Variance Estimation Research Articles

Smokeless tobacco and periodontitis: A systematic review with meta-analysis.

The present systematic review with meta-analysis aimed to investigate the global association between smokeless tobacco (SLT) use and periodontitis, considering significant effect size variation based on the income levels of countries. We searched seven databases to identify studies that assessed the prevalence of periodontitis in adult SLT users compared to non-users. The quality of studies was evaluated using the 10-item risk-of-bias tool, and publication bias was addressed through the trim-and-fill method. Sensitivity analysis utilized the leave-one-out approach. Meta-analysis and meta-regression, stratified by country income, SLT type, and smoking status, employed robust variance estimation. From an initial pool of 484 studies, 29 studies met the selection criteria and were subjected to qualitative synthesis. Subsequently, data from 19 studies were included in the meta-analysis. SLT users exhibited a nearly threefold greater likelihood of periodontitis compared to non-users (OR = 2.99; 95% CI: 2.10, 4.27; p < .01). The pooled estimate did not vary significantly based on the type of SLT used or concurrent smoking. However, the odds of periodontitis varied according to the economic level of the country; the pooled estimate was higher in high-income countries (OR = 1.69; 95% CI: 1.20, 2.37; p < .01) and even higher in lower-middle-income and lower-income countries (OR = 3.91; 95% CI: 2.66, 5.77; p < .01). Smokeless tobacco users have a higher likelihood of developing periodontitis. This study underscores global disparities in the SLT-periodontitis relationship, highlighting the need for targeted interventions, particularly in economically challenged areas where SLT use is largely unregulated.

The intersection between migration, HIV, and contraceptive use in Uganda: a cross-sectional population-based study.

Low use of modern methods of contraception has been linked to HIV seropositivity and to migration, but few studies have evaluated the intersection of both risk factors with contraceptive use. We analyzed cross-sectional data from sexually active female participants aged 15 to 49years in the Rakai Community Cohort Study (RCCS) between 2011 and 2013. The RCCS is an open population-based census and individual survey in south-central Uganda. Recent in-migrants (arrival within approximately 1.5years) into RCCS communities were identified at time of household census. The primary outcome was unsatisfied demand for a modern contraceptive method (injectable, oral pill, implant, or condom), which was defined as non-use of a modern contraceptive method among female participants who did not want to become pregnant in the next 12months. Poisson regression models with robust variance estimators were used to identify associations and interactions between recent migration and HIV serostatus on unsatisfied contraceptive demand. There were 3,417 sexually active participants with no intention of becoming pregnant in the next year. The mean age was 30 (± 8) years, and 17.3% (n = 591) were living with HIV. Overall, 43.9% (n = 1,500) were not using any modern contraceptive method. Recent in-migrants were somewhat more likely to have unsatisfied contraceptive demand as compared to long-term residents (adjusted prevalence risk ratio [adjPRR] = 1.14; 95% confidence interval [95%CI]: 1.02-1.27), whereas participants living with HIV were less likely to have unsatisfied contraceptive demand relative to HIV-seronegative participants (adjPRR = 0.80; 95%CI = 0.70-0.90). When stratifying on migration and HIV serostatus, we observed the highest levels of unsatisfied contraceptive demand among in-migrants living with HIV (48.7%); however, in regression analyses, interaction terms between migration and HIV serostatus were not statistically significant. Unsatisfied contraceptive demand was high in this rural Ugandan setting. Being an in-migrant, particularly among those living with HIV, was associated with higher unsatisfied contraceptive demand.

Invited Commentary: Mixing multiple imputation and bootstrapping for variance estimation.

In 2023, Martinez et al. examined trends in the inclusion, conceptualization, operationalization and analysis of race and ethnicity among studies published in US epidemiology journals. Based on a random sample of papers (N=1,050) published from 1995-2018, the authors describe the treatment of race, ethnicity, and ethnorace in the analytic sample (N=414, 39% of baseline sample) over time. Between 32% and 19% of studies in each time stratum lacked race data; 61% to 34% lacked ethnicity data. The review supplies stark evidence of the routine omission and variability of measures of race and ethnicity in epidemiologic research. Informed by public health critical race praxis (PHCRP), this commentary discusses the implications of four problems the findings suggest pervade epidemiology: 1) a general lack of clarity about what race and ethnicity are; 2) the limited use of critical race or other theory; 3) an ironic lack of rigor in measuring race and ethnicity; and, 4) the ordinariness of racism and white supremacy in epidemiology. The identified practices reflect neither current publication guidelines nor the state of the knowledge on race, ethnicity and racism; therefore, we conclude by offering recommendations to move epidemiology toward more rigorous research in an increasingly diverse society.

Temporal Relationship Between Serum Neurofilament Light Chain and Radiologic Disease Activity in Patients With Multiple Sclerosis.

Serum neurofilament light chain (sNfL) levels correlate with multiple sclerosis (MS) disease activity, but the dynamics of this correlation are unknown. We evaluated the relationship between sNfL levels and radiologic MS disease activity through monthly assessments during the 24-week natalizumab treatment interruption period in RESTORE (NCT01071083). In the RESTORE trial, participants with relapsing forms of MS who had received natalizumab for ≥12 months were randomized to either continue or stop natalizumab and followed with MRI and blood draws every 4 weeks to week 28 and again at week 52 The sNfL was measured, and its dynamics were correlated with the development of gadolinium-enhancing (Gd+) lesions. Log-linear trend in sNfL levels were modeled longitudinally using generalized estimating equations with robust variance estimator from baseline to week 28. Of 175 patients enrolled in RESTORE, 166 had serum samples for analysis. Participants with Gd+ lesions were younger (37.7 vs 43.1, p = 0.001) and had lower Expanded Disability Status Scale scores at baseline (2.7 vs 3.4, p = 0.017) than participants without Gd+ lesions. sNfL levels increased in participants with Gd+ lesions (n = 65) compared with those without (n = 101, mean change from baseline to maximum sNfL value, 12.1 vs 3.2 pg/mL, respectively; p = 0.003). As the number of Gd+ lesions increased, peak median sNfL change also increased by 1.4, 3.0, 4.3, and 19.6 pg/mL in the Gd+ lesion groups of 1 (n = 12), 2-3 (n = 18), 4-9 (n = 21), and ≥10 (n = 14) lesions, respectively. However, 46 of 65 (71%) participants with Gd+ lesions did not increase above the 95th percentile threshold of the group without Gd+ lesions. The initial increase of sNfL typically trailed the first observation of Gd+ lesions, and the peak increase in sNfL was a median [interquartile range] of 8 [0, 12] weeks after the first appearance of the Gd+ lesion. Although sNfL correlated with the presence of Gd+ lesions, most participants with Gd+ lesions did not have elevations in sNfL levels. These observations have implications for the use and interpretation of sNfL as a biomarker for monitoring MS disease activity in controlled trials and clinical practice.

The Alternating Direction Method of Multipliers for Sufficient Dimension Reduction

The minimum average variance estimation (MAVE) method has proven to be an effective approach to sufficient dimension reduction. In this study, we apply the computationally efficient optimization algorithm named alternating direction method of multipliers (ADMM) to a particular approach (MAVE or minimum average variance estimation) to the problem of sufficient dimension reduction (SDR). Under some assumptions, we prove that the iterative sequence generated by ADMM converges to some point of the associated augmented Lagrangian function. Moreover, that point is stationary. It also presents some numerical simulations on synthetic data to demonstrate the computational efficiency of the algorithm.

Random Forest Spatial Interpolation Techniques for Crop Yield Estimation at District Level

General Crop Estimation Surveys (GCES) based on Crop Cutting Experiments (CCEs) are conducted for estimation of crop yield following random sampling approach for almost all major crops. About 13 lakh CCEs are conducted every year which has now increased rapidly due to the Pradhan Mantri Fasal Bima Yojana (PMFBY) which is yield based insurance scheme. As suggested by Ministry of Agriculture and Farmers’ Welfare (MoA&amp;FW), this number needs to be reduced drastically by developing sampling procedures based on the use of advanced technologies and advanced survey techniques for crop yield estimation. In this study, an attempt has been made to develop crop yield estimation procedures using Random Forest Spatial Interpolation (RFSI) technique including the spatial variables like spatial distance and nearest neighbours as covariates. RFSI is one of the most adaptable and user-friendly interpolation techniques, as well as one of the fastest in large training datasets. Estimates of yield of wheat were obtained for all the six tehsils of Barabanki district using the estimator under stratified two stage sampling technique. The district level estimates were also obtained by pooling area under wheat crop in each tehsil along with the district level estimate of crop yield, estimate of variance, estimate of standard error (SE) and percentage SE (%SE) of these estimates were also computed in order to make comparison. The results of this study suggest that the estimates derived using RFSI are comparable to kriging and superior to inverse distance weighting (IDW) for the prediction of yield at unknown locations using distance and nearest neighbours.

Almost Sure Central Limit Theorem for Error Variance Estimator in Pth-Order Nonlinear Autoregressive Processes

In this paper, under some suitable assumptions, using the Taylor expansion, Borel–Cantelli lemma and the almost sure central limit theorem for independent random variables, the almost sure central limit theorem for error variance estimator in the pth-order nonlinear autoregressive processes with independent and identical distributed errors was established. Four examples, first-order autoregressive processes, self-exciting threshold autoregressive processes, threshold-exponential AR progresses and multilayer perceptrons progress, are given to verify the results.

Calibration Estimation of Population Total in Two-Stage Sampling Design under unavailability of Population Level Auxiliary Information for the selected PSUs

Calibration approach is a popular technique in sample surveys which incorporates auxiliary information in the estimation process assuming that population aggregates of auxiliary variable are available. Many often under two-stage sampling design, such population aggregates of auxiliary variable, i.e., population mean or total are unavailable and under such situations, estimation of population total has been limited to the use of two phase sampling. Therefore, in the present study, efficient estimator of population total is developed under two-stage sampling design when population aggregates of auxiliary variable are unavailable for the selected psu’s. The calibrated estimator is developed using the information on known population aggregates of additional auxiliary variable which is less linearly related to the study variable through two step calibration approach. The approximate variance and variance estimator of the proposed calibrated estimator has also been developed. Empirical evaluations using both real and simulated data shows the superior performance of the developed calibrated estimator in comparison to the existing estimators.

Statistical methods to control for confounders in rare disease settings that use external control

ABSTRACT In the drug development for rare disease, the number of treated subjects in the clinical trial is often very small, whereas the number of external controls can be relatively large. There is no clear guidance on choosing an appropriate statistical method to control baseline confounding in this situation. To fill this gap, we conduct extensive simulations to evaluate the performance of commonly used matching and weighting methods as well as the more recently developed targeted maximum likelihood estimation (TMLE) and cardinality matching in small sample settings, mimicking the motivating data from a pediatric rare disease. Among the methods examined, the performance of coarsened exact matching (CEM) and TMLE are relatively robust under various model specifications. CEM is only feasible when the number of controls far exceeds the number of treated, whereas TMLE has better performance with less extreme treatment allocation ratios. Our simulations suggest bootstrap is useful for variance estimation in small samples after matching.

Propensity score matching for estimating a marginal hazard ratio.

Propensity score matching is commonly used to draw causal inference from observational survival data. However, its asymptotic properties have yet to be established, and variance estimation is still open to debate. We derive the statistical properties of the propensity score matching estimator of the marginal causal hazard ratio based on matching with replacement and a fixed number of matches. We also propose a double-resampling technique for variance estimation that takes into account the uncertainty due to propensity score estimation prior to matching.

An Estimator of Population Variance Using Multi-Auxiliary Information

In this article, an estimator of population variance using multi-auxiliary information has been proposed. It is seen that under certain conditions, the proposed estimator is less biased and more efficiency than existing estimators. Theoretical results are supported numerically. Moreover, a simulation study also has been made.

Oil Quality Prediction in Olive Oil by Near-Infrared Spectroscopy: Applications in Olive Breeding

The oxidative stability index (OSI) and fatty acid (FA) composition of extra virgin olive oils (EVOOs) are key parameters in the characterization of new varieties in breeding programs. Their determination through traditional methods (Rancimat and gas chromatography, respectively) is expensive and time-consuming. Therefore, there is a need to develop rapid and cost-effective analytical procedures. This study aimed to evaluate the potential use of near-infrared spectroscopy (NIRS) for analyzing OSI and FA composition in EVOOs. A total of 318 samples sourced from different origins were evaluated using both FT-NIR MPA and MicroNIR instruments in transmittance mode, with wavelengths ranging from 1100 to 2500 nm and 908 to 1676 nm, respectively. Different accuracies were obtained in the models developed for the different evaluated traits, with simpler models (using a lower number of latent variables) for the MPA analyzer in all cases. Additionally, consistent results between instruments for the partitioning of the variance and heritability estimation, and the reliable ranking of genotypes were obtained from one of the sample sets tested. In summary, models derived from PLS regression using spectroscopic data of both instruments demonstrated promising results in determining these EVOO traits, facilitating their evaluation and selection of genotypes, particularly in breeding programs.

Negative variance components and intercept-slope correlations greater than one in magnitude: How do such "non-regular" random intercept and slope models arise, and what should be done when they do?

Statistical models with random intercepts and slopes (RIAS models) are commonly used to analyze longitudinal data. Fitting such models sometimes results in negative estimates of variance components or estimates on parameter space boundaries. This can be an unlucky chance occurrence, but can also occur because certain marginal distributions are mathematically identical to those from RIAS models with negative intercept and/or slope variance components and/or intercept-slope correlations greater than one in magnitude. We term such parameters "pseudo-variances" and "pseudo-correlations," and the models "non-regular." We use eigenvalue theory to explore how and when such non-regular RIAS models arise, showing: (i) A small number of measurements, short follow-up, and large residual variance increase the parameter space for which data (with a positive semidefinite marginal variance-covariance matrix) are compatible with non-regular RIAS models. (ii) Non-regular RIAS models can arise from model misspecification, when non-linearity in fixed effects is ignored or when random effects are omitted. (iii) A non-regular RIAS model can sometimes be interpreted as a regular linear mixed model with one or more additional random effects, which may not be identifiable from the data. (iv) Particular parameterizations of non-regular RIAS models have no generality for all possible numbers of measurements over time. Because of this lack of generality, we conclude that non-regular RIAS models can only be regarded as plausible data-generating mechanisms in some situations. Nevertheless, fitting a non-regular RIAS model can be acceptable, allowing unbiased inference on fixed effects where commonly recommended alternatives such as dropping the random slope result in bias.

Hybrid class of robust type estimators for variance estimation using mean and variance of auxiliary variable

This study is completed for the estimation of unknown population variance for the variable of mean and variance of interest. To accomplish this task, a new generalized class of robust kind of variance estimators proposed utilizing known descriptives of auxiliary variable, for example, Mid-range, Hodges-Lehmann Mean, Tri-mean, deciles mean, coefficient of skewness, interquartile range, first quartile, coefficient of kurtosis, semi-interquartile average, inter decile range and Mean, etc. These conventional measures of auxiliary variable improve the accuracy of the suggested class under simple random sampling without replacement (SRSWOR) scheme. The properties such as the bias, mean square errors (MSE), and least MSE of the suggested class are derived up to first order of approximation. The superiority conditions of the developed class of estimators over existing estimators are also made out theoretically. Finally, numerical representation is also completed for the motivations behind the article. The usual variance estimator is considered as a benchmark for comparing all considered estimators in numerical illustration. The results have been indicated that the suggested class is performing better than the usual variance estimator and all other thoughts about existing estimators.

Prenatal ochratoxin A exposure, birth outcomes and infant growth in rural Burkina Faso: a human biomonitoring sub-study from the MISAME-III trial

Mycotoxin exposure during pregnancy has been associated with adverse birth outcomes and poor infant growth in low- and middle-income countries. We assessed multiple biomarkers and metabolites of exposure to mycotoxins during pregnancy and their associations with birth outcomes and infant growth in 305 pregnant participants, between 30 and 34 completed weeks of gestation, in rural Burkina Faso. In this study, whole blood microsamples were analyzed for mycotoxin concentrations using ultra-performance liquid chromatography coupled to tandem mass spectrometry. Unadjusted and adjusted associations between mycotoxin exposure, and birth outcomes and infant growth at 6 months were estimated using linear regression models for continuous outcomes and linear probability models with robust variance estimation for binary outcomes. Infant growth trajectories from birth to 6 months were compared by exposure status using mixed-effects models with random intercept for the individual infant and random slope for the infant’s age. Ochratoxin A (OTA) exposure was detected in 50.8% of the study participants, with aflatoxin G1, aflatoxin M1, cyclopiazonic acid, deoxynivalenol and T-2-toxin being detected in the range of 0.33% and 2.31% of the population. We found no statistically significant ( p ≥ 0.05) associations between OTA exposure, and birth outcomes and infant growth. Despite this, the findings indicate a significant presence of ochratoxin A among pregnant participants. Public health policies and nutrition-sensitive interventions must ensure that OTA exposure is reduced in Burkina Faso.

Variance estimation for sequential Monte Carlo algorithms: A backward sampling approach

Variance estimation for sequential Monte Carlo algorithms: A backward sampling approach

VACCINE ADHERENCE AND ADVERSE EVENTS OF THE SARS-COV VACCINE IN PATIENTS WITH INFLAMMATORY BOWEL DISEASE

to assess adherence to and the adverse effects of the SARS-COV vaccine in patients with Inflammatory Bowel Disease (IBD). This is an observational, analytical, cross-sectional study. Sociodemographic and clinical data, SARS-COV vaccine Data, medications for IBD with use during the vaccination period, and adverse events during the vaccination period were collected. Carried out Logistic regressions with robust variance estimation to estimate the Odds Ratio with the respective 95% Confidence Intervals (95%CI) to assess the factors associated with non-serious adverse effects following vaccine doses as outcome variables. 194 patients participated, with vaccine compliance of 78.3% for three doses of any vaccine (n=152). Local symptoms and mild systemic symptoms predominated, regardless of the type of vaccine. The first dose of the SARS-COV vaccine with AstraZeneca had a higher percentage of patients with vaccine symptoms. AstraZeneca vaccine increased the chance of non-serious adverse effects in IBD patients by 2.65 times (95% CI: 1.38-5.08; p=0.003), regardless of age, gender, physical activity, excess weight, use of disease-modifying drugs, immunobiological and corticosteroids. CoronaVac vaccine was associated with asymptomatic patients at the first dose and reduced the chance of adverse effects by 0.28 times (OR: 0.284; 95%CI: 0.13-0.62; p=0.002). Local symptoms and mild systemic symptoms predominated, regardless of the type of vaccine. Using CoronaVac in the first dose reduced the chances of adverse effects, while AstraZeneca increased the risk of adverse effects.

Optimizing Variance Estimation in Stratified Random Sampling through a Log-Type Estimator for Finite Populations

In this research, a logarithmic-type estimator was formulated for estimating the finite population variance in stratified random sampling. By ensuring that the sampling process is symmetrically conducted across the population, biases can be minimized, and the sample is more likely to be representative of the population as a whole. We conducted a comprehensive numerical study and simulation study to evaluate the performance of the proposed estimator. The mean squared error values were computed for both our proposed estimator and several existing ones, including the standard unbiased variance estimator, difference-type estimator, and other considered estimators. The results of the numerical study and simulation study demonstrated that the proposed log-type estimator outperforms the other considered estimators in terms of MSE and percentage relative efficiency. Graphical representations of the results are also provided to illustrate the efficiency of the proposed estimator. Based on the findings of this study, we conclude that the proposed log-type estimator is a valuable addition to the existing literature on variance estimation in stratified random sampling. It provides a more efficient and accurate estimate of the population variance, which can be beneficial for various statistical applications.

Distributional imputation for the analysis of censored recurrent events.

Longitudinal clinical trials for which recurrent events endpoints are of interest are commonly subject to missing event data. Primary analyses in such trials are often performed assuming events are missing at random, and sensitivity analyses are necessary to assess robustness of primary analysis conclusions to missing data assumptions. Control-based imputation is an attractive approach in superiority trials for imposing conservative assumptions on how data may be missing not at random. A popular approach to implementing control-based assumptions for recurrent events is multiple imputation (MI), but Rubin's variance estimator is often biased for the true sampling variability of the point estimator in the control-based setting. We propose distributional imputation (DI) with corresponding wild bootstrap variance estimation procedure for control-based sensitivity analyses of recurrent events. We apply control-based DI to a type I diabetes trial. In the application and simulation studies, DI produced more reasonable standard error estimates than MI with Rubin's combining rules in control-based sensitivity analyses of recurrent events.

Awareness of having hypertension, diabetes and dyslipidaemia among US adults: The 2011-2018 NHANES data.

This study aimed to investigate awareness of having hypertension, diabetes and dyslipidaemia and their associated factors among US adults. Data from the National Health and Nutrition Examination Survey, including 21,399 adults aged ⩾20 years (pregnant women excluded) collected between 2011 and 2018, were used. Blood pressure was measured using a Baumanometer calibrated mercury true gravity wall model sphygmomanometer. Serum total cholesterol levels were measured using enzymatic assays. The percentage of haemoglobin A1C (HbA1c), which reflects long-term blood glucose levels, was measured and used to identify diabetes. Participants self-reported whether they were told by a doctor that they have hypertension, dyslipidaemia and diabetes. Awareness was defined as alignment between objective and self-reported measures for having the conditions. Sampling weights and the Taylor series linearisation variance estimation method were used in the analyses. The findings showed that 64.06% of people with hypertension, 54.71% of those with dyslipidaemia and 78.40% of those with diabetes were aware of having the respective condition. Age, sex and health insurance were associated with awareness of having all three conditions, but marital status was not associated with any outcome. Weight status was associated with awareness of having hypertension and dyslipidaemia, whereas ethnicity was associated with awareness of having hypertension and diabetes. Relative family income was only associated with awareness of having hypertension. Large proportions of US adults with hypertension, dyslipidaemia and diabetes are not aware of having the conditions. Interventions targeting groups at higher risk of being unaware of these conditions are needed.