Addressing the estimation of standard errors in fixed effects meta-analysis.
Ontology highlight
ABSTRACT:
SUBMITTER: Dominguez Islas C
PROVIDER: S-EPMC6001579 | biostudies-literature | 2018 May
REPOSITORIES: biostudies-literature
Publications
Addressing the estimation of standard errors in fixed effects meta-analysis.
Domínguez Islas Clara C Rice Kenneth M KM
Statistics in medicine 20180325 11
Standard methods for fixed effects meta-analysis assume that standard errors for study-specific estimates are known, not estimated. While the impact of this simplifying assumption has been shown in a few special cases, its general impact is not well understood, nor are general-purpose tools available for inference under more realistic assumptions. In this paper, we aim to elucidate the impact of using estimated standard errors in fixed effects meta-analysis, showing why it does not go away in la ...[more]
Similar Datasets
Project description:Issue: The estimation of foetal weight is an integral part of prenatal care and obstetric routine. In spite of its known susceptibility to errors in cases of underweight or overweight babies, important obstetric decisions depend on it. In the present contribution we have examined the accuracy and error distribution of 35 weight estimation formulae within the normal weight range of 2500-4000 g. The aim of the study was to identify the weight estimation formulae with the best possible correspondence to the requirements of clinical routine. Materials and Methods: 35 clinically established weight estimation formulae were analysed in 3416 foetuses with weights between 2500 and 4000 g. For this we determined and compared the mean percentage error (MPE), the mean absolute percentage error (MAPE), and the proportions of estimates within the error ranges of 5, 10, 20 and 30 %. In addition, separate regression lines were calculated for the relationship between estimated and actual birth weights for the weight range 2500-4000 g. The formulae were thus examined for possible inhomogeneities. Results: The lowest MPE were achieved with the Hadlock III and V formulae (0.8 %, STW 9.2 % or, respectively, -0.8 %, STW 10.0 %). The lowest absolute error (6.6 %) as well as the most favourable frequency distribution in cases below 5 % and 10 % error (43.9 and 77.5) were seen for the Halaska formula. In graphic representations of the regression lines, 16 formulae revealed a weight overestimation in the lower weight range and an underestimation in the upper range. 14 formulae gave underestimations and merely 5 gave overestimations over the entire tested weight range. Conclusion: The majority of the tested formulae gave underestimations of the actual birth weight over the entire weight range or at least in the upper weight range. This result supports the current strategy of a two-stage weight estimation in which a formula is first chosen after a pre-estimation of the weight range.
| S-EPMC5123881 | biostudies-literature
Project description:A reference interval provides a basis for physicians to determine whether a measurement is typical of a healthy individual. It can be interpreted as a prediction interval for a new individual from the overall population. However, a reference interval based on a single study may not be representative of the broader population. Meta-analysis can provide a general reference interval based on the overall population by combining results from multiple studies. Methods for estimating the reference interval from a random effects meta-analysis have been recently proposed to incorporate the within and between-study variation, but a random effects model may give imprecise estimates of the between-study variation with only few studies. In addition, the normal distribution of underlying study-specific means, and equal within-study variance assumption in these methods may be inappropriate in some settings. In this article, we aim to estimate the reference interval based on the fixed effects model assuming study effects are unrelated, which is useful for a meta-analysis with only a few studies (e.g., ≤5). We propose a mixture distribution method only assuming parametric distributions (e.g., normal) for individuals within each study and integrating them to form the overall population distribution. This method is compared to an empirical method only assuming a parametric overall population distribution. Simulation studies have shown that both methods can estimate a reference interval with coverage close to the targeted value (i.e., 95%). Meta-analyses of women daytime urination frequency and frontal subjective postural vertical measurements are reanalyzed to demonstrate the application of our methods.
| S-EPMC8924903 | biostudies-literature
Project description:ObjectivesWe discuss the problem of computing the standard errors of functions involving estimated parameters and provide the relevant computer code for three different computational approaches using two popular computer packages.Study designWe show how to compute the standard errors of several functions of interest: the predicted value of the dependent variable for a particular subject, and the effect of a change in an explanatory variable on the predicted value of the dependent variable for an individual subject and average effect for a sample of subjects. EMPIRICAL APPLICATION: Using a publicly available dataset, we explain three different methods of computing standard errors: the delta method, Krinsky–Robb, and bootstrapping. We provide computer code for Stata 12 and LIMDEP 10/NLOGIT 5.ConclusionsIn most applications, choice of the computational method for standard errors of functions of estimated parameters is a matter of convenience. However, when computing standard errors of the sample average of functions that involve both estimated parameters and nonstochastic explanatory variables, it is important to consider the sources of variation in the function's values.
| S-EPMC3976195 | biostudies-literature
Project description:BACKGROUND: When multiple endpoints are of interest in evidence synthesis, a multivariate meta-analysis can jointly synthesise those endpoints and utilise their correlation. A multivariate random-effects meta-analysis must incorporate and estimate the between-study correlation (rhoB). METHODS: In this paper we assess maximum likelihood estimation of a general normal model and a generalised model for bivariate random-effects meta-analysis (BRMA). We consider two applied examples, one involving a diagnostic marker and the other a surrogate outcome. These motivate a simulation study where estimation properties from BRMA are compared with those from two separate univariate random-effects meta-analyses (URMAs), the traditional approach. RESULTS: The normal BRMA model estimates rhoB as -1 in both applied examples. Analytically we show this is due to the maximum likelihood estimator sensibly truncating the between-study covariance matrix on the boundary of its parameter space. Our simulations reveal this commonly occurs when the number of studies is small or the within-study variation is relatively large; it also causes upwardly biased between-study variance estimates, which are inflated to compensate for the restriction on rhoB. Importantly, this does not induce any systematic bias in the pooled estimates and produces conservative standard errors and mean-square errors. Furthermore, the normal BRMA is preferable to two normal URMAs; the mean-square error and standard error of pooled estimates is generally smaller in the BRMA, especially given data missing at random. For meta-analysis of proportions we then show that a generalised BRMA model is better still. This correctly uses a binomial rather than normal distribution, and produces better estimates than the normal BRMA and also two generalised URMAs; however the model may sometimes not converge due to difficulties estimating rhoB. CONCLUSION: A BRMA model offers numerous advantages over separate univariate synthesises; this paper highlights some of these benefits in both a normal and generalised modelling framework, and examines the estimation of between-study correlation to aid practitioners.
| S-EPMC1800862 | biostudies-other
Project description:When conducting a meta-analysis of a continuous outcome, estimated means and standard deviations from the selected studies are required in order to obtain an overall estimate of the mean effect and its confidence interval. If these quantities are not directly reported in the publications, they must be estimated from other reported summary statistics, such as the median, the minimum, the maximum, and quartiles.We propose a simulation-based estimation approach using the Approximate Bayesian Computation (ABC) technique for estimating mean and standard deviation based on various sets of summary statistics found in published studies. We conduct a simulation study to compare the proposed ABC method with the existing methods of Hozo et al. (2005), Bland (2015), and Wan et al. (2014).In the estimation of the standard deviation, our ABC method performs better than the other methods when data are generated from skewed or heavy-tailed distributions. The corresponding average relative error (ARE) approaches zero as sample size increases. In data generated from the normal distribution, our ABC performs well. However, the Wan et al. method is best for estimating standard deviation under normal distribution. In the estimation of the mean, our ABC method is best regardless of assumed distribution.ABC is a flexible method for estimating the study-specific mean and standard deviation for meta-analysis, especially with underlying skewed or heavy-tailed distributions. The ABC method can be applied using other reported summary statistics such as the posterior mean and 95 % credible interval when Bayesian analysis has been employed.
| S-EPMC4542106 | biostudies-literature
Project description:As DRAM technology continues to evolve towards smaller feature sizes and increased densities, faults in DRAM subsystem are becoming more severe. Current servers mostly use CHIPKILL based schemes to tolerate up-to one/two symbol errors per DRAM beat. Such schemes may not detect multiple symbol errors arising due to faults in multiple devices and/or data-bus, address bus. In this article, we introduce Single Symbol Correction Multiple Symbol Detection (SSCMSD)-a novel error handling scheme to correct single-symbol errors and detect multi-symbol errors. Our scheme makes use of a hash in combination with Error Correcting Code (ECC) to avoid silent data corruptions (SDCs). We develop a novel scheme that deploys 32-bit CRC along with Reed-Solomon code to implement SSCMSD for a ×4 based DDR4 system. Simulation based experiments show that our scheme effectively guards against device, data-bus and address-bus errors only limited by the aliasing probability of the hash. Our novel design enabled us to achieve this without introducing additional READ latency. We need 19 chips per rank, 76 data bus-lines and additional hash-logic at the memory controller.
| S-EPMC7959593 | biostudies-literature
Project description:Computational models used in the estimation of thermodynamic quantities of large chemical systems often require approximate energy models that rely on parameterization and cancellation of errors to yield agreement with experimental measurements. In this work, we show how energy function errors propagate when computing statistical mechanics-derived thermodynamic quantities. Assuming that each microstate included in a statistical ensemble has a measurable amount of error in its calculated energy, we derive low-order expressions for the propagation of these errors in free energy, average energy, and entropy. Through gedanken experiments we show the expected behavior of these error propagation formulas on hypothetical energy surfaces. For very large microstate energy errors, these low-order formulas disagree with estimates from Monte Carlo simulations of error propagation. Hence, such simulations of error propagation may be required when using poor potential energy functions. Propagated systematic errors predicted by these methods can be removed from computed quantities, while propagated random errors yield uncertainty estimates. Importantly, we find that end-point free energy methods maximize random errors and that local sampling of potential energy wells decreases random error significantly. Hence, end-point methods should be avoided in energy computations and should be replaced by methods that incorporate local sampling. The techniques described herein will be used in future work involving the calculation of free energies of biomolecular processes, where error corrections are expected to yield improved agreement with experiment.
| S-EPMC3568774 | biostudies-literature
Project description:BackgroundPrevious functional magnetic resonance imaging studies indicated that acupuncture could activate the brain regions in patients with migraine. However, these studies showed inconsistent results. This activation likelihood estimation (ALE) meta-analysis aimed to investigate the consistent activated change of brain regions between pre- and post-acupuncture treatment in migraineurs.MethodsWe conducted a literature search in PubMed, Embase, Web of Science, the Cochrane Library, the China National Knowledge Infrastructure, the Chinese Science and Technology Periodical Database, the Wanfang Database, and the Chinese Biomedical Literature Database from their inception to 18 August, 2022, to obtain articles assessing the functional magnetic resonance imaging changes of acupuncture for migraine. Two investigators independently performed literature selection, data extraction, and quality assessment. The methodological quality was assessed with a modified version of the checklist. The reporting quality of interventions among included studies was evaluated by the Revised Standards for Reporting Interventions in Clinical Trials of Acupuncture (STRICTA). Our meta-analysis was conducted according to the GingerALE software. The Jackknife sensitivity analysis was used to assess the robustness of the results.Results14 articles were finally included according to the eligible criteria. Regarding the immediate effect of acupuncture on migraine, the ALE meta-analysis demonstrated that the deactivation regions were mainly located in the superior frontal gyrus, and middle frontal gyrus (uncorrected P < 0.001). The ALE meta-analysis of the cumulative effect showed that the activation regions were the thalamus, superior frontal gyrus, posterior lobe of the cerebellum, insula, middle frontal gyrus, precentral gyrus, anterior cingulate, and the deactivation brain regions were located in the transverse temporal gyrus, postcentral gyrus, superior temporal gyrus, anterior cingulate, parahippocampal gyrus, inferior parietal lobule, and inferior occipital gyrus (uncorrected P < 0.001).ConclusionAcupuncture could activate multiple brain areas related with the regulation of pain conduction, processing, emotion, cognition, and other brain regions in patients with migraine. In the future, the combination of multiple imaging technologies could be a new approach to deeply investigate the central mechanism of acupuncture for migraine.
| S-EPMC9911686 | biostudies-literature
Project description:In meta-analysis based on continuous outcome, estimated means and corresponding standard deviations from the selected studies are key inputs to obtain a pooled estimate of the mean and its confidence interval. We often encounter the situation that these quantities are not directly reported in the literatures. Instead, other summary statistics are reported such as median, minimum, maximum, quartiles, and study sample size. Based on available summary statistics, we need to estimate estimates of mean and standard deviation for meta-analysis. We developed an R Shiny code based on approximate Bayesian computation (ABC), ABCMETA, to deal with this situation. In this article, we present an interactive and user-friendly R Shiny application for implementing the proposed method (named ABCMETAapp). In ABCMETAapp, users can choose an underlying outcome distribution other than the normal distribution when the distribution of the outcome variable is skewed or heavy tailed. We show how to run ABCMETAapp with examples. ABCMETAapp provides an R Shiny implementation. This method is more flexible than the existing analytical methods since estimation can be based on five different distributions (Normal, Lognormal, Exponential, Weibull, and Beta) for the outcome variable.
| S-EPMC8596912 | biostudies-literature
Project description:BackgroundFormalin-fixed, paraffin-embedded (FFPE) tissue is the gold standard in pathology tissue storage, representing the largest collections of patient material. Their reliable use for DNA analyses could open a trove of potential samples for research and are currently being recognised as a viable source material for bacterial analysis. There are several key features which limit bacterial-related data generation from this material: (i) DNA damage inherent to the fixing process, (ii) low bacterial biomass that increases the vulnerability to contamination and exacerbates the host DNA effects and (iii) lack of suitable DNA extraction methods, leading to data bias. The development and systematic use of reliable standards is a key priority for microbiome research. More than perhaps any other sample type, FFPE material urgently requires the development of standards to ensure the validity of results and to promote reproducibility.ResultsTo address these limitations and concerns, we have developed the Protoblock as a biological standard for FFPE tissue-based research and method optimisation. This is a novel system designed to generate bespoke mock FFPE 'blocks' with a cell content that is user-defined and which undergoes the same treatment conditions as clinical FFPE tissues. The 'Protoblock' features a mix of formalin-fixed cells, of known number, embedded in an agar matrix which is solidified to form a defined shape that is paraffin embedded. The contents of various Protoblocks populated with mammalian and bacterial cells were verified by microscopy. The quantity and condition of DNA purified from blocks was evaluated by qPCR, 16S rRNA gene amplicon sequencing and whole genome sequencing. These analyses validated the capability of the Protoblock system to determine the extent to which each of the three stated confounding features impacts on eventual analysis of cellular DNA present in FFPE samples.ConclusionThe Protoblock provides a representation of biological material after FFPE treatment. Use of this standard will greatly assist the stratification of biological variations detected into those legitimately resulting from experimental conditions, and those that are artefacts of the processed nature of the samples, thus enabling users to relate the outputs of laboratory analyses to reality. Video Abstract.
| S-EPMC7443293 | biostudies-literature