Relative Efficiency of the Case-Control to the Cohort Design in Epidemiology

In deriving the efficiency of the stratified to the simple random sample design in survey research, the critical link between the designs is the population analysis of variance. Analogously, the efficiency of the case-control to the cohort design in epidemiologic research can be derived using Bayes' theorem as the essential connection between the designs. Prior information on the odds of the disease is also required. A numerical example using data from Fleiss' text is used to illustrate the result.     

An alternative experimental case–control design for genetic association studies on bovine mastitis

The possibility of using genetic control strategies to increase disease resistance to infectious diseases relies on the identification of markers to include in the breeding plans. Possible incomplete exposure of mastitis-free (control) animals, however, is a major issue to find relevant markers in genetic association studies for infectious diseases. Usually, designs based on elite dairy sires are used in association studies, but an epidemiological case–control strategy, based on cows repeatedly field-tested could be an alternative for disease traits. To test this hypothesis, genetic association results obtained in the present work from a cohort of Italian Holstein cows tested for mastitis over time were compared with those from a previous genome-wide scan on Italian Holstein sires genotyped with 50k single nucleotide polymorphisms for de-regressed estimated breeding values for somatic cell counts (SCCs) on Bos taurus autosome (BTA6) and BTA14. A total of 1121 cows were selected for the case–control approach (cases=550, controls=571), on a combination of herd level of SCC incidence and of within herd individual level of SCC. The association study was conducted on nine previously identified markers, six on BTA6 and four on BTA14, using the R statistical environment with the ‘qtscore’ function of the GenABEL package, on high/low adjusted linear score as a binomial trait. The results obtained in the cow cohort selected on epidemiological information were in agreement with those obtained from the previous sire genome-wide association study (GWAS). Six out of the nine markers showed significant association, four on BTA14 (rs109146371, rs109234250, rs109421300, rs109162116) and two on BTA6 (rs110527224 and rs42766480). Most importantly, using mastitis as a case study, the current work further validated the alternative use of historical field disease data in case–control designs for genetic analysis of infectious diseases in livestock.     

An LRP8 variant is associated with familial and premature coronary artery disease and myocardial infarction

Our previous genomewide linkage scan of 428 nuclear families (GeneQuest) identified a significant genetic susceptibility locus for premature myocardial infarction (MI) on chromosome 1p34-36. We analyzed candidate genes in the locus with a population-based association study involving probands with premature coronary artery disease (CAD) and/or MI from the GeneQuest families (381 cases) and 560 controls without stenosis detectable by coronary angiography. A nonconservative substitution, R952Q, in LRP8 was significantly associated with susceptibility to premature CAD and/or MI by use of both population-based and family-based designs. Three additional white populations were used for follow-up replication studies: another independent cohort of CAD- and/or MI-affected families (GeneQuest II: 441 individuals from 22 pedigrees), an Italian cohort with familial MI (248 cases) and 308 Italian controls, and a separate Cleveland GeneBank cohort with sporadic MI (1,231 cases) and 560 controls. The association was significantly replicated in two independent populations with a family history of CAD and/or MI, the GeneQuest II family-based replication cohort and the Italian cohort, but not in the population with sporadic disease. The R952Q variant of LRP8 increased activation of p38 mitogen-activated protein kinase by oxidized low-density lipoprotein. This extensive study, involving multiple independent populations, provides the first evidence that genetic variants in LRP8 may contribute to the development of premature and familial CAD and MI.     

Cohort case-control design and analysis for clustered failure-time data

Cohort case-control design is an efficient and economical design to study risk factors for disease incidence or mortality in a large cohort. In the last few decades, a variety of cohort case-control designs have been developed and theoretically justified. These designs have been exclusively applied to the analysis of univariate failure-time data. In this work, a cohort case-control design adapted to multivariate failure-time data is developed. A risk set sampling method is proposed to sample controls from nonfailures in a large cohort for each case matched by failure time. This method leads to a pseudolikelihood approach for the estimation of regression parameters in the marginal proportional hazards model (Cox, 1972, Journal of the Royal Statistical Society, Series B 34, 187-220), where the correlation structure between individuals within a cluster is left unspecified. The performance of the proposed estimator is demonstrated by simulation studies. A bootstrap method is proposed for inferential purposes. This methodology is illustrated by a data example from a child vitamin A supplementation trial in Nepal (Nepal Nutrition Intervention Project-Sarlahi, or NNIPS).     

Retrospective versus prospective score tests for genetic association with case‐control data

Since the seminal work of Prentice and Pyke, the prospective logistic likelihood has become the standard method of analysis for retrospectively collected case‐control data, in particular for testing the association between a single genetic marker and a disease outcome in genetic case‐control studies. In the study of multiple genetic markers with relatively small effects, especially those with rare variants, various aggregated approaches based on the same prospective likelihood have been developed to integrate subtle association evidence among all the markers considered. Many of the commonly used tests are derived from the prospective likelihood under a common‐random‐effect assumption, which assumes a common random effect for all subjects. We develop the locally most powerful aggregation test based on the retrospective likelihood under an independent‐random‐effect assumption, which allows the genetic effect to vary among subjects. In contrast to the fact that disease prevalence information cannot be used to improve efficiency for the estimation of odds ratio parameters in logistic regression models, we show that it can be utilized to enhance the testing power in genetic association studies. Extensive simulations demonstrate the advantages of the proposed method over the existing ones. A real genome‐wide association study is analyzed for illustration.     

Combined association,segregation and aggregation analysis on case-control family data

Recently genetic epidemiologists have begun using case-control family study designs to investigate the role of genetic and environmental risk factors in disease etiology. The objective of these studies is to assess the association of environmental factors with the disease trait; to characterize the disease genes using segregation analysis; and to quantify the residual familial aggregation after controlling for environmental and genetic factors. Typically these objectives are achieved by conducting separate studies and analysis. This paper describes an estimating equation based approach for a combined association, segregation and aggregation analysis on data from case-control family studies. Simulations indicate that the method performs well in a variety of settings. The method is illustrated using simulated family history data made available to participants in a recent Genetic Analysis Workshop.     

Combining data from 2 nested case-control studies of overlapping cohorts to improve efficiency

Researchers subject to time and budget constraints may conductsmall nested case–control studies with individually matchedcontrols to help optimize statistical power. In this paper,we show how precision can be improved considerably by combiningdata from a small nested case–control study with datafrom a larger nested case–control study of a differentoutcome in the same or overlapping cohort. Our approach is basedon the inverse probability weighting concept, in which the log-likelihoodcontribution of each individual observation is weighted by theinverse of its probability of inclusion in either study. Weillustrate our approach using simulated data and an applicationwhere we combine data sets from 2 nested case–controlstudies to investigate risk factors for anorexia nervosa ina cohort of young women in Sweden.     

Comparison of cohort and nested case-control designs for estimating the effect of time-varying drug exposure on the risk of adverse event in the presence of ties

Cohort and nested case-control (NCC) designs are frequently used in pharmacoepidemiology to assess the associations of drug exposure that can vary over time with the risk of an adverse event. Although it is typically expected that estimates from NCC analyses are similar to those from the full cohort analysis, with moderate loss of precision, only few studies have actually compared their respective performance for estimating the effects of time-varying exposures (TVE). We used simulations to compare the properties of the resulting estimators of these designs for both time-invariant exposure and TVE. We varied exposure prevalence, proportion of subjects experiencing the event, hazard ratio, and control-to-case ratio and considered matching on confounders. Using both designs, we also estimated the real-world associations of time-invariant ever use of menopausal hormone therapy (MHT) at baseline and updated, time-varying MHT use with breast cancer incidence. In all simulated scenarios, the cohort-based estimates had small relative bias and greater precision than the NCC design. NCC estimates displayed bias to the null that decreased with a greater number of controls per case. This bias markedly increased with higher proportion of events. Bias was seen with Breslow's and Efron's approximations for handling tied event times but was greatly reduced with the exact method or when NCC analyses were matched on confounders. When analyzing the MHT-breast cancer association, differences between the two designs were consistent with simulated data. Once ties were taken correctly into account, NCC estimates were very similar to those of the full cohort analysis.     

The power of genomic control

Although association analysis is a useful tool for uncovering the genetic underpinnings of complex traits, its utility is diminished by population substructure, which can produce spurious association between phenotype and genotype within population-based samples. Because family-based designs are robust against substructure, they have risen to the fore of association analysis. Yet, if population substructure could be ignored, this robustness can come at the price of power. Unfortunately it is rarely evident when population substructure can be ignored. Devlin and Roeder recently have proposed a method, termed "genomic control" (GC), which has the robustness of family-based designs even though it uses population-based data. GC uses the genome itself to determine appropriate corrections for population-based association tests. Using the GC method, we contrast the power of two study designs, family trios (i.e., father, mother, and affected progeny) versus case-control. For analysis of trios, we use the TDT test. When population substructure is absent, we find GC is always more powerful than TDT; furthermore, contrary to previous results, we show that as a disease becomes more prevalent the discrepancy in power becomes more extreme. When population substructure is present, however, the results are more complex: TDT is more powerful when population substructure is substantial, and GC is more powerful otherwise. We also explore general issues of power and implementation of GC within the case-control setting and find that, economically, GC is at least comparable to and often less expensive than family-based methods. Therefore, GC methods should prove a useful complement to family-based methods for the genetic analysis of complex traits.     

Construction of resolvable spatial row-column designs

Resolvable row-column designs are widely used in field trials to control variation and improve the precision of treatment comparisons. Further gains can often be made by using a spatial model or a combination of spatial and incomplete blocking components. Martin, Eccleston, and Gleeson presented some general principles for the construction of robust spatial block designs which were addressed by spatial designs based on the linear variance (LV) model. In this article we define the two-dimensional form of the LV model and investigate extensions of the Martin et al. principles for the construction of resolvable spatial row-column designs. The computer construction of efficient spatial designs is discussed and some comparisons made with designs constructed assuming an autoregressive variance structure.     

Combining evidence for association from transmission disequilibrium and case-control studies using single-nucleotide polymorphisms

The aim of the present analysis is to combine evidence for association from the two most commonly used designs in genetic association analysis, the case-control design and the transmission disequilibrium test (TDT) design. The cases here are affected offspring from nuclear families and are used in both the case-control and TDT designs. As a result, inference from these designs is not independent. We applied a simple logistic regression method for combining evidence for association from case-control and TDT designs to single-nucleotide polymorphism data purchased on a region on chromosome 3, replicate 1 of the Aipotu population. Combining the evidence from the case-control and TDT designs yielded a 5-10% reduction in the standard errors of the relative risk estimates. The authors did not know the results before the analyses were conducted.     

The association of a SNP upstream of INSIG2 with body mass index is reproduced in several but not all cohorts

A SNP upstream of the INSIG2 gene, rs7566605, was recently found to be associated with obesity as measured by body mass index (BMI) by Herbert and colleagues. The association between increased BMI and homozygosity for the minor allele was first observed in data from a genome-wide association scan of 86,604 SNPs in 923 related individuals from the Framingham Heart Study offspring cohort. The association was reproduced in four additional cohorts, but was not seen in a fifth cohort. To further assess the general reproducibility of this association, we genotyped rs7566605 in nine large cohorts from eight populations across multiple ethnicities (total n = 16,969). We tested this variant for association with BMI in each sample under a recessive model using family-based, population-based, and case-control designs. We observed a significant (p < 0.05) association in five cohorts but saw no association in three other cohorts. There was variability in the strength of association evidence across examination cycles in longitudinal data from unrelated individuals in the Framingham Heart Study Offspring cohort. A combined analysis revealed significant independent validation of this association in both unrelated (p = 0.046) and family-based (p = 0.004) samples. The estimated risk conferred by this allele is small, and could easily be masked by small sample size, population stratification, or other confounders. These validation studies suggest that the original association is less likely to be spurious, but the failure to observe an association in every data set suggests that the effect of SNP rs7566605 on BMI may be heterogeneous across population samples.     

Statistical Challenges in Sequence-Based Association Studies with Population- and Family-Based Designs

Over the past few years, association analysis has become the primary tool for finding genes that underlie complex traits. Both population-based and family-based designs are commonly used designs in genetic association studies. Recent technological advances in exome and whole genome sequencing afford the next generation of sequence-based association studies. We review here recent developments in statistical methodology and remaining challenges related to sequence-based association studies with both population-based and family-based designs.     

The quadratic cumulative odds regression model for scored ordinal outcomes: application to alcohol dependence

In this paper, we develop new regression models for the analysis of scored ordinal data (i.e. ordinal outcomes where the categories are assigned numeric values). The novel feature of these models is that they enable one to capture and identify nonlinear aspects of the relationship between an ordinal clinical measurement (used for disease diagnosis) and risk factors. These nonlinearities may be useful in generating hypotheses about the risk factor's role in the etiologic process as well as suggesting how to design future studies of the risk factor. We apply our model to study the effects of race, gender, and family history on alcohol dependence among a cohort of lifetime drinkers from the 1992 National Longitudinal Alcohol Epidemiologic Survey.     

Efficient study designs for test of genetic association using sibship data and unrelated cases and controls

Linkage mapping of complex diseases is often followed by association studies between phenotypes and marker genotypes through use of case-control or family-based designs. Given fixed genotyping resources, it is important to know which study designs are the most efficient. To address this problem, we extended the likelihood-based method of Li et al., which assesses whether there is linkage disequilibrium between a disease locus and a SNP, to accommodate sibships of arbitrary size and disease-phenotype configuration. A key advantage of our method is the ability to combine data from different family structures. We consider scenarios for which genotypes are available for unrelated cases, affected sib pairs (ASPs), or only one sibling per ASP. We construct designs that use cases only and others that use unaffected siblings or unrelated unaffected individuals as controls. Different combinations of cases and controls result in seven study designs. We compare the efficiency of these designs when the number of individuals to be genotyped is fixed. Our results suggest that (1) when the disease is influenced by a single gene, the one sibling per ASP-control design is the most efficient, followed by the ASP-control design, and familial cases contribute more association information than singleton cases; (2) when the disease is influenced by multiple genes, familial cases provide more association information than singleton cases, unless the effect of the locus being tested is much smaller than at least one other untested disease locus; and (3) the case-control design can be useful for detecting genes with small effect in the presence of genes with much larger effect. Our findings will be helpful for researchers designing and analyzing complex disease-association studies and will facilitate genotyping resource allocation.     

Identification of a novel risk locus for multiple sclerosis at 13q31.3 by a pooled genome-wide scan of 500,000 single nucleotide polymorphisms

Multiple sclerosis is a chronic inflammatory demyelinating disease of the central nervous system with an important genetic component and strongest association driven by the HLA genes. We performed a pooling-based genome-wide association study of 500,000 SNPs in order to find new loci associated with the disease. After applying several criteria, 320 SNPs were selected from the microarrays and individually genotyped in a first and independent Spanish Caucasian replication cohort. The 8 most significant SNPs validated in this cohort were also genotyped in a second US Caucasian replication cohort for confirmation. The most significant association was obtained for SNP rs3129934, which neighbors the HLA-DRB/DQA loci and validates our pooling-based strategy. The second strongest association signal was found for SNP rs1327328, which resides in an unannotated region of chromosome 13 but is in linkage disequilibrium with nearby functional elements that may play important roles in disease susceptibility. This region of chromosome 13 has not been previously identified in MS linkage genome screens and represents a novel risk locus for the disease.     

复杂疾病全基因组关联研究进展——遗传统计分析

2005年, Science杂志首次报道了有关人类年龄相关性黄斑变性的全基因组关联研究, 此后有关肥胖、2型糖尿病、冠心病、阿尔茨海默病等一系列复杂疾病的全基因组关联研究被陆续报道, 这一阶段被称为人类全基因组关联研究的第一次浪潮。文章分别介绍了全基因组关联研究统计分析的方法、软件和应用实例; 比较了关联分析中多重检验的P值调整方法, 包括Bonferroni、递减的Bonferroni校正法、模拟运算法和控制错误发现率的方法; 还讨论了人群混杂对关联分析结果可能产生的影响及原理, 以及全基因组关联研究中控制人群混杂的方法的研究进展和应用实例。在全基因组关联研究的第一次浪潮中, 应用经典的遗传统计方法发现了许多基因-表型之间的关联并且能够对这些关联做出解释, 其中包括许多基因组中的未知基因和染色体区域。然而, 全基因组关联研究的继续发展需要进一步阐述基因组内基因之间相互作用、基因-基因之间的复杂作用网络与环境因素的相互作用在复杂疾病发生中的作用, 现有的统计分析方法肯定不能满足需要, 开发更为高级的统计分析方法势在必行。最后, 文章还给出了全基因组关联研究统计分析软件的相关网站信息。     

Genome-Wide Association Study Heterogeneous Cohort Homogenization via Subject Weight Knock-Down

Population structure can be a source of both false-positive and false-negative findings in a genome-wide association study. This article proposes an approach that helps to reduce the false-positives. It consists of homogenizing the diseased/healthy phenotype ratio across the cohort, by decreasing the statistical weight of selected individuals. After homogenization, the cohort is statistically handled as if originating from a single well-mixed population. The method was applied to homogenize a Parkinson''s disease genome-wide association study cohort.     

Investigation of the Locus Near MC4R With Childhood Obesity in Americans of European and African Ancestry

Recently a modest, but consistently, replicated association was demonstrated between obesity and the single‐nucleotide polymorphism (SNP), rs17782313, 3′ of the MC4R locus as a consequence of a meta‐analysis of genome‐wide association (GWA) studies of the disease in white populations. We investigated the association in the context of the childhood form of the disease utilizing data from our ongoing GWA study in a cohort of 728 European‐American (EA) obese children (BMI ≥95th percentile) and 3,960 EA controls (BMI <95th percentile), as well as 1,008 African‐American (AA) obese children and 2,715 AA controls. rs571312, rs10871777, and rs476828 (perfect surrogates for rs17782313) yielded odds ratios in the EA cohort of 1.142 (P = 0.045), 1.137 (P = 0.054), and 1.145 (P = 0.042); however, there was no significant association with these SNPs in the AA cohort. When investigating all 30 SNPs present on the Illumina BeadChip at this locus, again there was no evidence for association in AA cases when correcting for the number of tests employed. As such, variants 3′ to the MC4R locus present on the genotyping platform utilized confer a similar magnitude of risk of obesity in white children as to their adult white counterparts but this observation did not extend to AAs.     

Causal mediation analysis in nested case-control studies using conditional logistic regression

The paper proposes an approach to causal mediation analysis in nested case-control study designs, often incorporated with countermatching schemes using conditional likelihood, and we compare the method's performance to that of mediation analysis using the Cox model for the full cohort with a continuous or dichotomous mediator. Simulation studies are conducted to assess our proposed method and investigate the efficiency relative to the cohort. We illustrate the method using actual data from two studies of potential mediation of radiation risk conducted within the Adult Health Study cohort of atomic-bomb survivors. The performance becomes comparable to that based on the full cohort, illustrating the potential for valid mediation analysis based on the reduced data obtained through the nested case-control design.