Selecting controls for assessing interaction in nested case-control studies

BACKGROUND: Two methods for selecting controls in nested case-control studies--matching on X and counter matching on X--are compared when interest is in interaction between a risk factor X measured in the full cohort and another risk factor Z measured only in the case-control sample. This is important because matching provides efficiency gains relative to random sampling when X is uncommon and the interaction is positive (greater than multiplicative), whereas counter matching is generally efficient compared to random sampling. METHODS: Matching and counter matching were compared to each other and to random sampling of controls for dichotomous X and Z. Comparison was by simulation, using as an example a published study of radiation and other risk factors for breast cancer in the Japanese atomic-bomb survivors, and by asymptotic relative efficiency calculations for a wide range of parameters specifying the prevalence of X and Z as well as the levels of correlation and interaction between them. Focus was on analyses utilizing general models for the joint risk of X and Z. RESULTS: Counter-matching performed better than matching or random sampling in terms of efficiency for inference about interaction in the case of a rare risk factor X and uncorrelated risk factor Z. Further, more general, efficiency calculations demonstrated that counter-matching is generally efficient relative to matched case-control designs for studying interaction. CONCLUSIONS: Because counter-matched designs may be analyzed using standard statistical methods and allow investigation of confounding of the effect of X, whereas matched designs require a non-standard approach when fitting general risk models and do not allow investigating the adjusted risk of X, it is concluded that counter-matching on X can be a superior alternative to matching on X in nested case-control studies of interaction when X is known at the time of case-control sampling.     

Potential misinterpretation of the case-only study to assess gene-environment interaction.

Novel epidemiologic study designs are often required to assess gene-environment interaction. A design using only cases, without controls, is one of several approaches that have been proposed as more efficient alternatives to the typical random sampling of cases and controls. However, it has not been pointed out that a case-only analysis estimates a different interaction parameter than does a traditional case-control analysis: The latter typically estimates departure from multiplicative population odds or rate ratios, depending on the method of control selection, while the former estimates departure from multiplicative risk ratios if genotype and environmental exposure are not associated in the population. These parameters are approximately equal if the disease risk is small at all levels of the study variables. The authors quantify the impact of allowing for higher disease risk among gene carriers, a relevant situation when the gene under study is highly penetrant. Their findings show that the cross-product ratio computed from case-only data may be substantially smaller than the odds ratio computed from case-control data and may therefore underestimate either the population odds or the rate ratio. Thus, to avoid misinterpretation of interaction parameters estimated from case-only data, the definition of multiplicative interaction should be made explicit.     

不同研究设计的相对危险度估计

根据不同的研究设计，在队列研究中可分别估计累积发病率比（ＣＩＲ）、发病密度比（ＩＤＲ）和发病比值（ＩＯＲ）。当疾病罕见时，这３个比之值很接近。在病例对照研究中，是否可用暴露比值比（ＥＯＲ）作为ＣＩＲ或ＩＤＲ的无偏估计，取决于抽样方法和疾病率高低。本文还讨论了在不同抽样设计下ＥＯＲ与ＣＩＲ或ＩＤＲ之间的关系。     

To use or not to use the odds ratio in epidemiologic analyses?

This paper argues that the use of the odds ratio parameter in epidemiology needs to be considered with a view to the specific study design and the types of exposure and disease data at hand. Frequently, the odds ratio measure is being used instead of the risk ratio or the incidence-proportion ratio in cohort studies or as an estimate for the incidence-density ratio in case-referent studies. Therefore, the analyses of epidemiologic data have produced biased estimates and the presentation of results has been misleading. However, the odds ratio can be relinquished as an effect measure for these study designs; and, the application of the case-base sampling approach permits the incidence ratio and difference measures to be estimated without any untenable assumptions. For the Poisson regression, the odds ratio is not a parameter of interest; only the risk or rate ratio and difference are relevant. For the conditional logistic regression in matched case-referent studies, the odds ratio remains useful, but only when it is interpreted as an estimate of the incidence-density ratio. Thus the odds ratio should, in general, give way to the incidence ratio and difference as the measures of choice for exposure effect in epidemiology.     

Case-control analysis with a continuous outcome variable

It is not uncommon for a continuous outcome variable Y to be dichotomized and analysed using logistic regression. Moser and Coombs (Statist. Med. 2004; 23:1843-1860) provide a method for converting the output from a standard linear regression analysis using the original continuous outcome Y to give much more efficient inferences about the same odds-ratio parameters being estimated by the logistic regression. However, these results apply only to prospective studies. This paper follows up Moser and Coombs by providing an efficient linear-model-based solution for data collected using case-control studies. Gains in statistical efficiency of up to 240 per cent are obtained even with small to moderate odds ratios. Differences in design efficiency between case-control and prospective sampling designs are found to be much smaller, however, when linear-model-based analyses are being used than they are when logistic regression analyses are being used.     

CAPL: a novel association test using case-control and family data and accounting for population stratification

The recent successes of GWAS based on large sample sizes motivate combining independent datasets to obtain larger sample sizes and thereby increase statistical power. Analysis methods that can accommodate different study designs, such as family-based and case-control designs, are of general interest. However, population stratification can cause spurious association for population-based association analyses. For family-based association analysis that infers missing parental genotypes based on the allele frequencies estimated in the entire sample, the parental mating-type probabilities may not be correctly estimated in the presence of population stratification. Therefore, any approach to combining family and case-control data should also properly account for population stratification. Although several methods have been proposed to accommodate family-based and case-control data, all have restrictions. Most of them require sampling a homogeneous population, which may not be a reasonable assumption for data from a large consortium. One of the methods, FamCC, can account for population stratification and uses nuclear families with arbitrary number of siblings but requires parental genotype data, which are often unavailable for late-onset diseases. We extended the family-based test, Association in the Presence of Linkage (APL), to combine family and case-control data (CAPL). CAPL can accommodate case-control data and families with multiple affected siblings and missing parents in the presence of population stratification. We used simulations to demonstrate that CAPL is a valid test either in a homogeneous population or in the presence of population stratification. We also showed that CAPL can have more power than other methods that combine family and case-control data.     

Sampling of subpopulations in two‐stage surveys

Many health and other surveys aim to produce statistics on small subpopulations, such as specific ethnic groups or the indigenous population of a country. In most countries, there is no reliable sampling frame of the subpopulations of interest, hence it is necessary to sample from the general population, which can be very expensive. A range of issues and strategies for sampling rare subpopulations is reviewed. The most common approaches in practice are the use of a large screening sample, and disproportionate sampling by strata. Optimal sample designs have been derived for the case of one‐stage sampling, but most household interview surveys use two or more stages of selection. This paper develops optimal designs for two‐stage sampling, where there is auxiliary information on subpopulation numbers for each primary sampling unit. Various alternative designs are evaluated using a simulated population derived from the New Zealand Census. Copyright © 2009 John Wiley & Sons, Ltd.     

Estimating interaction between genetic and environmental risk factors: efficiency of sampling designs within a cohort

Large prospective cohorts originally assembled to study environmental risk factors are increasingly exploited to study gene-environment interactions. Given the cost of genetic studies in large samples, being able to select a subsample for genotyping that contains most of the information from the cohort would lead to substantial savings. We consider nested case-control and case-cohort sampling designs with and without stratification and compare their efficiency relative to the entire cohort for estimating the effects of genetic and environmental risk factors and their interactions. Asymptotic calculations show that the relative efficiency of the case-cohort and nested case-control designs implementing the same sampling stratification are similar over a range of scenarios for the relationships among genes, environmental exposures, and disease status. Sampling equal numbers of exposed and unexposed subjects improves efficiency when the exposure is rare. The case-cohort designs had a slight advantage in simulations of sampling designs within the Framingham Offspring Study, using the interaction between apolipoprotein E and smoking on the risk of coronary heart disease as an example. It was possible to estimate the interaction effect with precision close to that of the full cohort when using case-cohort or nested case-control samples containing fewer than half the subjects of the cohort.     

流行病学研究中的样本代表性问题(一)

研究的样本代表性一直是流行病学领域内具有争议性的话题之一。本文首先对流行病学样本代表性进行定义，同时对研究人群、源人群、目标人群、内部真实性、外部真实性等相关流行病学基本概念进行阐述。在此基础上，本文深入分析了横断面研究、队列研究、病例对照研究、干预性研究四种主要流行病学研究设计中样本代表性的价值和实际可行性。总而言之，除针对人群疾病或健康现状的横断面研究外，多数涉及病因或干预效果推断的流行病学研究不应过度强调样本代表性。     

Expected value of sample information calculations in medical decision modeling.

There has been an increasing interest in using expected value of information (EVI) theory in medical decision making, to identify the need for further research to reduce uncertainty in decision and as a tool for sensitivity analysis. Expected value of sample information (EVSI) has been proposed for determination of optimum sample size and allocation rates in randomized clinical trials. This article derives simple Monte Carlo, or nested Monte Carlo, methods that extend the use of EVSI calculations to medical decision applications with multiple sources of uncertainty, with particular attention to the form in which epidemiological data and research findings are structured. In particular, information on key decision parameters such as treatment efficacy are invariably available on measures of relative efficacy such as risk differences or odds ratios, but not on model parameters themselves. In addition, estimates of model parameters and of relative effect measures in the literature may be heterogeneous, reflecting additional sources of variation besides statistical sampling error. The authors describe Monte Carlo procedures for calculating EVSI for probability, rate, or continuous variable parameters in multi parameter decision models and approximate methods for relative measures such as risk differences, odds ratios, risk ratios, and hazard ratios. Where prior evidence is based on a random effects meta-analysis, the authors describe different ESVI calculations, one relevant for decisions concerning a specific patient group and the other for decisions concerning the entire population of patient groups. They also consider EVSI methods for new studies intended to update information on both baseline treatment efficacy and the relative efficacy of 2 treatments. Although there are restrictions regarding models with prior correlation between parameters, these methods can be applied to the majority of probabilistic decision models. Illustrative worked examples of EVSI calculations are given in an appendix.     

Comparing paired vs non-paired statistical methods of analyses when making inferences about absolute risk reductions in propensity-score matched samples

Propensity-score matching allows one to reduce the effects of treatment-selection bias or confounding when estimating the effects of treatments when using observational data. Some authors have suggested that methods of inference appropriate for independent samples can be used for assessing the statistical significance of treatment effects when using propensity-score matching. Indeed, many authors in the applied medical literature use methods for independent samples when making inferences about treatment effects using propensity-score matched samples. Dichotomous outcomes are common in healthcare research. In this study, we used Monte Carlo simulations to examine the effect on inferences about risk differences (or absolute risk reductions) when statistical methods for independent samples are used compared with when statistical methods for paired samples are used in propensity-score matched samples. We found that compared with using methods for independent samples, the use of methods for paired samples resulted in: (i) empirical type I error rates that were closer to the advertised rate; (ii) empirical coverage rates of 95 per cent confidence intervals that were closer to the advertised rate; (iii) narrower 95 per cent confidence intervals; and (iv) estimated standard errors that more closely reflected the sampling variability of the estimated risk difference. Differences between the empirical and advertised performance of methods for independent samples were greater when the treatment-selection process was stronger compared with when treatment-selection process was weaker. We recommend using statistical methods for paired samples when using propensity-score matched samples for making inferences on the effect of treatment on the reduction in the probability of an event occurring.     

Designing and analysing case-control studies to exploit independence of genotype and exposure

Genetic susceptibility and environmental exposures play a synergistic role in the aetiology of many diseases. We consider a case-control study of a rare disease in relation to a categorical exposure and a genetic factor under the assumption that the genotype and the exposure occur independently in the population under study. Using a logistic model for risk, we describe maximum likelihood methods based on log-linear models that explicitly impose the independence assumption, something the usual logistic regression analyses cannot do. The estimator of the genotype–exposure interaction effect depends only on data from cases. Estimators for genotype and for exposure effects depend also on data from controls, but only through their respective marginal totals. All three estimators have smaller variance than they would were independence not enforced. These results have important implications for design: (i) Case-only studies can efficiently estimate gene-by-environment interactions. (ii) Studies where controls are genotyped anonymously can estimate genotype, exposure, and interaction effects as efficiently as designs where genotype and exposure data are linked. This feature addresses a growing concern of human subjects review boards. (iii) Exposure and interaction effects, but not genotype effects, can be estimated from studies where genetic information is only collected from cases (although one can recover the genotype effect if external gene prevalence data exist). Such designs have the compensatory benefit that the response rate (hence, validity) is higher when controls are not subjected to intrusive tissue sampling. However, the independence assumption can be checked only with linked genotype and exposure data for some controls. We illustrate the methods by applying them to recent study of cleft palate in relation to maternal cigarette smoking and to a variant of the transforming growth factor alpha gene in the child. © 1997 by John Wiley & Sons, Ltd.     

Kin-cohort evaluation of relative risks of genetic variants

The kin-cohort study, an embedded retrospective cohort of first-degree relatives of probands, has been used frequently in conjunction with case-control studies to evaluate penetrance of high-risk genotypes. However, in principle, the kin-cohort approach can be used in conjunction with a case-control study in which family history data are carefully recorded to examine and validate the observed association of any candidate genotype with disease. Its utility for this purpose is governed by the relative efficiency of the kin-cohort analysis to test hypotheses about the relevant genotype-disease odds ratios, which in turn depends strongly on various factors, notably the magnitude of the disease incidence rate and the prevalence of the aberrant genotype. In this study, we compare the efficiencies of the two analytic approaches, case-control vs. kin-cohort, and illustrate the results in the context of cancer epidemiology. The kin-cohort analysis can provide substantial, important supplementary information for relatively common cancers such as breast, lung, colorectal, and prostate, and is also relatively efficient in general for the evaluation of the stronger and more rare genetic risk factors. Genet Epidemiol 24:220-229, 2003.     

应用Meta分析有效估计人群归因危险度百分比的探讨

人群归因危险度百分比 (populationattributableriskproportion ,PARP)是总体人群中某种疾病归因于某种因素的暴露所引起的发病 (死亡 )占全部发病 (死亡 )的百分比 ,反映该因素所引起的发病 (死亡 )占全部发病 (死亡 )的比重。通过PARP可了解各危险因素对人群中某疾病的发病所产生的影响 ,亦即消除某危险因素后 ,所产生的对预防该疾病的效果将占有多大比重。它能够为卫生政策的制订提供依据 ,有着重要的公共卫生的实际意义。目前常用的估计PARP的方法有两种 :一种是利用全国人群抽样调查获得的人群总暴露率来估计 ;另一种是利用某地区…     

Analysis of epidemiologic case-base studies for binary data

Recent developments in statistical methods for epidemiology have revived the application of sampling techniques in the design and analysis of cohort studies. The 'case-base' design involves sampling of both the cases and the cohort base of the study. This paper reviews some data-analytic imperfections of the approaches to risk ratio estimation, and modifies and advances a consistent likelihood-based procedure-analogous to Miettinen and Nurminen's proposal for a full cohort design--for interval estimation (and also point estimation and significance testing) in the context of binary case-base data. First, the procedure avoids the use of Taylor-series approximations to derive variance estimators for non-linear functions of parameters. Second, the asymptotic condition effects a simple computational expression for the chi-square function of risk ratios that is universally applicable to small samples. The statistical modelling underlying the method allows inferences about risk ratios without the assumption of rare disease either for the general population or for a particular base. The paper also extends the analysis to encompass stratified data. Finally, a numerical evaluation evinced the accurate small-sample properties of the proposed method.     

Renal cell cancer among paperboard printing workers.

A physician's alert prompted us to investigate workers' cancer risk at a paperboard printing manufacturer. We conducted a retrospective cohort mortality study of all 2,050 persons who had worked at the facility for more than 1 day, calculated standardized incidence ratios (SIRs) for bladder and renal cell cancer, and conducted a nested case-control study for renal cell cancer. Standardized mortality ratios (SMRs) from all causes [SMR = 1.0, 95% confidence interval (CI) = 0.9-1.2] and all cancers (SMR = 0.6, 95% CI = 0.3-1.0) were not greater than expected. One bladder cancer and one renal cell cancer were included in the mortality analysis. Six incident renal cell cancers were observed, however, compared with less than two renal cell cancers expected (SIR = 3.7, 95% CI = 1.4-8.1). Based on a nested case-control analysis, the risk of renal cell cancer was associated with overall length of employment but was not limited to any single department or work process. Although pigments containing congeners of dichlorobenzidine and o-toluidine had been used at the plant, environmental sampling could not confirm any current exposure. Several limitations and a potential selection bias limit the inferences that can be drawn.     

Risk of nonfatal venous thromboembolism with oral contraceptives containing norgestimate or desogestrel compared with oral contraceptives containing levonorgestrel

CONTEXT: Previous studies have reported that users of the "third-generation" oral contraceptives (OCs) containing the progestins gestodene and desogestrel have about twice the risk for venous thromboembolism (VTE) compared to users of older OCs containing levonorgestrel. Estimates of the risk for VTE among users of norgestimate-containing OCs compared to other OCs, however, are lacking. OBJECTIVE: The purpose of this study is to obtain quantitative information on the risk of nonfatal VTE in women using OCs containing either norgestimate or desogestrel in comparison with women taking OCs containing levonorgestrel. DESIGN, SETTING AND PARTICIPANTS: Based on information from PharMetrics, a United States-based company that collects and records information on claims paid by managed care plans, we used a nested case-control study design to estimate relative risks of nonfatal VTE among 15- to 39-year-old current users of OCs containing norgestimate with 35 microg of ethinyl estradiol (EE), desogestrel with 30 microg of EE or levonorgestrel with 30 microg of EE, both monophasic and triphasic preparations, during the period January 2000 to March 2005. Cases were women with a well-documented VTE of uncertain origin that was diagnosed in current users of a study drug. Up to four controls were closely matched to each case by age and calendar time, and odds ratios (ORs) were calculated using conditional logistic regression comparing the risk of VTE among users of the three contraceptives. We also estimated and compared the incidence rates for all three OCs. RESULTS: Based on 281 newly diagnosed idiopathic cases of VTE and 1055 controls, we found that the adjusted ORs for nonfatal VTE comparing norgestimate- or desogestrel-containing OC users to users of levonorgestrel-containing OCs were 1.1 [95% confidence interval (CI), 0.8-1.6] and 1.7 (95% CI, 1.1-2.4), respectively. The incidence rates of VTE were 30.6 (95% CI, 25.5-36.5), 53.5 (95% CI, 42.9-66.0) and 27.1 (95% CI, 21.1-34.3) per 100,000 woman-years for users of norgestimate-, desogestrel- and levonorgestrel-containing OCs, respectively. The incidence rate ratios for norgestimate-containing OCs compared to levonorgestrel-containing OCs and desogestrel-containing OCs compared to levonorgestrel-containing OCs were 1.1 (95% CI, 0.8-1.5) and 2.0 (95% CI, 1.4-2.7), respectively. CONCLUSIONS: The risk of nonfatal VTE among users of desogestrel-containing OCs is significantly elevated compared to that of levonorgestrel-containing OCs. The risk of VTE in users of norgestimate-containing OCs was closely similar to that of users of levonorgestrel-containing OCs.     

Limitations of the case-exposure study

The case-exposure study, described in 1983, is a modification of the case-control study that allows estimation of risk ratios without need for a rare-disease assumption. For a fixed population the approach is the same as the case-base or case-cohort study. Two sampling strategies have been described for extending application of the case-exposure approach to stationary populations, although the validity of the extension was not proven. As shown here, the two sampling strategies for stationary populations can provide valid estimates of the risk ratio, but restrictive assumptions related to stationarity may limit application of the strategy. We also show that the case-exposure study depends on the stationarity assumptions in a different, possibly more restrictive way than some other case-control approaches. We emphasize limitations of the case-exposure study that may preclude its use, particularly if the period of risk is long.     

Tuberculosis incidence and the seroprevalence of HIV infection in the United States.

Unlike other available studies, this article not only recognizes a human immunodeficiency virus-tuberculosis (HIV-TB) correlation but also shows its strength, statistical significance, and the method of calculation. These results increase comparability of the data on the HIV-TB relationship and allow more accurate analysis for developing better control of the two diseases. This population-based study compares data by states on HIV seroprevalence among 1,799,771 military recruits and the incidence of TB in the general population of the United States in 1987. The corresponding correlation coefficient was 0.8 (P less than .001). Although military recruits do not constitute a random sampling of the population at risk for TB, the high correlation suggests that HIV infection contributes substantially to population-based rates of TB.     

Validity of the cross-sectional study for the ascertainment of nosocomial infection risk factors

This study was carried out in order to assess the validity of the pure cross-sectional study in the ascertainment of nosocomial infection risk-factors. The results yielded by two designs (cross-sectional and case-control) are compared. A cross-sectional design was performed in a tertiary hospital. 592 patients were studied, 38 of whom were nosocomially infected. The clinical information on all the patients included in this design was reviewed after hospital discharge. A matched case-control study was nested in the population cross-sectionally surveyed. 66 cases (28 additional patients developed a hospital infection) and 132 controls were selected. Odds ratios (ORs) for the risk factors analyzed by both designs were compared. There were no significant differences between the estimates yielded by both designs; however, a trend of lower OR estimates for the cross-sectional study was seen, which may be important for risk factors not strongly related to (low relative risk) nosocomial infection. Several factors which might account for the results observed (random error, bias introduced by matching) are discussed. It is suggested that pure cross-sectional designs for the study of risk factors of nosocomial infection may introduce a negative (toward-the-null) bias.