Imputed Datasets Research Articles

Evaluating the median p-value method for assessing the statistical significance of tests when using multiple imputation

Rubin’s Rules are commonly used to pool the results of statistical analyses across imputed samples when using multiple imputation. Rubin’s Rules cannot be used when the result of an analysis in an imputed dataset is not a statistic and its associated standard error, but a test statistic (e.g. Student’s t-test). While complex methods have been proposed for pooling test statistics across imputed samples, these methods have not been implemented in many popular statistical software packages. The median p-value method has been proposed for pooling test statistics. The statistical significance level of the pooled test statistic is the median of the associated p-values across the imputed samples. We evaluated the performance of this method with nine statistical tests: Student’s t-test, Wilcoxon Rank Sum test, Analysis of Variance, Kruskal-Wallis test, the test of significance for Pearson’s and Spearman’s correlation coefficient, the Chi-squared test, the test of significance for a regression coefficient from a linear regression and from a logistic regression. For each test, the empirical type I error rate was higher than the advertised rate. The magnitude of inflation increased as the prevalence of missing data increased. The median p-value method should not be used to assess statistical significance across imputed datasets.

Multiple imputation integrated to machine learning: predicting post-stroke recovery of ambulation after intensive inpatient rehabilitation

Good data quality is vital for personalising plans in rehabilitation. Machine learning (ML) improves prognostics but integrating it with Multiple Imputation (MImp) for dealing missingness is an unexplored field. This work aims to provide post-stroke ambulation prognosis, integrating MImp with ML, and identify the prognostic influential factors. Stroke survivors in intensive rehabilitation were enrolled. Data on demographics, events, clinical, physiotherapy, and psycho-social assessment were collected. An independent ambulation at discharge, using the Functional Ambulation Category scale, was the outcome. After handling missingness using MImp, ML models were optimised, cross-validated, and tested. Interpretability techniques analysed predictor contributions. Pre-MImp, the dataset included 54.1% women, 79.2% ischaemic patients, median age 80.0 (interquartile range: 15.0). Post-MImp, 368 non-ambulatory patients on 10 imputed datasets were used for training, 80 for testing. The random forest (the validation best-performing algorithm) obtained 75.5% aggregated balanced accuracy on the test set. The main predictors included modified Barthel index, Fugl-Meyer assessment/motricity index, short physical performance battery, age, Charlson comorbidity index/cumulative illness rating scale, and trunk control test. This is among the first studies applying ML, together with MImp, to predict ambulation recovery in post-stroke rehabilitation. This pipeline reliably exploits the potential of incomplete datasets for healthcare prognosis, identifying relevant predictors.

Determinants of Prevalence and Factors Associated with Anemia among Pregnant Women in Gambia: A Multivariate Analysis using DHS Data

Background Anemia is a prevalent public health concern in The Gambia and other developing nations. While previous studies relied on binary logistic regression, the Multinomial Survey Logistic Regression Model was developed to address complexities in survey design, including sampling weights, stratification, and clustering. Multiple Imputation with Chained Equations (MICE) was employed to address missing data in variables such as anemia level, contraceptive use, and malaria drug intake during pregnancy. Aim/Objective This study aims to determine the prevalence and factors of anemia among pregnant women across eight local government areas in The Gambia. The research focuses on Kanifing, Kuntaur, Mansakonko, Kerewan, Janjanbureh, Basse, Brikama, and Banjul. Given the limited research on anemia among pregnant women in The Gambia, there is a need for further investigation into this topic. Methods MICE were utilized to address missing data due to their validity in handling uncertainty. This method generates multiple reasonable datasets for imputation and analyzes them accurately. A survey logistic regression model was employed to assess data fitness, considering the complex sample design. Anemia levels were measured among 11,865 pregnant women aged 15 to 49 using a systematic questionnaire. The multinomial survey logistic regression model was utilized to account for sampling weights, stratification, and clustering, relaxing the premise of observation independence violated by conventional logistic regression. Results The study involved 11,865 expectant mothers, revealing mild, moderate, and severe cases of anemia with a prevalence of 44.6%. Factors associated with anemia included place of residence, local government area, contraceptive use, age, literacy level, and current pregnancy status. Conclusion Socio-economic, demographic, and environmental factors significantly impact anemia levels among reproductive women in The Gambia. Addressing factors such as place of residence, local government areas, contraceptive use, literacy level, and current pregnancy status is crucial in mitigating anemia risk. Government initiatives, such as improving access to sanitation facilities and offering free check-ups, could effectively address this public health concern.

Environmental chemical exposures and a machine learning-based model for predicting hypertension in NHANES 2003–2016

BackgroundHypertension is a common disease, often overlooked in its early stages due to mild symptoms. And persistent elevated blood pressure can lead to adverse outcomes such as coronary heart disease, stroke, and kidney disease. There are many risk factors that lead to hypertension, including various environmental chemicals that humans are exposed to, which are believed to be modifiable risk factors for hypertension.ObjectiveTo investigate the role of environmental chemical exposures in predicting hypertension.MethodsA total of 11,039 eligible participants were obtained from NHANES 2003–2016, and multiple imputation was used to process the missing data, resulting in 5 imputed datasets. 8 Machine learning algorithms were applied to the 5 imputed datasets to establish hypertension prediction models, and the average accuracy score, precision score, recall score, and F1 score were calculated. A generalized linear model was also built to predict the systolic and diastolic blood pressure levels.ResultsAll 8 algorithms had good predictions for hypertension, with Support Vector Machine (SVM) being the best, with accuracy, precision, recall, F1 scores and area under the curve (AUC) of 0.751, 0.699, 0.717, 0.708 and 0.822, respectively. The R2 of the linear model on the training and test sets was 0.28, 0.25 for systolic and 0.06, 0.05 for diastolic blood pressure.ConclusionsIn this study, relatively accurate prediction of hypertension was achieved using environmental chemicals with machine learning algorithms, demonstrating the predictive value of environmental chemicals for hypertension.

Predicting implementation of response to intervention in math using elastic net logistic regression.

The primary objective of this study was to identify variables that significantly influence the implementation of math Response to Intervention (RTI) at the school level, utilizing the ECLS-K: 2011 dataset. Due to missing values in the original dataset, a Random Forest algorithm was employed for data imputation, generating a total of 10 imputed datasets. Elastic net logistic regression, combined with nested cross-validation, was applied to each imputed dataset, potentially resulting in 10 models with different variables. Variables for the models derived from the imputed datasets were selected using four methods, leading to four candidate models for final selection. These models were assessed based on their performance of prediction accuracy, culminating in the selection of the final model that outperformed the others. Method50 and Methodcoef emerged as the most effective, achieving a balanced accuracy of 0.852. The ultimate model selected relevant variables that effectively predicted RTI. The predictive accuracy of the final model was also demonstrated by the receiver operating characteristic (ROC) plot and the corresponding area under the curve (AUC) value, indicating its ability to accurately forecast math RTI implementation in schools for the following year.

Chronic obstructive pulmonary disease, asthma, and mechanical ventilation are risk factors for dyspnea in patients with long COVID: A Japanese nationwide cohort study

BackgroundPatients often experience multiple prolonged symptoms following acute coronavirus disease 2019 (COVID-19) recovery, defined as long coronavirus disease (COVID). Patients with long COVID may experience dyspnea during acute and post-acute phases. Therefore, this study aimed to identify specific risk factors for dyspnea in patients with long COVID. MethodsHospitalized patients with COVID-19, aged ≥18 years, were enrolled in this multicenter cohort study conducted at 26 medical institutions across Japan. Clinical data during hospitalization and patient-reported outcomes after discharge at the 3, 6, and 12-month follow-ups were retrieved from medical records and paper-based or smartphone application-based questionnaires, respectively. ResultsGeneralized linear mixed model (GLMM) analysis of prolonged dyspnea at each time point during follow-up showed that this symptom was associated with chronic obstructive pulmonary disease (COPD) (odds ratio [OR], 2.74; 95% confidence interval [CI], 1.31–5.74), asthma (OR, 2.21; 95%CI, 1.17–4.16), and ventilator management (OR, 3.10; 95%CI, 1.65–5.83). In addition, patients with COPD (44.4%) and ventilator management (25.0%) were more frequently associated with delayed dyspnea onset. The generalized estimating equations analysis results with multiple imputed datasets, conducted as a sensitivity analysis, confirmed the adjusted GLMM analysis results. ConclusionsProlonged dyspnea was associated with COPD, asthma, and severe infection that required mechanical ventilation in the Japanese population with long COVID. Further investigation is needed to clarify its mechanism and develop prophylactic and therapeutic strategies for dyspnea in patients with long COVID.

Comparative Analysis of Imputation Methods for Enhancing Predictive Accuracy in Data Models

The presence of missing values within datasets can introduce a detrimental bias, significantly impeding the predictive algorithm's ability to discern patterns and accurately execute prediction. This paper aims to elucidate the intricacies of data imputation methods, providing a more profound understanding of prevalent imputation methods, including list-wise deletion (IGN), mean imputation (AVG), K-Nearest Neighbors (KNN), MissForest (MF), and Predictive Mean Matching (PMM). The dataset employed in this study consists of financial data about S&P 500 companies in the Compustat North America database. The training and validation dataset encompasses 1973 instances, consisting of data during the fourth quarter of 2009, the first quarter of 2010, and the third quarter of 2014. Within this set, 457 missing values were identified and imputed. The test dataset comprises 197 randomly selected instances from the fourth quarter of 2014, equivalent to ten percent of the total instances in the training dataset. The evaluation findings prominently position the dataset derived from MF imputation as the leading performer among all the imputed datasets. The insights derived from this study are intended to assist practitioners in making informed choices when selecting the most suitable data imputation method, particularly in the context of predictive modeling tasks.

Autoencoder imputation of missing heterogeneous data for Alzheimer's disease classification

AbstractMissing Alzheimer's disease (AD) data is prevalent and poses significant challenges for AD diagnosis. Previous studies have explored various data imputation approaches on AD data, but the systematic evaluation of deep learning algorithms for imputing heterogeneous and comprehensive AD data is limited. This study investigates the efficacy of denoising autoencoder‐based imputation of missing key features of heterogeneous data that comprised tau‐PET, MRI, cognitive and functional assessments, genotype, sociodemographic, and medical history. The authors focused on extreme (≥40%) missing at random of key features which depend on AD progression; identified as the history of a mother having AD, APoE ε4 alleles, and clinical dementia rating. Along with features selected using traditional feature selection methods, latent features extracted from the denoising autoencoder are incorporated for subsequent classification. Using random forest classification with 10‐fold cross‐validation, robust AD predictive performance of imputed datasets (accuracy: 79%–85%; precision: 71%–85%) across missingness levels, and high recall values with 40% missingness are found. Further, the feature‐selected dataset using feature selection methods, including autoencoder, demonstrated higher classification score than that of the original complete dataset. These results highlight the effectiveness and robustness of autoencoder in imputing crucial information for reliable AD prediction in AI‐based clinical decision support systems.

Miesize: Effect-size calculation in imputed data

In this article, I describe the miesize command for the calculation of effect sizes in imputed data. There may be situations where an effect size needs to be estimated for an intervention, an exposure, or a group membership variable but data on the independent or dependent variable are missing. Such missing data are commonly dealt with by multiply imputing plausible values. However, in this circumstance, the estimated effect size and associated standard errors will need to be pooled and estimated from the imputed dataset. The miesize command automates this process and calculates effect sizes for a binary variable from multiply imputed data in wide format. The estimates and standard errors (used to calculate the confidence intervals) are recombined using Rubin’s (1987, Multiple Imputation for Nonresponse in Surveys [Wiley]) rules. These rules are applied such that the average point estimate for the effect size is calculated from the imputed datasets. The pooled standard error, and hence confidence intervals, is calculated to account for both the variance between the imputed datasets and the variance within them. Pooled effect sizes and confidence intervals for Cohen’s (1988, Statistical Power Analysis for the Behavioral Sciences, 2nd ed. [Lawrence Erlbaum]) d, Hedges’s (1981, Journal of Educational Statistics 6: 107-128) g, and Glass’s ( Smith and Glass, 1977 , American Psychologist 32: 752-760) delta are provided by miesize.

The performance of prognostic models depended on the choice of missing value imputation algorithm: a simulation study

ObjectiveThe development of clinical prediction models is often impeded by the occurrence of missing values in the predictors. Various methods for imputing missing values before modelling have been proposed. Some of them are based on variants of multiple imputation by chained equations, while others are based on single imputation. These methods may include elements of flexible modelling or machine learning algorithms, and for some of them user-friendly software packages are available. The aim of this study was to investigate by simulation if some of these methods consistently outperform others in performance measures of clinical prediction models. Study Design and SettingWe simulated development and validation cohorts by mimicking observed distributions of predictors and outcome variable of a real data set. In the development cohorts, missing predictor values were created in 36 scenarios defined by the missingness mechanism and proportion of non-complete cases. We applied three imputation algorithms that were available in R software: mice, aregImpute and missForest. These algorithms differed in their use of linear or flexible models, or random forests, the way of sampling from the predictive posterior distribution, and the generation of a single or multiple imputed data sets. For multiple imputation we also investigated the impact of the number of imputations. Logistic regression models were fitted with the simulated development cohorts before (full data analysis) and after missing value generation (complete case analysis), and with the imputed data. Prognostic model performance was measured by the scaled Brier score, c-statistic, calibration intercept and slope, and by the mean absolute prediction error evaluated in validation cohorts without missing values. Performance of full data analysis was considered as ideal. ResultsNone of the imputation methods achieved the model’s predictive accuracy that would be obtained in case of no missingness. In general, complete case analysis yielded the worst performance, and deviation from ideal performance increased with increasing percentage of missingness and decreasing sample size. Across all scenarios and performance measures, aregImpute and mice, both with 100 imputations, resulted in highest predictive accuracy. Surprisingly aregImpute outperformed full data analysis in achieving calibration slopes very close to 1 across all scenarios and outcome models. The increase of mice’s performance with 100 compared to 5 imputations was only marginal. The differences between the imputation methods decreased with increasing sample sizes and decreasing proportion of non-complete cases. ConclusionIn our simulation study, model calibration was more affected by the choice of the imputation method than model discrimination. While differences in model performance after using imputation methods were generally small, multiple imputation methods as mice and aregImpute that can handle linear or nonlinear associations between predictors and outcome are an attractive and reliable choice in most situations.

The PROgnostic ModEl for chronic lung disease (PRO-MEL): development and temporal validation

BackgroundPatients with chronic lung diseases (CLDs), defined as progressive and life-limiting respiratory conditions, experience a heavy symptom burden as the conditions become more advanced, but palliative referral rates are low and late. Prognostic tools can help clinicians identify CLD patients at high risk of deterioration for needs assessments and referral to palliative care. As current prognostic tools may not generalize well across all CLD conditions, we aim to develop and validate a general model to predict one-year mortality in patients presenting with any CLD.MethodsA retrospective cohort study of patients with a CLD diagnosis at a public hospital from July 2016 to October 2017 was conducted. The outcome of interest was all-cause mortality within one-year of diagnosis. Potential prognostic factors were identified from reviews of prognostic studies in CLD, and data was extracted from electronic medical records. Missing data was imputed using multiple imputation by chained equations. Logistic regression models were developed using variable selection methods and validated in patients seen from January 2018 to December 2019. Discriminative ability, calibration and clinical usefulness of the model was assessed. Model coefficients and performance were pooled across all imputed datasets and reported.ResultsOf the 1000 patients, 122 (12.2%) died within one year. Patients had chronic obstructive pulmonary disease or emphysema (55%), bronchiectasis (38%), interstitial lung diseases (12%), or multiple diagnoses (6%). The model selected through forward stepwise variable selection had the highest AUC (0.77 (0.72–0.82)) and consisted of ten prognostic factors. The model AUC for the validation cohort was 0.75 (0.70, 0.81), and the calibration intercept and slope were − 0.14 (-0.54, 0.26) and 0.74 (0.53, 0.95) respectively. Classifying patients with a predicted risk of death exceeding 0.30 as high risk, the model would correctly identify 3 out 10 decedents and 9 of 10 survivors.ConclusionsWe developed and validated a prognostic model for one-year mortality in patients with CLD using routinely available administrative data. The model will support clinicians in identifying patients across various CLD etiologies who are at risk of deterioration for a basic palliative care assessment to identify unmet needs and trigger an early referral to palliative medicine.Trial registrationNot applicable (retrospective study).

Initial management of newly diagnosed WHO grade 2-3 adult meningioma following surgery: results from the Dutch Brain Tumour Registry (2016-2021).

Meningiomas classified as grade 2-3 according to the World Health Organisation (WHO) require combined surgery and in most cases radiotherapy (RT). Their initial management was evaluated using the Dutch Brain Tumour Registry. The study included 393 patients aged ≥ 18 years with newly diagnosed meningioma WHO grade 2-3 between 2016 and 2021. Factors associated with adjuvant RT < 6 months following surgery were identified using logistic regression analyses, thereby accounting for variation between CNS regional tumour boards through mixed-effect modelling. This variation was further assessed by funnel plots for case-mix adjusted ratios of RT across tumour boards. The association with patients' survival at 5 years was evaluated with inverse probability-weighted accelerated failure (Weibull) models. Analyses were performed on multiple imputed datasets (m = 10) to account for missing data. Adjuvant RT was administered to 22.2% (59/266) of patients with WHO grade 2 meningioma following a total resection, to 61.1% (58/95) following a partial resection, and to 68.8% (22/32) of patients with WHO grade 3 meningioma (61.5% after partial and 73.7% after total resection). RT was associated with grade 3, partial resection, bone invasion, and absence of multiple lesions. Management varied across tumour boards for grade 2 meningioma following total resection. Adjuvant RT was associated with survival benefit in case of grade 3 disease (hazard ratio: 0.40, 95%-confidence interval: 0.16-0.95, p = 0.04). This national review revealed variation across CNS regional tumour boards in the management of grade 2 meningioma following total resection, and demonstrated survival benefit of adjuvant RT in grade 3 meningioma.

Missing Values in Longitudinal Proteome Dynamics Studies: Making a Case for Data Multiple Imputation.

Temporal proteomics data sets are often confounded by the challenges of missing values. These missing data points, in a time-series context, can lead to fluctuations in measurements or the omission of critical events, thus hindering the ability to fully comprehend the underlying biomedical processes. We introduce a Data Multiple Imputation (DMI) pipeline designed to address this challenge in temporal data set turnover rate quantifications, enabling robust downstream analysis to gain novel discoveries. To demonstrate its utility and generalizability, we applied this pipeline to two use cases: a murine cardiac temporal proteomics data set and a human plasma temporal proteomics data set, both aimed at examining protein turnover rates. This DMI pipeline significantly enhanced the detection of protein turnover rate in both data sets, and furthermore, the imputed data sets captured new representation of proteins, leading to an augmented view of biological pathways, protein complex dynamics, as well as biomarker-disease associations. Importantly, DMI exhibited superior performance in benchmark data sets compared to single imputation methods (DSI). In summary, we have demonstrated that this DMI pipeline is effective at overcoming challenges introduced by missing values in temporal proteome dynamics studies.

Machine learning assessment of vildagliptin and linagliptin effectiveness in type 2 diabetes: Predictors of glycemic control.

Differential effects of linagliptin and vildagliptin may help us personalize treatment for Type 2 Diabetes Mellitus (T2DM). The current study compares the effect of these drugs on glycated hemoglobin (HbA1c) in an artificial neural network (ANN) model. Patients with T2DM who received either vildagliptin or linagliptin, with predefined exclusion criteria, qualified for the study. Two input variable datasets were constructed: with or without imputation for missing values. The primary outcome was HbA1c readings between 3 to 12 months or the reduction in HbA1c levels. The cohort comprised 191 individuals (92 vildagliptin and 99 linagliptin). Linagliptin group had significantly higher disease burden. For imputed dataset, HbA1c was lower with linagliptin at 3 to 12 months (7.442 ± 0.408 vs. 7.626 ± 0.408, P < 0.001). However, there was a small yet significant difference in HbA1c reduction favoring vildagliptin over linagliptin (-1.123 ± 0.033 vs. -1.111 ± 0.043, P < 0.001). LDL level, uric acid, and the drug group were identified as predictors for HbA1c levels. In the non-imputed dataset HbA1c at 3 to 12 months was lower with linagliptin (median ± IQR: 7.489 ± 0.467 vs. 7.634 ± 0.467, P-value < 0.001). However, both linagliptin and vildagliptin exhibited similar reductions in HbA1c levels (both median ± IQR of -1.07 ± 0.02). Predictors for HbA1c levels included eGFR level and the drug group. Linagliptin effectively lowers HbA1c levels more than vildagliptin including in patients with comorbidities. DPP4-I choice is a constant predictor of HbA1c in all models.

Post-endovascular therapy contrast extravasation in the mesial temporal region on dual-energy CT is associated with outcome in acute ischemic stroke patients

PurposePre- and post-endovascular treatment (EVT) imaging may aid in predicting functional outcomes in acute middle cerebral artery (MCA) ischemic stroke. Low post-EVT contrast extravasation (CE)-ASPECTS is associated with poor functional outcomes. Besides the MCA regions included in the ASPECTS score, CE may be seen in the mesial temporal (MT) region. In this study, we investigated the frequency and prognostic implication of MT-CE in acute ischemic stroke patients. MethodsPatients with an acute ischemic stroke due to anterior large vessel occlusion who received EVT and post-EVT DECT between 2010 and 2019 were included. Iodine overlay maps of DECT were assessed for the occurrence of CE, using the ASPECTS for occurrence in the MCA region and, calculating a CE-ASPECTS, for whether the MT region was involved. Multivariable linear and logistic regression were used to assess the relationship between involvement of MT-CE and 24-48h NIHSS, mRS, and mortality on a multiple imputed dataset. All models were adjusted significant variables in univariate analyses and for total CE-ASPECTS. Results501/651 patients met the inclusion criteria. MT-CE occurred in 97 (19 %) patients, and was more often present in patients with internal carotid artery occlusions. MT-CE was associated with higher NIHSS scores at 24-hours (aβ 2.2, 95 % CI 0.09-4.31), with increased risk of higher mRS scores (acOR 1.88, 95 % CI 1.16-3.06), and with increased risk of mortality (aOR 2.12, 95 % CI 1.16-3.86). ConclusionMT-CE is a common finding on post-EVT DECT and is an independent predictor for worse functional outcomes.

Randomized Trial of the Effectiveness of Videoconferencing-Based Versus Message-Based Psychotherapy on Depression.

The authors compared the engagement, clinical outcomes, and adverse events of text or voice message-based psychotherapy (MBP) versus videoconferencing-based psychotherapy (VCP) among adults with depression. The study used a sequential multiple-assignment randomized trial design with data drawn from phase 1 of a two-phase small business innovation research study. In total, 215 adults (ages ≥18 years) with depression received care from Talkspace, a digital mental health care company. Participants were initially randomly assigned to receive either asynchronous MBP or weekly VCP. All therapists provided evidence-based treatments such as cognitive-behavioral therapy. After 6 weeks of treatment, participants whose condition did not show a response on the Patient Health Questionnaire-9 or was rated as having not improved on the Clinical Global Impressions scale were randomly reassigned to receive either weekly VCP plus MBP or monthly VCP plus MBP. Longitudinal mixed-effects models with piecewise linear time trends applied to multiple imputed data sets were used to address missingness of data. Participants who were initially assigned to the MBP condition engaged with their therapists over more weeks than did participants in the VCP condition (7.8 weeks for MBP vs. 4.9 weeks for VCP; p<0.001). No meaningful differences were observed between the two groups in rates of change by 6 or 12 weeks for depression, anxiety, disability, or global ratings of improvement. Neither treatment resulted in any adverse events. MBP appears to be a viable alternative to VCP for treating adults with depression.

Effect of elevated depressive symptoms during adolescence on health-related quality of life in young adulthood-a six-year cohort study with repeated exposure measurements.

Depression is a major contributor of young people's burden of disease. In this study we aim to estimate the effect of elevated depressive symptoms on physical health-related quality of life. We used self-reported information from the prospective BELLA cohort study, which included adolescents selected from the general population in Germany. The baseline assessment (2003-2006) and the 1-, 2-, and 6-year follow-up waves provide the data basis. The baseline study population consisted of 1,460 adolescents between the ages of 12 and 17 who, according to their caregivers, did not suffer from depression. The primary outcome, as measured by the physical component score (PCS) of the SF-36 at a 6-year follow-up (range: 0-100), is physical health-related quality of life. The exposure of interest is depressive symptoms, as measured by the Center for Epidemiological Studies Depression Scale for Children (CES-DC) at baseline, 1-year follow-up and 2-year follow-ups (range: 0-60). We dichotomized the exposure into subthreshold (≤15) and elevated depressive symptoms (>15). For the main analyses we considered a cumulative index for elevated depressive symptoms across the three time points (range: 0-3). Considered confounders are sex, age, socioeconomic status, migrant background, social support, anxiety symptoms, physical activity, chronic diseases, and sleeping problems. We used multiple imputation to account for missing values. Within each imputed dataset, we applied inverse probability weighting (IPW) to estimate the effect of the cumulative index for elevated depressive symptoms at baseline, 1- and 2-year follow-up on physical health-related quality of life at 6-year follow-up. We derived 95% confidence intervals by bootstrapping. After adjusting with IPW, the effect of the cumulative index per one unit increase of elevated depressive symptoms on the physical component score was -1.71 (95% CI: -3.51 to -0.04). The adjusted effect estimates of single exposure of elevated depressive symptoms on physical health-related quality of life were -0.83 (95% CI: -3.69 to 1.87) at baseline, -2.96 (95% CI: -4.94 to -0.52) at 1-year follow-up and -1.32 (95% CI: -3.85 to 1.15) at 2-year follow-up. Findings suggest that elevated depressive symptoms during adolescence decrease physical health-related quality of life in young adulthood.

Effect of a Smartphone App (S-Check) on Actual and Intended Help-Seeking and Motivation to Change Methamphetamine Use Among Adult Consumers of Methamphetamine in Australia: Randomized Waitlist-Controlled Trial.

Interventions are required that address delays in treatment-seeking and low treatment coverage among people consuming methamphetamine. We aim to determine whether a self-administered smartphone-based intervention, the "S-Check app" can increase help-seeking and motivation to change methamphetamine use, and determine factors associated with app engagement. This study is a randomized, 28-day waitlist-controlled trial. Consenting adults residing in Australia who reported using methamphetamine at least once in the last month were eligible to download the app for free from Android or iOS app stores. Those randomized to the intervention group had immediate access to the S-Check app, the control group was wait-listed for 28 days before gaining access, and then all had access until day 56. Actual help-seeking and intention to seek help were assessed by the modified Actual Help Seeking Questionnaire (mAHSQ), modified General Help Seeking Questionnaire, and motivation to change methamphetamine use by the modified readiness ruler. χ2 comparisons of the proportion of positive responses to the mAHSQ, modified General Help Seeking Questionnaire, and modified readiness ruler were conducted between the 2 groups. Logistic regression models compared the odds of actual help-seeking, intention to seek help, and motivation to change at day 28 between the 2 groups. Secondary outcomes were the most commonly accessed features of the app, methamphetamine use, feasibility and acceptability of the app, and associations between S-Check app engagement and participant demographic and methamphetamine use characteristics. In total, 560 participants downloaded the app; 259 (46.3%) completed eConsent and baseline; and 84 (32.4%) provided data on day 28. Participants in the immediate access group were more likely to seek professional help (mAHSQ) at day 28 than those in the control group (n=15, 45.5% vs n=12, 23.5%; χ21=4.42, P=.04). There was no significant difference in the odds of actual help-seeking, intention to seek help, or motivation to change methamphetamine use between the 2 groups on the primary logistic regression analyses, while in the ancillary analyses, the imputed data set showed a significant difference in the odds of seeking professional help between participants in the immediate access group compared to the waitlist control group (adjusted odds ratio 2.64, 95% CI 1.19-5.83, P=.02). For participants not seeking help at baseline, each minute in the app increased the likelihood of seeking professional help by day 28 by 8% (ratio 1.08, 95% CI 1.02-1.22, P=.04). Among the intervention group, a 10-minute increase in app engagement time was associated with a decrease in days of methamphetamine use by 0.4 days (regression coefficient [β] -0.04, P=.02). The S-Check app is a feasible low-resource self-administered intervention for adults in Australia who consume methamphetamine. Study attrition was high and, while common in mobile health interventions, warrants larger studies of the S-Check app. Australian New Zealand Clinical Trials Registry ACTRN12619000534189; https://www.anzctr.org.au/Trial/Registration/TrialReview.aspx?id=377288&isReview=true.

Forecasting potential invaders to prevent future biological invasions worldwide.

The ever-increasing and expanding globalisation of trade and transport underpins the escalating global problem of biological invasions. Developing biosecurity infrastructures is crucial to anticipate and prevent the transport and introduction of invasive alien species. Still, robust and defensible forecasts of potential invaders are rare, especially for species without known invasion history. Here, we aim to support decision-making by developing a quantitative invasion risk assessment tool based on invasion syndromes (i.e., generalising typical attributes of invasive alien species). We implemented a workflow based on 'Multiple Imputation with Chain Equation' to estimate invasion syndromes from imputed datasets of species' life-history and ecological traits and macroecological patterns. Importantly, our models disentangle the factors explaining (i) transport and introduction and (ii) establishment. We showcase our tool by modelling the invasion syndromes of 466 amphibians and reptile species with invasion history. Then, we project these models to amphibians and reptiles worldwide (16,236 species [c.76% global coverage]) to identify species with a risk of being unintentionally transported and introduced, and risk of establishing alien populations. Our invasion syndrome models showed high predictive accuracy with a good balance between specificity and generality. Unintentionally transported and introduced species tend to be common and thrive well in human-disturbed habitats. In contrast, those with established alien populations tend to be large-sized, are habitat generalists, thrive well in human-disturbed habitats, and have large native geographic ranges. We forecast that 160 amphibians and reptiles without known invasion history could be unintentionally transported and introduced in the future. Among them, 57 species have a high risk of establishing alien populations. Our reliable, reproducible, transferable, statistically robust and scientifically defensible quantitative invasion risk assessment tool is a significant new addition to the suite of decision-support tools needed for developing a future-proof preventative biosecurity globally.

A novel method for settlement imputation and monitoring of earth-rockfill dams subjected to large-scale missing data

Various missing values inevitably exist in the settlement monitoring data of prolonged operation of earth-rockfill dams, which delays or even interrupts the dam structure analytical procedure. This study aims to develop a reliable method for addressing the missing data problem and monitoring earth-rockfill dam settlement behavior. First, an imputation model, support vector regression based on the finite element method and time input (FEMTSVR), is proposed and offers superior performance when capturing complex nonlinear mappings from environmental variables to settlement on small sample data. Herein, an improved particle swarm optimization algorithm (IPSO) is developed, realizing the nonlinear adjustment of weights and parameter reduction during hyperparameter optimization. Then, this study establishes a sequential prediction model based on gate recurrent unit (GRU) networks to monitor dam behavior on the imputed and complete dataset. Eventually, the proposed method is evaluated using real-world earth-rockfill dam monitoring data with the help of statistical indicators, demonstrating its efficiency in monitoring settlement data subjected to large-scale missingness. This study provides a robust database for dam structural health monitoring, while also providing a promising framework for the safety assessment of other civil or hydraulic engineering based on raw monitoring data.