Most screening tests for T2DM in use today were developed using multivariate regression methods that are often further simplified to allow transformation into a scoring formula. The increasing volume of electronically collected data opened the opportunity to develop more complex, accurate prediction models that can be continuously updated using machine learning approaches. This study compares machine learning-based prediction models (i.e. Glmnet, RF, XGBoost, LightGBM) to commonly used regression models for prediction of undiagnosed T2DM. The performance in prediction of fasting plasma glucose level was measured using 100 bootstrap iterations in different subsets of data simulating new incoming data in 6-month batches. With 6 months of data available, simple regression model performed with the lowest average RMSE of 0.838, followed by RF (0.842), LightGBM (0.846), Glmnet (0.859) and XGBoost (0.881). When more data were added, Glmnet improved with the highest rate (+ 3.4%). The highest level of variable selection stability over time was observed with LightGBM models. Our results show no clinically relevant improvement when more sophisticated prediction models were used. Since higher stability of selected variables over time contributes to simpler interpretation of the models, interpretability and model calibration should also be considered in development of clinical prediction models.

Emotional intelligence in nursing is of global interest. International studies identify that emotional intelligence influences nurses' work and relationships with patients. It is associated with compassion and care. Nursing students scored higher on measures of emotional intelligence compared to students of other study programmes. The level of emotional intelligence increases with age and tends to be higher in women. This study aims to measure the differences in emotional intelligence between nursing students with previous caring experience and those without; to examine the effects of gender on emotional intelligence scores; and to test whether nursing students score higher than engineering colleagues on emotional intelligence measures. A cross-sectional descriptive study design was used. The study included 113 nursing and 104 engineering students at the beginning of their first year of study at a university in Slovenia. Emotional intelligence was measured using the Trait Emotional Intelligence Questionnaire (TEIQue) and Schutte Self Report Emotional Intelligence Test (SSEIT). Shapiro-Wilk's test of normality was used to test the sample distribution, while the differences in mean values were tested using Student t-test of independent samples. Emotional intelligence was higher in nursing students (n = 113) than engineering students (n = 104) in both measures [TEIQue t = 3.972; p < 0.001; SSEIT t = 8.288; p < 0.001]. Although nursing female students achieved higher emotional intelligence scores than male students on both measures, the difference was not statistically significant [TEIQue t = -0.839; p = 0.403; SSEIT t = -1.159; p = 0.249]. EI scores in nursing students with previous caring experience were not higher compared to students without such experience for any measure [TEIQue t = -1.633; p = 0.105; SSEIT t = -0.595; p = 0.553]. Emotional intelligence was higher in nursing than engineering students, and slightly higher in women than men. It was not associated with previous caring experience.

Background: Physical activity is essential to maternal and infant health. Healthcare professionals should inform pregnant women about benefits of physical activity to prevent possible health issues. Those recommendations should elaborate on relevant contemporary evidence. The aim of this study was to review evidence-based recommendations for physical activity during pregnancy.Methods: A systematic search, analysis and synthesis of conducted randomised controlled trials (RCTs) was conducted from October 2021 to June 2022 in following databases: PubMed, CINAHL, ScienceDirect and Web of Science. Literature was searched using inclusion and exclusion criteria and following PRISMA recommendations.Results: Benefits for pregnant-women health and well-being were reported while performing aerobic exercise, lumbar stabilization and stretching exercise, water exercise, nerve and tendon-slip exercise, resistance training and strength training. For all exercise modalities it is recommended to perform moderate intensity activities during the whole time of pregnancy.Conclusions: This systematic literature review supplements current knowledge on physical activity of pregnant women. Exercise interventions are listed and suggested in an integrative model with physical-fitness components to contextualize and promote physical activity among pregnant women.

Use of conversational agents, like chatbots, avatars, and robots is increasing worldwide. Yet, their effectiveness in health care is largely unknown. The aim of this advanced review was to assess the use and effectiveness of conversational agents in various fields of health care. A literature search, analysis, and synthesis were conducted in February 2022 in PubMed and CINAHL. The included evidence was analyzed narratively by employing the principles of thematic analysis. We reviewed articles on artificial intelligence‐based question‐answering systems in health care. Most of the identified articles report its effectiveness; less is known about its use. We outlined study findings and explored directions of future research, to provide evidence‐based knowledge about artificial intelligence‐based question‐answering systems.This article is categorized under: Fundamental Concepts of Data and Knowledge > Human Centricity and User Interaction Application Areas > Health Care Technologies > Artificial Intelligence

Background: Poor oral hygiene is a great public health problem worldwide. Oral health care education is a public health priority as the maintenance of oral hygiene is integral to overall health. Maintaining optimal oral hygiene among children is challenging and can be supported by using relevant motivational approaches. Objective: The primary aim of this study was to identify mobile smartphone apps that include gamification features focused on motivating children to learn, perform, and maintain optimal oral hygiene. Methods: We searched six online app stores using four search terms (“oral hygiene game,” “oral hygiene gamification,” “oral hygiene brush game,” and “oral hygiene brush gamification”). We identified gamification features, identified whether apps were consistent with evidence-based dentistry, performed a quality appraisal with the Mobile App Rating Scale user version (uMARS), and quantified behavior scores (Behavior Change score, uMARS score, and Coventry, Aberdeen, and London-Refined [CALO-RE] score) using three different instruments that measure behavior change. Results: Of 612 potentially relevant apps included in the analysis, 17 met the inclusion criteria. On average, apps included 6.87 (SD 4.18) out of 31 possible gamification features. The most frequently used gamification features were time pressure (16/17, 94%), virtual characters (14/17, 82%), and fantasy (13/17, 76%). The most common oral hygiene evidence-based recommendation was brushing time (2-3 minutes), which was identified in 94% (16/17) of apps. The overall mean uMARS score for app quality was high (4.30, SD 0.36), with good mean subjective quality (3.79, SD 0.71) and perceived impact (3.58, SD 0.44). Sufficient behavior change techniques based on three taxonomies were detected in each app. Conclusions: The majority of the analyzed oral hygiene apps included gamification features and behavior change techniques to perform and maintain oral hygiene in children. Overall, the apps contained some educational content consistent with evidence-based dentistry and high-quality background for oral self-care in children; however, there is scope for improvement.