Miguel A. Padilla

Missing Data in Randomized Clinical Trials for Weight Loss: Scope of the Problem, State of the Field, and Performance of Statistical Methods

Background Dropouts and missing data are nearly-ubiquitous in obesity randomized controlled trails, threatening validity and generalizability of conclusions. Herein, we meta-analytically evaluate the extent of missing data, the frequency with which various analytic methods are employed to accommodate dropouts, and the performance of multiple statistical methods. Methodology/Principal Findings We searched PubMed and Cochrane databases (2000–2006) for articles published in English and manually searched bibliographic references. Articles of pharmaceutical randomized controlled trials with weight loss or weight gain prevention as major endpoints were included. Two authors independently reviewed each publication for inclusion. 121 articles met the inclusion criteria. Two authors independently extracted treatment, sample size, drop-out rates, study duration, and statistical method used to handle missing data from all articles and resolved disagreements by consensus. In the meta-analysis, drop-out rates were substantial with the survival (non-dropout) rates being approximated by an exponential decay curve (e−λt) where λ was estimated to be .0088 (95% bootstrap confidence interval: .0076 to .0100) and t represents time in weeks. The estimated drop-out rate at 1 year was 37%. Most studies used last observation carried forward as the primary analytic method to handle missing data. We also obtained 12 raw obesity randomized controlled trial datasets for empirical analyses. Analyses of raw randomized controlled trial data suggested that both mixed models and multiple imputation performed well, but that multiple imputation may be more robust when missing data are extensive. Conclusion/Significance Our analysis offers an equation for predictions of dropout rates useful for future study planning. Our raw data analyses suggests that multiple imputation is better than other methods for handling missing data in obesity randomized controlled trials, followed closely by mixed models. We suggest these methods supplant last observation carried forward as the primary method of analysis.

The HLA–DRB1 Shared Epitope Is Associated With Susceptibility to Rheumatoid Arthritis in African Americans Through European Genetic Admixture

OBJECTIVE To determine whether shared epitope (SE)-containing HLA-DRB1 alleles are associated with rheumatoid arthritis (RA) in African Americans and whether their presence is associated with higher degrees of global (genome-wide) genetic admixture from the European population. METHODS In this multicenter cohort study, African Americans with early RA and matched control subjects were analyzed. In addition to measurement of serum anti-cyclic citrullinated peptide (anti-CCP) antibodies and HLA-DRB1 genotyping, a panel of >1,200 ancestry-informative markers was analyzed in patients with RA and control subjects, to estimate the proportion of European ancestry. RESULTS The frequency of SE-containing HLA-DRB1 alleles was 25.2% in African American patients with RA versus 13.6% in control subjects (P = 0.00005). Of 321 patients with RA, 42.1% had at least 1 SE-containing allele, compared with 25.3% of 166 control subjects (P = 0.0004). The mean estimated percent European ancestry was associated with SE-containing HLA-DRB1 alleles in African Americans, regardless of disease status (RA or control). As reported in RA patients of European ancestry, there was a significant association of the SE with the presence of the anti-CCP antibody: 86 (48.9%) of 176 patients with anti-CCP antibody-positive RA had at least 1 SE allele, compared with 36 (32.7%) of 110 patients with anti-CCP antibody-negative RA (P = 0.01, by chi-square test). CONCLUSION HLA-DRB1 alleles containing the SE are strongly associated with susceptibility to RA in African Americans. The absolute contribution is less than that reported in RA among populations of European ancestry, in which approximately 50-70% of patients have at least 1 SE allele. As in Europeans with RA, the SE association was strongest in the subset of African American patients with anti-CCP antibodies. The finding of a higher degree of European ancestry among African Americans with SE alleles suggests that a genetic risk factor for RA was introduced into the African American population through admixture, thus making these individuals more susceptible to subsequent environmental or unknown factors that trigger the disease.

An Examination of the Association of Selected Toxic Metals with Total and Central Obesity Indices: NHANES 99-02

It is conceivable that toxic metals contribute to obesity by influencing various aspects of metabolism, such as by substituting for essential micronutrients and vital metals, or by inducing oxidative stress. Deficiency of the essential metal zinc decreases adiposity in humans and rodent models, whereas deficiencies of chromium, copper, iron, and magnesium increases adiposity. This study utilized the NHANES 99-02 data to explore the association between waist circumference and body mass index with the body burdens of selected toxic metals (barium, cadmium, cobalt, cesium, molybdenum, lead, antimony, thallium, and tungsten). Some of the associations were significant direct relationships (barium and thallium), and some of the associations were significant inverse relationships (cadmium, cobalt, cesium, and lead). Molybdenum, antimony, and tungsten had mostly insignificant associations with waist circumference and body mass index. This is novel result for most of the toxic metals studied, and a surprising result for lead because high stored lead levels have been shown to correlate with higher rates of diabetes, and obesity may be a key risk factor for developing diabetes. These associations suggest the possibility that environmental exposure to metals may contribute to variations in human weight gain/loss. Future research, such as prospective studies rather than the cross-sectional studies presented here, is warranted to confirm these findings.

Endocrine Disruptors and Obesity: An Examination of Selected Persistent Organic Pollutants in the NHANES 1999-2002 Data

Recent evidence suggests that endocrine disrupting chemicals (EDCs) may cause perturbations in endogenous hormonal regulation that predispose to weight gain. Using data from NHANES (1999–2002), we investigated the association between body mass index (BMI), waist circumference (WC) and selected persistent organic pollutants (POPs) via multiple linear regressions. Consistent interaction was found between gender, ln oxychlordane and ln p,p’ DDT. Also, we found an association between WC and ln oxychlordane and ln hpcdd in subjects with detectable levels of POPs, whereas an association between WC and ln p,p’ DDT was observed in all subjects. Furthermore, ln Ocdd showed an increase with higher WC and BMI, whereas, ln trans-nonachlor decreased with higher BMI. Hence, BMI and WC are associated with POPs levels, making the chemicals plausible contributors to the obesity epidemic.

Regional admixture mapping and structured association testing: conceptual unification and an extensible general linear model.

Individual genetic admixture estimates, determined both across the genome and at specific genomic regions, have been proposed for use in identifying specific genomic regions harboring loci influencing phenotypes in regional admixture mapping (RAM). Estimates of individual ancestry can be used in structured association tests (SAT) to reduce confounding induced by various forms of population substructure. Although presented as two distinct approaches, we provide a conceptual framework in which both RAM and SAT are special cases of a more general linear model. We clarify which variables are sufficient to condition upon in order to prevent spurious associations and also provide a simple closed form “semiparametric” method of evaluating the reliability of individual admixture estimates. An estimate of the reliability of individual admixture estimates is required to make an inherent errors-in-variables problem tractable. Casting RAM and SAT methods as a general linear model offers enormous flexibility enabling application to a rich set of phenotypes, populations, covariates, and situations, including interaction terms and multilocus models. This approach should allow far wider use of RAM and SAT, often using standard software, in addressing admixture as either a confounder of association studies or a tool for finding loci influencing complex phenotypes in species as diverse as plants, humans, and nonhuman animals.

Review and Evaluation of Methods Correcting for Population Stratification with a Focus on Underlying Statistical Principles

When two or more populations have been separated by geographic or cultural boundaries for many generations, drift, spontaneous mutations, differential selection pressures and other factors may lead to allele frequency differences among populations. If these ‘parental’ populations subsequently come together and begin inter-mating, disequilibrium among linked markers may span a greater genetic distance than it typically does among populations under panmixia [see glossary]. This extended disequilibrium can make association studies highly effective and more economical than disequilibrium mapping in panmictic populations since less marker loci are needed to detect regions of the genome that harbor phenotype-influencing loci. However, under some circumstances, this process of intermating (as well as other processes) can produce disequilibrium between pairs of unlinked loci and thus create the possibility of confounding or spurious associations due to this population stratification. Accordingly, researchers are advised to employ valid statistical tests for linkage disequilibrium mapping allowing conduct of genetic association studies that control for such confounding. Many recent papers have addressed this need. We provide a comprehensive review of advances made in recent years in correcting for population stratification and then evaluate and synthesize these methods based on statistical principles such as (1) randomization, (2) conditioning on sufficient statistics, and (3) identifying whether the method is based on testing the genotype-phenotype covariance (conditional upon familial information) and/or testing departures of the marginal distribution from the expected genotypic frequencies.

Most common single-nucleotide polymorphisms associated with rheumatoid arthritis in persons of European ancestry confer risk of rheumatoid arthritis in African Americans.

OBJECTIVE Large-scale genetic association studies have identified >20 rheumatoid arthritis (RA) risk alleles among individuals of European ancestry. The influence of these risk alleles has not been comprehensively studied in African Americans. We therefore sought to examine whether these validated RA risk alleles are associated with RA risk in an African American population. METHODS Twenty-seven candidate single-nucleotide polymorphisms (SNPs) were genotyped in 556 autoantibody-positive African Americans with RA and 791 healthy African American control subjects. Odds ratios (ORs) and 95% confidence intervals (95% CIs) for each SNP were compared with previously published ORs for RA patients of European ancestry. We then calculated a composite genetic risk score (GRS) for each individual based on the sum of all risk alleles. RESULTS Overlap of the ORs and 95% CIs between the European and African American populations was observed for 24 of the 27 candidate SNPs. Conversely, 3 of the 27 SNPs (CCR6 rs3093023, TAGAP rs394581, and TNFAIP3 rs6920220) demonstrated ORs in the opposite direction from those reported for RA patients of European ancestry. The GRS analysis indicated a small but highly significant probability that African American patients relative to control subjects were enriched for the risk alleles validated in European RA patients (P = 0.00005). CONCLUSION The majority of RA risk alleles previously validated for RA patients of European ancestry showed similar ORs in our population of African Americans with RA. Furthermore, the aggregate GRS supports the hypothesis that these SNPs are risk alleles for RA in the African American population. Future large-scale genetic studies are needed to validate these risk alleles and identify novel RA risk alleles in African Americans.

Development of the Cyberbullying Experiences Survey

The majority of cyberbullying studies have examined middle and high school students. The purpose of the present study was to develop a multifactor cyberbullying victimization and perpetration survey for use with an emerging adult population. The initial 88-item preliminary survey (44 victimization and 44 perpetration items) was administered to 538 college students (421 females). Exploratory factor analyses revealed four-factor (i.e., malice, public humiliation, unwanted contact, and deception) victimization and perpetration scales. A confirmatory factor analysis was then performed on the Cyberbullying Experiences Survey (CES) factor structure with a separate sample of 638 college students (446 females). Results indicated a final 21-item victimization scale and 20-item perpetration scale consisting of the same four factors. The CES has adequate internal consistency and convergent validity with other measures of cyberbullying and Internet harassment and may provide a promising multifactor method of measuring cyberbullying victimization and perpetration.

Commonality of functional annotation: a method for prioritization of candidate genes from genome-wide linkage studies†

Linkage studies of complex traits frequently yield multiple linkage regions covering hundreds of genes. Testing each candidate gene from every region is prohibitively expensive and computational methods that simplify this process would benefit genetic research. We present a new method based on commonality of functional annotation (CFA) that aids dissection of complex traits for which multiple causal genes act in a single pathway or process. CFA works by testing individual Gene Ontology (GO) terms for enrichment among candidate gene pools, performs multiple hypothesis testing adjustment using an estimate of independent tests based on correlation of GO terms, and then scores and ranks genes annotated with significantly-enriched terms based on the number of quantitative trait loci regions in which genes bearing those annotations appear. We evaluate CFA using simulated linkage data and show that CFA has good power despite being conservative. We apply CFA to published linkage studies investigating age-of-onset of Alzheimers disease and body mass index and obtain previously known and new candidate genes. CFA provides a new tool for studies in which causal genes are expected to participate in a common pathway or process and can easily be extended to utilize annotation schemes in addition to the GO.

Correcting for Measurement Error in Individual Ancestry Estimates in Structured Association Tests

We present theoretical explanations and show through simulation that the individual admixture proportion estimates obtained by using ancestry informative markers should be seen as an error-contaminated measurement of the underlying individual ancestry proportion. These estimates can be used in structured association tests as a control variable to limit type I error inflation or reduce loss of power due to population stratification observed in studies of admixed populations. However, the inclusion of such error-containing variables as covariates in regression models can bias parameter estimates and reduce ability to control for the confounding effect of admixture in genetic association tests. Measurement error correction methods offer a way to overcome this problem but require an a priori estimate of the measurement error variance. We show how an upper bound of this variance can be obtained, present four measurement error correction methods that are applicable to this problem, and conduct a simulation study to compare their utility in the case where the admixed population results from the intermating between two ancestral populations. Our results show that the quadratic measurement error correction (QMEC) method performs better than the other methods and maintains the type I error to its nominal level.