Excess Relative Risk as an Effect Measure in Case-Control Studies of Rare Diseases

Epidemiologists often use ratio-type indices (rate ratio, risk ratio and odds ratio) to quantify the association between exposure and disease. By comparison, less attention has been paid to effect measures on a difference scale (excess rate or excess risk). The excess relative risk (ERR) used primarily by radiation epidemiologists is of peculiar interest here, in that it involves both difference and ratio operations. The ERR index (but not the difference-type indices) is estimable in case-control studies. Using the theory of sufficient component cause model, the author shows that when there is no mechanistic interaction (no synergism in the sufficient cause sense) between the exposure under study and the stratifying variable, the ERR index (but not the ratio-type indices) in a rare-disease case-control setting should remain constant across strata and can therefore be regarded as a common effect parameter. By exploiting this homogeneity property, the related attributable fraction indices can also be estimated with greater precision. The author demonstrates the methodology (SAS codes provided) using a case-control dataset, and shows that ERR preserves the logical properties of the ratio-type indices. In light of the many desirable properties of the ERR index, the author advocates its use as an effect measure in case-control studies of rare diseases.     

Exact tests of Hardy-Weinberg equilibrium and homogeneity of disequilibrium across strata

Detecting departures from Hardy-Weinberg equilibrium (HWE) of marker-genotype frequencies is a crucial first step in almost all human genetic analyses. When a sample is stratified by multiple ethnic groups, it is important to allow the marker-allele frequencies to differ over the strata. In this situation, it is common to test for HWE by using an exact test within each stratum and then using the minimum P value as a global test. This approach does not account for multiple testing, and, because it does not combine information over strata, it does not have optimal power. Several approximate methods to combine information over strata have been proposed, but most of them sum over strata a measure of departure from HWE; if the departures are in different directions, then summing can diminish the overall evidence of departure from HWE. An exact stratified test is more appealing because it uses the probability of genotype configurations across the strata as evidence for global departures from HWE. We developed an exact stratified test for HWE for diallelic markers, such as single-nucleotide polymorphisms (SNPs), and an exact test for homogeneity of Hardy-Weinberg disequilibrium. By applying our methods to data from Perlegen and HapMap--a combined total of more than five million SNP genotypes, with three to four strata and strata sizes ranging from 23 to 60 subjects--we illustrate that the exact stratified test provides more-robust and more-powerful results than those obtained by either the minimum of exact test P values over strata or approximate stratified tests that sum measures of departure from HWE. Hence, our new methods should be useful for samples composed of multiple ethnic groups.     

Assessing Causal Mechanistic Interactions: A Peril Ratio Index of Synergy Based on Multiplicativity

The assessments of interactions in epidemiology have traditionally been based on risk-ratio, odds-ratio or rate-ratio multiplicativity. However, many epidemiologists fail to recognize that this is mainly for statistical conveniences and often will misinterpret a statistically significant interaction as a genuine mechanistic interaction. The author adopts an alternative metric system for risk, the ‘peril’. A peril is an exponentiated cumulative rate, or simply, the inverse of a survival (risk complement) or one plus an odds. The author proposes a new index based on multiplicativity of peril ratios, the ‘peril ratio index of synergy based on multiplicativity’ (PRISM). Under the assumption of no redundancy, PRISM can be used to assess synergisms in sufficient cause sense, i.e., causal co-actions or causal mechanistic interactions. It has a less stringent threshold to detect a synergy as compared to a previous index of ‘relative excess risk due to interaction’. Using the new PRISM criterion, many situations in which there is not evidence of interaction judged by the traditional indices are in fact corresponding to bona fide positive or negative synergisms.     

Heterogeneous treatment effects in stratified clinical trials with time‐to‐event endpoints

When analyzing clinical trials with a stratified population, homogeneity of treatment effects is a common assumption in survival analysis. However, in the context of recent developments in clinical trial design, which aim to test multiple targeted therapies in corresponding subpopulations simultaneously, the assumption that there is no treatment‐by‐stratum interaction seems inappropriate. It becomes an issue if the expected sample size of the strata makes it unfeasible to analyze the trial arms individually. Alternatively, one might choose as primary aim to prove efficacy of the overall (targeted) treatment strategy. When testing for the overall treatment effect, a violation of the no‐interaction assumption renders it necessary to deviate from standard methods that rely on this assumption. We investigate the performance of different methods for sample size calculation and data analysis under heterogeneous treatment effects. The commonly used sample size formula by Schoenfeld is compared to another formula by Lachin and Foulkes, and to an extension of Schoenfeld's formula allowing for stratification. Beyond the widely used (stratified) Cox model, we explore the lognormal shared frailty model, and a two‐step analysis approach as potential alternatives that attempt to adjust for interstrata heterogeneity. We carry out a simulation study for a trial with three strata and violations of the no‐interaction assumption. The extension of Schoenfeld's formula to heterogeneous strata effects provides the most reliable sample size with respect to desired versus actual power. The two‐step analysis and frailty model prove to be more robust against loss of power caused by heterogeneous treatment effects than the stratified Cox model and should be preferred in such situations.     

The analysis of stratified 2 x 2 contingency tables

We consider the problem of testing for independence against the consistent superiority of one treatment over another when the response variable is binary and is compared across two treatments in each of several strata. Specifically, we consider the randomized clinical trial setting. A number of issues arise in this context. First, should tables be combined if there are small or zero margins? Second, should one assume a common odds ratio across strata? Third, if the odds ratios differ across strata, then how does the standard test (based on a common odds ratio) perform? Fourth, are there other analyzes that are more appropriate for handling a situation in which the odds ratios may differ across strata? In addressing these issues we find that the frequently used Cochran–Mantel–Haenszel test may have a poor power profile, despite being optimal when the odds ratios are common. We develop novel tests that are analogous to the Smirnov, modified Smirnov, convex hull, and adaptive tests that have been proposed for ordered categorical data. (© 2006 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)     

Effects of stratification in the analysis of affected-sib-pair data: benefits and costs

The benefits and costs of stratification of affected-sib-pair (ASP) data were examined in three situations: (1) when there is no difference in identity-by-descent (IBD) allele sharing between stratified and unstratified ASP data sets; (2) when there is an increase in IBD allele sharing in one of the stratified groups; and (3) when the data are stratified on the basis of IBD allele-sharing status at one locus, and the stratified ASPs are then analyzed for linkage at a second locus. When there is no difference in IBD sharing between strata, a penalty is always paid for stratifying the data. The loss of power to detect linkage in the stratified ASP data sets is the result of multiple testing and the smaller sample size within individual strata. In the case in which etiologic heterogeneity (i.e., severity of phenotype, age at onset) represents genetic heterogeneity, the power to detect linkage can be increased by stratifying the ASP data. This benefit is obtained when there is sufficient IBD allele sharing and sample sizes. Once linkage has been established for a given locus, data can be stratified on the basis of IBD status at this locus and can be tested for linkage at a second locus. When the relative risk is in the vicinity of 1, the power to detect linkage at the second locus is always greater for the unstratified ASP data set. Even for values of the relative risk that diverge sufficiently from 1, with adequate sample sizes and IBD allele sharing, the benefits of stratifying ASP data are minimal.     

Stratified analysis in randomized trials with noncompliance

This article develops methods for stratified analyses of additive or multiplicative causal effect on binary outcomes in randomized trials with noncompliance. The methods are based on a weighted estimating function for an unbiased estimating function under randomization in each stratum. When known weights are used, the derived estimator is a natural extension of the instrumental variable estimator for stratified analyses, and test-based confidence limits are solutions of a quadratic equation in the causal parameter. Optimal weights that maximize asymptotic efficiency incorporate variability in compliance aspects across strata. An assessment based on asymptotic relative efficiency shows that a substantial enhancement in efficiency can be gained by using optimal weights instead of conventional ones, which do not incorporate the variability in compliance aspects across strata. Application to a field trial for coronary heart disease is provided.     

Describing Ordinal Odds Ratios for Stratified r × c Tables

For an r × ctable with ordinal responses, odds ratios are commonly used to describe the relationship between the row and column variables. This article shows two types of ordinal odds ratios where local‐global odds ratios are used to compare several groups on a c‐category ordinal response and a global odds ratio is used to measure the global association between a pair of ordinal responses. When there is a stratification factor, we consider Mantel‐Haenszel (MH) type estimators of these odds ratios to summarize the association from several strata. Like the ordinary MH estimator of the common odds ratio for several 2 × 2 contingency tables, the estimators are used when the association is not expected to vary drastically among the strata. Also, the estimators are consistent under the ordinary asymptotic framework in which the number of strata is fixed and also under sparse asymptotics in which the number of strata grows with the sample size. Compared to the maximum likelihood estimators, simulations find that the MH type estimators perform better especially when each stratum has few observations. This article provides variances and covariances formulae for the local‐global odds ratios estimators and applies the bootstrap method to obtain a standard error for the global odds ratio estimator. At the end, we discuss possible ways of testing the homogeneity assumption.     

Assessing numerical dependence in gene expression summaries with the jackknife expression difference

Statistical methods to test for differential expression traditionally assume that each gene's expression summaries are independent across arrays. When certain preprocessing methods are used to obtain those summaries, this assumption is not necessarily true. In general, the erroneous assumption of dependence results in a loss of statistical power. We introduce a diagnostic measure of numerical dependence for gene expression summaries from any preprocessing method and discuss the relative performance of several common preprocessing methods with respect to this measure. Some common preprocessing methods introduce non-trivial levels of numerical dependence. The issue of (between-array) dependence has received little if any attention in the literature, and researchers working with gene expression data should not take such properties for granted, or they risk unnecessarily losing statistical power.     

A climatic stratification of the environment of Europe

Aim To produce a statistical stratification of the European environment, suitable for stratified random sampling of ecological resources, the selection of sites for representative studies across the continent, and to provide strata for modelling exercises and reporting. Location A ‘Greater European Window’ with the following boundaries: 11° W, 32° E, 34° N, 72° N. Methods Twenty of the most relevant available environmental variables were selected, based on experience from previous studies. Principal components analysis (PCA) was used to explain 88% of the variation into three dimensions, which were subsequently clustered using an ISODATA clustering routine. The mean first principal component values of the classification variables were used to aggregate the strata into Environmental Zones and to provide a basis for consistent nomenclature. Results The Environmental Stratification of Europe (EnS) consists of 84 strata, which have been aggregated into 13 Environmental Zones. The stratification has a 1 km² resolution. Aggregations of the strata have been compared to other European classifications using the Kappa statistic, and show ‘good’ comparisons. The individual strata have been described using data from available environmental databases. The EnS is available for noncommercial use by applying to the corresponding author. Main conclusions The Environmental Stratification of Europe has been constructed using tried and tested statistical procedures. It forms an appropriate stratification for stratified random sampling of ecological resources, the selection of sites for representative studies across the continent and for the provision of strata for modelling exercises and reporting at the European scale.     

Competing risks regression for stratified data

For competing risks data, the Fine-Gray proportional hazards model for subdistribution has gained popularity for its convenience in directly assessing the effect of covariates on the cumulative incidence function. However, in many important applications, proportional hazards may not be satisfied, including multicenter clinical trials, where the baseline subdistribution hazards may not be common due to varying patient populations. In this article, we consider a stratified competing risks regression, to allow the baseline hazard to vary across levels of the stratification covariate. According to the relative size of the number of strata and strata sizes, two stratification regimes are considered. Using partial likelihood and weighting techniques, we obtain consistent estimators of regression parameters. The corresponding asymptotic properties and resulting inferences are provided for the two regimes separately. Data from a breast cancer clinical trial and from a bone marrow transplantation registry illustrate the potential utility of the stratified Fine-Gray model.     

Robustness of an odds-ratio test in a stratified group sequential trial with a binary outcome measure

Many clinical trials compare two or more treatment groups by using a binary outcome measure. For example, the goal could be to determine whether the frequency of pain episodes is significantly reduced in the treatment group (arm A) as compared to the control group (arm B). However, for ethical or regulatory reasons, group sequential designs are commonly employed. Then, based on a binomial distribution, the stopping boundaries for the interim analyses are constructed for assessing the difference in the response probabilities between the two groups. This is easily accomplished by using any of the standard procedures, e.g., those discussed by Jennison and Turnbull (2000), and using one of the most commonly used software packages, East (2000). Several factors are known to often affect the primary outcome of interest, but their true distributions are not known in advance. In addition, these factors may cause heterogeneous treatment responses among individuals in a group, and their exact effect size may be unknown. To limit the effect of such factors on the comparison of the two arms, stratified randomization is used in the actual conduct of the trial. Then, a stratified analysis based on the odds ratio proposed in Jennison and Turnbull (2000, pages 251-252) and consistent with the stratified design is undertaken. However, the stopping rules used for the interim analyses are those obtained for determining the differences in response rates in a design that was not stratified. The purpose of this paper is to assess the robustness of such an approach on the performance of the odds ratio test when the underlying distribution and effect size of the factors that influence the outcome may vary. The simulation studies indicate that, in general, the stratified approach offers consistently better results than does the unstratified approach, as long as the difference in the weighted average of the response probabilities across strata between the two groups remains closer to the hypothesized values, irrespective of the differences in the (allocation) distributions and heterogeneous response rate. However, if the response probabilities deviate significantly from the hypothesized values so that the difference in the weighted average is less than the hypothesized value, then the proposed study could be significantly underpowered.     

GOBLET: The Global Organisation for Bioinformatics Learning,Education and Training

In recent years, high-throughput technologies have brought big data to the life sciences. The march of progress has been rapid, leaving in its wake a demand for courses in data analysis, data stewardship, computing fundamentals, etc., a need that universities have not yet been able to satisfy—paradoxically, many are actually closing “niche” bioinformatics courses at a time of critical need. The impact of this is being felt across continents, as many students and early-stage researchers are being left without appropriate skills to manage, analyse, and interpret their data with confidence. This situation has galvanised a group of scientists to address the problems on an international scale. For the first time, bioinformatics educators and trainers across the globe have come together to address common needs, rising above institutional and international boundaries to cooperate in sharing bioinformatics training expertise, experience, and resources, aiming to put ad hoc training practices on a more professional footing for the benefit of all.     

Combining gene annotations and gene expression data in model-based clustering: weighted method

It has been increasingly recognized that incorporating prior knowledge into cluster analysis can result in more reliable and meaningful clusters. In contrast to the standard modelbased clustering with a global mixture model, which does not use any prior information, a stratified mixture model was recently proposed to incorporate gene functions or biological pathways as priors in model-based clustering of gene expression profiles: various gene functional groups form the strata in a stratified mixture model. Albeit useful, the stratified method may be less efficient than the global analysis if the strata are non-informative to clustering. We propose a weighted method that aims to strike a balance between a stratified analysis and a global analysis: it weights between the clustering results of the stratified analysis and that of the global analysis; the weight is determined by data. More generally, the weighted method can take advantage of the hierarchical structure of most existing gene functional annotation systems, such as MIPS and Gene Ontology (GO), and facilitate choosing appropriate gene functional groups as priors. We use simulated data and real data to demonstrate the feasibility and advantages of the proposed method.     

Effects of cluster sampling on epidemiologic analysis in population-based case-control studies

We consider population-based case-control designs in which controls are selected by one of three cluster sampling plans from the entire population at risk. The effects of cluster sampling on classical epidemiologic procedures are investigated, and appropriately modified procedures are developed. In particular, modified procedures for testing the homogeneity of odds ratios across strata, and for estimating and testing a common odds ratio are presented. Simulations that use the data from the 1970 Health Interview Survey as a population suggest that classical procedures may be fairly robust in the presence of cluster sampling. A more extreme example based on a mixed multinomial model clearly demonstrates that the classical Mantel-Haenszel (1959, Journal of the National Cancer Institute 22, 719-748) and Woolf-Haldane tests of no exposure effect may have sizes exceeding nominal levels and confidence intervals with less than nominal coverage under an alternative hypothesis. Classical estimates of odds ratios may also be biased with non-self-weighting cluster samples. The modified procedures we propose remedy these defects.     

Variance of Stratified Survey Estimators With Probability of Detection Adjustments

Abstract: Estimates of wildlife population sizes are frequently constructed by combining counts of observed animals from a stratified survey of aerial sampling units with an estimated probability of detecting animals. Unlike traditional stratified survey designs, stratum-specific estimates of population size will be correlated if a common detection model is used to adjust counts for undetected animals in all strata. We illustrate this concept in the context of aerial surveys, considering 2 cases: 1) a single-detection parameter is estimated under the assumption of constant detection probabilities, and 2) a logistic-regression model is used to estimate heterogeneous detection probabilities. Naïve estimates of variance formed by summing stratum-specific estimates of variance may result in significant bias, particularly if there are a large number of strata, if detection probabilities are small, or if estimates of detection probabilities are imprecise. (JOURNAL OF WILDLIFE MANAGEMENT 72(3):837–844; 2008)     

Planning and Assessment of Activity Budget Studies Employing Instantaneous Sampling

To obtain accurate estimates of activity budget parameters, samples must be unbiased and precise. Many researchers have considered how biased data may affect their ability to draw conclusions and examined ways to decrease bias in sampling efforts, but few have addressed the implications of not considering estimate precision. We propose a method to assess whether the number of instantaneous samples collected is sufficient to obtain precise activity budget parameter estimates. We draw on sampling theory to determine the number of observations per animal required to reach a desired bound on the error of estimation based on a stratified random sample, with individual animals acting as strata. We also discuss the optimal balance between the number of individuals sampled and the number of observations sampled per individual for a variety of sampling conditions. We present an empirical dataset on pronghorn (Antilocapra americana) as an example of the utility of the method. The required numbers of observation to reach precise estimates for pronghorn varied between common and rare behaviors, but precise estimates were achieved with <255 observations per individual for common behaviors. The two most apparent factors affecting the required number of observations for precise estimates were the number of individuals sampled and the complexity of the activity budget. This technique takes into account variation associated with individual activity budgets and population variation in activity budget parameter estimates, and helps to ensure that estimates are precise. The method can also be used for planning future sampling efforts.     

All SNPs Are Not Created Equal: Genome-Wide Association Studies Reveal a Consistent Pattern of Enrichment among Functionally Annotated SNPs

Recent results indicate that genome-wide association studies (GWAS) have the potential to explain much of the heritability of common complex phenotypes, but methods are lacking to reliably identify the remaining associated single nucleotide polymorphisms (SNPs). We applied stratified False Discovery Rate (sFDR) methods to leverage genic enrichment in GWAS summary statistics data to uncover new loci likely to replicate in independent samples. Specifically, we use linkage disequilibrium-weighted annotations for each SNP in combination with nominal p-values to estimate the True Discovery Rate (TDR = 1−FDR) for strata determined by different genic categories. We show a consistent pattern of enrichment of polygenic effects in specific annotation categories across diverse phenotypes, with the greatest enrichment for SNPs tagging regulatory and coding genic elements, little enrichment in introns, and negative enrichment for intergenic SNPs. Stratified enrichment directly leads to increased TDR for a given p-value, mirrored by increased replication rates in independent samples. We show this in independent Crohn''s disease GWAS, where we find a hundredfold variation in replication rate across genic categories. Applying a well-established sFDR methodology we demonstrate the utility of stratification for improving power of GWAS in complex phenotypes, with increased rejection rates from 20% in height to 300% in schizophrenia with traditional FDR and sFDR both fixed at 0.05. Our analyses demonstrate an inherent stratification among GWAS SNPs with important conceptual implications that can be leveraged by statistical methods to improve the discovery of loci.     

Hierarchical proportional hazards regression models for highly stratified data

In clinical trials conducted over several data collection centers, the most common statistically defensible analytic method, a stratified Cox model analysis, suffers from two important defects. First, identification of units that are outlying with respect to the baseline hazard is awkward since this hazard is implicit (rather than explicit) in the Cox partial likelihood. Second (and more seriously), identification of modest treatment effects is often difficult since the model fails to acknowledge any similarity across the strata. We consider a number of hierarchical modeling approaches that preserve the integrity of the stratified design while offering a middle ground between traditional stratified and unstratified analyses. We investigate both fully parametric (Weibull) and semiparametric models, the latter based not on the Cox model but on an extension of an idea by Gelfand and Mallick (1995, Biometrics 51, 843-852), which models the integrated baseline hazard as a mixture of monotone functions. We illustrate the methods using data from a recent multicenter AIDS clinical trial, comparing their ease of use, interpretation, and degree of robustness with respect to estimates of both the unit-specific baseline hazards and the treatment effect.