The labor experience and satisfaction with childbirth are affected by the care provided (external factors) and individual variables (internal factors). In this paper, we present a descriptive analysis that aims to indicate the strongest correlates of birth experience among a wide range of indicators. The study is a prospective, cross-sectional, self-report survey. It includes the experiences of women giving birth in public and private hospitals in Poland. The two main variables were birth experience and satisfaction with care. The analysis consists of three parts: data pre-processing and initial analysis, explorative investigation, and regression analysis. Among the 15 variables with the highest predictive value regarding birth experience were being informed by the medical personnel, communication, and birth environment. The most significant variables among 15 variables, with the highest predictive value regarding care, were those concerning support, information, and respectful care. The strongest predictor for both, birth experience and satisfaction with care, is the sense of information, with logit coefficients of 0.745 and 1.143, respectively, for birth experience and satisfaction (0.367 and 0.346 for standardized OLS coefficient). The findings demonstrate that by using explanatory variables, one can predict a woman’s description of her satisfaction with perinatal care received in the hospital. On the other hand, they do not have such a significant and robust influence on the birth experience examined by the variables. For both the birth experience and satisfaction with care, the sense of being informed is the highest predictor.

No abstract

<p>Deep learning-based models have been recently shown to be competitive with, or even outperform, state-of-the-art long range forecasting models, such as for projecting the El Niño-Southern Oscillation (ENSO). However, current deep learning models are based on convolutional neural networks which are difficult to interpret and can fail to model large-scale dependencies, such as teleconnections, that are particularly important for long range projections. Hence, we propose to explicitly model large-scale dependencies with Graph Neural Networks (GNN) to enhance explainability and improve the predictive skill of long lead time forecasts.</p><p>In preliminary experiments focusing on ENSO, our GNN model outperforms previous state-of-the-art machine learning based systems for forecasts up to 6 months ahead. The explicit modeling of information flow via edges makes our model more explainable, and it is indeed shown to learn a sensible graph structure from scratch that correlates with the ENSO anomaly pattern for a given number of lead months.</p><p> </p>

No abstract