Handle Missing Data Research Articles

Effectiveness of early pharmaceutical interventions in symptomatic COVID-19 patients: A randomized clinical trial.

We assessed the effectiveness of oral Hydroxychloroquine (HC), Azithromycin (AZ) and Oseltamivir (OS), alone or combined, among patients hospitalized with mildly symptomatic coronavirus infectious disease (COVID-19). Following the approval of the National Bioethics Committee and prospective registration (clinicaltrials.gov NCT04338698), a multicenter randomized clinical trial of adaptive design was conducted at 10 multispecialty hospitals in Pakistan. Patients were randomized into seven treatment groups. Starting April 15, 2020, consenting, eligible, otherwise healthy adult patients or those with co-morbidities under control, were recruited if they presented with mildly symptomatic COVID-19 (scored 3 on a 7-point ordinal scale anchored between 1 = not hospitalized, able to undertake normal activities, to 7 = death) confirmed by quantitative Real-Time Polymerase Chain Reaction (qRT-PCR). Two primary outcomes were assessed by day seven: Turning qRT-PCR negative; and clinical improvement of two points from the baseline. Outcome rates were compared using a chi-square test. Multiple imputations were applied to handle missing data. An interim data analysis was carried out on July 19, 2020, following which the study continued without treatment group changes. Data Safety and Monitoring Board advised to stop recruitment due to its futility on January 18, 2021. Of 471 patients randomized, a total of 426 (90.4%) completed the follow-up for primary outcomes. Based on imputed data analyses at day seven: Total qRT-PCR negative cases were 137/471 (29%, 95% CI 25.0 - 33.4). By day seven, a total of 111/471 (23.5%, 95% CI 19.8 - 27.6) showed clinical improvement. No serious or non-serious adverse event was reported. Among patients with mild COVID-19, there was no statistically significant difference in the effectiveness of oral antimalarial, antiviral, or antibiotic treatments.Clinicaltrials.gov ID: NCT04338698.

Periconceptional maternal homocysteine and birth weight outcomes in offspring: a prospective cohort study

Objective: To quantify the associations between periconceptional maternal homocysteine (HCY) and offspring's birth weight and risk of small for gestational age (SGA) infant. Methods: The 19 984 mother-child pairs in this prospective cohort study were recruited from the Shanghai preconception cohort; the infants were delivered from 1st September 2016 to 11th November 2022. A standardized questionnaire was used to collect the mothers' demographic information, medical history, dietary supplement use, and maternal complications during pregnancy, and their serum samples were collected. Serum HCY, folate, and vitamin B12 were measured using chemiluminescent microparticle immunoassay based on serum sample drawn at enrollment. Birth weight data were obtained from medical records. Multiple imputation methods were applied to handle missing data in key variables. Multivariable linear regression and Poisson regression models were used to analyze the relationship between maternal HCY concentration during the periconceptional period and the birth weight and SGA risk of the offspring. Results: A total of 9 452 pairs were enrolled preconceptionally and the remaining 10 532 pairs were enrolled in early pregnancy. The proportion of mothers whose pregnancy age was greater than 35 years was 9.2% (1 832/19 984), the proportion of primiparous women was 76.5% (15 283/19 984), the proportion of pre-pregnancy overweight and obesity was 14.0% (2 804/19 984), the proportion of using folic acid supplements before pregnancy was 21.4% (4 272/19 984), and the proportion of those who supplemented with folic acid during early pregnancy was 85.2% (8 976/10 532); gestational diabetes mellitus was in 6.2% (1 245/19 984), gestational hypertensive syndrome in 3.6% (711/19 984). The birth weight of the offspring was (3 297±468) g, and there were 1 962 SGA children (9.8%). The HCY concentration in the overall population in appropriate for gestational age during the periconceptional period was (7.9±3.2) μmol/L, with (8.3±3.7) μmol/L in the preconception subgroup and (7.3±2.4) μmol/L in the early pregnancy subgroup. After adjustment for the covariates of perinatal demographic information, adverse pregnancy outcomes, serum folate and vitamin B12, increased maternal periconceptional HCY was significantly associated with lower offspring birth weight (β=-2.30, 95%CI -4.43--0.16, P=0.035). Only the early pregnancy subgroup was significantly associated with lower offspring birth weight (β=-7.39, 95%CI-11.50--3.21, P<0.001). No association was found between peripregnancy HCY and offspring SGA risk. However, elevated HCY in early pregnancy was associated with an increased risk of SGA in the offspring (RR=1.05, 95%CI 1.01-1.08, P=0.002). Periconceptional vitamin B12 was a mediator of the association between HCY and offspring birth weight, accounting for 16.5%, 41.2% and 5.4% of its total effect in the overall periconceptional population, the pre-pregnancy subgroup and the early pregnancy subgroup, respectively. Conclusions: Maternal periconceptional HCY level is associated with lower birth weight in offspring, but not with the risk of SGA. Elevated maternal HCY in early pregnancy subgroup may be associated with increased risk of SGA in offspring.

Long-term survival outcomes of endoscopic therapy vs. surgical resection in patients with cardia gastrointestinal stromal tumor.

The ideal surgical approach for treating cardia gastrointestinal stromal tumor (GIST) is not clearly established. This study aimed to assess the long-term survival results among patients who received endoscopic therapy (ET) or surgical resection (SR) for cardia GIST. Cardia GIST patients from 2000 to 2019 were selected from the surveillance, epidemiology, and end result (SEER) database. Multiple imputation (MI) was applied to handle missing data, and propensity score matching (PSM) was carried out to mitigate selection bias during comparisons. Demographic and clinical characteristics' effects on overall survival (OS) and cancer-specific survival (CSS) were assessed using Kaplan-Meier analyses and multivariate Cox proportional hazard models. A total of 330 patients with cardia GIST were enrolled, including 47 (14.2%) patients with ET and 283 (85.8%) patients with SR. The 5-year OS and CSS rates in the ET and SR groups were comparable [before PSM, (OS) (76.1% vs. 81.2%, P = 0.722), (CSS) (95.0% vs. 89.3%, P = 0.186); after PSM, (OS) (75.4% vs. 85.4%, P = 0.540), (CSS) (94.9% vs. 92.0%, P = 0.099)]. Moreover, there was no significant difference between ET and SR in terms of long-term OS (hazard ratio [HR] 0.735, 95% confidence interval [CI] 0.422-1.282) and CSS (HR 1.560, 95% CI 0.543-4.481). Our study found no significant disparity in long-term survival outcomes between ET and SR in cardia GIST patients, implying that ET could be a valid surgical strategy for treating cardia GIST.

Everyday Discrimination, Neighborhood Perceptions, and Incidence of Activity Limitations Among Middle-Aged and Older African Americans.

This study aims to examine the relationship between everyday discrimination, neighborhood perceptions, and the incidence of daily activity limitations (i.e., activities of daily living [ADL] and instrumental activities of daily living [IADL]) among middle-aged and older African Americans. This study also examines whether neighborhood perceptions moderate the association between discrimination and the incidence of daily activity limitations. Data were from the 2006 to 2016 waves of the Health and Retirement Study. African Americans aged 50 or older free of ADL limitations (N=1,934) and IADL limitations (N=2,007) at baseline were selected. Cox proportional hazards regression models were performed to test the study aims. Multiple imputations were applied to handle missing data. One-unit increase in everyday discrimination was associated with a 25% (p < .05) higher risk of ADL limitation onset. Perceived neighborhood social cohesion and physical disadvantage moderated the association between discrimination and IADL limitation onset. Everyday discrimination represents a significant stressor that negatively affects older African Americans' performance of daily activities. Community-level efforts that improve neighborhood perceptions are needed to alleviate the negative effects of discrimination on the incidence of activity limitations.

Water Availability Forecasting Using Univariate and Multivariate Prophet Time Series Model for ACEA (European Automobile Manufacturers Association)

Time series is one of method to forecasting the data. The ACEA company has competition with opened the data in the Water Availability and uses the data to forecast. The dataset namely, Aquifers-Petrignano in Italy in water resources field has five parameters e.g. rainfall, temperature, depth to groundwater, drainage volume, and river hydrometry. In our research will be forecast the depth to groundwater data using univariate and multivariate approach of time series using Prophet Method. Prophet method is one of library which develop by Facebook team. We also use the other approach to making the data clean, or the data ready to forecast. We use handle missing data, transforming, differencing, decomposition time series, determine lag, stationary approach, and Augmented Dickey-Fuller (ADF). The all approach will be uses to make sure that the data not appearing the problem while we tried to forecast. In the other describe, we already get the results using univariate and multivariate Prophet method. The multivariate approach has presented the value of MAE 0.82 and RMSE 0.99, it’s better than while we forecast using univariate Prophet.

Development and validation of a prognostic model for leflunomide discontinuation with abnormal blood tests during long-term treatment: cohort study using data from the Clinical Practice Research Datalink Gold and Aurum.

ObjectiveTo develop and validate a prognostic model for LEF discontinuation with abnormal blood test results.MethodsData from the Clinical Practice Research Datalink Gold and Aurum were used for model development and external validation, respectively. Participants prescribed LEF between 1 January 2007 and 31 December 2019 were followed up from 6 months after the first general practitioner prescription to the earliest of date of outcome, death, 5 year follow-up or 31 December 2019. Candidate prognostic factors were ascertained using theory and data-driven approaches. Penalized Cox regression was performed to develop the risk equation, followed by internal validation using 500 bootstraps to correct for optimism. Multiple imputation was applied to handle missing data. Model performance was assessed in terms of calibration and discrimination.ResultsData for 1487 and 2329 participants contributing 3140 and 5246 person-years follow-up were included in the development and validation cohorts, respectively. Thirteen candidate predictors were included in the model. Epilepsy and either cytopenia or elevated liver enzymes during the first 6 months of shared-care LEF prescription were strong predictors of drug discontinuation with a hazard ratio of 4.39 (95% CI 1.74, 11.06) and 3.06 (2.15, 4.35), respectively. The unadjusted and optimism-adjusted calibration slope in development data was 1.00 (95% CI 0.75, 1.25) and 0.72 (95% CI 0.47, 0.97), respectively. The calibration slope in validation data was 0.91 (95% CI 0.74, 1.07). The model showed prognostic separation with an optimism-adjusted Royston D statistic of 0.73 (95% CI 0.44, 1.02).ConclusionWe have developed and externally validated an easy-to-use prognostic model that may be used to risk stratify monitoring for LEF toxicity and to make informed choices about risks when choosing treatments.

Evaluating and comparing profiles of burglaries developed using three statistical classification techniques: cluster analysis, multidimensional scaling, and latent class analysis

ABSTRACT While there are a variety of statistical classification techniques available, the most prominent in the behavioral sciences are Cluster Analysis (CA), Multidimensional Scaling (MDS), and Latent Class Analysis (LCA). Researchers often rely on person-oriented statistical classification techniques to identify and understand the latent heterogeneity in the complex individuals that we study. Using data on 405 randomly selected solved burglaries committed in Florida, this study is the first to conduct a head-to-head comparison of the benefits and weakness of each analysis and evaluate the resultant typologies and predictive validity when all three analyses are applied to the same dataset. Findings suggest that the number and nature of resultant subtypes differ depending on the statistical classification technique employed. We conclude that LCA is superior to MDS and CA due to its ability to objectively evaluate model fit and handle missing data, balance of parsimony and complexity in the results, and reliability and accuracy stemming from the first test of predictive validity among the three statistical classification techniques. Implications for future research and the application and testing of statistical classification techniques are also discussed.

SLE clinical trials: impact of missing data on estimating treatment effects

ObjectiveA common problem in clinical trials is missing data due to participant dropout and loss to follow-up, an issue which continues to receive considerable attention in the clinical research community....

A RankBoost-Based Data-Driven Method to Determine Maintenance Priority of Circuit Breakers

Prioritizing equipment maintenance activities has been an important and challenging task in developing an equipment maintenance strategy for electric power utility companies. This paper presents a RankBoost-based data-driven approach for determining the maintenance priority of circuit breakers. Compared with conventional methods, the proposed method can consider both equipment-level condition-monitoring parameters and system-level reliability impacting indices, automatically create the rational weights among the multiple parameters/indices, and handle missing and invalid data. The simulation results based on the data of 33 circuit breakers indicate that the maintenance priority list obtained using the proposed method is closer to reality than the ranking lists obtained using a subjective weighting method or only using system-level reliability indices.

Stochastic Approximation and Memory-Limited Subspace Tracking for Poisson Streaming Data

Poisson count data is ubiquitously encountered in applications such as optical imaging, social networks, and traffic monitoring, where the data is typically modeled after a Poisson distribution and presented in a streaming fashion. Therefore, it calls for techniques to efficiently extract and track the useful information embedded therein. We consider the problem of recovering and tracking the underlying Poisson rate, where the rate vectors are assumed to lie in an unknown low-dimensional subspace, from streaming Poisson data with possibly missing entries. The recovery of the underlying subspace is posed as an expected loss minimization problem under nonnegative constraints, where the loss function is a penalized Poisson log-likelihood function. A stochastic approximation (SA) algorithm is proposed and can be implemented in an online manner. Two theoretical results are established regarding the convergence of the SA algorithm. The SA algorithm is guaranteed almost surely to converge to the same point as the original expected loss minimization problem, and the estimate converges to a local minimum. To further reduce the memory requirement and handle missing data, the SA algorithm is modified via lower bounding the log-likelihood function in a form that is decomposable and can be implemented in a memory-limited manner without storing history data. Numerical experiments are provided to demonstrate the superior performance of the proposed algorithms, compared to existing algorithms. The memory-limited SA algorithm is shown to empirically yield similar performance as the original SA algorithm at a much lower memory requirement.

Using Game Theory to Handle Missing Data at Prediction Time of ID3 and C4.5 Algorithms

The raw material of our paper is a well known and commonly used type of supervised algorithms: decision trees. Using a training data, they provide some useful rules to classify new data sets. But a data set with missing values is always the bane of a data scientist. Even though decision tree algorithms such as ID3 and C4.5 (the two algorithms with which we are working in this paper) represent some of the simplest pattern classification algorithms that can be applied in many domains, but with the drawback of missing data the task becomes harder because they may have to deal with unknown values in two major steps: at training step and at prediction step. This paper is involved in the processing step of databases using trees already constructed to classify the objects of these data sets. It comes with the idea to overcome the disturbance of missing values using the most famous and the central concept of the game theory approach which is the Nash equilibrium.

Erratum to: Handling Missing Data in Within-Trial Cost-Effectiveness Analysis: A Review with Future Recommendations.

Erratum to: Handling Missing Data in Within-Trial Cost-Effectiveness Analysis: A Review with Future Recommendations.

A taxonomy of estimands for regulatory clinical trials with discontinuations

The National Research Council Panel on Handling Missing Data in Clinical Trials recommended that protocols for clinical trials 'explicitly define... causal estimands of primary interest'. In discussions with sponsors of clinical trials since the publication of the National Research Council report, the expression causal estimands has been the subject of confusion. It may not be entirely clear what the National Research Council panel meant, and in any case, it has not been clear how this recommendation might be put in practice. This paper's purpose is to say how the working group understands it and how we think it should be put in practice. We classify possible choices of estimand according to their usefulness for regulatory purposes in various clinical settings. Published 2015. This article is a U.S. Government work and is in the public domain in the USA.

Sensitivity analysis for missing data in regulatory submissions.

The National Research Council Panel on Handling Missing Data in Clinical Trials recommended that sensitivity analyses have to be part of the primary reporting of findings from clinical trials. Their specific recommendations, however, seem not to have been taken up rapidly by sponsors of regulatory submissions. The NRC report's detailed suggestions are along rather different lines than what has been called sensitivity analysis in the regulatory setting up to now. Furthermore, the role of sensitivity analysis in regulatory decision-making, although discussed briefly in the NRC report, remains unclear. This paper will examine previous ideas of sensitivity analysis with a view to explaining how the NRC panel's recommendations are different and possibly better suited to coping with present problems of missing data in the regulatory setting. It will also discuss, in more detail than the NRC report, the relevance of sensitivity analysis to decision-making, both for applicants and for regulators. Published 2015. This article is a U.S. Government work and is in the public domain in the USA.

Fallacies of last observation carried forward analyses.

Last observation carried forward is a common statistical approach to the analysis of longitudinal repeated measures data where some follow-up observations may be missing. In a last observation carried forward analysis, a missing follow-up visit value is replaced by (imputed as) that subject's previously observed value, that is, the last observation is carried forward. The combination of the observed and imputed data is then analyzed as though there were no missing data. There have been numerous statistical demonstrations of faults of this approach. In 2012, the National Research Council's Panel on Handling Missing Data in Clinical Trials issued a report that raised concerns with the use of last observation carried forward and described alternative methods that offer greater statistical validity. Nevertheless, the method persists, and its use is rampant. A search of the key word "last observation carried forward" using Google Scholar yielded "about 2480" published citations during 2014 alone, the overwhelming majority presenting the results of scientific studies. However, there has not been a simple explanation of the statistical deficiencies of last observation carried forward. Such a description is presented herein. A simple repeated measures model is described for quantitative observations at two times (e.g. 1 and 2 years), with complete values at 1 year that are used to impute by last observation carried forward the missing values at 2 years under the missing completely at random assumption. This results in a mixture distribution of observed and imputed values at 2 years with mean and variance that are a function of the mixture of the 1- and 2-year distributions. The expressions show that last observation carried forward is only unbiased when the distribution of the observed values at 1 year is exactly equal to the distribution of the missing values at 2 years, the latter, of course, being unknown. When the values at 2 years are not randomly missing, no simple expressions for the mean and variance of the mixture distribution are possible without additional unverifiable assumptions. All analyses using last observation carried forward are of questionable veracity, if not being outright specious (definition: appearing to be true but actually false). It is hoped that future studies will make a more vigorous attempt to minimize the amount of missing data and that more valid statistical analyses will be employed in cases where missing data occur. Last observation carried forward should not be employed in any analyses.

Enabling network inference methods to handle missing data and outliers.

BackgroundThe inference of complex networks from data is a challenging problem in biological sciences, as well as in a wide range of disciplines such as chemistry, technology, economics, or sociology. The quantity and quality of the data greatly affect the results. While many methodologies have been developed for this task, they seldom take into account issues such as missing data or outlier detection and correction, which need to be properly addressed before network inference.ResultsHere we present an approach to (i) handle missing data and (ii) detect and correct outliers based on multivariate projection to latent structures. The method, called trimmed scores regression (TSR), enables network inference methods to analyse incomplete datasets by imputing the missing values coherently with the latent data structure. Furthermore, it substitutes the faulty values in a dataset by proper estimations. We provide an implementation of this approach, and show how it can be integrated with any network inference method as a preliminary data curation step. This functionality is demonstrated with a state of the art network inference method based on mutual information distance and entropy reduction, MIDER.ConclusionThe methodology presented here enables network inference methods to analyse a large number of incomplete and faulty datasets that could not be reliably analysed so far. Our comparative studies show the superiority of TSR over other missing data approaches used by practitioners. Furthermore, the method allows for outlier detection and correction.Electronic supplementary materialThe online version of this article (doi:10.1186/s12859-015-0717-7) contains supplementary material, which is available to authorized users.

Incorporating temporal EHR data in predictive models for risk stratification of renal function deterioration

Predictive models built using temporal data in electronic health records (EHRs) can potentially play a major role in improving management of chronic diseases. However, these data present a multitude of technical challenges, including irregular sampling of data and varying length of available patient history. In this paper, we describe and evaluate three different approaches that use machine learning to build predictive models using temporal EHR data of a patient.The first approach is a commonly used non-temporal approach that aggregates values of the predictors in the patient’s medical history. The other two approaches exploit the temporal dynamics of the data. The two temporal approaches vary in how they model temporal information and handle missing data. Using data from the EHR of Mount Sinai Medical Center, we learned and evaluated the models in the context of predicting loss of estimated glomerular filtration rate (eGFR), the most common assessment of kidney function.Our results show that incorporating temporal information in patient’s medical history can lead to better prediction of loss of kidney function. They also demonstrate that exactly how this information is incorporated is important. In particular, our results demonstrate that the relative importance of different predictors varies over time, and that using multi-task learning to account for this is an appropriate way to robustly capture the temporal dynamics in EHR data. Using a case study, we also demonstrate how the multi-task learning based model can yield predictive models with better performance for identifying patients at high risk of short-term loss of kidney function.

Bayesian Functional Data Clustering for Temporal Microarray Data

We propose a Bayesian procedure to cluster temporal gene expression microarray profiles, based on a mixed-effect smoothing-spline model, and design a Gibbs sampler to sample from the desired posterior distribution. Our method can determine the cluster number automatically based on the Bayesian information criterion, and handle missing data easily. When applied to a microarray dataset on the budding yeast, our clustering algorithm provides biologically meaningful gene clusters according to a functional enrichment analysis.

Predicting Hospital Mortality Among Injured Children Using a National Trauma Database

The purpose of this study was to develop a model that accurately predicts mortality among injured children based on components of the initial patient evaluation and that is generalizable to diverse acute care settings. Important predictive variables obtained in an emergency setting are frequently missing in even large national databases, limiting their effectiveness for developing predictions. In this study, a model predicting pediatric trauma mortality was developed using a national database and methods to handle missing data that may avoid biases that can occur restricting analyses to complete cases. Records of pediatric patients included in the National Pediatric Trauma Registry (NPTR) between 1996 and 1999 were used as a training set in a logistic regression model to predict hospital mortality using vital signs, Glasgow Coma Scale (GCS) score, and intubation status. Multiple imputation was applied to handle missing data. The model was tested using independent data from the NPTR and National Trauma Data Bank (NTDB). Complete case analysis identified only GCS-eye and intubation status as predictors of mortality. A model based on complete case analysis had good discrimination (c-index = 0.784) and excellent calibration (Hosmer-Lemeshow c-statistic, 6.8) (p > 0.05). Using multiple imputation, three additional predictors of mortality (systolic blood pressure, pulse, and GCS-motor) were identified and improved model performance was observed. The model developed using multiple imputation had excellent discrimination (c-index, 0.947-0.973) in both test datasets. Calibration was better in the NPTR testing set than in the NTDB (Hosmer-Lemeshow c-statistic, 9.2 for NPTR [p > 0.05] and 258.2 for NTDB [p < 0.05]). At a probability cutoff that minimized misclassification in the training set, the false-negative and false-negative rates of the model were better than those obtained with either the Revised Trauma Score (RTS) or Pediatric Trauma Score using data from the NPTR testing set. Although the false-positive rates were lower with the RTS using data from the NTDB, the false-negative rates of the proposed model and the RTS were similar in this test dataset. Using multiple imputation to handle missing data, a model predicting pediatric trauma mortality was developed that compared favorably with existing trauma scores. Application of these methods may produce predictive trauma models that are more statistically reliable and applicable in clinical practice.

Associations between health-related quality of life and demographics and health risks. Results from Rhode Island's 2002 behavioral risk factor survey.

BackgroundHealth-Related Quality of Life (HRQOL) has received much attention in recent years. HRQOL indicators have been used to track population trends, identify health disparities, and monitor progress in achieving national health objectives for 2010. Prior studies have examined health risks and HRQOL at the national level as well as at the state level. This paper examines multiple indicators of HRQOL by demographic characteristics and selected health behaviors for Rhode Island adults.MethodsData from Rhode Island's 2002 Behavioral Risk Factor Surveillance System (BRFSS), a random digit dialled telephone survey, were used for this study. The state wide sample contained a total of 3,843 respondents ages 18 and older. Multiple Imputation (MI) was applied to handle missing data, and data were modelled for each of 10 HRQOL indicators using multivariable logistic regression.ResultsBy examining HRQOL through a multivariable approach we identified the strongest predictors for multiple indicators of poor HRQOL as well as predictors for specific indicators of poor HRQOL. Predictors for multiple indicators of poor HRQOL were: disability, inability to work, unemployment, lower income, lack of exercise, asthma, and smoking (specifically associated with poor mental health).ConclusionUsing multiple measures of HRQOL can help to assess the burden of poor health in a population, identify subgroups with unmet HRQOL needs, inform the development of targeted interventions, and monitor changes in a population's HRQOL over time. Use of these HRQOL measures in longitudinal and intervention studies is needed to increase our understanding of the causal relationships between demographics, health risk behaviors, and HRQOL.