment tool (ogRAT). Machine learning also identified more cancer patients than the ogRAT: 11.0% more with little to no effect on false positives, or up to 25.0% more with a slight increase in false positives (for Logistic Regression, results threshold-dependent). Feature contribution estimates and individual prediction explanations indicated clinical relevance. We conclude that machine learning could improve primary care cancer risk-assessment tools, potentially helping clinicians to identify additional cancer cases earlier. This could, in turn, improve survival outcomes. [ABSTRACT FROM AUTHOR]

In this article, we present a secondary analysis of a descriptive phenomenological study that we conducted in the United Kingdom exploring nurses' experiences of working with patients with substance dependence and pain. Our aim was to focus upon the ethical issues that emerged in the empirical data and so we used the Four Principles of Biomedical Ethics plus attention to scope to guide and inform our analysis. We present six key themes: trust, paternalism, coercion, failure to respect autonomy, advocacy and withholding. We discuss how these themes intersect with the four principles plus scope to illuminate practice and the ethical issues that emerge when managing this patient population's pain. We recommend that clinicians adopt a collaborative approach to managing pain for patients with substance dependence that they remain aware of the power differentials inherent within the clinical setting and ensure that communication and teamwork remain at the forefront of decisions. Clinicians need access to ethical guidance to inform their practice decisions and clinical ethics support services could provide one solution. [ABSTRACT FROM AUTHOR]

Abstract: The prevalence and burden of pain has long been reported as problematic. Comprehensive pain education in undergraduate programmes is essential for developing knowledgeable, skilled and effective healthcare professionals. This cross-sectional survey describes the nature, content and learning strategies for pain curricula in undergraduate healthcare programmes in major universities in the United Kingdom (UK). Document analysis also highlighted gaps in pain-related standards from professional regulators and a higher education quality assurance body. The sample consisted of 19 higher education institutions delivering 108 programmes across dentistry, medicine, midwifery, nursing, occupational therapy, pharmacy, physiotherapy and veterinary science. Seventy-four (68.5%) questionnaires were returned averaging 12.0h of pain content with physiotherapy and veterinary science students receiving the highest input. Pain education accounted for less than 1% of programme hours for some disciplines. Traditional teaching methods dominated (e.g. lectures 87.8%) and only two programmes had fully implemented the International Association for the Study of Pain’s (IASP) curricula. Minimal pain-related standards were found from professional regulators and the quality assurance documents. Pain education is variable across and within disciplines and interprofessional learning is minimal. Published curricula for pain education have been available for over 20years but are rarely employed and pain is not a core part of regulatory and quality assurance standards for health professions. The hours of pain education is woefully inadequate given the prevalence and burden of pain. Recommendations include the introduction of pain-related educational standards across all professions, greater integration of pain content in undergraduate programmes and interprofessional approaches to the topic. [Copyright &y& Elsevier]

Limited research exists regarding proximity of vehicles to "vulnerable road users" and effects of "conspicuity equipment". Previous studies stated over 60% of horse/rider combinations experience near-miss traffic collisions in any one year, whereas use of fluorescent/reflective (FR) equipment did not decrease occurrences. Similarly, research into vehicle proximity to bicycles reported no significance in cyclists wearing FR. This study undertook a similar field experiment using a horse/rider combination wearing different conspicuity equipment to test proximity distance. Two horses with similar height, age, base coat color, and temperament were ridden using four conspicuity measures 60 times each (n = 240) along a straight 0.8 km lane. Measures were chosen by random crossover with two selected for each separate trip. These were an FR "tabard", a novel black/white pattern (BW), flashing lights on a helmet (L), and control of mid-blue (C). A novel proximity meter was used on the saddle pommel and distance/approach either front or rear were logged into a voice recorder. Analysis reported a significant difference between all measures and proximities observed (P = .000). Mann-Whitney tests concluded significant differences in proximity in C versus L and BW (P = .000; P = .000), respectively. FR versus C showed no significant difference in proximity (P = .005). L versus BW reported no significant difference in proximity (W = 3640.0, P = .958). Results may suggest wearing conspicuity equipment of L or BW outperforms both C and FR. Further study is recommended to test these findings, and it is pertinent to still consider suggesting the use of FR equipment due to previous positive findings with other road users., (Copyright © 2019 Elsevier Inc. All rights reserved.)