RESUMEN

Animal research often involves measuring the outcomes of interest multiple times on the same animal, whether over time or for different exposures. These repeated outcomes measured on the same animal are correlated due to animal-specific characteristics. While this repeated measures data can address more complex research questions than single-outcome data, the statistical analysis must take into account the study design resulting in correlated outcomes, which violate the independence assumption of standard statistical methods (e.g. a two-sample t-test, linear regression). When standard statistical methods are incorrectly used to analyze correlated outcome data, the statistical inference (i.e. confidence intervals and p-values) will be incorrect, with some settings leading to null findings too often and others producing statistically significant findings despite no support for this in the data. Instead, researchers can leverage approaches designed specifically for correlated outcomes. In this article, we discuss common study designs that lead to correlated outcome data, motivate the intuition about the impact of improperly analyzing correlated outcomes using methods for independent data, and introduce approaches that properly leverage correlated outcome data.

RESUMEN

Many clinical trials involve partially clustered data, where some observations belong to a cluster and others can be considered independent. For example, neonatal trials may include infants from single or multiple births. Sample size and analysis methods for these trials have received limited attention. A simulation study was conducted to (1) assess whether existing power formulas based on generalized estimating equations (GEEs) provide an adequate approximation to the power achieved by mixed effects models, and (2) compare the performance of mixed models vs GEEs in estimating the effect of treatment on a continuous outcome. We considered clusters that exist prior to randomization with a maximum cluster size of 2, three methods of randomizing the clustered observations, and simulated datasets with uninformative cluster size and the sample size required to achieve 80% power according to GEE-based formulas with an independence or exchangeable working correlation structure. The empirical power of the mixed model approach was close to the nominal level when sample size was calculated using the exchangeable GEE formula, but was often too high when the sample size was based on the independence GEE formula. The independence GEE always converged and performed well in all scenarios. Performance of the exchangeable GEE and mixed model was also acceptable under cluster randomization, though under-coverage and inflated type I error rates could occur with other methods of randomization. Analysis of partially clustered trials using GEEs with an independence working correlation structure may be preferred to avoid the limitations of mixed models and exchangeable GEEs.

RESUMEN

BACKGROUND: Previous intervention studies among night workers mainly focused on single interventions and found inconclusive evidence for effectiveness. A comprehensive intervention approach that includes individual and environmental components has been argued as important. Gaining insight into contributing factors for the implementation of interventions for night workers and effectiveness is important to distinguish between theory and programme failure. OBJECTIVES: To evaluate the effects and implementation of the PerfectFit@Night intervention to improve sleep, fatigue and recovery of night workers in healthcare, using the RE-AIM framework, which assesses reach, effectiveness, adoption, implementation and maintenance of interventions. DESIGN: A prospective pre-post study design, with two measurements before and three and six months after the intervention. SETTING: Twelve different departments of a university hospital in the Netherlands. PARTICIPANTS: Healthcare workers working night shifts (n = 210). METHODS: PerfectFit@Night consisted of environmental (provision of a powernap bed and healthy food, and workshop healthy rostering) and individual elements (e-learning and sleep coaching) and was implemented for three months in a phased manner. Questionnaires, logbooks and interview data were used. Effects of the intervention on sleep, fatigue and recovery were evaluated with mixed-effects models, and implementation factors of reach, adoption, implementation and maintenance were evaluated. RESULTS: Night shift-related insomnia (-11 %-points, 95 % CI: -19 %, -4 % at three months), need for recovery (ß: -2.45, 95 % CI: -4.86, -0.03 at six months) and fatigue (OR: 0.46, 95 % CI: 0.25, 0.86 at six months) decreased significantly after the intervention. No changes were found for subjective sleep quality and sleep duration. Barriers and facilitators for implementation were identified for each intervention element at individual (e.g., dietary preferences), organisational (e.g., responsibilities at work) and workplace levels (e.g., location of power nap bed), and for the intervention itself (e.g., useful information in e-learning). Although satisfaction was high and continuation was preferred, embedding of the intervention in the daily routine was limited. Facilitators for future implementation include a positive attitude towards the intervention, clear guidelines regarding intervention elements, appointment of night workers as ambassadors, and suitable conditions in terms of work demands and for the intervention elements. CONCLUSIONS: The multi-faceted PerfectFit@Night intervention reduced insomnia, fatigue and need for recovery in night workers in healthcare. The most important facilitators to improve the implementation of PerfectFit@Night exist at the organisational level (e.g., positive attitude within the culture and suitable work demands). Combining effect and implementation evaluation is crucial to identify barriers and facilitators that hamper or enhance intervention effects. TRIAL REGISTRATION: The study was registered in the Netherlands Trial Register on 17 January 2021 (trial number NL9224).

RESUMEN

Foraging efficiency is key to animal fitness. Consequently, animals evolved a variety of kinematic, morphological, physiological, and behavioral adaptations for efficient locomotion to reduce energy expenditure while moving to find, capture, and consume prey. Often suited to specific habitat and prey types, these adaptations correspond to the terrain or substrate the animal moves through. In aquatic systems, adaptations focus on overcoming drag, buoyancy, and hydrostatic forces. Buoyancy both benefits and hinders diving animals; in particular, shallow divers constantly contend with the costs of overcoming buoyancy to dive and maintain position. Pacific Coast Feeding Group (PCFG) gray whales forage in shallow habitats where they work against buoyancy to dive and feed using various foraging tactics. Bubble blasts (underwater exhalations) have been observed during several foraging tactics performed by PCFG whales. As exhalations aid buoyancy regulation in other diving animals, we hypothesize that bubble blasts are performed by longer, more buoyant whales in shallower water and that bubble blasts increase dive duration while accounting for size and tactic. We test our hypotheses using Bayesian linear mixed effects models and a 7-year dataset of drone footage containing concurrent individual morphological and behavioral data. We find that while headstanding - a stationary, head-down tactic - bubble blasts are performed by longer, more buoyant whales and extend the dive duration, whereas whales using forward-swimming tactics are less likely to bubble blast. Our results suggest that PCFG gray whales may use bubble blasts as a behavioral adaption to mitigate the cost of energetically expensive tactics in their shallow habitat foraging niche.

RESUMEN

For topical, dermatological drug products, an in vitro option to determine bioequivalence (BE) between test and reference products is recommended. In particular, in vitro permeation test (IVPT) data analysis uses a reference-scaled approach for two primary endpoints, cumulative penetration amount (AMT) and maximum flux (Jmax), which takes the within donor variability into consideration. In 2022, the Food and Drug Administration (FDA) published a draft IVPT guidance that includes statistical analysis methods for both balanced and unbalanced cases of IVPT study data. This work presents a comprehensive evaluation of various methodologies used to estimate critical parameters essential in assessing BE. Specifically, we investigate the performance of the FDA draft IVPT guidance approach alongside alternative empirical and model-based methods utilizing mixed-effects models. Our analyses include both simulated scenarios and real-world studies. In simulated scenarios, empirical formulas consistently demonstrate robustness in approximating the true model, particularly in effectively addressing treatment-donor interactions. Conversely, the effectiveness of model-based approaches heavily relies on precise model selection, which significantly influences their results. The research emphasizes the importance of accurate model selection in model-based BE assessment methodologies. It sheds light on the advantages of empirical formulas, highlighting their reliability compared to model-based approaches and offers valuable implications for BE assessments. Our findings underscore the significance of robust methodologies and provide essential insights to advance their understanding and application in the assessment of BE, employed in IVPT data analysis.

RESUMEN

The case of the deepwater redfish (Sebastes mentella) in the Gulf of St. Lawrence (GSL) is a compelling example of drastic fluctuations in annual recruitment strength, characteristic of spasmodic stocks. After three decades of low abundance, the emergence of three consecutive strong year classes in 2011-2013 resulted in an unprecedented increase in biomass. In spasmodic stocks such as GSL redfish, strong year classes sustain both the biomass and catch for decades. Therefore, understanding the growth dynamics of these cohorts is essential. In the present study, we reconstructed the annual growth rates of redfish using otolith increment-based annual chronology and investigated the drivers of growth variation in redfish strong year classes of the early 2010s and early 1980s. Stock biomass was identified as the main extrinsic driver of redfish growth, suggesting intense competition for food at high conspecific density. Warming of deep waters in the GSL, where adult redfish settle, positively correlated with individual growth. However, recent warming of the cold intermediate layer showed a negative correlation with redfish growth, likely related to the shrinking of the habitat this water mass provides for various redfish cold-water prey rather than to a direct effect of temperature. Reconstruction of redfish annual growth trajectories from birth to capture emphasized the importance of carryover effects in the growth potential of strong year classes. This work provided an important first outlook of the factors driving growth variation in GSL redfish spasmodic stock and explored midterm consequences of density-dependent pressures on biological parameters of the population.

RESUMEN

It remains unclear how ambient air pollution may affect the prevalence of obstructive ventilatory dysfunction (OVD) among workers. We aim to assess the association of a comprehensive set of ambient air pollutants with OVD prevalence in workers and to explore the potential interactive effects of the occupational factors. This is a population-based cross-sectional study among 305,022 participants from the Guangdong Province, China. Mixed-effects models were used to obtain differences in the OVD risk associated with a 10 µg/m3 increase in ambient air pollution. We found that for each 10 µg/m3 increase in PM2.5, PM10, PM coarse, O3, and NO2 concentrations, the odds ratio (OR) for OVD in workers is 1.324 (95 % confidence interval (CI), 1.282-1.367), 1.292 (95 % CI, 1.268-1.315),1.666 (95 % CI, 1.614-1.719), 1.153 (95 % CI, 1.142-1.165), and 1.023 (95 % CI, 1.012-1.033). We observed that young participants (18-38 years old), women, participants with longer years of service (>48 months), participants working in large enterprises, professional skills workers, and production and manufacturing workers have higher estimated effects. In addition, we also found that workers exposed to high temperatures have higher estimated effects under air pollutants exposure, while workers exposed to noise have higher estimated effects under PM2.5, PM10, NO2, and O3 exposure. Workers exposed to dust have a lower risk of developing OVD under exposure to ambient air pollutants compared to those not exposed. Our results indicate that ambient air pollution increases the risk of OVD in workers. Moreover, air pollutants exhibit a greater estimated effect among workers exposed to high temperatures or noise. Our research findings highlight the importance of fully considering the impact of ambient air pollution on protecting the respiratory health of workers.

Asunto(s)

RESUMEN

PURPOSE: In exposure-response analyses of oral targeted anticancer agents, longitudinal plasma trough concentrations are often aggregated into a single value even though plasma trough concentrations can vary over time due to dose adaptations, for example. The aim of this study was to compare joint models to conventional exposure-response analyses methods with the application of alectinib as proof-of-concept. METHODS: Joint models combine longitudinal pharmacokinetic data and progression-free survival data to infer the dependency and association between the two datatypes. The results from the best joint model and the standard and time-dependent cox proportional hazards models were compared. To normalize the data, alectinib trough concentrations were normalized using a sigmoidal transformation to transformed trough concentrations (TTC) before entering the models. RESULTS: No statistically significant exposure-response relationship was observed in the different Cox models. In contrast, the joint model with the current value of TTC in combination with the average TTC over time did show an exposure-response relationship for alectinib. A one unit increase in the average TTC corresponded to an 11% reduction in progression (HR, 0.891; 95% confidence interval, 0.805-0.988). CONCLUSION: Joint models are able to give insights in the association structure between plasma trough concentrations and survival outcomes that would otherwise not be possible using Cox models. Therefore, joint models should be used more often in exposure-response analyses of oral targeted anticancer agents.

Asunto(s)

RESUMEN

Autism spectrum disorder (autism) is a prevalent neurodevelopmental condition characterized by early emerging impairments in social behavior and communication. EEG represents a powerful and non-invasive tool for examining functional brain differences in autism. Recent EEG evidence suggests that greater intra-individual trial-to-trial variability across EEG responses in stimulus-related tasks may characterize brain differences in autism. Traditional analysis of EEG data largely focuses on mean trends of the trial-averaged data, where trial-level analysis is rarely performed due to low neural signal to noise ratio. We propose to use nonlinear (shape-invariant) mixed effects (NLME) models to study intra-individual inter-trial EEG response variability using trial-level EEG data. By providing more precise metrics of response variability, this approach could enrich our understanding of neural disparities in autism and potentially aid the identification of objective markers. The proposed multilevel NLME models quantify variability in the signal's interpretable and widely recognized features (e.g., latency and amplitude) while also regularizing estimation based on noisy trial-level data. Even though NLME models have been studied for more than three decades, existing methods cannot scale up to large data sets. We propose computationally feasible estimation and inference methods via the use of a novel minorization-maximization (MM) algorithm. Extensive simulations are conducted to show the efficacy of the proposed procedures. Applications to data from a large national consortium find that children with autism have larger intra-individual inter-trial variability in P1 latency in a visual evoked potential (VEP) task, compared to their neurotypical peers.

Asunto(s)

RESUMEN

Stepped wedge trials (SWTs) are a type of cluster randomized trial that involve repeated measures on clusters and design-induced confounding between time and treatment. Although mixed models are commonly used to analyze SWTs, they are susceptible to misspecification particularly for cluster-longitudinal designs such as SWTs. Mixed model estimation leverages both "horizontal" or within-cluster information and "vertical" or between-cluster information. To use horizontal information in a mixed model, both the mean model and correlation structure must be correctly specified or accounted for, since time is confounded with treatment and measurements are likely correlated within clusters. Alternative non-parametric methods have been proposed that use only vertical information; these are more robust because between-cluster comparisons in a SWT preserve randomization, but these non-parametric methods are not very efficient. We propose a composite likelihood method that focuses on vertical information, but has the flexibility to recover efficiency by using additional horizontal information. We compare the properties and performance of various methods, using simulations based on COVID-19 data and a demonstration of application to the LIRE trial. We found that a vertical composite likelihood model that leverages baseline data is more robust than traditional methods, and more efficient than methods that use only vertical information. We hope that these results demonstrate the potential value of model-based vertical methods for SWTs with a large number of clusters, and that these new tools are useful to researchers who are concerned about misspecification of traditional models.

Asunto(s)

RESUMEN

BACKGROUND: Since the 1960's, mercury (Hg) contamination of the aquatic environment of Asubpeeschoseewagong Anishinabek (Grassy Narrows First Nation) territories has impacted the community members' traditions, culture, livelihood, diet and health. Despite decreasing Hg exposure over time, a recent study suggested that long-term exposure contributed to later-life symptom clusters of nervous system dysfunction. Here, the objective was to evaluate, 5 years later, the prevalence and progression of these symptoms and examine the contribution of long-term, past Hg exposure. METHODS: The symptom questionnaire, applied in the 2016/17 Grassy Narrows Community Health Assessment (GN-CHA) (Time 1), was re-administered in the 2021/22 Niibin study (Time 2). A total of 85 adults (median age: 47y; range: 29-75y) responded at both times. Paired statistics were used to test the differences (Time 2 - Time 1) in self-reported symptom frequencies. The symptom clustering algorithm, derived from the entire study group of the GN-CHA (n = 391), which had yielded 6 clusters, was applied at Time 1 and 2. Equivalent hair Hg measurements (HHg) between 1970 and 1997 were used in Longitudinal Mixed Effects Models (LMEM), with a sub-group with ≥ 10 repeated HHg mesurements (age > 40y), to examine its associations with symptom cluster scores and their progression. RESULTS: For most symptoms, paired analyses (Time 2 - Time 1) showed a significant increase in persons reporting " very often" or "all the time", and in the mean Likert scores for younger and older participants (< and ≥ 50y). The increase in cluster scores was not associated with age or sex, except for sensory impairment where a greater increase in symptom frequency was observed for younger persons. LMEM showed that, for the sub-group, long-term past Hg exposure was associated with most cluster scores at both times, and importantly, for all clusters, with their rate of increase over time (Time 2 - Time 1). CONCLUSIONS: The persistence of reported symptoms and their increase in frequency over the short 5-year period underline the need for adequate health care services. Results of the sub-group of persons > 40y, whose HHg reflects exposure over the 28-year sampling period, suggest that there may be a progressive impact of Hg on nervous system dysfunction.

Asunto(s)

RESUMEN

The relative contributions of genetic variation and experience in shaping the morphology of the adolescent brain are not fully understood. Using longitudinal data from 11,665 subjects in the ABCD Study, we fit vertex-wise variance components including family effects, genetic effects, and subject-level effects using a computationally efficient framework. Variance in cortical thickness and surface area is largely attributable to genetic influence, whereas sulcal depth is primarily explained by subject-level effects. Our results identify areas with heterogeneous distributions of heritability estimates that have not been seen in previous work using data from cortical regions. We discuss the biological importance of subject-specific variance and its implications for environmental influences on cortical development and maturation.

Asunto(s)

RESUMEN

Exposure to increased temperatures during early development can lead to phenotypic plasticity in morphology, physiology, and behavior across a range of ectothermic animals. In addition, maternal effects are known to be important contributors to phenotypic variation in offspring. Whether the 2 factors interact to shape offspring morphology and behavior is rarely explored. This is critical because climate change is expected to impact both incubation temperature and maternal stress and resource allocation. Using a fully factorial design, and Bayesian multivariate mixed models, we explored how the manipulation of early thermal environment and yolk-quantity in eggs affected the morphology, performance, and antipredator behavior of 2 sympatric Australian skink species (Lampropholis delicata and L. guichenoti). We found that juveniles from the hot treatment were larger than those on the cold treatment in L. guichenoti but not L. delicata. Using repeated behavioral measures for individual lizards, we found an interaction between incubation temperature and maternal investment in performance, with running speed being affected in a species-specific way by the treatment. We predicted that changes in performance should influence antipredator responses. In support of this prediction, we found that maternal investment impacted antipredator behavior, with animals from the yolk-reduced and cold treatment resuming activity faster after a simulated predatory attack in L. delicata. However, the prediction was not supported in L. guichenoti. Our results highlight the importance of exploring the multifaceted role that environments play across generations to understand how different anthropogenic factors will impact wildlife in the future.

RESUMEN

Background: Immune reconstitution following the initiation of combination antiretroviral therapy (cART) significantly impacts the prognosis of individuals infected with human immunodeficiency virus (HIV). Our previous studies have indicated that the baseline CD4+ T cells count and percentage before cART initiation are predictors of immune recovery in TB-negative children infected with HIV, with TB co-infection potentially causing a delay in immune recovery. However, it remains unclear whether these predictors consistently impact immune reconstitution during long-term intensive cART treatment in TB-negative/positive children infected with HIV. Results: We confirmed that the baseline CD4+ T cell count is a significant predictor of immune recovery following long-term intensive cART treatment among children aged 5 to 18 years. Children with lower CD4+ T cell count prior cART initiation did not show substantial immunological recovery during the follow-up period. Interestingly, children who were co-infected with TB and had higher baseline CD4+ T cell count eventually achieved good immunological recovery comparable to the TB-negative HIV-infected children. Hence, the baseline CD4+ T cell count at the onset of treatment serves as a reliable predictor of immunological reconstitution in HIV-infected children with or without TB co-infection. Taken together, this follow-up study validates our previous findings and further establishes that initiating cART early alongside early HIV testing can help prevent the diminished CD4+ T cell count associated with inadequate immunological reconstitution.

RESUMEN

Sleep serves vital physiological functions, yet how sleep in wild animals is influenced by environmental conditions is poorly understood. Here we use high-resolution biologgers to investigate sleep in wild animals over ecologically relevant time scales and quantify variability between individuals under changing conditions. We developed a robust classification for accelerometer data and measured multiple dimensions of sleep in the wild boar (Sus scrofa) over an annual cycle. In support of the hypothesis that environmental conditions determine thermoregulatory challenges, which regulate sleep, we show that sleep quantity, efficiency and quality are reduced on warmer days, sleep is less fragmented in longer and more humid days, while greater snow cover and rainfall promote sleep quality. Importantly, this longest and most detailed analysis of sleep in wild animals to date reveals large inter- and intra-individual variation. Specifically, short-sleepers sleep up to 46% less than long-sleepers but do not compensate for their short sleep through greater plasticity or quality, suggesting they may pay higher costs of sleep deprivation. Given the major role of sleep in health, our results suggest that global warming and the associated increase in extreme climatic events are likely to negatively impact sleep, and consequently health, in wildlife, particularly in nocturnal animals.

Asunto(s)

RESUMEN

The future performance of the widely abundant European beech (Fagus sylvatica L.) across its ecological amplitude is uncertain. Although beech is considered drought-sensitive and thus negatively affected by drought events, scientific evidence indicating increasing drought vulnerability under climate change on a cross-regional scale remains elusive. While evaluating changes in climate sensitivity of secondary growth offers a promising avenue, studies from productive, closed-canopy forests suffer from knowledge gaps, especially regarding the natural variability of climate sensitivity and how it relates to radial growth as an indicator of tree vitality. Since beech is sensitive to drought, we in this study use a drought index as a climate variable to account for the combined effects of temperature and water availability and explore how the drought sensitivity of secondary growth varies temporally in dependence on growth variability, growth trends, and climatic water availability across the species' ecological amplitude. Our results show that drought sensitivity is highly variable and non-stationary, though consistently higher at dry sites compared to moist sites. Increasing drought sensitivity can largely be explained by increasing climatic aridity, especially as it is exacerbated by climate change and trees' rank progression within forest communities, as (co-)dominant trees are more sensitive to extra-canopy climatic conditions than trees embedded in understories. However, during the driest periods of the 20th century, growth showed clear signs of being decoupled from climate. This may indicate fundamental changes in system behavior and be early-warning signals of decreasing drought tolerance. The multiple significant interaction terms in our model elucidate the complexity of European beech's drought sensitivity, which needs to be taken into consideration when assessing this species' response to climate change.

Asunto(s)

RESUMEN

Growth curve models are popular tools for studying the development of a response variable within subjects over time. Heterogeneity between subjects is common in such models, and researchers are typically interested in explaining or predicting this heterogeneity. We show how generalized linear mixed-effects model (GLMM) trees can be used to identify subgroups with different trajectories in linear growth curve models. Originally developed for clustered cross-sectional data, GLMM trees are extended here to longitudinal data. The resulting extended GLMM trees are directly applicable to growth curve models as an important special case. In simulated and real-world data, we assess performance of the extensions and compare against other partitioning methods for growth curve models. Extended GLMM trees perform more accurately than the original algorithm and LongCART, and similarly accurate compared to structural equation model (SEM) trees. In addition, GLMM trees allow for modeling both discrete and continuous time series, are less sensitive to (mis-)specification of the random-effects structure and are much faster to compute.

Asunto(s)

RESUMEN

Background: Racial and ethnic disparities in sleep quality and cognitive health are increasingly recognized, yet little is understood about their associations among Chinese older adults living in the United States. This study aims to examine the relationships between sleep parameters and cognitive functioning in this population, utilizing data from the Population Study of Chinese Elderly in Chicago (PINE). Methods: This observational study utilized a two-wave panel design as part of the PINE, including 2,228 participants aged 65 years or older, self-identified as Chinese, who completed interviews at two time points. Cognitive functioning was assessed using a battery of tests on perceptual speed, episodic memory, working memory, and mental status. Sleep quality was assessed using Pittsburgh sleep quality index (PSQI) with four aspects: subjective sleep quality, sleep latency, sleep efficiency, and sleep duration at night. Insomnia was assessed using four items from the Women's Health Initiative Insomnia Rating Scale. Mixed-effects regression models were estimated to assess the predictive effects of sleep parameters on baseline cognitive functioning and the rate of cognitive change over time. Results: Significant negative associations were observed between poor sleep quality and baseline cognitive functioning across various domains, although these initial negative associations diminished over time. More insomnia problems were related to poorer perceptual speed and episodic memory. Long sleep latency, or a long time to sleep onset, was associated with worse functioning across all domains except mental status. Sleep efficiency showed inconsistent associations with various cognitive domains, while sleep duration showed no significant relation to any domains. Conclusions: These findings suggest that poor sleep quality indicators serve as early markers of cognitive impairments. Hence, targeted interventions aimed at improving sleep quality could potentially enhance cognitive health outcomes.

RESUMEN

BACKGROUND: Longitudinal data in health informatics studies often present challenges due to sparse observations from each subject, limiting the application of contemporary deep learning for prediction. This issue is particularly relevant in predicting birthweight, a crucial factor in identifying conditions such as macrosomia and large-for-gestational age (LGA). Previous approaches have relied on empirical formulas for estimated fetal weights (EFWs) from ultrasound measurements and mixed-effects models for interim predictions. METHOD: The proposed novel supervised longitudinal learning procedure features a three-step approach. First, EFWs are generated using empirical formulas from ultrasound measurements. Second, nonlinear mixed-effects models are applied to create augmented sequences of EFWs, spanning daily gestational timepoints. This augmentation transforms sparse longitudinal data into a dense parallel sequence suitable for training recurrent neural networks (RNNs). A tailored RNN architecture is then devised to incorporate the augmented sequential EFWs along with non-sequential maternal characteristics. RESULTS: The RNNs are trained on augmented data to predict birthweights, which are further classified for macrosomia and LGA. Application of this supervised longitudinal learning procedure to the Successive Small-for-Gestational-Age Births study yields improved performance in classification metrics. Specifically, sensitivity, area under the receiver operation characteristic curve, and Youden's Index demonstrate enhanced results, indicating the effectiveness of the proposed approach in overcoming sparsity challenges in longitudinal health informatics data. CONCLUSIONS: The integration of mixed-effects models for temporal data augmentation and RNNs on augmented sequences shows effective in accurately predicting birthweights, particularly in the context of identifying excessive fetal growth conditions.

Asunto(s)

RESUMEN

The CoLab score was developed and externally validated to rule out COVID-19 among suspected patients presenting at the emergency department. We hypothesized a within-patient decrease in the CoLab score over time in an intensive care unit (ICU) cohort. Such a decrease would create the opportunity to potentially rule out the need for isolation when the infection is overcome. Using linear mixed-effects models, data from the Maastricht Intensive Care COVID (MaastrICCht) cohort were used to investigate the association between time and the CoLab score. Models were adjusted for sex, APACHE II score, ICU mortality, and daily SOFA score. The CoLab score decreased by 0.30 points per day (95% CI - 0.33 to - 0.27), independent of sex, APACHE II, and Mortality. With increasing SOFA score over time, the CoLab score decreased more strongly (- 0.01 (95% CI - 0.01 to - 0.01) additional decrease per one-point increase in SOFA score.) The CoLab score decreased in ICU patients on mechanical ventilation for COVID-19, with a one-point reduction per three days, independent of sex, APACHE II, and ICU mortality, and somewhat stronger with increasing multi-organ failure over time. This suggests that the CoLab score would decrease below a threshold where COVID-19 can be excluded.

Asunto(s)