Analysis of the yearly transition function in measles disease modeling

Globally, there were an estimated 9.8 million measles cases and 207 500 measles deaths in 2019. As the effort to eliminate measles around the world continues, modeling remains a valuable tool for public health decision-makers and program implementers. This study presents a novel approach to the use of a yearly transition function that formulates mathematically the vaccine schedules for different age groups while accounting for the effects of the age of vaccination, the timing of vaccination, and disease seasonality on the yearly number of measles cases in a country. The methodology presented adds to an existing modeling framework and expands its analysis, making its utilization more adjustable for the user and contributing to its conceptual clarity. This article also adjusts for the temporal interaction between vaccination and exposure to disease, applying adjustments to estimated yearly counts of cases and the number of vaccines administered that increase population immunity. These new model features provide the ability to forecast and compare the effects of different vaccination timing scenarios and seasonality of transmission on the expected disease incidence. Although the work presented is applied to the example of measles, it has potential relevance to modeling other vaccine-preventable diseases.

Active learning strategies for robotic tactile texture recognition tasks

Accurate texture classification empowers robots to improve their perception and comprehension of the environment, enabling informed decision-making and appropriate responses to diverse materials and surfaces. Still, there are challenges for texture classification regarding the vast amount of time series data generated from robots' sensors. For instance, robots are anticipated to leverage human feedback during interactions with the environment, particularly in cases of misclassification or uncertainty. With the diversity of objects and textures in daily activities, Active Learning (AL) can be employed to minimize the number of samples the robot needs to request from humans, streamlining the learning process. In the present work, we use AL to select the most informative samples for annotation, thus reducing the human labeling effort required to achieve high performance for classifying textures. We also use a sliding window strategy for extracting features from the sensor's time series used in our experiments. Our multi-class dataset (e.g., 12 textures) challenges traditional AL strategies since standard techniques cannot control the number of instances per class selected to be labeled. Therefore, we propose a novel class-balancing instance selection algorithm that we integrate with standard AL strategies. Moreover, we evaluate the effect of sliding windows of two-time intervals (3 and 6 s) on our AL Strategies. Finally, we analyze in our experiments the performance of AL strategies, with and without the balancing algorithm, regarding f1-score, and positive effects are observed in terms of performance when using our proposed data pipeline. Our results show that the training data can be reduced to 70% using an AL strategy regardless of the machine learning model and reach, and in many cases, surpass a baseline performance. Finally, exploring the textures with a 6-s window achieves the best performance, and using either Extra Trees produces an average f1-score of 90.21% in the texture classification data set.

Offering an auto-play feature likely increases total gambling activity at online slot-machines: preliminary evidence from an interrupted time series experiment at a real-life online casino

Auto-play is a ubiquitous feature in online casino gambling and virtual slot machines especially, allowing gamblers to initiate spin sequences of pre-set length and value. While theoretical accounts diverge on the hypothesized causal effect on gambling behavior of using the auto-play feature, observational findings show that this feature is used to a higher degree by problem and/or high-intensity gamblers, suggesting that banning this feature may constitute a global responsible gambling measure. Direct, experimental research on causal effects of offering auto-play at online casinos is however lacking. Here, we report the findings of an interrupted time series experiment, conducted at a real-life online casino in Sweden, in which the auto-play feature was made available during a pre-set duration on 40 online slot machines, with 40 matched slots serving as control. Aggregated time series on daily betted amount, spins and net losses were analyzed using a structural Bayesian framework that compared observed developments during the peri-intervention period to modeled counterfactual estimates. Results suggest that offering an auto-play feature on online casinos likely increases total gambling activity in terms of betted amount (approx.+ 7-9%) and (perhaps) number of spins (approx. +3%) but has no effect on net losses. Limitations of studying auto-play effects on a population-level, as well as the complexities of banning this feature within a complex ecosystem of non-perfect channelization to licensed providers, are discussed, including suggestions for future research.

AS, Hatlestad-Hall C, Yazidi A, Hammer HL, Hebold Haraldsen IRJ. Introducing Region Based Pooling for handling a varied number of EEG channels for deep learning models

Introduction

A challenge when applying an artificial intelligence (AI) deep learning (DL) approach to novel electroencephalography (EEG) data, is the DL architecture's lack of adaptability to changing numbers of EEG channels. That is, the number of channels cannot vary neither in the training data, nor upon deployment. Such highly specific hardware constraints put major limitations on the clinical usability and scalability of the DL models.

Methods

In this work, we propose a technique for handling such varied numbers of EEG channels by splitting the EEG montages into distinct regions and merge the channels within the same region to a region representation. The solution is termed Region Based Pooling (RBP). The procedure of splitting the montage into regions is performed repeatedly with different region configurations, to minimize potential loss of information. As RBP maps a varied number of EEG channels to a fixed number of region representations, both current and future DL architectures may apply RBP with ease. To demonstrate and evaluate the adequacy of RBP to handle a varied number of EEG channels, sex classification based solely on EEG was used as a test example. The DL models were trained on 129 channels, and tested on 32, 65, and 129-channels versions of the data using the same channel positions scheme. The baselines for comparison were zero-filling the missing channels and applying spherical spline interpolation. The performances were estimated using 5-fold cross validation.

Results

For the 32-channel system version, the mean AUC values across the folds were: RBP (93.34%), spherical spline interpolation (93.36%), and zero-filling (76.82%). Similarly, on the 65-channel system version, the performances were: RBP (93.66%), spherical spline interpolation (93.50%), and zero-filling (85.58%). Finally, the 129-channel system version produced the following results: RBP (94.68%), spherical spline interpolation (93.86%), and zero-filling (91.92%).

Conclusion

In conclusion, RBP obtained similar results to spherical spline interpolation, and superior results to zero-filling. We encourage further research and development of DL models in the cross-dataset setting, including the use of methods such as RBP and spherical spline interpolation to handle a varied number of EEG channels.

Cancer mortality in childhood and adolescence: analysis of trends and spatial distribution in the 133 intermediate Brazilian regions grouped by macroregions

OBJECTIVE

To assess the magnitude, trend, and spatial patterns of childhood and adolescent cancer mortality between 1996 and 2017 in 133 Brazilian intermediate regions by using socioeconomic and healthcare services indicators.

METHODS

This is an ecological study for analyzing the trend of mortality from cancer in childhood and adolescence through time series. Data on deaths were extracted from the Brazilian Mortality Information System. Data on population were extracted from the 1991, 2000, and 2010 demographic censuses of the Brazilian Institute of Geography and Statistics, with interpolation for intercensal years. Time series were delineated for mortality by type of cancer in each intermediate region. Such regions were grouped by macroregions to present the results. The calculation and interpretation of mortality trends use the Prais-Winsten autoregression procedure.

RESULTS

Mortality rates for all neoplasms were higher in the Northern region (7.79 deaths per 100 thousand population), while for leukemias, they were higher in the Southern region (1.61 deaths per 100 thousand population). In both regions, mortality was higher in boys and in the 0-4 age group. The trend was decreasing (annual percent change [APC] - -2.11 [95%CI: -3.14; - 1.30]) for all neoplasms in the Brazilian regions and stationary (APC - -0.43 [95%CI: -1.61; 2.12]) for leukemias in the analyzed period.

CONCLUSION

The mortality rate for all neoplasms showed higher values in regions with smaller numbers of ICU beds in the public healthcare system.

Depression and Anxiety Symptoms Among Cisgender Gay and Bisexual Men During the Onset of the COVID-19 Pandemic: Time Series Analysis of a US National Cohort Study

BACKGROUND

The onset of the COVID-19 pandemic in the United States in March 2020 caused a dramatic change in the way many people lived. Few aspects of daily life were left undisrupted by the pandemic's onset as well as the accompanying policies to control the spread of the disease. Previous research has found that the pandemic may have significantly impacted the mental health of lesbian, gay, bisexual, transgender, and queer (LGBTQ) individuals-potentially more so than other individuals. However, the pandemic did not affect all areas of the United States at the same time, and there may be regional variation in the impact of the onset of the pandemic on depressive symptoms among LGBTQ individuals.

OBJECTIVE

To assess regional variation of the impact of the pandemic, we conducted a time series analysis stratified by US geographic region to examine symptoms of depression and anxiety among a sample of primarily cisgender gay and bisexual men before and after the onset of the COVID-19 pandemic in the United States.

METHODS

In total, 5007 participants completed assessments as part of the Together 5000 study, an ongoing prospective cohort study. Depressive and anxiety symptoms were measured using the Patient Health Questionnaire-4. Patient Health Questionnaire-4 scores were graphed as a function of days from March 15, 2020. Locally estimated scatterplot smoothing trend lines were applied. A sieve-bootstrap Mann-Kendall test for monotonic trend was conducted to assess the presence and direction of trends in the scatterplots. We then compared the observed trends to those observed for 1 year prior (2018-2019) to the pandemic onset using data collected from the same sample.

RESULTS

Significant positive trends were detected for the Northeast (P=.03) and Midwest (P=.01) regions of the United States in the 2020 assessment, indicating that symptoms of anxiety and depression were increasing in the sample in these regions immediately prior to and during the onset of the pandemic. In contrast, these trends were not present in data from the 2018 to 2019 assessment window.

CONCLUSIONS

Symptoms of anxiety and depression increased among the study population in the Northeast and Midwest during the beginning months of the COVID-19 pandemic, but similar increase was not observed in the South and West regions. These trends were also not found for any region in the 2018 to 2019 assessment window. This may indicate region-specific trends in anxiety and depression, potentially driven by the burden of the pandemic and policies that varied from region to region. Future studies should consider geographic variation in COVID-19 spread and policies as well as explore potential mechanisms by which this could influence the mental health of LGBTQ individuals.

Assessing Internet Search Models in Predicting Daily New COVID-19 Cases and Deaths in South Korea

Search data were found to be useful variables for COVID-19 trend prediction. In this study, we aimed to investigate the performance of online search models in state space models (SSMs), linear regression (LR) models, and generalized linear models (GLMs) for South Korean data from January 20, 2020, to July 31, 2021. Principal component analysis (PCA) was run to construct the composite features which were later used in model development. Values of root mean squared error (RMSE), peak day error (PDE), and peak magnitude error (PME) were defined as loss functions. Results showed that integrating search data in the models for short- and long-term prediction resulted in a low level of RMSE values, particularly for SSMs. Findings indicated that type of model used highly impacts the performance of prediction and interpretability of the model. Furthermore, PDE and PME could be beneficial to be included in the evaluation of peaks.

Patterns of Animal Rabies Prevalence in Northern South Africa between 1998 and 2022

Rabies is endemic in South Africa and rabies cycles are maintained in both domestic and wildlife species. The significant number of canine rabies cases reported by the World Organization for Animal Health Reference Laboratory for Rabies at Onderstepoort suggests the need for increased research and mass dog vaccinations on specific targeted foci in the country. This study aimed to investigate the spatiotemporal distribution of animal rabies cases from 1998 to 2017 in northern South Africa and environmental factors associated with highly enzootic municipalities. A descriptive analysis was used to investigate temporal patterns. The Getis-Ord Gi statistical tool was used to exhibit low and high clusters. Logistic regression was used to examine the association between the predictor variables and highly enzootic municipalities. A total of 9580 specimens were submitted for rabies diagnosis between 1998 and 2022. The highest positive case rates were from companion animals (1733 cases, 59.71%), followed by livestock (635 cases, 21.88%) and wildlife (621 cases, 21.39%). Rabies cases were reported throughout the year, with the majority occurring in the mid-dry season. Hot spots were frequently in the northern and eastern parts of Limpopo and Mpumalanga. Thicket bush and grassland were associated with rabies between 1998 and 2002. However, between 2008 and 2012, cultivated commercial crops and waterbodies were associated with rabies occurrence. In the last period, plantations and woodlands were associated with animal rabies. Of the total number of municipalities, five consistently and repeatedly had the highest rabies prevalence rates. These findings suggest that authorities should prioritize resources for those municipalities for rabies elimination and management.

The contribution of gene flow, selection, and genetic drift to five thousand years of human allele frequency change

Genomic time series from experimental evolution studies and ancient DNA datasets offer us a chance to directly observe the interplay of various evolutionary forces. We show how the genome-wide variance in allele frequency change between two time points can be decomposed into the contributions of gene flow, genetic drift, and linked selection. In closed populations, the contribution of linked selection is identifiable because it creates covariances between time intervals, and genetic drift does not. However, repeated gene flow between populations can also produce directionality in allele frequency change, creating covariances. We show how to accurately separate the fraction of variance in allele frequency change due to admixture and linked selection in a population receiving gene flow. We use two human ancient DNA datasets, spanning around 5,000 years, as time transects to quantify the contributions to the genome-wide variance in allele frequency change. We find that a large fraction of genome-wide change is due to gene flow. In both cases, after correcting for known major gene flow events, we do not observe a signal of genome-wide linked selection. Thus despite the known role of selection in shaping long-term polymorphism levels, and an increasing number of examples of strong selection on single loci and polygenic scores from ancient DNA, it appears to be gene flow and drift, and not selection, that are the main determinants of recent genome-wide allele frequency change. Our approach should be applicable to the growing number of contemporary and ancient temporal population genomics datasets.

Quantifying changes in individual-specific template-based representations of center-of-mass dynamics during walking with ankle exoskeletons using Hybrid-SINDy

Ankle exoskeletons alter whole-body walking mechanics, energetics, and stability by altering center-of-mass (CoM) motion. Controlling the dynamics governing CoM motion is, therefore, critical for maintaining efficient and stable gait. However, how CoM dynamics change with ankle exoskeletons is unknown, and how to optimally model individual-specific CoM dynamics, especially in individuals with neurological injuries, remains a challenge. Here, we evaluated individual-specific changes in CoM dynamics in unimpaired adults and one individual with post-stroke hemiparesis while walking in shoes-only and with zero-stiffness and high-stiffness passive ankle exoskeletons. To identify optimal sets of physically interpretable mechanisms describing CoM dynamics, termed template signatures, we leveraged hybrid sparse identification of nonlinear dynamics (Hybrid-SINDy), an equation-free data-driven method for inferring sparse hybrid dynamics from a library of candidate functional forms. In unimpaired adults, Hybrid-SINDy automatically identified spring-loaded inverted pendulum-like template signatures, which did not change with exoskeletons (p > 0.16), except for small changes in leg resting length (p < 0.001). Conversely, post-stroke paretic-leg rotary stiffness mechanisms increased by 37-50% with zero-stiffness exoskeletons. While unimpaired CoM dynamics appear robust to passive ankle exoskeletons, how neurological injuries alter exoskeleton impacts on CoM dynamics merits further investigation. Our findings support Hybrid-SINDy's potential to discover mechanisms describing individual-specific CoM dynamics with assistive devices.

A Time-Varying Dynamic Partial Credit Model to Analyze Polytomous and Multivariate Time Series Data

The accessibility to electronic devices and the novel statistical methodologies available have allowed researchers to comprehend psychological processes at the individual level. However, there are still great challenges to overcome as, in many cases, collected data are more complex than the available models are able to handle. For example, most methods assume that the variables in the time series are measured on an interval scale, which is not the case when Likert-scale items were used. Ignoring the scale of the variables can be problematic and bias the results. Additionally, most methods also assume that the time series are stationary, which is rarely the case. To tackle these disadvantages, we propose a model that combines the partial credit model (PCM) of the item response theory framework and the time-varying autoregressive model (TV-AR), which is a popular model used to study psychological dynamics. The proposed model is referred to as the time-varying dynamic partial credit model (TV-DPCM), which allows to appropriately analyze multivariate polytomous data and nonstationary time series. We test the performance and accuracy of the TV-DPCM in a simulation study. Lastly, by means of an example, we show how to fit the model to empirical data and interpret the results.

Spatial and temporal analysis of juvenile blacktip reef shark (Carcharhinus melanopterus) demographies identifies critical habitats

Reef shark species have undergone sharp declines in recent decades, as they inhabit coastal areas, making them an easy target in fisheries (i.e., sharks are exploited globally for their fins, meat, and liver oil) and exposing them to other threats (e.g., being part of by-catch, pollution, and climate change). Reef sharks play a critical role in coral reef ecosystems, where they control populations of smaller predators and herbivorous fishes either directly via predation or indirectly via behavior, thus protecting biodiversity and preventing potential overgrazing of corals. The urgent need to conserve reef shark populations necessitates a multifaceted approach to policy at local, federal, and global levels. However, monitoring programmes to evaluate the efficiency of such policies are lacking due to the difficulty in repeatedly sampling free-ranging, wild shark populations. Over nine consecutive years, we monitored juveniles of the blacktip reef shark (Carcharhinus melanopterus) population around Moorea, French Polynesia, and within the largest shark sanctuary globally, to date. We investigated the roles of spatial (i.e., sampling sites) and temporal variables (i.e., sampling year, season, and month), water temperature, and interspecific competition on shark density across 10 coastal nursery areas. Juvenile C. melanopterus density was found to be stable over 9 years, which may highlight the effectiveness of local and likely federal policies. Two of the 10 nursery areas exhibited higher juvenile shark densities over time, which may have been related to changes in female reproductive behavior or changes in habitat type and resources. Water temperatures did not affect juvenile shark density over time as extreme temperatures proven lethal (i.e., 33°C) in juvenile C. melanopterus might have been tempered by daily variation. The proven efficiency of time-series datasets for reef sharks to identify critical habitats (having the highest juvenile shark densities over time) should be extended to other populations to significantly contribute to the conservation of reef shark species.

Application of Time-Series Analysis and Expert Judgment in Modeling and Forecasting Blood Donation Trends in Zimbabwe

Background. Blood cannot be artificially manufactured, and there is currently no substitute for human blood. The supply of blood in transfusion facilities requires constant and timely collection of blood from donors. Modeling and forecasting trends in blood collections are critical for determining both the current and future capacity requirements and appropriate models of adequate blood provision. Objectives. The objective of this study is to determine blood collection or donation patterns and develop time-series models that can be updated and refined in predicting future blood donations in Zimbabwe when given the historical data. Materials and Methods. Monthly blood donation data for the period 2009 to 2019 were collected retrospectively from the National Blood Service Zimbabwe database. Time-series models (i.e., the Seasonal Autoregressive Integrated Moving Average [SARIMA] and Error, Trend and Seasonal [ETS]) models were applied and compared. The models were chosen because of their ability to handle the seasonality and other time-series components evident in the blood donation data. Expert opinions and experience were used in selecting the models and in making inferences in the analysis. Results. Time-series plots of blood donations showed seasonal patterns, with significant drops in blood donations in months associated with Zimbabwe's school holidays (April, August, and December) and public holidays. During these holidays, there is a reduced number of school donors, while at about the same time, there is increasing blood demand as a result of road accidents. Model identification procedures established the SARIMA ( 1 , 1 , 2 ) ( 0 , 1 , 1 ) 12 model as the appropriate model for forecasting total blood donation in Zimbabwe. The results and forecasts show an upward trend in blood donations. According to the accuracy measures used, the SARIMA model outperforms the ETS model. Conclusions. Expert knowledge in the blood donation process, coupled with statistical models, can help explain trends exhibited in blood donation data in Zimbabwe. These findings help the blood authorities plan for blood donor campaign drives. The findings are key indicators of where to allocate more resources toward blood donation and when to collect more blood units. The increasing blood donation projections ensure a stable blood bank inventory in the near future.

Highlights

A SARIMA model can be used to predict the flow of blood donations in Zimbabwe.The seasonal blood donation pattern peaks in the months of March, June/July, and September.The donations troughs are in the months of April, August, December, and January. These are the months coinciding with school holidays in Zimbabwe.Both the SARIMA and ETS models provided similar forecasts, but measures of fit and expert knowledge gave a slight preference to the SARIMA ( 1 , 1 , 2 ) ( 0 , 1 , 1 ) 12 model in predicting the flow of blood donations in Zimbabwe.These model results are useful for guiding allocation of blood donation resources and blood donation drive timing.

In-Advance Prediction of Pressure Ulcers via Deep-Learning-Based Robust Missing Value Imputation on Real-Time Intensive Care Variables

Pressure ulcers (PUs) are a prevalent skin disease affecting patients with impaired mobility and in high-risk groups. These ulcers increase patients' suffering, medical expenses, and burden on medical staff. This study introduces a clinical decision support system and verifies it for predicting real-time PU occurrences within the intensive care unit (ICU) by using MIMIC-IV and in-house ICU data. We develop various machine learning (ML) and deep learning (DL) models for predicting PU occurrences in real time using the MIMIC-IV and validate using the MIMIC-IV and Kangwon National University Hospital (KNUH) dataset. To address the challenge of missing values in time series, we propose a novel recurrent neural network model, GRU-D++. This model outperformed other experimental models by achieving the area under the receiver operating characteristic curve (AUROC) of 0.945 for the on-time prediction and AUROC of 0.912 for 48h in-advance prediction. Furthermore, in the external validation with the KNUH dataset, the fine-tuned GRU-D++ model demonstrated superior performances, achieving an AUROC of 0.898 for on-time prediction and an AUROC of 0.897 for 48h in-advance prediction. The proposed GRU-D++, designed to consider temporal information and missing values, stands out for its predictive accuracy. Our findings suggest that this model can significantly alleviate the workload of medical staff and prevent the worsening of patient conditions by enabling timely interventions for PUs in the ICU.

A novel approach for sports injury risk prediction: based on time-series image encoding and deep learning

The rapid development of big data technology and artificial intelligence has provided a new perspective on sports injury prevention. Although data-driven algorithms have achieved some valuable results in the field of sports injury risk assessment, the lack of sufficient generalization of models and the inability to automate feature extraction have made it challenging to deploy research results in the real world. Therefore, this study attempts to build an injury risk prediction model using a combination of time-series image encoding and deep learning algorithms to address this issue better. This study used the time-series image encoding approach for feature construction to represent relationships between values at different moments, including Gramian Angular Summation Field (GASF), Gramian Angular Difference Field (GADF), Markov Transition Field (MTF), and Recurrence Plot (RP). Deep Convolutional Auto-Encoder (DCAE) learned the image-encoded data for representation to obtain features with good discrimination, and the classifier was performed using Deep Neural Network (DNN). The results from five repeated experiments show that the GASF-DCAE-DNN model is overall better in the training (AUC: 0.985 ± 0.001, Gmean: 0.930 ± 0.007, Sensitivity: 0.997 ± 0.003, Specificity: 0.868 ± 0.013) and test sets (AUC: 0.891 ± 0.026, Gmean: 0.830 ± 0.027, Sensitivity: 0.816 ± 0.039, Specificity: 0.845 ± 0.022), with good discriminative power, robustness, and generalization ability. Compared with the best model reported in the literature, the AUC, Gmean, Sensitivity, and Specificity of the GASF-DCAE-DNN model were higher by 23.9%, 27.5%, 39.7%, and 16.2%, respectively, which confirmed the validity and practicability of the model in injury risk prediction. In addition, differences in injury risk patterns between the training and test sets were identified through shapley additivity interpretation. It was also found that the training volume was an essential factor that affected injury risk prediction. The model proposed in this study provides a powerful injury risk prediction tool for future sports injury prevention practice.

Fisher-Shannon Investigation of the Effect of Nonlinearity of Discrete Langevin Model on Behavior of Extremes in Generated Time Series

Diverse forms of nonlinearity within stochastic equations give rise to varying dynamics in processes, which may influence the behavior of extreme values. This study focuses on two nonlinear models of the discrete Langevin equation: one with a fixed diffusion function (M1) and the other with a fixed marginal distribution (M2), both characterized by a nonlinearity parameter. Extremes are defined according to the run theory with thresholds based on percentiles. The behavior of inter-extreme times and run lengths is examined by employing Fisher's Information Measure and the Shannon Entropy. Our findings reveal a clear relationship between the entropic and informational measures and the nonlinearity of model M1-these measures decrease as the nonlinearity parameter increases. Similar relationships are evident for the M2 model, albeit to a lesser extent, even though the background data's marginal distribution remains unaffected by this parameter. As thresholds increase, both the values of Fisher's Information Measure and the Shannon Entropy also increase.

Impact of the Spanish smoke-free laws on cigarette sales by brands, 2000-2021: Evidence from a club convergence approach

INTRODUCTION

In January 2006, the Spanish government enacted a tobacco control law that banned the advertising, promotion and sponsorship of tobacco. In January 2011, further legislation on this matter was adopted to provide a more restrictive specification of the ban. In this study, we analyze the effect produced on cigarette sales by these two prohibitions. We address this problem using a cluster time-series analysis to test whether the sales of cigarettes by brands have been homogenized with the prohibition of advertising, promotion, and sponsorship.

METHODS

The data source used was the official data on legal sales of cigarettes by brands in Spain, from January 2005 to December 2021 (excluding the Canary Islands and the Autonomous Communities of the cities of Ceuta and Melilla). To achieve our objective, we used log(t) test statistics to check if there is global convergence in the three selected periods according to the regulatory changes that have occurred in Spain (2005-2021, 2005-2010 and 2011-2021). Second, once absolute convergence is rejected, we applied a clustering algorithm to test for the existence of subgroup convergence.

RESULTS

The cigarette brands that have been marketed during the period 2005-2021 (n=40), can only be grouped into three groups according to the behavior of their sales. When we focus on the period 2005-2010 (n=74), cigarette brands are grouped into five groups according to their sales behavior. Finally, the cigarette brands marketed during the period 2011-2021 (n=67) are grouped into three groups according to the temporal evolution of their sales. These results suggest a greater homogenization of cigarette sales after the application of the law of January 2011.

CONCLUSIONS

Act 42/2010 (total ban on tobacco advertising, promotion, and sponsorship actions) was associated with greater homogenization of cigarette sales than the application of Act 28/2005 (partial ban). This finding supports what is established in the previous literature that indicates that Act 42/2010 provided a more restrictive specification of the ban than Act 28/2005.

Forecasting local hospital bed demand for COVID-19 using on-request simulations

Accurate forecasting of hospital bed demand is crucial during infectious disease epidemics to avoid overwhelming healthcare facilities. To address this, we developed an intuitive online tool for individual hospitals to forecast COVID-19 bed demand. The tool utilizes local data, including incidence, vaccination, and bed occupancy data, at customizable geographical resolutions. Users can specify their hospital's catchment area and adjust the initial number of COVID-19 occupied beds. We assessed the model's performance by forecasting ICU bed occupancy for several university hospitals and regions in Germany. The model achieves optimal results when the selected catchment area aligns with the hospital's local catchment. While expanding the catchment area reduces accuracy, it improves precision. However, forecasting performance diminishes during epidemic turning points. Incorporating variants of concern slightly decreases precision around turning points but does not significantly impact overall bed occupancy results. Our study highlights the significance of using local data for epidemic forecasts. Forecasts based on the hospital's specific catchment area outperform those relying on national or state-level data, striking a better balance between accuracy and precision. These hospital-specific bed demand forecasts offer valuable insights for hospital planning, such as adjusting elective surgeries to create additional bed capacity promptly.

Response trait diversity and species asynchrony underlie the diversity-stability relationship in Romanian bird communities

Biodiversity-stability relationships have frequently been studied in ecology, with the recent integration of traits to explain community stability over time. Classical theory underlying the biodiversity-stability relationship posits that different species' responses to the environment should stabilise community-level properties (e.g. biomass or abundance) through compensatory dynamics. However, functional response traits, which aim to predict how species respond to environmental change, are still rarely integrated into studies of ecological stability. Such traits should mechanistically drive community stability, both in terms of community abundance (functional variability) and composition (compositional variability). In turn, whether and how functional or compositional stability scales to affect temporal variation in functional effect traits (a proxy for ecosystem functioning) remains largely unknown, but is key to consistent ecosystem functioning under environmental change. Here, we explore the diversity-stability relationship in bird communities using annual survey data across 98 sites in central Romania, in combination with global trait databases and structural equation models. We show that higher response trait diversity promotes compositional variability directly, and functional variability indirectly via species asynchrony. In turn, functional variability impacts the temporal stability of effect trait diversity. Multiple facets of diversity and community stability differ between natural forests and agricultural or human-dominated survey sites, and the relationship between response diversity and functional variability is mediated by land cover. Further integration of response-and-effect trait frameworks into studies of community stability will enhance understanding of the drivers of biodiversity change, allowing targeted conservation decision-making with a focus on stable ecosystem functioning in the face of global environmental change.

Trends in mental health inequalities for people with disability, Australia 2003 to 2020

OBJECTIVE

Cross-sectional studies have demonstrated that people with disability have substantial inequalities in mental health compared to people without disability. However, it is not known if these inequalities have changed over time. This study compared the mental health of people with and without disability annually from 2003 to 2020 to investigate time trends in disability-related mental health inequalities.

METHODS

We use annual data (2003-2020) of the Household, Income and Labour Dynamics in Australia Survey. Mental health was measured using the five-item Mental Health Inventory. For each wave, we calculated population-weighted age-standardised estimates of mean Mental Health Inventory scores for people with and without disability and calculated the mean difference in Mental Health Inventory score to determine inequalities. Analyses were stratified by age, sex and disability group (sensory or speech, physical, intellectual or learning, psychological, brain injury or stroke, other).

RESULTS

From 2003 to 2020, people with disability had worse mental health than people without disability, with average Mental Health Inventory scores 9.8 to 12.1 points lower than for people without disability. For both people with and without disability, Mental Health Inventory scores decreased, indicating worsening mental health, reaching the lowest point for both groups in 2020. For some subpopulations, including young females and people with intellectual disability, brain injury or stroke, mental health inequalities worsened.

CONCLUSION

This paper confirms that people with disability experience worse mental health than people without disability. We add to previous findings by demonstrating that disability-related inequalities in mental health have been sustained for a long period and are worsening in some subpopulations.

Short-term effects of air pollution exposure on the risk of preterm birth in Xi'an, China

INTRODUCTION

Long-term exposure to air pollution is known to be harmful to preterm birth (PTB), but little is known about the short-term effects. This study aims to quantify the short-term effect of particulate matter with aerodynamic diameter ≤ 2.5 μm (PM_2.5), ≤10 μm (PM₁₀) and nitrogen dioxide (NO₂) on PTB.

MATERIALS AND METHODS

A total of 18,826 singleton PTBs were collected during the study period. Poisson regression model combined with the distributed lag non-linear model was applied to evaluate the short-term effects of PTBs and air pollutants.

RESULTS

Maternal exposure to NO₂ was significantly associated increased risk of PTB at Lag1 (RR: 1.025, 95%CI: 1.003-1.047). In the moving average model, maternal exposure to NO₂ significantly increased the risk of PTB at Lag01 (RR: 1.029, 95%CI: 1.004-1.054). In the cumulative model, maternal exposure to NO₂ significant increased the risk of PTB at Cum01 (RR:1.026, 95%CI: 1.002-1.051), Cum02 (RR: 1.030, 95%CI: 1.003-1.059), and Cum03 (RR: 1.033, 95%CI: 1.002-1.066). The effects of PM_2.5, PM₁₀ and NO₂ on PTB were significant and greater in the cold season than the warm season.

CONCLUSIONS

Maternal exposure to NO₂, PM_2.5 and PM₁₀ before delivery has a significant risk for PTB, particularly in the cold season.Key messagesMaternal exposure to NO₂ was significant associated with an increased risk of preterm birth at the day 1 before delivery.Particle matter (PM_2.5 and PM₁₀) showed a significant short-term effect on preterm birth in the cold season.The effects of air pollutants on preterm birth was greater in the cold season compared with the warm season.

CALDA: Improving Multi-Source Time Series Domain Adaptation With Contrastive Adversarial Learning

Unsupervised domain adaptation (UDA) provides a strategy for improving machine learning performance in data-rich (target) domains where ground truth labels are inaccessible but can be found in related (source) domains. In cases where meta-domain information such as label distributions is available, weak supervision can further boost performance. We propose a novel framework, CALDA, to tackle these two problems. CALDA synergistically combines the principles of contrastive learning and adversarial learning to robustly support multi-source UDA (MS-UDA) for time series data. Similar to prior methods, CALDA utilizes adversarial learning to align source and target feature representations. Unlike prior approaches, CALDA additionally leverages cross-source label information across domains. CALDA pulls examples with the same label close to each other, while pushing apart examples with different labels, reshaping the space through contrastive learning. Unlike prior contrastive adaptation methods, CALDA requires neither data augmentation nor pseudo labeling, which may be more challenging for time series. We empirically validate our proposed approach. Based on results from human activity recognition, electromyography, and synthetic datasets, we find utilizing cross-source information improves performance over prior time series and contrastive methods. Weak supervision further improves performance, even in the presence of noise, allowing CALDA to offer generalizable strategies for MS-UDA.

Sample sizes for estimating the sensitivity of a monitoring system that generates repeated binary outcomes with autocorrelation

Sample size formulas are provided to determine how many events and how many patient care units are needed to estimate the sensitivity of a monitoring system. The monitoring systems we consider generate time series binary data that are autocorrelated and clustered by patient care units. Our application of interest is an automated hand hygiene monitoring system that assesses whether healthcare workers perform hand hygiene when they should. We apply an autoregressive order 1 mixed effects logistic regression model to determine sample sizes that allow the sensitivity of the monitoring system to be estimated at a specified confidence level and margin of error. This model overcomes a major limitation of simpler approaches that fail to provide confidence intervals with the specified levels of confidence when the sensitivity of the monitoring system is above 90%.

Physical Activity Pattern of Adults With Metabolic Syndrome Risk Factors: Time-Series Cluster Analysis

Background

Physical activity plays a crucial role in maintaining a healthy lifestyle, and wrist-worn wearables, such as smartwatches and smart bands, have become popular tools for measuring activity levels in daily life. However, studies on physical activity using wearable devices have limitations; for example, these studies often rely on a single device model or use improper clustering methods to analyze the wearable data that are extracted from wearable devices.

Objective

This study aimed to identify methods suitable for analyzing wearable data and determining daily physical activity patterns. This study also explored the association between these physical activity patterns and health risk factors.

Methods

People aged >30 years who had metabolic syndrome risk factors and were using their own wrist-worn devices were included in this study. We collected personal health data through a web-based survey and measured physical activity levels using wrist-worn wearables over the course of 1 week. The Time-Series Anytime Density Peak (TADPole) clustering method, which is a novel time-series method proposed recently, was used to identify the physical activity patterns of study participants. Additionally, we defined physical activity pattern groups based on the similarity of physical activity patterns between weekdays and weekends. We used the χ2 or Fisher exact test for categorical variables and the 2-tailed t test for numerical variables to find significant differences between physical activity pattern groups. Logistic regression models were used to analyze the relationship between activity patterns and health risk factors.

Results

A total of 47 participants were included in the analysis, generating a total of 329 person-days of data. We identified 2 different types of physical activity patterns (early bird pattern and night owl pattern) for weekdays and weekends. The physical activity levels of early birds were less than that of night owls on both weekdays and weekends. Additionally, participants were categorized into stable and shifting groups based on the similarity of physical activity patterns between weekdays and weekends. The physical activity pattern groups showed significant differences depending on age (P=.004) and daily energy expenditure (P<.001 for weekdays; P=.003 for weekends). Logistic regression analysis revealed a significant association between older age (≥40 y) and shifting physical activity patterns (odds ratio 8.68, 95% CI 1.95-48.85; P=.007).

Conclusions

This study overcomes the limitations of previous studies by using various models of wrist-worn wearables and a novel time-series clustering method. Our findings suggested that age significantly influenced physical activity patterns. It also suggests a potential role of the TADPole clustering method in the analysis of large and multidimensional data, such as wearable data.