			Home About us Contact

Spurious Associations (spurious + association)

Distribution by Scientific Domains

Medical Sciences	55%

Selected Abstracts

Spurious associations in oral epidemiological research: the case of dental flossing and obesity

JOURNAL OF CLINICAL PERIODONTOLOGY, Issue 8 2006
P. P. Hujoel
Abstract Background: Individuals with increased oral health awareness may also have increased general health awareness, and vice versa. Such associations between oral and general health awareness has the potential to induce spurious associations in oral epidemiological research. Objective: To assess the extent to which oral self-care patterns and general health awareness are confounded, we investigated the association between flossing and obesity, two lifestyle factors that are unlikely to be causally related. Methods: A cross-sectional study of 1497 individuals presenting for an initial periodontal exam by the specialist. Self-reported flossing behaviors and body mass index (BMI) categories were related using logistic regression models. Results: After adjustment for confounding variables, lack of daily flossing was associated in a dose-dependent way with morbid obesity (odds ratio (OR), 20.3; 95% confidence interval (CI), 2.7,154.0), obesity (OR, 2.1; 95% CI, 1.5,2.9), and being overweight (OR, 1.7; 95% CI, 1.3,2.2). When restricting to never smokers, a significant relationship between obesity and lack of flossing remained. Conclusion: The strong associations between two causally unrelated oral and general lifestyle characteristics indicate that simplistic epidemiologic methodology is unlikely to provide insights into causal mechanisms of oral diseases or oral-systemic relationships. [source]

Robust Quantitative Trait Association Tests in the Parent-Offspring Triad Design: Conditional Likelihood-Based Approaches

ANNALS OF HUMAN GENETICS, Issue 2 2009
J.-Y. Wang
Summary Association studies, based on either population data or familial data, have been widely applied to mapping of genes underlying complex diseases. In family-based association studies, using case-parent triad families, the popularly used transmission/disequilibrium test (TDT) was proposed for avoidance of spurious association results caused by other confounders such as population stratification. Originally, the TDT was developed for analysis of binary disease data. Extending it to allow for quantitative trait analysis of complex diseases and for robust analysis of binary diseases against the uncertainty of mode of inheritance has been thoroughly discussed. Nevertheless, studies on robust analysis of quantitative traits for complex diseases received relatively less attention. In this paper, we use parent-offspring triad families to demonstrate the feasibility of establishment of the robust candidate-gene association tests for quantitative traits. We first introduce the score statistics from the conditional likelihoods based on parent-offspring triad data under various genetic models. By applying two existing robust procedures we then construct the robust association tests for analysis of quantitative traits. Simulations are conducted to evaluate empirical type I error rates and powers of the proposed robust tests. The results show that these robust association tests do exhibit robustness against the effect of misspecification of the underlying genetic model on testing powers. [source]

Genotypic Association Analysis Using Discordant-Relative-Pairs

ANNALS OF HUMAN GENETICS, Issue 1 2009
T. Yan
Summary In practice, family-based design has been widely used in disease-gene association analysis. The major advantage of such design is that it is not subject to spurious association due to population structure such as population stratification (PS) and admixture. A disadvantage is that parental genotypes are hard to obtain if the disease is late onset for which a discordant-relative-pair design is useful. Designs of such kind include full-sib-pair, half-sib-pair, first-cousin-pair, and so on. The closer the relatedness of the pair, the less possible that they are subject to population stratification. On the other hand, the association test using close relative-pairs may be less powerful due to over-matching. Trade-off between these two factors (population structure and over-matching) is the major concern of this study. Some tests, namely McNemar's test, matched Cochran-Armitage trend tests (CATTs), matched maximum efficient robust test (MERT), and Bhapkar's test, are used for testing disease-gene association based on relative-pair designs. These tests are shown to be valid in the presence of PS but not admixture. Numerical studies show that the McNemar's test, additive CATT, MERT, and Bhapkar's test are robust in power, but none of them is uniformly more powerful than the others. In most simulations, the power of any of the tests increases as the pair is more distant. The proposed methods are applied to two real examples. [source]

Genomic Control for Association Studies under Various Genetic Models

BIOMETRICS, Issue 1 2005
Gang Zheng
Summary Case,control studies are commonly used to study whether a candidate allele and a disease are associated. However, spurious association can arise due to population substructure or cryptic relatedness, which cause the variance of the trend test to increase. Devlin and Roeder derived the appropriate variance inflation factor (VIF) for the trend test and proposed a novel genomic control (GC) approach to estimate VIF and adjust the test statistic. Their results were derived assuming an additive genetic model and the corresponding VIF is independent of the candidate allele frequency. We determine the appropriate VIFs for recessive and dominant models. Unlike the additive test, the VIFs for the optimal tests for these two models depend on the candidate allele frequency. Simulation results show that, when the null loci used to estimate the VIF have allele frequencies similar to that of the candidate gene, the GC tests derived for recessive and dominant models remain optimal. When the underlying genetic model is unknown or the null loci and candidate gene have quite different allele frequencies, the GC tests derived for the recessive or dominant models cannot be used while the GC test derived for the additive model can be. [source]

Early onset of alcohol use and health problems: spurious associations and prevention

ADDICTION, Issue 11 2003
THOMAS C. HARFORD
No abstract is available for this article. [source]

The generalizability of the Buss,Perry Aggression Questionnaire

INTERNATIONAL JOURNAL OF METHODS IN PSYCHIATRIC RESEARCH, Issue 3 2007
József Gerevich
Abstract Aggressive and hostile behaviours and anger constitute an important problem across cultures. The Buss,Perry Aggression Questionnaire (AQ), a self-rating scale was published in 1992, and has quickly become the gold-standard for the measurement of aggression. The AQ scale has been validated extensively, but the validation focused on various narrowly selected populations, typically, on samples of college students. Individuals, however, who are at risk of displaying aggressive and hostile behaviours may come from a more general population. Therefore, it is important to investigate the scale's properties in such a population. The objective of this study was to examine the factorial structure and the psychometric properties of the AQ scale in a nationally representative sample of the Hungarian adult population. A representative sample of 1200 subjects was selected by a two-step procedure. The dimensionality and factorial composition of the AQ scale was investigated by exploratory and confirmatory factor analyses. Since spurious associations and increased factorial complexity can occur when the analysis fails to consider the inherently categorical nature of the item level data, this study, in contrast to most previous studies, estimated the correlation matrices subjected to factor analysis using the polychoric correlations. The resulting factors were validated via sociodemographic characteristics and psychopathological scales obtained from the respondents. The results showed that based on the distribution of factor loadings and factor correlations, in the entire nationally representative sample of 1200 adult subjects, from the original factor structure three of the four factors (Physical and Verbal Aggression and Hostility) showed a good replication whereas the fourth factor (Anger) replicated moderately well. Replication further improved when the sample was restricted in age, i.e. the analysis focused on a sample representing the younger age group, comparable to that used in the original Buss,Perry study. Similar to the Buss,Perry study, and other investigations of the AQ scale, younger age and male gender were robustly related to physical aggression. In addition, level of verbal aggression was different between the two genders (with higher severity in males) whereas hostility and anger were essentially the same in both genders. In conclusion, the current study based on a representantive sample of adult population lends support to the use of the AQ scale in the general population. The authors suggest to exclude from the AQ the two inverse items because of the low reliability of these items with regard to their hypothesized constructs. Copyright © 2007 John Wiley & Sons, Ltd. [source]

Fine mapping and detection of the causative mutation underlying Quantitative Trait Loci

JOURNAL OF ANIMAL BREEDING AND GENETICS, Issue 5 2010
E. Uleberg
Summary The effect on power and precision of including the causative SNP amongst the investigated markers in Quantitative Trait Loci (QTL) mapping experiments was investigated. Three fine mapping methods were tested to see which was most efficient in finding the causative mutation: combined linkage and linkage disequilibrium mapping (LLD); association mapping (MARK); a combination of LLD and association mapping (LLDMARK). Two simulated data sets were analysed: in one set, the causative SNP was included amongst the markers, while in the other set the causative SNP was masked between markers. Including the causative SNP amongst the markers increased both precision and power in the analyses. For the LLD method the number of correctly positioned QTL increased from 17 for the analysis without the causative SNP to 77 for the analysis including the causative SNP. The likelihood of the data analysis increased from 3.4 to 13.3 likelihood units for the MARK method when the causative SNP was included. When the causative SNP was masked between the analysed markers, the LLD method was most efficient in detecting the correct QTL position, while the MARK method was most efficient when the causative SNP was included as a marker in the analysis. The LLDMARK method, combining association mapping and LLD, assumes a QTL as the null hypothesis (using LLD method) and tests whether the ,putative causative SNP' explains significantly more variance than a QTL in the region. Thus, if the putative causative SNP does not only give an Identical-By-Descent (IBD) signal, but also an Alike-In-State (AIS) signal, LLDMARK gives a positive likelihood ratio. LLDMARK detected less than half as many causative SNPs as the other methods, and also had a relatively high false discovery rate when the QTL effect was large. LLDMARK may however be more robust against spurious associations, because the regional IBD is largely corrected for by fitting a QTL effect in the null hypothesis model. [source]

Spurious associations in oral epidemiological research: the case of dental flossing and obesity

Evaluating the Ability of Tree-Based Methods and Logistic Regression for the Detection of SNP-SNP Interaction

ANNALS OF HUMAN GENETICS, Issue 3 2009
M. García-Magariños
Summary Most common human diseases are likely to have complex etiologies. Methods of analysis that allow for the phenomenon of epistasis are of growing interest in the genetic dissection of complex diseases. By allowing for epistatic interactions between potential disease loci, we may succeed in identifying genetic variants that might otherwise have remained undetected. Here we aimed to analyze the ability of logistic regression (LR) and two tree-based supervised learning methods, classification and regression trees (CART) and random forest (RF), to detect epistasis. Multifactor-dimensionality reduction (MDR) was also used for comparison. Our approach involves first the simulation of datasets of autosomal biallelic unphased and unlinked single nucleotide polymorphisms (SNPs), each containing a two-loci interaction (causal SNPs) and 98 ,noise' SNPs. We modelled interactions under different scenarios of sample size, missing data, minor allele frequencies (MAF) and several penetrance models: three involving both (indistinguishable) marginal effects and interaction, and two simulating pure interaction effects. In total, we have simulated 99 different scenarios. Although CART, RF, and LR yield similar results in terms of detection of true association, CART and RF perform better than LR with respect to classification error. MAF, penetrance model, and sample size are greater determining factors than percentage of missing data in the ability of the different techniques to detect true association. In pure interaction models, only RF detects association. In conclusion, tree-based methods and LR are important statistical tools for the detection of unknown interactions among true risk-associated SNPs with marginal effects and in the presence of a significant number of noise SNPs. In pure interaction models, RF performs reasonably well in the presence of large sample sizes and low percentages of missing data. However, when the study design is suboptimal (unfavourable to detect interaction in terms of e.g. sample size and MAF) there is a high chance of detecting false, spurious associations. [source]