Search

BackgroundLate HIV diagnosis remains frequent among the gay, bisexual, and other men who have sex with men (GBMSM) population across Europe. HIV self-sampling could help remove barriers and facilitate access to testing for this high-risk population. ObjectiveWe assessed the capacity of HIV self-sampling to increase the testing frequency among GBMSM living in Denmark, Germany, Greece, Portugal, Romania, and Spain, and evaluated the role of new technologies in the result communication phase. MethodsWe analyzed a convenience sample of 5019 GBMSM with prior HIV testing experience who were recruited during 2016 through gay dating websites. We estimated the proportion of GBMSM who reported that the availability of self-sampling would result in an increase of their current testing frequency. We constructed a Poisson regression model for each country to calculate prevalence ratios and 95% CIs of factors associated with an increase of testing frequency as a result of self-sampling availability. ResultsOverall, 59% (between country range 54.2%-77.2%) of the participants considered that they would test more frequently for HIV if self-sampling was available in their country. In the multivariate analysis, the increase of testing frequency as a result of self-sampling availability was independently associated with reporting a higher number of unprotected anal intercourse events in all countries except for Greece. Independent associations were also observed among GBMSM who were not open about their sex life in Germany, Greece, Portugal, and Spain; those with a lower number of previous HIV tests in Denmark, Greece, Portugal, and Spain; and for those that took their last test more than 3 months previously in Germany, Portugal, Romania, and Spain. In addition, 58.4% (range 40.5%-73.6%) of the participants indicated a preference for learning their result through one-way interaction methods, mainly via email (25.6%, range 16.8%-35.2%) and through a secure website (20.3%, range 7.3%-23.7%). Almost two thirds (65%) of GBMSM indicated preferring one of these methods even if the result was reactive. ConclusionsAvailability of HIV self-sampling kits as an additional testing methodology would lead to a much-needed increase of testing frequency, especially for the hidden, high-risk, and undertested GBMSM population. Online-based technologies without any personal interaction were preferred for the communication of the results, even for reactive results.

Aims: Many European healthcare providers struggle to adopt multidisciplinary, integrated care pathways for people with heart failure (HF) as recommended by the European Society of Cardiology. PRO‐HF (Program to Optimize Heart Failure Patient Pathways) was developed to help clinicians identify strengths, gaps, and shortcomings in their HF pathways and support tailored interventions to optimize pathways and enhance patient care. We report initial findings from baseline assessments of HF pathway characteristics and challenges from 10 hospitals in six European countries (France, Ireland, Portugal, Spain, The Netherlands, and United Kingdom). Methods and results: Baseline assessments were holistic appraisals of full HF services to calibrate current status and development needs and assist management teams in prioritizing improvement projects. Assessments were performed using a comprehensive checklist of measures covering the HF patient journey from diagnosis to ongoing follow‐up. These included a digital survey sent to full HF care teams and one‐to‐one interviews. The digital survey focused on four key areas (HF outpatient clinic; remote patient management; efficient device implantation and inpatient pathways; and network maximization) and 16 dimensions of excellence. Priority areas and themes for action identified in baseline assessments were (i) provision of HF specialist care; (ii) data capture and analysis; (iii) institutional care protocols; (iv) hospital‐wide strategies; and (v) multidisciplinary teams (MDTs). Suboptimal specialist care of emergency inpatients was an issue at all hospitals and prioritized at 8/10. Availability and accessibility of data on patients, activities, and outcomes was an issue at all hospitals and prioritized by 4/10. A lack of clear protocols, templates, and tools for some HF activities created variability in patient care (e.g., HF specialist consultations, diagnostic testing, follow‐up appointments, medications, and device eligibility) and inefficient use of clinician time. This made it difficult to initiate new technologies (e.g., remote patient monitoring) due to the risk of overburdening staff. MDTs were frequently understaffed. Multiple interventions were identified to address gaps and shortcomings that could be tailored to specific needs of individual hospitals (e.g., inpatient pathway optimization, creation/optimization of HF outpatient clinics, development of an HF performance dashboard, enhancement of protocol adherence, streamlining cardiac resynchronisation therapy pathways, and MDT coordination). Conclusions: PRO‐HF provides a valuable opportunity to identify gaps and significant shortcomings in HF pathways in European hospitals. Preliminary findings from hospitals that have initiated suggested changes to address these challenges are encouraging, though longer‐term follow‐up from more hospitals is needed to confirm the impact of PRO‐HF on HF pathway optimization and patient care. [ABSTRACT FROM AUTHOR]

Several studies have demonstrated the ability of long short-term memory (LSTM) machine-learning-based modeling to outperform traditional spatially lumped process-based modeling approaches for streamflow prediction. However, due mainly to the structural complexity of the LSTM network (which includes gating operations and sequential processing of the data), difficulties can arise when interpreting the internal processes and weights in the model. Here, we propose and test a modification of LSTM architecture that is calibrated in a manner that is analogous to a hydrological system. Our architecture, called "HydroLSTM", simulates the sequential updating of the Markovian storage while the gating operation has access to historical information. Specifically, we modify how data are fed to the new representation to facilitate simultaneous access to past lagged inputs and consolidated information, which explicitly acknowledges the importance of trends and patterns in the data. We compare the performance of the HydroLSTM and LSTM architectures using data from 10 hydro-climatically varied catchments. We further examine how the new architecture exploits the information in lagged inputs, for 588 catchments across the USA. The HydroLSTM-based models require fewer cell states to obtain similar performance to their LSTM-based counterparts. Further, the weight patterns associated with lagged input variables are interpretable and consistent with regional hydroclimatic characteristics (snowmelt-dominated, recent rainfall-dominated, and historical rainfall-dominated). These findings illustrate how the hydrological interpretability of LSTM-based models can be enhanced by appropriate architectural modifications that are physically and conceptually consistent with our understanding of the system. [ABSTRACT FROM AUTHOR]

High-resolution, spatially distributed process-based (PB) simulators are widely employed in the study of complex watershed processes and their responses to a changing climate. However, calibrating these simulators to observed data remains a significant challenge due to several persistent issues including: (1) intractability stemming from the computational demands and complex responses of simulators, which renders infeasible calculation of the conditional probability of parameters and data, and (2) uncertainty stemming from the choice of simplified model representations of complex natural hydrologic processes. Here we demonstrate how Simulation-Based Inference (SBI) can help address both these challenges for parameter estimation. SBI uses a learned mapping between parameter space and observed data to estimate parameters for generation of calibrated model simulations. To demonstrate the potential of SBI in hydrologic modelling, we conduct a set of synthetic experiments to infer two common physical parameters, Manning's coefficient and hydraulic conductivity, using a representation of a snowmelt-dominated catchment in Colorado, USA. We introduce novel deep learning (DL) components to the SBI approach, including an 'emulator' as a surrogate for the process-based simulator to rapidly explore parameter responses. We also employ a density-based neural network to represent the joint probability of parameters and data without strong assumptions about its functional form. While addressing intractability, we also show that where uncertainty about model structure is significant, SBI can yield unreliable parameter estimates. Approaches to adopting the SBI framework to cases where model structure(s) may not be adequate are introduced using a performance-weighting approach. [ABSTRACT FROM AUTHOR]

Several studies have demonstrated the ability of Long Short-Term Memory (LSTM) machine learning based modeling to outperform traditional spatially-lumped process-based modeling approaches for streamflow prediction. However, due mainly to the structural complexity of the LSTM network (which includes gating operations and sequential processing of the data), difficulties can arise when interpreting the internal processes and weights in the model. Here, we propose and test a modification of LSTM architecture that is calibrated in a manner that is analogous to a hydrological system. Our architecture, called HydroLSTM, simulates the sequential updating of the Markovian storage while the gating operation has access to historical information. Specifically, we modify how data is fed to the new representation to facilitate simultaneous access to past lagged inputs and consolidated information, which explicitly acknowledges the importance of trends and patterns in the data. We compare the performance of the HydroLSTM and LSTM architectures using data from 10 hydro-climatically varied catchments. We further examine how the new architecture exploits the information in lagged inputs, for 588 catchments across the USA. The HydroLSTM-based models require fewer cell states to obtain similar performance to their LSTM-based counterparts. Further, the weight patterns associated with lagged input variables are interpretable and consistent with regional hydroclimatic characteristics (snowmelt-dominated, recent rainfall-dominated, and historical rainfall-dominated). These findings illustrate how the hydrological interpretability of LSTM-based models can be enhanced by appropriate architectural modifications that are physically and conceptually consistent with our understanding of the system. [ABSTRACT FROM AUTHOR]