Uncertainty and probability in neonatal end-of-life decision-making: analysing real-time conversations between healthcare professionals and families of critically ill newborns.

BACKGROUND: A significant number of critically ill neonates face potentially adverse prognoses and outcomes, with some of them fulfilling the criteria for perinatal palliative care. When counselling parents about the critical health condition of their child, neonatal healthcare professionals require extensive skills and competencies in palliative care and communication. Thus, this study aimed to investigate the communication patterns and contents between neonatal healthcare professionals and parents of neonates with life-limiting or life-threatening conditions regarding options such as life-sustaining treatment and palliative care in the decision-making process. METHODS: A qualitative approach to analysing audio-recorded conversations between neonatal team and parents. Eight critically ill neonates and a total of 16 conversations from two Swiss level III neonatal intensive care units were included. RESULTS: Three main themes were identified: the weight of uncertainty in diagnosis and prognosis, the decision-making process, and palliative care. Uncertainty was observed to impede the discussion about all options of care, including palliative care. Regarding decision-making, neonatologists oftentimes conveyed to parents that this was a shared endeavour. However, parental preferences were not ascertained in the conversations analysed. In most cases, healthcare professionals were leading the discussion and parents expressed their opinion reactively to the information or options received. Only few couples proactively participated in decision-making. The continuation of therapy was often the preferred course of action of the healthcare team and the option of palliative care was not mentioned. However, once the option for palliative care was raised, the parents' wishes and needs regarding the end-of-life care of their child were obtained, respected, and implemented by the team. CONCLUSION: Although shared decision-making was a familiar concept in Swiss neonatal intensive care units, parental involvement in the decision-making process illustrated a somewhat different and complex picture. Strict adherence to the concept of certainty might impede the process of decision-making, thereby not discussing palliation and missing opportunities to include parental values and preferences.

Clinical significance of microscopic polyangiitis with interstitial lung disease and bronchiectasis: probability of preexisting comorbidities.

BACKGROUND: The association between pulmonary involvement and microscopic polyangiitis (MPA) has been increasingly recognized in recent years. Whether interstitial lung disease (ILD) and bronchiectasis (BE) are disease manifestations of MPA, preexisting comorbidities or important complications remains unclear. The purpose of this study was to determine the clinical characteristics and prognosis of MPA with pulmonary involvement to further guide clinical management. METHODS: The data for 97 patients with a definitive diagnosis of MPA were retrospectively reviewed. The MPA diagnosis was based on the 2012 revised Chapel Hill Consensus Conference (CHCC) criteria. The baseline clinical information and laboratory parameters were collected and analysed at each patient's initial diagnosis. RESULTS: Forty-seven out of the 97 (48.5%) patients who were diagnosed with MPA presented with pulmonary involvement, including 37 patients with ILD, 12 patients with BE and two patients with diffuse alveolar haemorrhage (DAH). ILD and BE antedated MPA in 56.76% and 75.00% of the patients, respectively. Compared with that in the MPA-BE group, the serum LDH level (222.86 ± 68.19 vs. 171.58 ± 31.43, p = .016) in the MPA-ILD group was significantly higher. In the multivariate Cox analysis, elevated serum creatinine (HR 4.08, confidence interval (CI) 1.38-12.05, p = .011) was an independent risk factor for shorter survival in MPA patients with pulmonary involvement, and treatment with glucocorticoid pulse cyclophosphamide therapy (HR 0.095, 95% CI 0.019-0.47, p = .004) was independently associated with prolonged survival. Among the patients in the MPA-ILD group, acute exacerbations of ILD (HR 4.55 CI 1.16-17.86, p = .029) and elevated serum creatinine (HR 4.95, CI 1.39-17.54, p = .014) were independently associated with a poor prognosis, and treatment with glucocorticoids (HR 0.057, 95% CI 0.012-0.28, p < .001) was independently associated with significant prolongation of survival. CONCLUSIONS: Patients with MPA have a high prevalence of pulmonary involvement, and ILD is the most common subtype of MPA. ILD and BE can be considered preexisting comorbidities of MPA. Elevated serum creatinine was associated with shorter survival. However, remission induction regimens with glucocorticoids and/or immunosuppressants may improve this outcome.

Non-cognitive skills and labour market performance of immigrants.

This paper investigates how non-cognitive skills relate to the relative labour market performance of immigrants. Using the German Socio-Economic Panel (SOEP) and the Five-Factor Model of personality as a proxy for the non-cognitive skills, we show that these skills matter for the labour market integration of immigrants in the host country. We use two comparison benchmarks. Compared to an average native, immigrants' non-cognitive skills, e.g., extroversion or emotional stability, can lead to 5-15 percentage points lower lifetime employment probability disadvantage implying a better overall integration. Comparing immigrants and natives with the same type and level of non-cognitive skills suggests that returns of extroversion and openness to experience are higher among immigrants, leading to 3-5 percentage points lower lifetime employment probability disadvantage. These results are robust with respect to self-selection, non-random returns to the home country, stability of personality, and estimators. Our detailed analysis suggests that non-cognitive skills (especially extroversion) are substitutes for the standard human capital measures (e.g., formal education and training) among low educated immigrants, while there is no significant relative return of non-cognitive skills among highly educated immigrants.

Improved reinforcement learning path planning algorithm integrating prior knowledge.

In order to realize the optimization of autonomous navigation of mobile robot under the condition of partial environmental knowledge known. An improved Q-learning reinforcement learning algorithm based on prior knowledge is proposed to solve the problem of slow convergence and low learning efficiency in mobile robot path planning. Prior knowledge is used to initialize the Q-value, so as to guide the agent to move toward the target direction with a greater probability from the early stage of the algorithm, eliminating a large number of invalid iterations. The greedy factor ε is dynamically adjusted based on the number of times the agent successfully reaches the target position, so as to better balance exploration and exploitation and accelerate convergence. Simulation results show that the improved Q-learning algorithm has a faster convergence rate and higher learning efficiency than the traditional algorithm. The improved algorithm has practical significance for improving the efficiency of autonomous navigation of mobile robots.

An age-of-infection model with both symptomatic and asymptomatic infections.

We formulate a general age-of-infection epidemic model with two pathways: the symptomatic infections and the asymptomatic infections. We then calculate the basic reproduction number [Formula: see text] and establish the final size relation. It is shown that the ratio of accumulated counts of symptomatic patients and asymptomatic patients is determined by the symptomatic ratio f which is defined as the probability of eventually becoming symptomatic after being infected. We also formulate and study a general age-of-infection model with disease deaths and with two infection pathways. The final size relation is investigated, and the upper and lower bounds for final epidemic size are given. Several numerical simulations are performed to verify the analytical results.

The causality-chance binomial: explanatory principle of scientific knowledge in medicine?

This essay questions mathematical truths as an explanatory principle of the medical scientific knowledge. It analyzes, in the first place, the current concept of normality based on a distribution of probabilistic values and its limitations and mistakes to capture the complexity of the human condition are highlighted. The closed systems (gambling) origin of the theory of probabilities and the binomial causality-chance are compared with open systems typical of the complexity of the vital process, and their extreme differences are argued. The nonsense of depositing in the causality-chance binomial the meaning of associations between events typical of the complexity of human life in health and disease is highlighted. The characteristics of mechanistic causality (punctual, homogeneous, linear, unidirectional and fixed), which equates the organism with a machine and is the only accepted scientific explanation of events of human life, are confronted with those of contextual causality (diffuse, heterogeneous, hierarchical, multidirectional and changing), which specifies various interacting causal orders that shape of human condition: the historical, the social, the political, the economical, the cultural or the biological that represents a scrutinizing and penetrating look at the complexity of human beings. It concludes the superiority of contextual causality over mechanistic causality that opens up explanatory possibilities of the vital events that are usually put away as "effects of chance". This integrative approach to the human complexity can enrich and strengthen the clinical method that is now degraded and at risk of extinction.

Este ensayo cuestiona las verdades matemáticas como principio explicativo del conocimiento científico médico. Se analiza, en primer término, el concepto de normalidad actual basado en una distribución de valores probabilísticos, y se destacan sus limitaciones y equívocos para captar la complejidad de la condición humana. Los sistemas cerrados (juegos de azar), origen de la teoría de las probabilidades y del binomio causalidad-azar, se comparan con los sistemas abiertos propios del proceso vital y se argumentan sus diferencias extremas. Se destaca el despropósito de depositar en el binomio causalidad-azar el significado de asociaciones entre sucesos propios de la complejidad de la vida humana en salud y enfermedad. Se confrontan las características de la causalidad mecanicista (puntual, lineal, unidireccional, homogénea y fija), que equipara al organismo con una máquina y es la única explicación científica aceptada del acontecer de la vida humana, con las de la causalidad contextual (difusa, heterogénea, jerárquica, multidireccional y cambiante), que especifica diversos órdenes causales interactuantes que dan forma a la condición humana: el histórico, el social, el político, el económico, el cultural o el biológico, que representa una mirada escrutadora y penetrante de la complejidad de los seres humanos. Se concluye la superioridad de la causalidad contextual sobre la mecanicista, que abre posibilidades explicativas de sucesos vitales que suelen arrumbarse como «efectos del azar¼. Esta aproximación integradora a la complejidad humana puede enriquecer y fortalecer el método clínico, hoy degradado y en riesgo de extinción.

A two-level copula joint model for joint analysis of longitudinal and competing risks data.

In this article, we propose a two-level copula joint model to analyze clinical data with multiple disparate continuous longitudinal outcomes and multiple event-times in the presence of competing risks. At the first level, we use a copula to model the dependence between competing latent event-times, in the process constructing the submodel for the observed event-time, and employ the Gaussian copula to construct the submodel for the longitudinal outcomes that accounts for their conditional dependence; these submodels are glued together at the second level via the Gaussian copula to construct a joint model that incorporates conditional dependence between the observed event-time and the longitudinal outcomes. To have the flexibility to accommodate skewed data and examine possibly different covariate effects on quantiles of a non-Gaussian outcome, we propose linear quantile mixed models for the continuous longitudinal data. We adopt a Bayesian framework for model estimation and inference via Markov Chain Monte Carlo sampling. We examine the performance of the copula joint model through a simulation study and show that our proposed method outperforms the conventional approach assuming conditional independence with smaller biases and better coverage probabilities of the Bayesian credible intervals. Finally, we carry out an analysis of clinical data on renal transplantation for illustration.

Dosage balance acts as a time-dependent selective barrier to subfunctionalization.

BACKGROUND: Gene duplication is an important process for genome expansion, sometimes allowing for new gene functions to develop. Duplicate genes can be retained through multiple processes, either for intermediate periods of time through processes such as dosage balance, or over extended periods of time through processes such as subfunctionalization and neofunctionalization. RESULTS: Here, we built upon an existing subfunctionalization Markov model by incorporating dosage balance to describe the interplay between subfunctionalization and dosage balance to explore selective pressures on duplicate copies. Our model incorporates dosage balance using a biophysical framework that penalizes the fitness of genetic states with stoichiometrically imbalanced proteins. These imbalanced states cause increased concentrations of exposed hydrophobic surface areas, which cause deleterious mis-interactions. We draw comparison between our Subfunctionalization + Dosage-Balance Model (Sub + Dos) and the previous Subfunctionalization-Only (Sub-Only) Model. This comparison includes how the retention probabilities change over time, dependent upon the effective population size and the selective cost associated with spurious interaction of dosage-imbalanced partners. We show comparison between Sub-Only and Sub + Dos models for both whole-genome duplication and small-scale duplication events. CONCLUSION: These comparisons show that following whole-genome duplication, dosage balance serves as a time-dependent selective barrier to the subfunctionalization process, by causing an overall delay but ultimately leading to a larger portion of the genome retained through subfunctionalization. This higher percentage of the genome that is ultimately retained is caused by the alternative competing process, nonfunctionalization, being selectively blocked to a greater extent. In small-scale duplication, the reverse pattern is seen, where dosage balance drives faster rates of subfunctionalization, but ultimately leads to a smaller portion of the genome retained as duplicates. This faster rate of subfunctionalization is because the dosage balance of interacting gene products is negatively affected immediately after duplication and the loss of a duplicate restores the stoichiometric balance. Our findings provide support that the subfunctionalization of genes that are susceptible to dosage balance effects, such as proteins involved in complexes, is not a purely neutral process. With stronger selection against stoichiometrically imbalanced gene partners, the rates of subfunctionalization and nonfunctionalization slow; however, this ultimately leads to a greater proportion of subfunctionalized gene pairs.

Learning-based sensitivity analysis and feedback design for drug delivery of mixed therapy of cancer in the presence of high model uncertainties.

In this paper, a methodology is proposed that enables to analyze the sensitivity of the outcome of a therapy to unavoidable high dispersion of the patient specific parameters on one hand and to the choice of the parameters that define the drug delivery feedback strategy on the other hand. More precisely, a method is given that enables to extract and rank the most influent parameters that determine the probability of success/failure of a given feedback therapy for a given set of initial conditions over a cloud of realizations of uncertainties. Moreover predictors of the expectations of the amounts of drugs being used can also be derived. This enables to design an efficient stochastic optimization framework that guarantees safe contraction of the tumor while minimizing a weighted sum of the quantities of the different drugs being used. The framework is illustrated and validated using the example of a mixed therapy of cancer involving three combined drugs namely: a chemotherapy drug, an immunology vaccine and an immunotherapy drug. Finally, in this specific case, it is shown that dash-boards can be built in the 2D-space of the most influent state components that summarize the outcomes' probabilities and the associated drug usage as iso-values curves in the reduced state space.

Chunking Versus Transitional Probabilities: Differentiating Between Theories of Statistical Learning.

There are two main approaches to how statistical patterns are extracted from sequences: The transitional probability approach proposes that statistical learning occurs through the computation of probabilities between items in a sequence. The chunking approach, including models such as PARSER and TRACX, proposes that units are extracted as chunks. Importantly, the chunking approach suggests that the extraction of full units weakens the processing of subunits while the transitional probability approach suggests that both units and subunits should strengthen. Previous findings using sequentially organized, auditory stimuli or spatially organized, visual stimuli support the chunking approach. However, one limitation of prior studies is that most assessed learning with the two-alternative forced-choice task. In contrast, this pre-registered experiment examined the two theoretical approaches in sequentially organized, visual stimuli using an online self-paced task-arguably providing a more sensitive index of learning as it occurs-and a secondary offline familiarity judgment task. During the self-paced task, abstract shapes were covertly organized into eight triplets (ABC) where one in every eight was altered (BCA) from the canonical structure in a way that disrupted the full unit while preserving a subunit (BC). Results from the offline familiarity judgment task revealed that the altered triplets were perceived as highly familiar, suggesting the learned representations were relatively flexible. More importantly, results from the online self-paced task demonstrated that processing for subunits, but not unit-initial stimuli, was impeded in the altered triplet. The pattern of results is in line with the chunking approach to statistical learning and, more specifically, the TRACX model.

How overconfidence bias influences suboptimality in perceptual decision making.

In perceptual decision making, it is often found that human observers combine sensory information and prior knowledge suboptimally. Typically, in detection tasks, when an alternative is a priori more likely to occur, observers choose it more frequently to account for the unequal base rate but not to the extent they should, a phenomenon referred to as "conservative decision bias" (i.e., observers do not shift their decision criterion enough). One theoretical explanation of this phenomenon is that observers are overconfident in their ability to interpret sensory information, resulting in overweighting the sensory information relative to prior knowledge. Here, we derived formally this candidate model, and we tested it in a visual discrimination task in which we manipulated the prior probabilities of occurrence of the stimuli. We measured confidence in decisions and decision criterion placement in two separate experimental sessions for the same participants (N = 69). Both overconfidence bias and conservative decision bias were found in our data, but critically the link that was predicted between these two quantities was absent. Our data suggested instead that when informed about the a priori probability, overconfident participants put less effort into processing sensory information. These findings offer new perspectives on the role of overconfidence bias to explain suboptimal decisions. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Temporal dynamics in mental health symptoms and loneliness during the COVID-19 pandemic in a longitudinal probability sample: a network analysis.

Figuring out which symptoms are central for symptom escalation during the COVID-19 pandemic is important for targeting prevention and intervention. Previous studies have contributed to the understanding of the course of psychological distress during the pandemic, but less is known about key symptoms of psychological distress over time. Going beyond a pathogenetic pathway perspective, we applied the network approach to psychopathology to examine how psychological distress unfolds in a period of maximum stress (pre-pandemic to pandemic onset) and a period of repeated stress (pandemic peak to pandemic peak). We conducted secondary data analyses with the Understanding Society data (N = 17,761), a longitudinal probability study in the UK with data before (2019), at the onset of (April 2020), and during the COVID-19 pandemic (November 2020 & January 2021). Using the General Health Questionnaire and one loneliness item, we computed three temporal cross-lagged panel network models to analyze psychological distress over time. Specifically, we computed (1) a pre-COVID to first incidence peak network, (2) a first incidence peak to second incidence peak network, and (3) a second incidence peak to third incidence peak network. All networks were highly consistent over time. Loneliness and thinking of self as worthless displayed a high influence on other symptoms. Feeling depressed and not overcoming difficulties had many incoming connections, thus constituting an end-product of symptom cascades. Our findings highlight the importance of loneliness and self-worth for psychological distress during COVID-19, which may have important implications in therapy and prevention. Prevention and intervention measures are discussed, as single session interventions are available that specifically target loneliness and worthlessness to alleviate mental health problems.

Redefining Possible: Combining Phylogenomic and Supersparse Data in Frogs.

The data available for reconstructing molecular phylogenies have become wildly disparate. Phylogenomic studies can generate data for thousands of genetic markers for dozens of species, but for hundreds of other taxa, data may be available from only a few genes. Can these two types of data be integrated to combine the advantages of both, addressing the relationships of hundreds of species with thousands of genes? Here, we show that this is possible, using data from frogs. We generated a phylogenomic data set for 138 ingroup species and 3,784 nuclear markers (ultraconserved elements [UCEs]), including new UCE data from 70 species. We also assembled a supermatrix data set, including data from 97% of frog genera (441 total), with 1-307 genes per taxon. We then produced a combined phylogenomic-supermatrix data set (a "gigamatrix") containing 441 ingroup taxa and 4,091 markers but with 86% missing data overall. Likelihood analysis of the gigamatrix yielded a generally well-supported tree among families, largely consistent with trees from the phylogenomic data alone. All terminal taxa were placed in the expected families, even though 42.5% of these taxa each had >99.5% missing data and 70.2% had >90% missing data. Our results show that missing data need not be an impediment to successfully combining very large phylogenomic and supermatrix data sets, and they open the door to new studies that simultaneously maximize sampling of genes and taxa.

Phylogenomic comparative methods: Accurate evolutionary inferences in the presence of gene tree discordance.

Phylogenetic comparative methods have long been a mainstay of evolutionary biology, allowing for the study of trait evolution across species while accounting for their common ancestry. These analyses typically assume a single, bifurcating phylogenetic tree describing the shared history among species. However, modern phylogenomic analyses have shown that genomes are often composed of mosaic histories that can disagree both with the species tree and with each other-so-called discordant gene trees. These gene trees describe shared histories that are not captured by the species tree, and therefore that are unaccounted for in classic comparative approaches. The application of standard comparative methods to species histories containing discordance leads to incorrect inferences about the timing, direction, and rate of evolution. Here, we develop two approaches for incorporating gene tree histories into comparative methods: one that constructs an updated phylogenetic variance-covariance matrix from gene trees, and another that applies Felsenstein's pruning algorithm over a set of gene trees to calculate trait histories and likelihoods. Using simulation, we demonstrate that our approaches generate much more accurate estimates of tree-wide rates of trait evolution than standard methods. We apply our methods to two clades of the wild tomato genus Solanum with varying rates of discordance, demonstrating the contribution of gene tree discordance to variation in a set of floral traits. Our approaches have the potential to be applied to a broad range of classic inference problems in phylogenetics, including ancestral state reconstruction and the inference of lineage-specific rate shifts.

Large scale sequence alignment via efficient inference in generative models.

Finding alignments between millions of reads and genome sequences is crucial in computational biology. Since the standard alignment algorithm has a large computational cost, heuristics have been developed to speed up this task. Though orders of magnitude faster, these methods lack theoretical guarantees and often have low sensitivity especially when reads have many insertions, deletions, and mismatches relative to the genome. Here we develop a theoretically principled and efficient algorithm that has high sensitivity across a wide range of insertion, deletion, and mutation rates. We frame sequence alignment as an inference problem in a probabilistic model. Given a reference database of reads and a query read, we find the match that maximizes a log-likelihood ratio of a reference read and query read being generated jointly from a probabilistic model versus independent models. The brute force solution to this problem computes joint and independent probabilities between each query and reference pair, and its complexity grows linearly with database size. We introduce a bucketing strategy where reads with higher log-likelihood ratio are mapped to the same bucket with high probability. Experimental results show that our method is more accurate than the state-of-the-art approaches in aligning long-reads from Pacific Bioscience sequencers to genome sequences.

Multi-vehicle safety functions for freeway weaving segments using lane-level traffic data.

This study develops Safety Performance Functions (SPFs) for freeway weaving segments. Due to the coexistence of three different movements including through, merging, and divering traffic, the probability of crashes in weaving segments is higher compared to other segment types. Further, the traffic flow in this section is the most unstable. Hence, to analyze detailed traffic conditions, this study utilized lane-level traffic data. The SPFs were developed using the Poisson Lognormal (PLN) regression model technique. The results showed that different traffic parameters were significant based on the types of crashes. For the rear-end crashes model, more general traffic conditions of the weaving segment were found to be significantly associated with the crash frequency such as the natural logarithm of average speed of through lanes. Nevertheless, for the sideswipe and angle crashes models, the traffic variables which are directly related to the weaving movements were selected as significant factors such as the off-ramp volume ratio, and standard deviation of speed of the rightmost lane. The results presented in this study can be meaningful in that they can serve as a basis for the weaving segments related safety evaluation studies. In addition, the developed models' results can be a great source to establish operational strategies to improve traffic safety on freeway weaving segments.

Electroencephalography in young onset dementia.

BACKGROUND: Young onset dementia (YOD) is a major diagnostic and management problem. METHODS: We set out to explore if electroencephalography (EEG) might be useful in the diagnosis of young onset Alzheimer's disease (YOAD) and young onset frontotemporal dementia (YOFTD). The ARTEMIS project is a 25-year prospective study of YOD based in Perth, Western Australia. 231 participants were included: YOAD: n = 103, YOFTD: n = 28, controls: n = 100. EEGs were performed prospectively, with 30-minute recording time for each subject, without knowledge of diagnosis or other diagnostic data. RESULTS: 80.9% of patients with YOD had abnormal EEGs (P < 0.00001). Slow wave changes were more frequent in YOAD that YOFTD (P < 0.00001), but no difference in the frequency of epileptiform activity (P = 0.32), with 38.8% of YOAD and 28.6% of YOFTD patients having epileptiform activity. Slow wave changes were more generalized in YOAD (P = 0.001). Slow wave changes and epileptiform activity were not sensitive to the diagnosis of YOD, but highly specific (97-99%). The absence of slow wave changes and epileptiform activity had a 100% negative predictive value and likelihood radio 0.14 and 0.62 respectively, meaning that those without slow wave changes or epileptiform activity had low probability of having YOD. No relationship was established between EEG findings and the patient's presenting problem. Eleven patients with YOAD developed seizures during the study, and only one with YOFTD. CONCLUSIONS: The EEG is highly specific for the diagnosis of YOD with the absence of slow wave changes and epileptiform phenomena making the diagnosis unlikely, with 100% negative predictive value and with low probability for the dementia diagnosis.

Inter-species stimulus enhancement: herring gulls (Larus argentatus) mimic human food choice during foraging.

Herring gulls (Larus argentatus) are one of few species that thrive in anthropogenic landscapes, and their familiarity with people makes them an excellent target for studies of inter-species social cognition. Urban gulls pay attention to human behaviour in food-related contexts and, thus, we set out to investigate whether such cues can influence a gull's attention to and choice of potential food items in their environment. Herring gulls were given free choice of two differently coloured anthropogenic food items in the presence of a demonstrator, who was either sitting still or eating food from an item that matched one of the presented ones. We found that a demonstrator eating significantly increased the likelihood of a gull pecking one of the presented items. Furthermore, 95% of pecks were directed toward the presented food item that colour-matched the demonstrator's food item. The results showed gulls were able to use human cues for stimulus enhancement and foraging decisions. Given the relatively recent history of urbanization in herring gulls, this cross-species social information transfer could be a by-product of the cognitive flexibility inherent in kleptoparasitic species.

MSMpred: interactive modelling and prediction of individual evolution via multistate models.

BACKGROUND: Modelling the course of a disease regarding severe events and identifying prognostic factors is of great clinical relevance. Multistate models (MSM) can be used to describe diseases or processes that change over time using different states and the transitions between them. Specifically, they are useful to analyse a disease with an increasing degree of severity, that may precede death. The complexity of these models changes depending on the number of states and transitions taken into account. Due to that, a web tool has been developed making easier to work with those models. RESULTS: MSMpred is a web tool created with the shiny R package that has two main features: 1) to allow to fit a MSM from specific data; 2) to predict the clinical evolution for a given subject. To fit the model, the data to be analysed must be upload in a prespecified format. Then, the user has to define the states and transitions as well as the covariates (e.g., age or gender) involved in each transition. From this information, the app returns histograms or barplots, as appropriate, to represent the distributions of the selected covariates and boxplots to show the patient' length of stay (for uncensored data) in each state. To make predictions, the values of selected covariates from a new subject at baseline has to be provided. From these inputs, the app provides some indicators of the subject's evolution such as the probability of 30-day death or the most likely state at a fixed time. Furthermore, visual representations (e.g., the stacked transition probabilities plot) are given to make predictions more understandable. CONCLUSIONS: MSMpred is an intuitive and visual app that eases the work of biostatisticians and facilitates to the medical personnel the interpretation of MSMs.

"I wish I had known what I was getting into": a qualitative study exploring the experiences of Canadians who study medicine abroad.

BACKGROUND: An increasing number of Canadians are choosing to study medicine abroad (CSA); however, many CSA are not fully informed of the challenges that exist in returning to Canada to practice and relatively little information is known on the topic. This study explores CSA experiences in choosing to study abroad and their attempts to navigate a return to Canada to practice medicine. METHODS: We conducted semi-structured qualitative interviews with CSA who were attending medical school abroad, waiting to obtain or in a post-graduate residency program, or practicing in Canada. We asked participants about their decision to study medicine abroad and choice of school, medical school experiences, activities they engaged in to increase their likelihood of returning to Canada, perceived barriers and facilitators, and alternative plans if they were unable to return to Canada to practice. Interviews were transcribed and analyzed using a thematic analysis approach. RESULTS: Fourteen CSA participated in an interview. Expedited timelines (i.e., direct entry from high school) and a lack of competitiveness for medical school in Canada were the main justifications for CSAs' decision to study abroad and a number of key factors (e.g., location, reputation) influenced their choice of school. Participants reported not fully anticipating the challenges associated with obtaining residency in Canada. CSA relied upon a variety of informal and formal supports and employed numerous methods to increase their likelihood of returning to Canada. CONCLUSIONS: Studying medicine abroad remains a popular choice for Canadians; however, many trainees are unaware of the challenges associated with returning to Canada to practice. More information on this process as well as the quality of these medical schools is needed for Canadians considering this option.