Paola Garcia

We present PhrasIS, a benchmark dataset composed of natural occurring Phrase pairs with Inference and Similarity annotations for the evaluation of semantic representations. The described dataset fills the gap between word and sentence-level datasets, allowing to evaluate compositional models at a finer granularity than sentences. Contrary to other datasets, the phrase pairs are extracted from naturally occurring text in image captions and news headlines. All the text fragments have been annotated by experts following a rigorous process also described in the manuscript achieving high inter annotator agreement. In this work we analyse the dataset, showing the relation between inference labels and similarity scores. With 10K phrase pairs split in development and test, the dataset is an excellent benchmark for testing meaning representation systems.

Background Moral distress appears when a healthcare professional is not able to carry out actions in accordance with their professional ethical standards. The Moral Distress Scale-Revised is the most widely used to assess levels of moral distress, but it is not validated in Spanish. The aim of the study is to validate the Spanish version of the Moral Distress Scale – utilised within a sample of Spanish healthcare professionals treating COVID–19 patients. Methods The original (english) and the portuguese and french versions of the scale were translated into spanish by native or bilingual researchers and reviewed by an academic expert in ethics and moral philosophy as well as by a clinical expert. Research design: Descriptive cross-sectional study carried out using a self-reporting online survey. The data was collected between June- November 2020. A total of 661 professionals responded to the survey (N = 2873). Participants: healthcare professionals with more than two weeks of experience treating COVID–19 patients at the end of their life and working in the public sector of the Balearic Islands Health Service (Spain). Analyses included descriptive statistics, competitive confirmatory factor analysis, evidence on criterion-related validity and estimates of reliability. The study was approved by the Research Ethics Committee at the University of Balearic Islands. Results An unidimensional model in which a general factor of moral distress explained by 11 items of the Spanish version of the MDS–R scale was an adequate representation of the data: χ 2 (44) = 113.492 (p 0.001); Comparative Fit Index = 0.965; Root Mean Square Error of Approximation = 0.079[0.062,0.097]; and Standarized Root Mean-Square = 0.037. Evidence of reliability was excellent: Cronbach’s alpha = 0.886 and McDonald’s omega = 0.910. Moral distress was related to discipline, with nurses having statistically significant higher levels than physicians. Additionally, moral distress successfully predicted professional quality of life, with higher levels of moral distress being related to poorer quality of life. Conclusions The Spanish version of Moral Distress Scale–Revised can be used as a reliable and valid measurement tool for the evaluation of moral distress experienced by health professionals. This tool will be highly useful for managers and applicable to a variety of healthcare professionals and settings.