Dose Response Analysis Using Robust Covariance Estimation

Dementia frequently goes undetected in community settings, particularly among socially disadvantaged populations. Here, we estimated the prevalence of underdiagnosed dementia across diverse sociodemographic determinants of health in the Health and Aging Brain Study-Health Disparities (HABS-HD), a community-based cohort of adults recruited through community outreach in Fort Worth, Texas. We estimated age-specific probabilities of underdiagnosis using Poisson regression models with a log link, including age and sex as covariates. Robust (sandwich) variance estimators were used to obtain standard errors and 95% confidence intervals (CI). Group differences or trends for continuous measures were assessed using robust variance estimates. The prevalence of underdiagnosed dementia was higher among individuals without physician access (98.1% vs. 78.1%, p<.0001), non-English speakers (97.9% vs. 76.8%, p<.0001), and the uninsured (91.5% vs. 79.5%, p=.03). Black and Hispanic participants also showed higher prevalence (85.8% and 90.9%) compared to non-Hispanic White participants (64.9%; p=.02 and p=.002, respectively). Each additional year of education was associated with a 2.5% lower risk of underdiagnosis (p<.0001). No differences were observed by sex, marital status, income or social support. Our results highlight that several sociodemographic factors contribute to the likelihood of living with undiagnosed dementia.

In his article in the March 1948 number of the Tournai of Experimental Education, H. L. Stein1 suggested how the problem of establishing meas* ures of central tendency and variability for a school system could be met. The solution offer ed for finding a measure of central tendency is of practical interest, although an indication of the extent of the bias caused by heterogeneity of variance in the schools would have been wel come. However, the solution given for an un biased estimate of population variance is prob ably not unbiased. The formula quoted from Hoel gives an unbiased estimate of the within schools variance, and since this variance has been shown by Stein to be heterogeneous, i.e. varying by an amount greater than would be expected by chance, there is little point in es timating it. A more exact procedure for ob taining an unbiased estimate of the population variance is given below; but all such proced ures must ignore the heterogeneity of school variances.

Several studies have evaluated the effects of the dietary application of exogenous alpha-amylase preparations (AMA) as a strategy to increase total tract starch digestibility (TTSD) and milk yield (MY) in dairy cows, but the results have been inconsistent. Thus, the objective of this study was to evaluate the effects of the dietary application of AMA on the performance, digestibility, and rumen fermentation of lactating dairy cows using a meta-analytic method. A total of 18 peer-reviewed manuscripts (N = 32 treatment comparisons) from 2003 to 2019 were systematically identified following the PRISMA method. The weighted raw mean differences between dietary AMA and control treatments were compared with a robust variance estimation. Likewise, diet characteristics like crude protein (CP) content, NDF content, starch content, days in milk (DIM), experimental design (Latin square and continuous), and AMA dose (0 to 732 Kilo Novo units [KNU]/kg TMR) were used as covariates in a meta-regression, subgrouping, and dose-response analysis. Compared to the control, dietary AMA increased (P < 0.05) DM digestibility (69.32% vs. 68.30%), TTSD (94.62% vs. 94.10%), milk protein concentration and yield (3.11% vs. 3.08%; 1.14 vs. 1.10 kg/d) and tended to increase (P = 0.09) fat-corrected milk (35.96 vs. 35.10 kg/d), but no effects were observed on DM intake (22.99 vs. 22.90 kg/d) and feed efficiency (1.50 vs. 1.48). Dietary AMA tended (P = 0.10) to reduce rumen pH (6.27 vs. 6.30). Both the enzyme dose and DIM strongly influenced (P < 0.05) the effects of AMA on digestibility and performance. The dose-response analysis revealed that feeding 600 KNU/kg to high-producing early lactation (< 70 DIM) dairy cows increased FCM and milk protein. Accounting for the type of experimental design was associated with a lower between-studies-variance among comparisons. Overall, this meta-analysis supports the hypothesis that dietary AMA supplementation is associated with a better lactational performance in dairy cows. However, these effects are only suitable for high-producing early lactation dairy cows.

BackgroundCross-sectional studies with binary outcomes analyzed by logistic regression are frequent in the epidemiological literature. However, the odds ratio can importantly overestimate the prevalence ratio, the measure of choice in these studies. Also, controlling for confounding is not equivalent for the two measures. In this paper we explore alternatives for modeling data of such studies with techniques that directly estimate the prevalence ratio.MethodsWe compared Cox regression with constant time at risk, Poisson regression and log-binomial regression against the standard Mantel-Haenszel estimators. Models with robust variance estimators in Cox and Poisson regressions and variance corrected by the scale parameter in Poisson regression were also evaluated.ResultsThree outcomes, from a cross-sectional study carried out in Pelotas, Brazil, with different levels of prevalence were explored: weight-for-age deficit (4%), asthma (31%) and mother in a paid job (52%). Unadjusted Cox/Poisson regression and Poisson regression with scale parameter adjusted by deviance performed worst in terms of interval estimates. Poisson regression with scale parameter adjusted by χ2 showed variable performance depending on the outcome prevalence. Cox/Poisson regression with robust variance, and log-binomial regression performed equally well when the model was correctly specified.ConclusionsCox or Poisson regression with robust variance and log-binomial regression provide correct estimates and are a better alternative for the analysis of cross-sectional studies with binary outcomes than logistic regression, since the prevalence ratio is more interpretable and easier to communicate to non-specialists than the odds ratio. However, precautions are needed to avoid estimation problems in specific situations.

Comparison of SYBR green I and lactate dehydrogenase antimalarial in vitro assay in Plasmodium falciparum field isolates

Finite-sample improved confidence intervals based on the estimating equation theory for the modified Poisson and least-squares regressions

Intercontinental Deceased-donor Kidney Transplantation in Dutch Antilleans: Favorable Outcomes Despite Prolonged Ischemia.

A simple variance estimator for product-limit survival functions is demonstrated for survival times with nested errors. Such data arise whenever survival times are observed within clusters of related observations. Greenwood's formula, which assumes independent observations, is not appropriate in this situation. A robust variance estimator is developed using Taylor series linearized values and the between-cluster variance estimator commonly used in multi-stage sample surveys. A simulation study shows that the between-cluster variance estimator is approximately unbiased and yields confidence intervals that maintain the nominal level for several patterns of correlated survival times. The simulation study also shows that Greenwood's formula underestimates the variance when the survival times are positively correlated within a cluster and yields confidence intervals that are too narrow. Extension to life table methods is also discussed.

Persistent infection with high-risk human papillomavirus (HR-HPV) is a major cause of cervical precancerous lesions and cervical cancer. This study aimed to evaluate the effectiveness of Qingdu Suppository (QDS) in promoting HPV clearance and to identify predictors of treatment response. This was a prospective observational cohort with retrospective analysis. A total of 60 patients with persistent HR-HPV infection were included (30 QDS and 30 non-QDS). The primary endpoint was HPV clearance, defined as 2 consecutive negative PCR results ≥6 months apart. Intention-to-treat (ITT, n=66, with lost-to-follow-up treated as persistent infection) and per-protocol (PP, n=60) analyses were performed. Inverse probability of treatment weighting (IPTW) was applied, and robust (sandwich) variance estimators were used to obtain valid confidence intervals. In the IPTW-weighted analysis, QDS was associated with higher HPV clearance than non-QDS (55.2% vs. 3.4%, p<.001). HPV clearance was significantly higher in the QDS group than in the non-QDS group (55.2% vs. 3.4%, p<.001). QDS remained an independent predictor of clearance (adjusted OR=33.1, 95% CI: 4.5-243.6, p=.001), with consistent benefits across subgroups. Using a strict endpoint of 2 consecutive negatives, QDS was associated with higher HPV clearance. Results were consistent across analytic approaches but should be interpreted with caution, given the limited sample size and nonrandomized design. QDS may represent a promising adjunctive therapy, but its efficacy and long-term benefits require confirmation in larger multicenter randomized controlled trials.

A comprehensive review of group level model performance in the presence of heteroscedasticity: Can a single model control Type I errors in the presence of outliers?

Introduction: Inflammation plays a important role in determining the prognosis of heart failure. Recent studies have demonstrated that both established and novel inflammatory biomarkers—such as the neutrophil-to-lymphocyte ratio (NLR), monocyte-to-lymphocyte ratio (MLR), platelet-to-lymphocyte ratio (PLR), Systemic Immune-Inflammation Index (SII), and, in particular, the Systemic Inflammation Response Index (SIRI)—serve as important predictors of mortality in heart failure patients. Nonetheless, data from developing countries like Vietnam remain limited. Hypothesis: Inflammatory markers are associated with 30-day all-cause mortality in patients hospitalized with acute heart failure (AHF). Methods: We performed a prospective cohort study at Can Tho Central General Hospital—a tertiary care center in Vietnam—from May 2024 through April 2025. Consecutive adults (≥18 years) admitted with acute heart failure (NT-proBNP &gt;300 pg/mL) who provided informed consent were enrolled, while those with active infection, immunosuppressant use within the prior three months, chronic liver disease, or active malignancy were excluded. Survivors were followed for 30 days after discharge, with all-cause mortality as the primary endpoint. Relative risks were derived from a modified Poisson log-linear regression using a robust (sandwich) variance estimator, adjusting for baseline demographics and comorbidities. Results: Among 411 AHF patients (mean age 69.6 ± 12.6 years; 47.4 % men), 262 (63.7 %) completed 30-day follow-up, during which 56 deaths occurred (all-cause mortality 21.4 %). Non-survivors were older, carried a greater comorbidity burden, and exhibited higher median levels of NLR, MLR, PLR, SII, and SIRI than survivors. In multivariable Poisson models adjusted for age, sex, coronary artery disease, prior heart failure, hypertension, and diabetes, individuals in the highest biomarker quartile had markedly increased 30-day mortality risk compared with those in the lowest quartile: NLR (RR 3.6; 95 % CI 1.6–8.3), MLR (RR 6.1; 95 % CI 2.2–16.5), PLR (RR 2.3; 95 % CI 1.2–4.4), SII (RR 2.4; 95 % CI 1.2–4.8), and SIRI (RR 4.3; 95 % CI 1.8–10.4). Conclusion: Inflammatory markers independently predict 30-day mortality in Vietnamese AHF patients; adding them to risk models may enhance short-term prognosis, while its longer-term predictive value warrants further investigation.

&lt;p&gt;Comparing the means of independent groups is a concern when the assumptions of normal-ity and variance homogeneity are violated. Robust means modeling (RMM) was proposed as an alternative to ANOVA-type procedures when the assumptions of normality and vari-ance homogeneity are violated. The purpose of this study is to compare the Type I error and power rates of RMM to the trimmed Welch procedure. A Monte Carlo study was used to investigate RMM and the trimmed Welch procedure under several conditions of nonnor-mality and variance heterogeneity. The results suggest that the trimmed Welch provides a better balance of Type I error control and power than RMM&lt;/p&gt;

Pooling data, when justified, is advantageous for estimating the true parameter. In this paper the problem of estimating the coefficient of variation is considered when it is a priori suspected that two coefficients of variation are the same. Various estimators based on pretest and shrinkage rules are considered. A comparison through the Simulated Mean Squared Error (SMSE) criterion is carried out among various proposed estimators of the target coefficient of variation. The relative simulated efficiencies of the restricted, shrinkage restricted and shrinkage pretest estimators are studied. It is found that the proposed estimators are quite robust when the sample sizes are not too large. The result of Monte Carlo study indicates that the proposed shrinkage pretest estimator is efficient than the usual estimator in a wider range.

Results on genetic variability in some of the morphometric characters on head and face, body girths and skin folds, based on 45 MZ and 67 like-sex DZ twin pairs, are presented. The data were subjected to a method which eliminated possible biases in the estimated genetic variances that could result from heterogeneity of total variances between zygosities of the 17 head and face measurements, heterogeneity was observed for only bizygomatic diameter. Head breadth means differed between MZ and DZ twins, indicating bias in the trait's genetic variance analysis. The results indicated a significant genetic component in these morphometric traits. For 11 girth and skinfold measurements, the t'-test based on hierarchical structure of twin data, also failed to reveal any appreciable difference between the mean values of MZ and DZ twins. Heterogeneity of total variance between zygosities was observed for three skinfold measurements, ie, biceps, triceps and suprailiac. The girth measurements, however, did not reveal any heterogeneity of total variances between the zygosities. The estimates of genetic variance revealed stronger genetic component for the girths than for the skinfolds.

BackgroundThe effectiveness of malaria vector control interventions is often evaluated using cluster randomized trials (CRT) with outcomes assessed using repeated cross-sectional surveys. A key requirement for appropriate design and analysis of longitudinal CRTs is accounting for the intra-cluster correlation coefficient (ICC). In addition to exchangeable correlation (constant ICC over time), correlation structures proposed for longitudinal CRT are block exchangeable (allows a different within- and between-period ICC) and exponential decay (allows between-period ICC to decay exponentially). More flexible correlation structures are available in statistical software packages and, although not formally proposed for longitudinal CRTs, may offer some advantages. Our objectives were to empirically explore the impact of these correlation structures on treatment effect inferences, identify gaps in the methodological literature, and make practical recommendations.MethodsWe obtained data from a parallel-arm CRT conducted in Tanzania to compare four different types of insecticide-treated bed-nets. Malaria prevalence was assessed in cross-sectional surveys of 45 households in each of 84 villages at baseline, 12-, 18- and 24-months post-randomization. We re-analyzed the data using mixed-effects logistic regression according to a prespecified analysis plan but under five different correlation structures as well as a robust variance estimator under exchangeable correlation and compared the estimated correlations and treatment effects. A proof-of-concept simulation was conducted to explore general conclusions.ResultsThe estimated correlation structures varied substantially across different models. The unstructured model was the best-fitting model based on information criteria. Although point estimates and confidence intervals for the treatment effect were similar, allowing for more flexible correlation structures led to different conclusions based on statistical significance. Use of robust variance estimators generally led to wider confidence intervals. Simulation results showed that under-specification can lead to coverage probabilities much lower than nominal levels, but over-specification is more likely to maintain nominal coverage.ConclusionMore flexible correlation structures should not be ruled out in longitudinal CRTs. This may be particularly important in malaria trials where outcomes may fluctuate over time. In the absence of robust methods for selecting the best-fitting correlation structure, researchers should examine sensitivity of results to different assumptions about the ICC and consider robust variance estimators.