Overestimation of risk ratios by odds ratios in trials and cohort studies: alternatives to logistic regression.
Ontology highlight
ABSTRACT:
SUBMITTER: Knol MJ
PROVIDER: S-EPMC3348192 | biostudies-literature | 2012 May
REPOSITORIES: biostudies-literature
Publications
Overestimation of risk ratios by odds ratios in trials and cohort studies: alternatives to logistic regression.
Knol Mirjam J MJ Le Cessie Saskia S Algra Ale A Vandenbroucke Jan P JP Groenwold Rolf H H RH
CMAJ : Canadian Medical Association journal = journal de l'Association medicale canadienne 20111212 8
Similar Datasets
Project description:BackgroundIn epidemiological studies researchers use logistic regression as an analytical tool to study the association of a binary outcome to a set of possible exposures.MethodsUsing a simulation study we illustrate how the analytically derived bias of odds ratios modelling in logistic regression varies as a function of the sample size.ResultsLogistic regression overestimates odds ratios in studies with small to moderate samples size. The small sample size induced bias is a systematic one, bias away from null. Regression coefficient estimates shifts away from zero, odds ratios from one.ConclusionIf several small studies are pooled without consideration of the bias introduced by the inherent mathematical properties of the logistic regression model, researchers may be mislead to erroneous interpretation of the results.
| S-EPMC2724427 | biostudies-literature
Project description:BackgroundThe odds ratio (OR) is used as an important metric of comparison of two or more groups in many biomedical applications when the data measure the presence or absence of an event or represent the frequency of its occurrence. In the latter case, researchers often dichotomize the count data into binary form and apply the well-known logistic regression technique to estimate the OR. In the process of dichotomizing the data, however, information is lost about the underlying counts which can reduce the precision of inferences on the OR.MethodsWe propose analyzing the count data directly using regression models with the log odds link function. With this approach, the parameter estimates in the model have the exact same interpretation as in a logistic regression of the dichotomized data, yielding comparable estimates of the OR. We prove analytically, using the Fisher information matrix, that our approach produces more precise estimates of the OR than logistic regression of the dichotomized data. We also show the gains in precision using simulation studies and real-world datasets. We focus on three related distributions for count data: geometric, Poisson, and negative binomial.ResultsIn simulation studies, confidence intervals for the OR were 56-65% as wide (geometric model), 75-79% as wide (Poisson model), and 61-69% as wide (negative binomial model) as the corresponding interval from a logistic regression produced by dichotomizing the data. When we analyzed existing datasets using our approach, we found that confidence intervals for the OR could be up to 64% shorter (36% as wide) compared to if the data had been dichotomized and analyzed using logistic regression.ConclusionsMore precise estimates of the OR can be obtained directly from the count data by using the log odds link function. This analytic approach is easy to implement in software packages that are capable of fitting generalized linear models or of maximizing user-defined likelihood functions.
| S-EPMC6195979 | biostudies-literature
Project description:BackgroundLogistic regression models are widely used to evaluate the association between a binary outcome and a set of covariates. However, when there are few study participants at the outcome and covariate levels, the models lead to bias of the odds ratio (OR) estimated using the maximum likelihood (ML) method. This bias is known as sparse data bias, and the estimated OR can yield impossibly large values because of data sparsity. However, this bias has been ignored in most epidemiological studies.MethodsWe review several methods for reducing sparse data bias in logistic regression. The primary aim is to evaluate the Bayesian methods in comparison with the classical methods, such as the ML, Firth's, and exact methods using a simulation study. We also apply these methods to a real data set.ResultsOur simulation results indicate that the bias of the OR from the ML, Firth's, and exact methods is considerable. Furthermore, the Bayesian methods with hyper-ɡ prior modeling of the prior covariance matrix for regression coefficients reduced the bias under the null hypothesis, whereas the Bayesian methods with log F-type priors reduced the bias under the alternative hypothesis.ConclusionThe Bayesian methods using log F-type priors and hyper-ɡ prior are superior to the ML, Firth's, and exact methods when fitting logistic models to sparse data sets. The choice of a preferable method depends on the null and alternative hypothesis. Sensitivity analysis is important to understand the robustness of the results in sparse data analysis.
| S-EPMC10165217 | biostudies-literature
Project description:Missing data are a commonly occurring threat to the validity and efficiency of epidemiologic studies. Perhaps the most common approach to handling missing data is to simply drop those records with 1 or more missing values, in so-called "complete records" or "complete case" analysis. In this paper, we bring together earlier-derived yet perhaps now somewhat neglected results which show that a logistic regression complete records analysis can provide asymptotically unbiased estimates of the association of an exposure of interest with an outcome, adjusted for a number of confounders, under a surprisingly wide range of missing-data assumptions. We give detailed guidance describing how the observed data can be used to judge the plausibility of these assumptions. The results mean that in large epidemiologic studies which are affected by missing data and analyzed by logistic regression, exposure associations may be estimated without bias in a number of settings where researchers might otherwise assume that bias would occur.
| S-EPMC4597800 | biostudies-literature
Project description:BackgroundThe logistic regression analysis proposed by Schouten et al. (Stat Med. 1993;12:1733-1745) has been a standard method in current statistical analysis of case-cohort studies, and it enables effective estimation of risk ratio from selected subsamples, with adjustment of potential confounding factors. Schouten et al. (1993) also proposed the standard error estimate of the risk ratio estimator can be calculated by the robust variance estimator, and this method has been widely adopted.Methods and resultsWe show that the robust variance estimator does not account for the duplications of case and subcohort samples and generally has certain bias, i.e., inaccurate confidence intervals and P-values are possibly obtained. To address the invalid statistical inference problem, we provide an alternative bootstrap-based valid variance estimator. Through simulation studies, the bootstrap method consistently provided more precise confidence intervals compared with those provided by the robust variance method, while retaining adequate coverage probabilities.ConclusionThe robust variance estimator has certain bias, and inadequate conclusions might be deduced from the resultant statistical analyses. The proposed bootstrap variance estimator can provide more accurate and precise interval estimates. The bootstrap method would be an alternative effective approach in practice to provide accurate evidence.
| S-EPMC10701255 | biostudies-literature
Project description:Many studies have reported the association between the matrix metalloproteinase (MMP) polymorphisms and lung cancer susceptibility, but the results were inconclusive. We conducted a meta-analysis, using a comprehensive strategy based on the logistic regression and a model-free approach, to derive a more precise estimation of the relationship between MMP1, MMP2, MMP9 and MMP13 polymorphisms with lung cancer risk. A total of 22 case-control studies including 8202 cases and 7578 controls were included in this meta-analysis. For MMP1-1607 1G/2G, increased lung cancer risk was found among Asians in additive model(OR = 1.34, 95%CI:1.18-1.53) and with model-free approach(ORG = 1.41, 95%CI:1.21-1.65). For MMP2-1306 C/T and -735 C/T, based on the model-free approach, a significantly reduced risk was found in Asians(MMP2-1306 C/T:ORG = 0.49,95%CI:0.42-0.57; MMP2-735 C/T: ORG = 0.71, 95%CI:0.61-0.84). For MMP9-1562 C/T, a significantly increased risk was found among Asians(OR = 2.73, 95%CI:1.74-4.27) with model-free approach. For MMP13-77A/G, there was no association between this polymorphism and lung cancer risk in the recessive model(OR = 1.02, 95%CI:0.83-1.26) and with the model-free approach(ORG = 0.95, 95%CI:0.76-1.17). Therefore, this meta-analysis suggests that the MMP1-1607 1G/2G, MMP2-1306 C/T, MMP2-735 C/T, MMP9 -1562 C/T polymorphisms were risk factors for lung cancer among Asians, while MMP13 -77A/G polymorphism was not associated with lung cancer risk.
| S-EPMC4510488 | biostudies-other
Project description:Case-control studies are an important part of the epidemiologic literature, yet confusion remains about how to interpret estimates from different case-control study designs. We demonstrate that not all case-control study designs estimate odds ratios. On the contrary, case-control studies in the literature often report odds ratios as their main parameter even when using designs that do not estimate odds ratios. Only studies using specific case-control designs should report odds ratios, whereas the case-cohort and incidence-density sampled case-control studies must report risk ratio and incidence rate ratios, respectively. This also applies to case-control studies conducted in open cohorts, which often estimate incidence rate ratios. We also demonstrate the misinterpretation of case-control study estimates in a small sample of highly cited case-control studies in general epidemiologic and medical journals. We therefore suggest that greater care be taken when considering which parameter is to be reported from a case-control study.
| S-EPMC7850067 | biostudies-literature
Project description:BackgroundMixed linear models (MLM) have been widely used to account for population structure in case-control genome-wide association studies, the status being analyzed as a quantitative phenotype. Chen et al. proved in 2016 that this method is inappropriate in some situations and proposed GMMAT, a score test for the mixed logistic regression (MLR). However, this test does not produces an estimation of the variants' effects. We propose two computationally efficient methods to estimate the variants' effects. Their properties and those of other methods (MLM, logistic regression) are evaluated using both simulated and real genomic data from a recent GWAS in two geographically close population in West Africa.ResultsWe show that, when the disease prevalence differs between population strata, MLM is inappropriate to analyze binary traits. MLR performs the best in all circumstances. The variants' effects are well evaluated by our methods, with a moderate bias when the effect sizes are large. Additionally, we propose a stratified QQ-plot, enhancing the diagnosis of p values inflation or deflation when population strata are not clearly identified in the sample.ConclusionThe two proposed methods are implemented in the R package milorGWAS available on the CRAN. Both methods scale up to at least 10,000 individuals. The same computational strategies could be applied to other models (e.g. mixed Cox model for survival analysis).
| S-EPMC7684894 | biostudies-literature
Project description:Odds ratios or other effect sizes estimated from genome scans are upwardly biased, because only the top-ranking associations are reported, and moreover only if they reach a defined level of significance. No unbiased estimate exists based on data selected in this fashion, but replication studies are routinely performed that allow unbiased estimation of the effect sizes. Estimation based on replication data alone is inefficient in the sense that the initial scan could, in principle, contribute information on the effect size. We propose an unbiased estimator combining information from both the initial scan and the replication study, which is more efficient than that based just on the replication. Specifically, we adjust the standard combined estimate to allow for selection by rank and significance in the initial scan. Our approach explicitly allows for multiple associations arising from a scan, and is robust to mis-specification of a significance threshold. We require replication data to be available but argue that, in most applications, estimates of effect sizes are only useful when associations have been replicated. We illustrate our approach on some recently completed scans and explore its efficiency by simulation.
| S-EPMC2726957 | biostudies-literature
Project description:Psychophysical "reverse-correlation" methods allow researchers to gain insight into the perceptual representations and decision weighting strategies of individual subjects in perceptual tasks. Although these methods have gained momentum, until recently their development was limited to experiments involving only two response categories. Recently, two approaches for estimating decision weights in m-alternative experiments have been put forward. One approach extends the two-category correlation method to m > 2 alternatives; the second uses multinomial logistic regression (MLR). In this article, the relative merits of the two methods are discussed, and the issues of convergence and statistical efficiency of the methods are evaluated quantitatively using Monte Carlo simulations. The results indicate that, for a range of values of the number of trials, the estimated weighting patterns are closer to their asymptotic values for the correlation method than for the MLR method. Moreover, for the MLR method, weight estimates for different stimulus components can exhibit strong correlations, making the analysis and interpretation of measured weighting patterns less straightforward than for the correlation method. These and other advantages of the correlation method, which include computational simplicity and a close relationship to other well-established psychophysical reverse-correlation methods, make it an attractive tool to uncover decision strategies in m-alternative experiments.
| S-EPMC3505214 | biostudies-other