Data Imputation Research Articles

Optimizing Multi-Omics Data Imputation with NMF and GAN Synergy.

Integrating multiple omics datasets can significantly advance our understanding of disease mechanisms, physiology, and treatment responses. However, a major challenge in multi-omics studies is the disparity in sample sizes across different datasets, which can introduce bias and reduce statistical power. To address this issue, we propose a novel framework, OmicsNMF, designed to impute missing omics data and enhance disease phenotype prediction. OmicsNMF integrates Generative Adversarial Networks (GANs) with Non-Negative Matrix Factorization (NMF). NMF is a well-established method for uncovering underlying patterns in omics data, while GANs enhance the imputation process by generating realistic data samples. This synergy aims to more effectively address sample size disparity, thereby improving data integration and prediction accuracy. For evaluation, we focused on predicting breast cancer subtypes using the imputed data generated by our proposed framework, OmicsNMF. Our results indicate that OmicsNMF consistently outperforms baseline methods. We further assessed the quality of the imputed data through survival analysis, revealing that the imputed omics profiles provide significant prognostic power for both overall survival and disease-free status. Overall, OmicsNMF effectively leverages GANs and NMF to impute missing samples while preserving key biological features. This approach shows potential for advancing precision oncology by improving data integration and analysis. Source code is available at: https://github.com/compbiolabucf/OmicsNMF.

Data discretization impact on deep learning for missing value imputation of continuous data

In various fields of information examination, for example, AI, profoundly getting the hang of missing information is a typical issue. Missing qualities should be tended to since they can adversely affect the exactness and adequacy of prescient models. This research investigates how data discretization affects deep learning methods for filling the missing values in datasets with continuous features. They provide a unique method for imputing missing values using deep neural networks (DNNs) called extravagant expectation maximization-deep neural network (EEM-DNN). This approach discretizes continuous features into separate intervals initially. This is justified by treating the issue of missing value imputation as a classification work, with the missing values being considered a distinct class. A DNN, designed explicitly for imputation, is then trained using the discretized data. The expectation maximization concepts are incorporated into the network architecture, and as a result, the network iteratively improves its imputation predictions. They run comprehensive experiments on several datasets from different fields to gauge the efficacy of the suggested strategy. The effectiveness of EEM-DNN is compared to that of other imputation approaches, such as traditional imputation techniques and deep learning methods without data discretization. Our findings show that data discretization significantly enhances imputation accuracy. In terms of imputation accuracy and prediction performance on downstream tasks, the EEM-DNN method regularly performs better than alternative methods. It also examines if various discretization techniques affect the overall imputation process. They find that the trade-off between bias and variance in imputed data depends on the discretization method selected. This highlights the significance of choosing a suitable discretization approach depending on the unique properties of the dataset.

Impact of sociodemographic disparities on sarcopenia, telomere length, and mortality in patients with liver disease in the US population.

Sarcopenia is common in patients with liver disease and both sarcopenia and short telomeres are associated with mortality, however their relationship in patients with liver disease remains unknown. A cohort of 16,072 adults from the National Health and Nutrition Examination Survey from 1999 to 2006 was analyzed. Liver disease was defined by aminotransferases and classified into etiology-based categories. Sarcopenia was defined by dual-energy x-ray absorptiometry. All analyses were conducted separately on each multiple imputation data set and combined via Rubin's rules. P-values for group comparisons were calculated by testing logistic regression parameter estimates. Cox proportional hazards regression was used for mortality analysis with mortality data available until 2015. Sarcopenia was present in 9.5% of patients with liver disease. Age, race, income, education, physical inactivity, and certain medical comorbidities were associated with sarcopenia. Patients with liver disease and sarcopenia had significantly shorter telomeres than patients with liver disease without sarcopenia when unadjusted for age. The interaction between telomere length and sarcopenia was significantly associated with all-cause mortality. The implications of telomere length on all-cause mortality in patients with liver disease varied by age and sarcopenia status. Shorter telomeres appear to be more highly associated with increased mortality in older patients without sarcopenia.

Multiple Data Imputation Methods Advance Risk Analysis and Treatability of Co-occurring Inorganic Chemicals in Groundwater.

Accurately assessing and managing risks associated with inorganic pollutants in groundwater is imperative. Historic water quality databases are often sparse due to rationale or financial budgets for sample collection and analysis, posing challenges in evaluating exposure or water treatment effectiveness. We utilized and compared two advanced multiple data imputation techniques, AMELIA and MICE algorithms, to fill gaps in sparse groundwater quality data sets. AMELIA outperformed MICE in handling missing values, as MICE tended to overestimate certain values, resulting in more outliers. Field data sets revealed that 75% to 80% of samples exhibited no co-occurring regulated pollutants surpassing MCL values, whereas imputed values showed only 15% to 55% of the samples posed no health risks. Imputed data unveiled a significant increase, ranging from 2 to 5 times, in the number of sampling locations predicted to potentially exceed health-based limits and identified samples where 2 to 6 co-occurring chemicals may occur and surpass health-based levels. Linking imputed data to sampling locations can pinpoint potential hotspots of elevated chemical levels and guide optimal resource allocation for additional field sampling and chemical analysis. With this approach, further analysis of complete data sets allows state agencies authorized to conduct groundwater monitoring, often with limited financial resources, to prioritize sampling locations and chemicals to be tested. Given existing data and time constraints, it is crucial to identify the most strategic use of the available resources to address data gaps effectively. This work establishes a framework to enhance the beneficial impact of funding groundwater data collection by reducing uncertainty in prioritizing future sampling locations and chemical analyses.

Benchmarking machine learning missing data imputation methods in large-scale mental health survey databases

Databases tied to mental and behavioral health surveys suffer from the issue of missing data when participants skip the entire survey, which affects the data quality and sample size. These missing data patterns were investigated and the imputation performance was evaluated in Simons Foundations Powering Autism Research for Knowledge, a large-scale autism cohort consists of over 117,000 participants. Four common methods were assessed – Multiple imputation by chained equations (MICE), K-nearest neighbors (KNN), MissForest, and multiple imputation with denoising autoencoders (MIDAS). In a complete subset of 15,196 autism participants, three types of missingness patterns were simulated. We observed that MIDAS and KNN performed the best as the random missingness rate increased and when blockwise missingness was simulated. The average computational times were each 10 min for MIDAS and KNN, 35 min for MissForest, and 290 min for MICE. MIDAS and KNN both provide promising imputation performance in mental and behavioral health survey data that exhibit blockwise missingness patterns.

Electronic Health Record Data Quality and Performance Assessments: Scoping Review.

Electronic health records (EHRs) have an enormous potential to advance medical research and practice through easily accessible and interpretable EHR-derived databases. Attainability of this potential is limited by issues with data quality (DQ) and performance assessment. This review aims to streamline the current best practices on EHR DQ and performance assessments as a replicable standard for researchers in the field. PubMed was systematically searched for original research articles assessing EHR DQ and performance from inception until May 7, 2023. Our search yielded 26 original research articles. Most articles had 1 or more significant limitations, including incomplete or inconsistent reporting (n=6, 30%), poor replicability (n=5, 25%), and limited generalizability of results (n=5, 25%). Completeness (n=21, 81%), conformance (n=18, 69%), and plausibility (n=16, 62%) were the most cited indicators of DQ, while correctness or accuracy (n=14, 54%) was most cited for data performance, with context-specific supplementation by recency (n=7, 27%), fairness (n=6, 23%), stability (n=4, 15%), and shareability (n=2, 8%) assessments. Artificial intelligence-based techniques, including natural language data extraction, data imputation, and fairness algorithms, were demonstrated to play a rising role in improving both dataset quality and performance. This review highlights the need for incentivizing DQ and performance assessments and their standardization. The results suggest the usefulness of artificial intelligence-based techniques for enhancing DQ and performance to unlock the full potential of EHRs to improve medical research and practice.

Identification of novel hypertension biomarkers using explainable AI and metabolomics

BackgroundThe global incidence of hypertension, a condition of elevated blood pressure, is rising alarmingly. According to the World Health Organization’s Qatar Hypertension Profile for 2023, around 33% of adults are affected by hypertension. This is a significant public health concern that can lead to serious health complications if left untreated. Metabolic dysfunction is a primary cause of hypertension. By studying key biomarkers, we can discover new treatments to improve the lives of those with high blood pressure.AimsThis study aims to use explainable artificial intelligence (XAI) to interpret novel metabolite biosignatures linked to hypertension in Qatari Population.MethodsThe study utilized liquid chromatography-mass spectrometry (LC/MS) method to profile metabolites from biosamples of Qatari nationals diagnosed with stage 1 hypertension (n = 224) and controls (n = 554). Metabolon platform was used for the annotation of raw metabolite data generated during the process. A comprehensive series of analytical procedures, including data trimming, imputation, undersampling, feature selection, and biomarker discovery through explainable AI (XAI) models, were meticulously executed to ensure the accuracy and reliability of the results.ResultsElevated Vanillylmandelic acid (VMA) levels are markedly associated with stage 1 hypertension compared to controls. Glycerophosphorylcholine (GPC), N-Stearoylsphingosine (d18:1/18:0)*, and glycine are critical metabolites for accurate hypertension prediction. The light gradient boosting model yielded superior results, underscoring the potential of our research in enhancing hypertension diagnosis and treatment. The model’s classification metrics: accuracy (78.13%), precision (78.13%), recall (78.13%), F1-score (78.13%), and AUROC (83.88%) affirm its efficacy. SHapley Additive exPlanations (SHAP) further elucidate the metabolite markers, providing a deeper understanding of the disease’s pathology.ConclusionThis study identified novel metabolite biomarkers for precise hypertension diagnosis using XAI, enhancing early detection and intervention in the Qatari population.

A hybrid model for missing traffic flow data imputation based on clustering and attention mechanism optimizing LSTM and AdaBoost

Reliable traffic flow data is not only crucial for traffic management and planning, but also the foundation for many intelligent applications. However, the phenomenon of missing traffic flow data often occurs, so we propose an imputation model for missing traffic flow data to overcome the randomness and instability bands of traffic flow. First, k-means clustering is used to classify road segments with traffic flow belonging to the same pattern into a group to utilize the spatial characteristics of roads fully. Then, the LSTM networks optimized with an attention mechanism are used as the base learner to extract the temporal dependence of the traffic flow. Finally, the AdaBoost algorithm is used to integrate all the LSTM-attention networks into a reinforced learner to impute the missing data. To validate the effectiveness of the proposed model, we use the PeMS dataset for validation, we impute the data with missing data rate from 10 to 60% under three missing modes, and we use multiple baseline models for comparison, which confirms that our proposed model improves the stability and accuracy of imputing the missing data of the traffic flow with different scenarios.

Integrating Data Imputation and Augmentation with Interpretable Machine Learning for Efficient Strength Prediction of Fly Ash-Based Alkali-Activated Concretes

Fly ash-based alkali-activated concrete (AAC) is renowned for its superior mechanical performance and sustainability, presenting an attractive alternative to traditional Portland cement concrete. Despite these advantages, the broad compositional range of AACs presents challenges in precisely tailoring material properties. In this context, machine learning (ML) offers promising prospects to streamline and fast-track the development of advanced materials design strategies by predicting mechanical properties from compositional variations. Effective ML model development, however, hinges on the availability of a comprehensive, high-quality dataset. Previous studies often relied on literature-derived datasets, which typically include outliers, noise, and missing values, potentially leading to biased predictions. Moreover, limited dataset sizes could undermine the robustness of the models. Traditional ML methods applied to AACs also tend to lack interpretability. To address these issues, this paper utilizes several data imputation methods and Generative Adversarial Networks (GANs) for data augmentation, effectively doubling the dataset size. Following this, ML algorithms such as Random Forest (RF), Extreme Gradient Boosting (XGBoost), and Neural Networks (NNs) are leveraged to predict compressive strength. The NN model, especially when enhanced by k-nearest neighbors (kNN) imputation (k = 5), demonstrated superior predictive accuracy compared to RF and XGBoost models. Further, SHAP (SHapley Additive exPlanations) analysis reveals key determinants of compressive strength, such as water content, SiO2, and curing conditions. Visualizations such as SHAP violin and river flow plots further elucidated feature contributions and property distributions. Overall, this study provides a robust framework for exploring composition-strength relationships in AACs, advancing the design of these environment-friendly materials.

A self-help mobile messaging intervention to improve subthreshold depressive symptoms among older adults in a socioeconomically deprived region of Brazil (PRODIGITAL): a pragmatic, two-arm randomised controlled trial

Subthreshold depression is a risk factor for major depression and is associated with increased morbidity and mortality, especially in older adults. There is emerging evidence that digital interventions, including self-help interventions, may reduce depressive symptoms. We aimed to evaluate the effectiveness of a mobile messaging intervention at reducing subthreshold depressive symptoms among older adults in Brazil. PRODIGITAL was a single blind, two-arm, individually randomised controlled trial conducted in 46 primary care clinics in the city of Guarulhos, Brazil. Individuals aged 60+ years were contacted by phone following a randomly ordered list for a screening assessment. Those who presented with anhedonia and/or depressed mood (Patient Health Questionnaire (PHQ)-2≥1), and who subsequently scored between 5 and 9 on the PHQ-9 were invited to participate. The intervention arm received the 'Viva Vida' digital self-help intervention consisting of automated multi-media messages sent via WhatsApp. Forty-eight audio and visual messages based on psychoeducation and behavioural activation were automatically delivered over six weeks. The control arm received a single message containing information about depression. The primary outcome was the difference in mean PHQ-9 scores between treatment arms at the three-month follow-up. All primary analyses were performed according to allocated arm with imputed data. The trial is registered with ReBEC, RBR-6c7ghfd. Participants were recruited between 8 September 2021 and 19 August 2022. Of the 454 participants enrolled, 223 were randomised to the intervention arm, 231 to the control arm. Participants' mean age was 65.3 years (SD 5.0) and 64.0% (n=292) were female. A total of 385 (84.8%) completed the three-month follow-up assessment; no difference in mean PHQ-9 scores between the treatment arms was observed (adjusted difference:-0.61; 95% CI:-1.75, 0.53; p=0.29). These results demonstrate that the Viva Vida digital self-help intervention did not help to improve subthreshold depressive symptoms amongst older adults. Further research is needed to understand why this self-help intervention was not effective in this population, and to explore how it might be adapted to achieve this goal. São Paulo Research Foundation and UK Joint Global Health Trials.

Teleconsultation on patients with type 2 diabetes in the Brazilian public health system: a randomised, pragmatic, open-label, phase 2, non-inferiority trial (TELECONSULTA diabetes trial)

This study addresses the rising burden of type 2 diabetes mellitus, and explores the potential of teleconsultation, as an alternative for diabetes management. The primary objective was to test the hypothesis that teleconsultation is non-inferior to face-to-face consultation in terms of glycaemic control measured as glycated haemoglobin (HbA1c) (non-inferiority margin for the upper confidence interval for the difference between groups of 0,5% in HbA1c) for type 2 diabetes mellitus patients referred from Primary Healthcare to Specialized Care within the SUS. TELECONSULTA, is a randomized, pragmatic, phase 2, single-centre, open-label, non-inferiority trial conducted in Joinville, Brazil. A total of 278 participants diagnosed with type 2 diabetes were randomized through mandatory teleconsulting services from primary care health units. The randomization was 1:1 to teleconsultation or face to face consultation. The study was registered at the Brazilian Clinical Trial Register-REBEC, under the code RBR-8gpgyd. Study status is "Completed". This study included 278 participants in the intention-to-treat (ITT) analysis. The median age was 61 (54-68) years, 167 (60%) were women. The between-groups comparative average reduction in HbA1c was-0.6% (90% CI-1.0;-0.1) at 3-months and-0.5% (90% CI-0.9; 0.0) at 6-months in Modified Intention-to-Treat (mITT) population with imputed data, showing the non-inferiority of teleconsultation. Results with no missing data imputation and in the per protocol population were similar. The frequency of hypoglycaemia and other adverse events was well balanced between groups. The results underscore the transformative potential of telemedicine in addressing the complexities of diabetes management within the framework of a universal healthcare system, contributing with valuable insights for healthcare policymakers and practitioners seeking innovative solutions to tackle the growing diabetes epidemic. This study was funded by the Brazilian Ministry of Health, through the Unified Health System-Institutional Development Support Program (PROADI-SUS).

Missing Data Imputation for Industrial Time Series With Adaptive Median Iteration Based on Generative Adversarial Networks

Missing Data Imputation for Industrial Time Series With Adaptive Median Iteration Based on Generative Adversarial Networks

Do mindfulness interventions cause harm? Findings from the Learning to Apply Mindfulness to Pain (LAMP) Pragmatic Clinical Trial.

Although mindfulness-based interventions (MBIs) are widely used in clinical and nonclinical settings, there has been little systematic study of their potential risks. To address this gap, we examined differences in psychological and physical worsening among participants in the usual care and intervention conditions of a 3-group, randomized pragmatic trial (Learning to Apply Mindfulness to Pain [LAMP]) that tested the effectiveness of 2 approaches to delivering MBIs to patients with chronic pain. The sample consisted of 374 male and 334 female patients with chronic pain enrolled in the LAMP trial who completed a 10-week follow-up survey, 61% of whom had a mental health diagnosis. Psychological and physical worsening was assessed by a checklist asking whether participants experienced specific symptoms since beginning the study. We used multivariable logistic regression models with imputed data to determine whether predicted probabilities of increased symptoms differed between usual care and the 2 MBIs. Participants in usual care were more likely to report experiencing increased psychological and physical worsening than were those in the MBIs, including an increase in disturbing memories; sadness, anxiousness, and fatigue; isolation and loneliness; and feeling more upset than usual when something reminded them of the past. MBIs do not appear to cause harm, in terms of increased symptoms, for this population of patients with chronic pain and high levels of mental health comorbidities. Preregistration with an analysis plan at www.ClinicalTrials.gov: NCT04526158. Patient enrollment began December 4, 2020.

Missing Data Imputation in Balanced Construction for Incomplete Block Designs

Observational data with massive sample sizes are often distributed on many local machines. From an experimental design perspective, investigators often desire to identify the effect of new treatments (even ML algorithms) on many blocks of experimental data. With time requirements or budget constraints, assigning all treatments to each block is not always feasible. This creates incomplete responses with respect to a randomized complete block design (RCBD). These incomplete responses are missing by design. However, whether they can be estimated with missing imputation methods is not well understood. Thus, it is challenging to correctly identify the treatment effects with missing data. To this end, this paper provides a method for imputation and analysis of the responses with missing data. The proposed method consists of three steps: Reconstruction, Imputation, and `Complete’-data Analysis (RICA). The incomplete responses are imputed with the expectation-maximization (EM) algorithm. The RCBD model is then fitted by the resulting dataset. The identifiability result suggests that the missing may be nonignorable for each block, but the whole data of an incomplete design are missing by design when the design is balanced. Theoretical results on relative efficiency also inform us when the missingness should be imputed for incomplete designs with the role of balanced variance. Applications on real-world data verify the efficacy of this method.

Network Attack Classification using Neural Network-Based Imputation Technique

Rapid technological developments have changed access to information significantly, especially in telecommunications. This growth creates new threats, such as network attacks, so detection becomes critical for network security. Leveraging machine learning algorithms to detect threats is promising, with effectiveness largely dependent on selecting relevant features optimized by the bat algorithm. Data imputation is critical in preparing data sets, and neural network-based imputation techniques demonstrate outstanding performance, achieving accuracy rates of 99.4% on validation data and 99.3% on test data. This method consistently maintains precision, recall, and scores around 98%. Models using this method also approach perfection in classifying normal and neptune labels. This imputation method can also be applied to other model architectures using autoML. Alternative models such as Light GBM, XGBoost, Random Forest, Extra Trees, and Weighted Ensemble L2 also exhibit exceptional accuracy, exceeding 99.8%.

Factors associated with 30-day mortality in patients receiving mechanical circulatory support for cardiogenic shock

Abstract Introduction Mechanical circulatory support (MCS) is often deployed in patients with refractory cardiogenic shock complicating acute myocardial infarction (AMI-CS). Nevertheless, the mortality rate in these patients remains utterly high, around 50%. Purpose The aim of this study was to identify patient characteristics associated with 30-day mortality in patients receiving MCS for AMI-CS. Methods In the Netherlands, all patients undergoing percutaneous coronary intervention (PCI) are prospectively entered in the Netherlands Heart Registration (NHR). Additional data were retrieved for all CS patients between 2017-2021 in 14 Dutch hospitals, including MCS usage. Survivors and non-survivors amongst MCS treated patients were compared to identify patient characteristics associated with 30-day mortality. In addition, multivariate logistic regression analysis on imputed data was performed to quantify these associations in adjustment for confounders. Results MCS was deployed in 516 (23.3%) of all AMI-CS patients. MCS patients had a mean age of 64.2 (± 12.5) years and 74.6% were male. An intra-aortic balloon pump (IABP) was used in 49.0% of the patients. Impella, extracorporeal membrane oxygenation (ECMO) and multiple devices were used in 18.2%, 13.2% and 18.4% respectively. The overall 30-day mortality rate was 55.0%. At baseline, MCS survivors were younger (62.8 vs 65.2, p=0.03), had diabetes less often (18.4 vs. 28.0, p=0.02), and presented with a higher mean arterial pressure (MAP) (77.9 vs. 71.7, p<0.01). An out-of-hospital cardiac arrest (OHCA) preceded in 34.3% of the survivors versus 42.0% of the non-survivors (p=0.09). Survivors also had lower admission levels of lactate (5.0 vs. 7.8, p<0.01), glucose (12.1 vs. 15.2, p<0.01) and creatinine (94 vs. 115, p<0.01). Multivessel disease (MVD) was present less frequently in survivors compared to non-survivors (62.1% vs. 74.2%, p<0.01). In-hospital cardiac arrest (IHCA) occurred less often in survivors both prior to PCI (8.2% vs. 18.0%, p<0.01) and during PCI (22.8% vs. 34.6%, p<0.01). After multivariate logistic regression, higher age (OR 1.04, 95%CI: 1.02-1.05), resuscitation prior to (OR 2.11, 95 %CI: 1.09-4.09) or during PCI (OR 1.58, 95% CI: 1.01-2.47) and the presence of MVD (OR 2.07, 95% CI: 1.32-3.25) were independently associated with 30-day mortality in AMI-CS patients who received MCS. Conclusion In AMI-CS patients who received MCS, older age, the occurrence of an IHCA prior to or during PCI, and the presence of MVD were independently associated with an increased 30-day mortality.Figure 1.Table 1.

Nutritional Supplement with Fermented Soy in Patients Under Active Surveillance for Low-Risk or Intermediate-Risk Prostate Cancer: Results from the PRAEMUNE Trial.

Background/Objectives: To investigate the effect of a fermented soy supplement during 18 months in patients under active surveillance (AS) for low-risk and selected favorable intermediate-risk prostate cancer (PCa), with an emphasis on PSA modulation. Methods: Low-risk patients with ISUP grade 1, clinical stage cT1 or cT2a, PSA < 10 ng/mL and favorable intermediate-risk patients with ISUP grade 2 (<10% pattern 4), clinical stage T2b-c, PSA 10-20 ng/mL. The primary outcome was PSA response defined as maximum PSA rise less than or equal to 0.87 ng/mL after 1 year, based on the weighted average of PSA velocity (PSAV) in previous studies in similar populations. Secondary outcomes were disease progression, adverse histology on repeat biopsy or switch to active therapy. In addition, primary and secondary outcomes with imputed data were also determined as sensitivity analyses, using Mann-Whitney U or Chi-squared tests. Results: Overall, 92 (61.3%) of 150 patients showed a PSA level response. This was more evident in ISUP 1 patients and resulted in fewer follow-up MRIs and fewer control biopsies, as well as a fewer number of positive control biopsies with statistical significance in the imputed dataset. Obtaining a PSA response was numerically associated with less initiated therapy. Conclusions: a fermented soy supplement in patients under AS for low-risk and selected favorable intermediate-risk PCa could be useful in selecting patients who may remain under AS or who may need to switch to active therapy.

Examining the link between age of alcohol initiation, adolescent alcohol use and adult substance use

Abstract Background Understanding the relationship between adolescent alcohol use and adult substance use is important for prevention programmes. Existing evidence presents divergent findings and potential mechanisms are debated. Methods We estimated associations between the age of alcohol initiation and alcohol use at age 17 and substance use at age 20 in a nationally representative cohort (n = 4554). Weighted logistic regression models were fitted using Generalised Estimating Equations in multiply imputed and complete case data. Results Our analysis identified that 27% of participants reported first alcohol use at ≤ 14 years old, with 37% engaging in risky alcohol use by age 17. At age 20, 14% reported hazardous alcohol use, while 38% used tobacco, 24% cannabis, and 28% other drugs. The age at which alcohol was initiated and risky alcohol use at 17 years old were both strongly associated with alcohol, tobacco, cannabis and other illicit drug use at 20 years old. Sequential change in effect estimates coupled with non-overlapping confidence intervals suggest a dose-response relationship between age of alcohol initiation and tobacco and other drug use at age 20. Conclusions First use of alcohol at 14 or younger was common. Earlier alcohol initiation and risky alcohol use are each associated with substantially elevated risk for substance use at 20 years old even after accounting for common liability factors. Acknowledgement: Growing Up in Ireland (GUI) is funded by the Department of Children, Equality, Disability, Integration and Youth (DCEDIY). It is managed by DCEDIY in association with the Central Statistics Office (CSO). Results in this report are based on analyses of data from Research Microdata Files provided by the CSO. Neither the CSO nor DCEDIY take any responsibility for the views expressed or the outputs generated from these analyses. Key messages • Delaying alcohol initiation and preventing risky alcohol use during adolescence may reduce levels of substance use in young adults. • Self-reported early use of alcohol remains common.

Examining factors associated with cocaine use in late adolescence and early adulthood

Abstract Background In 2022, cocaine was the leading problem drug for which individuals sought treatment in Ireland, overtaking opioids for the first time. Despite increasing concerns about cocaine, there is limited longitudinal research examining factors associated with cocaine use in the general youth population. We describe levels of use and associated factors at 17 and 20 years old Methods Past-year cocaine use was studied at age 17 (N = 5965) and 20 (N = 4679) in a nationally representative cohort. Individual, family, social and environmental exposure variables were selected a priori. Weighted logistic regression models were fitted using Generalised Estimating Equations to multiply imputed and complete case data. Results Cocaine use was reported by 220 (4%) at age 17, and 1072 (23%) at age 20. Both individual and community factors predict use. Earlier alcohol initiation and having friends who use cannabis were strongly associated with increased odds of cocaine use at both ages. Those with parents who reported that neighbourhood intoxication or drug-taking was common had elevated odds of cocaine use at age 17. Conclusions Past-year cocaine use at 20 years old in this cohort is substantially higher than other international cohort studies. Levels of cocaine use increase significantly between 17 and 20 years old. Acknowledgement: Growing Up in Ireland (GUI) is funded by the Department of Children, Equality, Disability, Integration and Youth (DCEDIY). It is managed by DCEDIY in association with the Central Statistics Office (CSO). Results in this report are based on analyses of data from Research Microdata Files provided by the CSO. Neither the CSO nor DCEDIY take any responsibility for the views expressed or the outputs generated from these analyses. Key messages • Our study underscores significant cocaine use rates and correlations with early substance initiation and neighbourhood factors. • This emphasizes the necessity of a comprehensive, multilevel strategy for preventing cocaine use.

AVOTREX: A Global Dataset of Extinct Birds and Their Traits

ABSTRACTMotivationHuman activities have been reshaping the natural world for tens of thousands of years, leading to the extinction of hundreds of bird species. Past research has provided evidence of extinction selectivity towards certain groups of species, but trait information is lacking for the majority of clades, especially for prehistoric extinctions identified only through subfossil remains. This incomplete knowledge potentially obscures the structure of natural communities, undermining our ability to infer changes in biodiversity across space and time, including trends in functional and phylogenetic diversity. Biases in currently available trait data also limit our ability to identify drivers and processes of extinction. Here we present AVOTREX, an open‐access database of species traits for all birds known to have gone extinct in the last 130,000 years. This database provides detailed morphological information for 610 extinct species, along with a pipeline to build phylogenetic trees that include these extinct species.Main Types of Variables ContainedFor each extinct bird species, we provide information on the taxonomy, geographic location, and period of extinction. We also present data on island endemicity, flight ability, and body mass, as well as standard measurements of external (matching the AVONET database of extant birds) and skeletal morphology from museum specimens where available. To ensure comprehensive morphological data coverage, we estimate all missing morphological measurements using a data imputation technique based on machine learning. Finally, we provide an R package to graft all extinct species onto a global phylogeny of extant species (BirdTree).Spatial Location and GrainGlobal.Time Period and GrainAll known globally extinct bird species from 130,000 years ago up until 2024.Major Taxa and Level of MeasurementBirds (Class Aves), species level.Software FormatSpreadsheets (.csv) stored in Dryad.