Quantitative Covariates Research Articles

Simultaneous Inference Using Multiple Marginal Models.

This tutorial describes single-step low-dimensional simultaneous inference with a focus on the availability of adjusted p values and compatible confidence intervals for more than just the usual mean value comparisons. The basic idea is, first, to use the influence of correlation on the quantile of the multivariate t-distribution: the higher the less conservative. In addition, second, the estimability of the correlation matrix using the multiple marginal models approach (mmm) using multiple models in the class of linear up to generalized linear mixed models. The underlying maxT-test using mmm is discussed by means of several real data scenarios using selected R packages. Surprisingly, different features are highlighted, among them: (i) analyzing different-scaled, correlated, multiple endpoints, (ii) analyzing multiple correlated binary endpoints, (iii) modeling dose as qualitative factor and/or quantitative covariate, (iv) joint consideration of several tuning parameters within the poly-k trend test, (v) joint testing of dose and time, (vi) considering several effect sizes, (vii) joint testing of subgroups and overall population in multiarm randomized clinical trials with correlated primary endpoints, (viii) multiple linear mixed effect models, (ix) generalized estimating equations, and (x) nonlinear regression models.

Microbiome confounders and quantitative profiling challengepredicted microbial targets in colorectal cancer development

Despite substantial progress in cancer microbiome research, recognizedconfounders and advances in absolute microbiome quantification remain underused;this raises concerns regarding potential spurious associations. Here we study thefecal microbiota of 589 patients at different colorectal cancer (CRC) stages andcompare observations with up to 15 published studies (4,439 patients and controlstotal). Using quantitative microbiome profiling based on 16S ribosomal RNA ampliconsequencing, combined with rigorous confounder control, we identified transit time,fecal calprotectin (intestinal inflammation) and body mass index as primarymicrobial covariates, superseding variance explained by CRC diagnostic groups.Well-established microbiome CRC targets, such as Fusobacterium nucleatum, did not significantly associate with CRCdiagnostic groups (healthy, adenoma and carcinoma) when controlling for thesecovariates. In contrast, the associations of Anaerococcusvaginalis, Dialister pneumosintes,Parvimonas micra, Peptostreptococcus anaerobius, Porphyromonasasaccharolytica and Prevotellaintermedia remained robust, highlighting their future targetpotential. Finally, control individuals (age 22–80 years, mean57.7 years, standard deviation 11.3) meeting criteria for colonoscopy (forexample, through a positive fecal immunochemical test) but without colonic lesionsare enriched for the dysbiotic Bacteroides2 enterotype, emphasizing uncertainties indefining healthy controls in cancer microbiome research. Together, these resultsindicate the importance of quantitative microbiome profiling and covariate controlfor biomarker identification in CRC microbiome studies.

Activity behaviors and relative changes in activity patterns recorded by precision technology were associated with diarrhea status in individually housed calves

The objective of this case-control study was to quantify if there was an association of daily activity behaviors and relative changes in activity patterns (lying time, lying bouts, step count, activity index) with diarrhea status in preweaned dairy calves. Individually housed calves sourced from auction were health scored daily for signs of diarrhea (fecal consistency loose or watery for 2 consecutive d) for the 28 d after arrival. Calves with diarrhea were pair matched to healthy controls (n = 13 matched by arrival date, arrival weight, and diagnosis d to diarrheic calves). Mixed linear regression models were used to evaluate the association of diarrhea status, and the diarrhea status by day interaction with activity behaviors (d -3 to d 4) and relative changes in activity patterns (d -3 to d 4) relative to diagnosis of a diarrhea bout. The serum Brix % at arrival and daily THI from the calf barn were explored as quantitative covariates, with d as a repeated measure. The baseline for relative changes in activity patterns was set at 100% on d 0. Diarrheic calves were less active; they averaged less steps (119.1 ± 18.81 steps/d) than healthy calves (227.4 ± 18.81 steps/d, LSM ± SEM). Diarrheic calves also averaged lower activity indices (827.34 ± 93.092 daily index) than healthy calves (1396.32 ± 93.092 daily index). There was also a diarrhea status by day interaction for lying time on d -3, with diarrheic calves spending more time lying (20.80 ± 0.300 h/d) than healthy calves (19.25 ± 0.300 h/d). For relative changes in activity patterns, there was a diarrhea status by day interaction on d -2, where diarrheic calves had greater relative changes in step counts (diarrhea 634.85 ± 87.581% vs. healthy 216.51 ± 87.581%) and activity index (diarrhea 316.83 ± 35.692% vs. healthy 150.68 ± 35.692%). Lying bouts were not associated with diarrhea status. These results show that diarrheic calves were more lethargic, and they had relative changes in activity patterns -2 d before clinical signs of diarrhea. Future research should explore the potential of an activity alert which positively indicates an individually housed calf at-risk for a diarrhea bout using deviations from relative changes in individual calf activity patterns.

Permutation Tests for Assessing Potential Non-Linear Associations between Treatment Use and Multivariate Clinical Outcomes

In many psychometric applications, the relationship between the mean of an outcome and a quantitative covariate is too complex to be described by simple parametric functions; instead, flexible nonlinear relationships can be incorporated using penalized splines. Penalized splines can be conveniently represented as a linear mixed effects model (LMM), where the coefficients of the spline basis functions are random effects. The LMM representation of penalized splines makes the extension to multivariate outcomes relatively straightforward. In the LMM, no effect of the quantitative covariate on the outcome corresponds to the null hypothesis that a fixed effect and a variance component are both zero. Under the null, the usual asymptotic chi-square distribution of the likelihood ratio test for the variance component does not hold. Therefore, we propose three permutation tests for the likelihood ratio test statistic: one based on permuting the quantitative covariate, the other two based on permuting residuals. We compare via simulation the Type I error rate and power of the three permutation tests obtained from joint models for multiple outcomes, as well as a commonly used parametric test. The tests are illustrated using data from a stimulant use disorder psychosocial clinical trial.

Can glutathione be a biomarker for suicide risk in women 18 months postpartum?

Suicide risk is prominent among the problems affecting populations, mainly due to the broad family, psychosocial and economic impact. Most individuals at suicidal risk have some mental disorder. There is considerable evidence that psychiatric disorders are accompanied by the activation of neuro-immune and neuro-oxidative pathways. The aim of the study is to evaluate the serum levels of oxidative stress biomarkers in women at risk of suicide after 18 months of postpartum. This is a case-control study, nested within a cohort study. From this cohort, 45 women [15 without mood disorders and 30 with mood disorders (Major depression and Bipolar disorder)] were selected at 18 months postpartum, the depression and suicide risk were assessed using the Mini-International Neuropsychiatric Interview Plus (MINI-Plus) instrument, module A and C, respectively. Blood was collected and stored for later analysis of the reactive species (DCFH), superoxide dismutase (SOD), and glutathione reduced (GSH). For data analysis, the SPSS program was used. To compare the nominal covariates with the outcome GSH levels, the Student's t-test or analysis of variance (ANOVA) was used. Spearman's correlation was performed for analysis between the quantitative covariates and the outcome. To analyze the interaction between the factors, multiple linear regression was performed. Bonferroni analysis was used as an additional/secondary result to visualize differences in glutathione levels according to risk severity. After the adjusted analysis, p-values < 0.05 were considered statistically significant. The percentage of suicide risk observed in our sample of women at 18 months postpartum was 24.4% (n = 11). After adjusting for the independent variables, only the presence of suicide risk remained associated with the outcome (β = 0.173; p = 0.007), low levels of GSH at 18 months after postpartum. Likewise, we verified the difference in GSH levels according to the degree of suicide risk, observing a significant association between the differences in glutathione means in the group of women with moderate to high risk compared to the reference group (no suicide risk) (p = 0.009). Our findings suggest that GSH may be a potential biomarker or etiologic factor in women at moderate to high risk of suicide.

The Efficient Covariate-Adaptive Design for high-order balancing of quantitative and qualitative covariates

The Efficient Covariate-Adaptive Design for high-order balancing of quantitative and qualitative covariates

New, diagnostic flicker test for optic neuritis shows specific stages following disease onset

BackgroundPreviously, results of the digital flicker test (DFT) have shown distinct patterns in acute optic neuritis (ON) and healthy eyes. We aimed to examine the diagnostic potential of the DFT in acute ON and to investigate the temporal development of the DFT response following ON while comparing with visual evoked potentials (VEP). MethodsThe DFT examines the subjective brightness of a flickering field, varied in 11 different frequencies from 0 to 60 Hz, compared to a steady field. Previous studies have indicated a pronounced darkness enhancement at medial frequencies in acute ON eyes. Darkness enhancement at medial frequencies was expressed as a quantitative covariate (DFTDE). Results were compared with healthy controls and follow-up measurements were compared with VEP. Results112 patients were examined <31 days of onset (median 14.0 days (IQR:12.25)). 104 of 112 patients showed an abnormal flicker test (sensitivity 93%). DFT was abnormal in 2 of 55 healthy controls. The DFT showed normalization in 34% at 3 months, 36.4% at 6 months and 71.4% at >8 months from ON onset compared to 13.3%, 22.4% and 28.6% for VEP. Changes to the pattern of the DFT results were shown at specific stages during and following ON. ConclusionsThe DFT is an easy-to-use and sensitive diagnostic test for acute ON. The flicker test shows a more pronounced temporal evolution following ON than VEP and may be of use monitoring the course of ON.

The Adaptability of Passion Scale for Tertiary Students in Japanese Online Courses in Indonesia

This study aimed at examining the opportunities for adapting the passion scale to Japanese literature students who carried out the online learning process in Indonesia. This study used quantitative covariance structure design as the application of item response theory. From a sample of 148 undergraduate students, originally from three different courses. Data were obtained by filling out a virtual questionnaire which adapted from the original Dual Model Passion scale. Since this scale had been developed abroad, the data were analysed quantitatively using confirmatory factor analysis and regression, this study replicated the similar analysis while examined passion by using five criteria such as time, like, value, passion, identity as used in the previous development process. In addition, the criteria for student characteristics (gender and age) and conditions (first year vs. advanced year) in attending online lectures by observing the relationship between the two main dimensions of passion). The results showed that first year students tend to have a higher harmonious passion (HP), while second year students had a higher obsessive passion (OP). In conclusion, several things needed to be considered by researchers and lecturers at tertiary institutions such as course and student characteristics, so that students' passion remained good even though they had to study online during the outbreak of the Covid-19.

Survival after stereotactic radiosurgery of recurrent glioblastomas in patients with radical resection of primary tumor

glioblastoma (GBM) is an aggressive tumor with high rate of recurrence and estimated survival of 15-18 months after diagnosis. Factors associated with longer survival of GBM patients are age &lt; 50 years, high performance status and radical resection of the primary tumor. The optimal treatment for recurrence/ progression of GBM has not yet been determined and remains a challenging issue. Stereotactic radiosurgery (SRS) is considered today as a therapeutic option for effective treatment of recurrent malignant gliomas. The aim of this retrospective study was to analyze the survival after SRS of the recurrent GBM in a cohort of 59 patients, which had a radical resection of the primary tumor. The cohort consisted of 59 patients (28 / 47.5% of women and 31 / 52.5% of men); the average age was 51 years (interval 24 - 81). SRS was performed by means of linear accelerator "Trilogy" (USA) (6 MeV) from 2014 to 2020 at the State Institution "Romodanov Neurosurgery Institute". In all cases, the diagnosis of grade 4 GBM according to the WHO classification was confirmed after neurosurgical procedures of the primary tumor. All 59 patients underwent the maximal safe removal of the primary tumor: in the vast majority of cases (54 / 91.5%) - in the perifocal area; in 5 / 8.5% of cases - subtotal. In all 59 cases, patients received adjuvant radiation therapy (total dose 60 Gy in 30 fractions); in 33 / 55.9% of cases radiotherapy was combined with concomitant alkylating chemotherapy (CHT) (Temozolomide 75 mg / m2). In 31 / 52.5% of patients, maintenance alkylating CHT was continued (Temozolomide 150-200 mg / m2). In most cases (51 / 86.4%) recurrent GBM (RGBM) was diagnosed by clinical and radiological signs; in 8 / 13.6% of patients - after repeated surgery. Overall survival (OS), recurrence/progression free survival (RFS) and survival after recurrence (SAR) represented the end-points of the study. The effect of the following quantitative and categorical factors (covariates) on the survival was studied: sex, age, performance status, combination of adjuvant RT with alkylating chemotherapy, neurosurgical procedures of RGBM, type of GBM recurrence, total dose of irradiation (BED11) and SRS dose (BED11), number of SRS fractions, volume of target in SRS, duration of RFS. The effect of RFS was studied in three independent groups: group I – RFS &lt; 10 months; group II – RFS from 10 to 20 months; group III – RFS &gt; 20 months. The survival was analyzed by Kaplan-Meier (KM) method. Log-rank test was used for analysis of the survival according to the binary predictors. The effect of several categorical factors on survival was analyzed by Pearson Chi-square test. The effect of the quantitative covariates on survival was studied by regression analysis in Cox proportional risk model. Hazard ratio was calculated with 95 % confidential intervals (CI). The analysis revealed the following. Median OS following SRS RGBM was 26.3 months (95 % CІ 17 – 45.5), median RFS was 12.9 months (95 % CІ 8.4 – 25.6), median SAR – 9.8 months (95 % CІ 6.7 – 24.4). Two-year OS in our study was 56 %. 6-month survival after SRS –77 %; one-year survival after SRS – 39 %, and two-year survival after SRS – 28 %. The significant impact of performance status (p = 0.00159), duration of recurrence-free period (p = 0.02711) and surgical resection of RGBM (р = 0.009391) on the OS was demonstrated. The best OS was shown for the patients with Karnofsky score 90, recurrence occurring after more than 20 months and previous surgical resection of RGBM. The effects of other factors on OS were not demonstrated. Such factors as age, sex, performance status, adjuvant RT with сoncomitant alkylating CHT, surgical resection of RGBM, type of recurrence, number of SRS fractions, BED11 in SRS RGBM, BED11 for overall courses of irradiation, SRS target volume demonstrated no effect on SAR. SRS is non-invasive method for RGBM treatment that allows for improving the survival without significant radiation toxicity. Primary biological properties of the tumor seem to be of priority in determining the survival of RGBM patients. Although irradiation of GBM is advantageous regarding the improvement of the survival, one could also speculate that re-irradiation of the recurrent malignant glioma triggers some changes in its biology neutralizing the potential effect of the survival factors that had predictive value before re-irradiation.

A class of categorization methods for credit scoring models

Credit scoring models are usually developed using logistic regression. For several reasons, professionals of this area frequently categorize the quantitative covariates before using them in the model. In this work, we introduce a class of methods for covariate categorization in regression models for binary response variables. Applications to real data and a Monte Carlo simulation study suggest that one of the methods of this class has a better predictive performance and a smaller computational cost than other methods available in the literature.

Fast Bayesian inference using Laplace approximations in nonparametric double additive location-scale models with right- and interval-censored data

Penalized B-splines are commonly used in additive models to describe smooth changes in a response with quantitative covariates. This is usually done through the conditional mean in the exponential family using generalized additive models with an indirect impact on other conditional moments. Another common strategy is to focus on several low-order conditional moments, leaving the full conditional distribution unspecified. Alternatively, a multi-parameter distribution could be assumed for the response with several of its parameters jointly regressed on covariates using additive expressions. The latter proposal for a right- or interval-censored continuous response with a highly flexible and smooth nonparametric density is considered. The focus is on location-scale models with additive terms in the conditional mean and standard deviation. Starting from recent results in the Bayesian framework, a fast converging algorithm is proposed to select penalty parameters from their marginal posteriors. It is based on Laplace approximations of the conditional posterior of the spline parameters. Simulations suggest that the estimators obtained in this way have excellent frequentist properties and superior efficiencies compared to approaches with a working Gaussian assumption. The methodology is illustrated by the analysis of right- and interval-censored income data.

Determine an optimal cut-point in presence of competing risks to categorize a continuous distribution: An application in studying outcomes of patients with cirrhosis listed for liver transplantation

Introduction Whether non-selective beta-blockers (NSBBs) are deleterious in patients with end-stage cirrhosis and refractory ascites has been widely debated. We hypothesized that only the subset of patients on the liver transplant waiting list who had impaired cardiac performance would be at increased risk of mortality if receiving NSBBs. Methods This study included 584 consecutive patients with cirrhosis evaluated for transplantation between 1999 and 2014. All patients had right heart catheterization with hemodynamic measurements at evaluation. Fifty percent received NSBBs. Refractory ascites was present in 33%. Cardiac performance was assessed by left ventricular stroke work index (LVSWI). Waiting list mortality without liver transplantation was explored using competing risk analysis trough Fine & Gray regression model. As describe in Contal et al. and Woo et al., we used the maximally selected rank statistic methods to determine an optimal cutpoint for a quantitative covariate with competing risks. There is no package available for this method so we implemented an R function and codes and description of the method are available in GitHub ( https://github.com/PManchon/Cutoff-Gray_DEBRC-Bichat ). Results LVSWI was significantly lower in patients with refractory ascites. In multivariate analysis using competing risk, refractory ascites, NSBBs (yes/no) and LVSWI (quantitative) were associated with waiting list mortality in the whole population, with a statistically significant interaction between NSBBs and LVSWI. To create a composite variable to ensure interpretability of the result we determine an optimal cutpoint for LVSWI. The most discriminant value was 64.1 g-m/m2 (corresponding to the maximally Gray statistic standardized at 0.86). In the final model, refractory ascites (subdistribution hazard ratio: 1.52; 95% CI: 1.01–2.28; P = 0.0083) and treatment by NSBBs with LVSWI Conclusion This study suggests that compromised cardiac performance is more common in patients with refractory ascites and that NSBBs are deleterious in cirrhotic patients with compromised cardiac performance. These results highlight the prognostic value of cardiac function in patients with end-stage cirrhosis.

Optimized homomorphic encryption solution for secure genome-wide association studies

BackgroundGenome-Wide Association Studies (GWAS) refer to observational studies of a genome-wide set of genetic variants across many individuals to see if any genetic variants are associated with a certain trait. A typical GWAS analysis of a disease phenotype involves iterative logistic regression of a case/control phenotype on a single-neuclotide polymorphism (SNP) with quantitative covariates. GWAS have been a highly successful approach for identifying genetic-variant associations with many poorly-understood diseases. However, a major limitation of GWAS is the dependence on individual-level genotype/phenotype data and the corresponding privacy concerns.MethodsWe present a solution for secure GWAS using homomorphic encryption (HE) that keeps all individual data encrypted throughout the association study. Our solution is based on an optimized semi-parallel GWAS compute model, a new Residue-Number-System (RNS) variant of the Cheon-Kim-Kim-Song (CKKS) HE scheme, novel techniques to switch between data encodings, and more than a dozen crypto-engineering optimizations.ResultsOur prototype can perform the full GWAS computation for 1,000 individuals, 131,071 SNPs, and 3 covariates in about 10 minutes on a modern server computing node (with 28 cores). Our solution for a smaller dataset was awarded co-first place in iDASH’18 Track 2: “Secure Parallel Genome Wide Association Studies using HE”.ConclusionsMany of the HE optimizations presented in our paper are general-purpose, and can be used in solving challenging problems with large datasets in other application domains.

Models to Assess the Association of a Semiquantitative Exposure With Outcomes.

A semiquantitative risk factor has 2 components: any exposure (yes/no) and the quantitative amount of exposure (if exposed). We describe the statistical properties of alternative analyses with such a risk factor using linear, logistic, or Cox proportional hazards models. Often analyses employ the amount exposed as a single quantitative covariate, including the nonexposed with value zero. However, this analysis provides a biased estimate of the exposure coefficient (slope) and we describe the magnitude of the bias. This bias can be eliminated by adding a binary covariate for exposed versus not to the model. This 2-factor analysis captures the full risk-factor effect on the outcome. However, the coefficient for any exposure versus not does not have a meaningful interpretation. Alternatively, when exposure values among those exposed are centered (by subtracting the mean), the estimate of this coefficient represents the difference in the outcome between those exposed versus not in aggregate. We also show that the biased model provides biased estimates of the coefficients for other covariates added to the model. Proper analysis of a semiquantitative risk factor should start with a 2-factor model, with centering, to assess the joint contributions of the 2 components of the risk-factor exposure. Properties of models were illustrated using data from a multisite study in North America (1983-2019).

Sensitive Assessment of Acute Optic Neuritis by a New, Digital Flicker Test

Background: The Aulhorn flicker test (AFT) previously showed promise in diagnosing acute optic neuritis (ON) albeit with suboptimal sensitivity. A new, digitalized version of the AFT (the DFT) has not previously been examined in acute ON. Objectives: To examine the sensitivity, specificity and reproducibility of the DFT in acute ON. Method: The DFT assesses the subjective brightness of a flickering field (1–60 Hz). In normal subjects, brightness enhancement occurs at intermediate frequencies, whereas in acute ON darkness enhancement (DE) is hypothesized. AFT and DFT measurements were obtained in acute ON patients (≤31 days from first symptom) with DE as a quantitative covariate. Reproducibility of the DFT end point was assessed in the form of an intraclass correlation. Results: 30 untreated first-time acute ON patients and 55 healthy controls were examined. AFT and DFT were performed 12.7 days (range: 4–30) following ON onset. The DFT showed a sensitivity of 0.93 (95% CI = 0.78–0.99) to a specificity of 0.96 (95% CI = 0.87–1.00). The AFT showed a sensitivity of 0.76 (95% CI = 0.56–0.90) to a specificity of 1.00 (95% CI = 0.93–1.00). No significant correlation was shown between DFT and visual acuity. The intraclass correlation of the DFT end point in healthy subjects was 0.84. Conclusions: We present a new DFT in acute ON displaying a high specificity of 0.96 and a sensitivity of 0.93. Our study indicates the DFT to be an accurate and easy-to-use tool in diagnosing acute ON, which may be especially helpful in atypical cases.

Consistency for the negative binomial regression with fixed covariate

We model an overdispersed count as a dependent measurement, by means of the Negative Binomial distribution. We consider a quantitative covariate that is fixed by design. The expectation of the dependent variable is assumed to be a known function of a linear combination involving the possibly multidimensional covariate and its coefficients. In the NB1-parametrization of the Negative Binomial distribution, the variance is a linear function of the expectation, inflated by the dispersion parameter, and the distribution not a generalized linear model. For the maximum likelihood estimator for all parameters we apply a general result of Bradley and Gart (Biometrika 49:205–214, 1962) to derive weak consistency and asymptotic normality and a technique in Fahrmeir and Kaufmann (Ann Stat 13:342–368, 1985) for strong consistency. To this end, we show (1) how to bound the logarithmic density by a function that is linear in the outcome of the dependent variable, independently of the parameter. Furthermore (2) the positive definiteness of the matrix related to the Fisher information is shown with the Cauchy–Schwarz inequality.

The relationship between clinical trial accrual volume and outcomes in acute myeloid leukemia: A SWOG/ECOG-ACRIN study (S0106 and E1900)

PurposeTo study whether institutional clinical trial accrual volume affects clinical outcomes of younger (age less than 61 years) patients with acute myeloid leukemia. Patients and methodsWe investigated the impact of clinical trial accrual on response rates, early mortality and survival in patients with AML enrolled between 2002 and 2009 into two parallel cooperative group clinical trials SWOG S0106/ECOG-ACRIN E1900. Institutions were classified as low- (LAIs) (≤ 9 enrolled patients) or high-accruing institutions (HAIs) (≥10 enrolled patients). Fisher’s exact text and logistic regression analysis were used to analyze the response and early mortality rates. The effect of accrual volume on survival was analyzed by log-rank tests and Cox regression models. ResultsA total of 1252 patients from 152 institutions were included in the final analyses. The median clinical trial registrations in HAIs was 19 patients (range, 10 to 92) versus 3 (range, 1 to 9) patients in LAIs. In multivariate analyses, HAIs, as a quantitative covariate, was associated with improved complete remission rates (odds ratio (OR) 1.08, p = 0.0051), but no improvement median overall survival (HR 0.97, p = 0.065) or median event-free (hazard ratio (HR) 0.97, p = 0.05). Early mortality rates were similar between cohorts and academic affiliation had no impact on response rates or survival. ConclusionClinical trial accrual volume, had an independent, albeit modest, impact on complete remission rates, but not on overall survival and event-free in younger patients with AML.

Risk Profiles for Microvascular and Macrovascular Disease at Baseline in the Glycemia Reduction Approaches in Diabetes—A Comparative Effectiveness (GRADE) Study

Microvascular and macrovascular complications drive morbidity, mortality, and health care costs in type 2 diabetes (T2D). The GRADE trial enrolled 5047 participants with T2D &amp;lt;10 years and treated with metformin only, with creatinine &amp;lt;1.4 (women) &amp;lt;1.5 (men) mg/dL, and HbA1c 6.8-8.6% at enrollment. The enrolled population was 36% female, 57±10 years old (mean±SD), had T2D duration 4.2±2.8 years, and HbA1c 8.0±1.0%. Baseline prevalence was 1.0% for retinopathy (self-report), 15.8% for moderately increased albuminuria (albumin/creatinine ratio 31-300 mg/g), 2.4% for decreased eGFR (30-60 ml/minutes/1.73m2), 41.8% for neuropathy (Michigan Neuropathy Screening Instrument clinical score &amp;gt;2), 5.1% for myocardial infarction (self-report); and 2.0% for stroke (self-report). The Table describes age- and sex-adjusted associations between individual risk factors and complications as odds ratios comparing the presence vs. absence of the complication. Notable features include associations of HbA1c, obesity and blood pressure with micro- but not macrovascular disease; association of triglycerides with microvascular disease only; and associations of smoking and HDL with both micro- and macrovascular disease. These data reveal unexpected patterns in the clustering of risk factors with complications in GRADE patients with short T2D duration. ACR≥30eGFR&amp;lt;60RetinopathyMNSI&amp;gt;2MIStrokeORpORpORpORpORpORpHBA1C at screening (%)1.0380.30660.6580.00300.8370.28541.0940.00170.9960.95910.9370.5605BMI (kg/m2)1.040&amp;lt;0.00011.0440.00290.9800.38861.045&amp;lt;0.00010.9960.71930.9960.8481Waist Circumference (cm)1.017&amp;lt;0.00011.0110.07570.9910.34301.022&amp;lt;0.00011.0000.93651.0010.7997Smoking former*1.0360.68770.8740.52510.4260.02931.0650.34731.5460.00431.1010.6799Smoking current*1.5000.00021.1540.64391.2300.57701.2380.01562.492&amp;lt;0.00012.1800.0036Systolic BP (mm Hg)1.021&amp;lt;0.00010.9950.47831.0200.02351.0020.19370.9990.82981.0100.0956Diastolic BP (mm Hg)1.022&amp;lt;0.00010.9940.58971.0160.27460.9980.60630.9940.41311.0110.2546LDLc (mg/dl) **1.0010.29020.9990.68031.0090.04370.9990.26620.9960.15261.0040.2398HDLc (mg/dl)0.9890.00210.9800.02551.0190.07870.9930.01740.9760.00031.0040.5854(log) Triglycerides (mg/dl)1.477&amp;lt;0.00011.5890.00640.7980.37341.1150.03151.1520.21420.8690.4324OR=odds ratio per unit increase in a quantitative covariate or for the stated category versus the complement.*Versus never smoking.** Further adjusted for statin exposure. Disclosure K.J. Mather: Research Support; Self; Merck &amp; Co., Inc., Sanofi-Aventis, Novo Nordisk A/S, Abbott. Advisory Panel; Self; Merck &amp; Co., Inc. R. Pop-Busui: Research Support; Self; AstraZeneca. I. Bebu: None. C. Baker: None. R.M. Cohen: Stock/Shareholder; Self; Bristol-Myers Squibb Company. J.P. Crandall: None. C. Desouza: Consultant; Self; Novo Nordisk A/S, Sanofi. Research Support; Self; Merck &amp; Co., Inc.. Advisory Panel; Self; Medscape. J.B. Green: Consultant; Self; Merck Sharp &amp; Dohme Corp.. Research Support; Self; AstraZeneca. Consultant; Self; Daiichi Sankyo Company, Limited. Research Support; Self; GlaxoSmithKline plc., Intarcia Therapeutics, Inc.. Consultant; Self; Boehringer Ingelheim Pharmaceuticals, Inc., Novo Nordisk Inc. M. Kirkman: Research Support; Self; Bayer AG, Theracos, Inc.. Consultant; Self; National Institutes of Health. Research Support; Self; National Institutes of Health, Centers for Disease Control and Prevention. H. Krause-Steinrauf: None. M.E. Larkin: None. J. Pettus: Advisory Panel; Self; Sanofi, Novo Nordisk Inc.. Consultant; Self; MannKind Corporation. Advisory Panel; Self; Insulet Corporation. Consultant; Self; Senseonics. E.R. Seaquist: Advisory Panel; Self; Eli Lilly and Company. Consultant; Self; Eli Lilly and Company, Sanofi, Zucera, InfoMed, 360 consulting. Other Relationship; Self; Novo Nordisk Inc.. E.Z. Soliman: None. E.B. Schroeder: None. D.J. Wexler: None. G. Research Group: Research Support; Self; National Institute of Diabetes and Digestive and Kidney Diseases, National Heart, Lung, and Blood Institute. Other Relationship; Self; Bristol-Myers Squibb Company, Merck &amp; Co., Inc., Novo Nordisk Inc., Sanofi-Aventis, Roche Diagnostics Corporation, Becton, Dickinson and Company, Centers for Disease Control and Prevention, National Diabetes Education Program.

Geological Modelling and Validation of Geological Interpretations via Simulation and Classification of Quantitative Covariates

This paper proposes a geostatistical approach for geological modelling and for validating an interpreted geological model, by identifying the areas of an ore deposit with a high probability of being misinterpreted, based on quantitative coregionalised covariates correlated with the geological categories. This proposal is presented through a case study of an iron ore deposit at a stage where the only available data are from exploration drill holes. This study consists of jointly simulating the quantitative covariates with no previous geological domaining. A change of variables is used to account for stoichiometric closure, followed by projection pursuit multivariate transformation, multivariate Gaussian simulation, and conditioning to the drill hole data. Subsequently, a decision tree classification algorithm is used to convert the simulated values into a geological category for each target block and realisation. The determination of the prior (ignoring drill hole data) and posterior (conditioned to drill hole data) probabilities of categories provides a means of identifying the blocks for which the interpreted category disagrees with the simulated quantitative covariates.

Bayesian inference for unidirectional misclassification of a binary response trait

When assessing association between a binary trait and some covariates, the binary response may be subject to unidirectional misclassification. Unidirectional misclassification can occur when revealing a particular level of the trait is associated with a type of cost, such as a social desirability or financial cost. The feasibility of addressing misclassification is commonly obscured by model identification issues. The current paper attempts to study the efficacy of inference when the binary response variable is subject to unidirectional misclassification. From a theoretical perspective, we demonstrate that the key model parameters possess identifiability, except for the case with a single binary covariate. From a practical standpoint, the logistic model with quantitative covariates can be weakly identified, in the sense that the Fisher information matrix may be near singular. This can make learning some parameters difficult under certain parameter settings, even with quite large samples. In other cases, the stronger identification enables the model to provide more effective adjustment for unidirectional misclassification. An extension to the Poisson approximation of the binomial model reveals the identifiability of the Poisson and zero-inflated Poisson models. For fully identified models, the proposed method adjusts for misclassification based on learning from data. For binary models where there is difficulty in identification, the method is useful for sensitivity analyses on the potential impact from unidirectional misclassification.