Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Statistical Power”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 649 records · Page 36Linked to original sources
PubMed1993

Essential fatty acids, plasma cholesterol, and fat-soluble vitamins in subjects with age-related maculopathy and matched control subjects.

A matched-control study of plasma retinol, alpha-tocopherol, carotenoid, and cholesterol concentrations and the polyunsaturated fatty acid content of plasma and erythrocyte phospholipids was undertaken in 65 elderly patients with age-related maculopathy and 65 control subjects matched for age and sex. Despite the high statistical power of the study and large variations between subjects in the variables under consideration, no significant differences were noted between patients and control subjects. However, several statistically significant differences were noted between male and female subjects independent of their classification with maculopathy or as controls and age: plasma cholesterol, total phospholipids, alpha-tocopherol, and beta-cryptoxanthin concentrations were higher in females than in males. The mean plasma cholesterol concentration for the upper tertile of the whole sample was 7.6 mmol/L. Plasma concentrations of total carotenoids, alpha-carotene, and beta-carotene, but not alpha-tocopherol, were significantly lower in smokers than in non-smokers. The results of this study do not provide any evidence in favor of changing the dietary intake of polyunsaturated fatty acids or fat-soluble vitamins to protect against age-related maculopathy.

Aged↗
PubMed2004

Systemic lupus erythematosus trials: successes and issues.

PURPOSE OF REVIEW: The purpose of this review is to discuss the most recent published clinical trials for systemic lupus erythematosus and to identify important issues that have arisen in association with the search for new therapies for systemic lupus erythematosus, as well as new regimens or indications for the use of "standard-of-care" agents such as corticosteroids and cyclophosphamide. RECENT FINDINGS: Important developments have occurred during the past 2 years as interest in this area has increased, largely because of the participation of pharmaceutical and biotechnical companies in the development and testing of novel agents for systemic lupus erythematosus. Several important large-scale, multicenter, randomized controlled trials have been completed, but none has yet resulted in a new, approved indication for systemic lupus erythematosus. Many issues in the identification of new therapeutic modalities remain. These include the fact that a majority of published reports include either small numbers of patients in controlled trials that lack statistical power to draw conclusions, or are uncontrolled anecdotal series or individual case reports. Among the larger controlled trials, a pervasive issue in the failure to reach statistical significance may be the initial study design. Inclusion of patients with mild and/or stable disease activity does not allow for an effect size sufficient to show differences in treatment arms without recruitment of very large numbers of subjects. Finally, several potentially important trials have been reported only in abstract form to date. Further assessment of the results must await formal publication of these studies.

Biological Products↗
PubMed2006

Testing differences between nested covariance structure models: Power analysis and null hypotheses.

For comparing nested covariance structure models, the standard procedure is the likelihood ratio test of the difference in fit, where the null hypothesis is that the models fit identically in the population. A procedure for determining statistical power of this test is presented where effect size is based on a specified difference in overall fit of the models. A modification of the standard null hypothesis of zero difference in fit is proposed allowing for testing an interval hypothesis that the difference in fit between models is small, rather than zero. These developments are combined yielding a procedure for estimating power of a test of a null hypothesis of small difference in fit versus an alternative hypothesis of larger difference.

Analysis of Variance↗
PubMed1987

Towards a reduction in publication bias.

Current practice results in the publication of many research studies in medical and related disciplines which may be criticised on the grounds of inadequate sample size and statistical power. Small studies continue to be carried out with little more than a blind hope of showing the desired effect. Nevertheless, papers based on such work are submitted for publication, especially if the results turn out to be statistically significant. There is confusion about what makes a result suitable for publication. Often there is a preference for statistically significant results at the peer review stage. Consequently published reports of small studies tend to contain too many false positive results and to exaggerate the true effects. The use of a criterion of a posteriori power does not eliminate the bias; a priori power is the criterion of choice. This could be implemented by peer review of study protocols at the planning stage by funding bodies and journals.

Periodicals as Topic↗
PubMed2005

Why carve up your continuous data?

Continuous data are commonplace in social, biophysical, and health research. For various reasons, researchers often carve up data into ordered chunks. Such data carving results in less information being carried by the data, a reduction or spurious increase in statistical power, and resultant Type I or Type II errors. We give examples of data carving in selected nursing literature, and illustrate how unnecessary categorization can produce erroneous statistical results. Finally, we propose credible alternatives to data carving.

Analysis of Variance↗
PubMed2003

The analysis of the influence of birth order and other factors in multiple birth data.

We compare three methods which can be used to analyse the influence of birth order and other factors on health outcomes in multiple birth data. We consider marginal models based on generalized estimating equations (GEE) and two kinds of conditional models; conditional logistic regression (CLR) and mixed effects models (MEM). Although the models may be written similarly, there are differences in both the interpretation and the numerical values assigned to the parameters. Our main conclusion is that GEE and MEM are preferable to CLR since they provide more flexibility in dealing with missing values and covariates. The choice between GEE and MEM is less obvious and depends on the data, the parameter of interest and statistical power.

Birth Order↗
PubMed2001

Genetic rat models of hypertension: relationship to human hypertension.

Experimental models of human disease are frequently used to investigate the pathophysiology of disease as well as the mechanisms of action of therapeutics. However, as long as models have been used there have been debates about the utility of experimental models and their applicability for human disease on the phenotypic and genomic level. The recent advances in molecular genetics and genomics have provided powerful tools to study the genetics of multifactorial diseases, such as hypertension. However, studies of such diseases in humans remain challenging in part due to lack of statistical power and genetic heterogeneity within patient populations. For hypertension, various rat models have been developed and used for the identification of susceptibility loci for genetic hypertension. With the advent of "comparative genomics," the application of genetic studies to both human and animal model systems allows for a new paradigm, where comparative genomics can be used to bridge between model utility and clinical relevance. This review discusses recent approaches in genetics to facilitate gene discovery for polygenic disorders with specific focus on how comparative mapping can be used to select target regions in the human genome for large-scale association studies and linkage disequilibrium testing in clinical populations.

Animals↗
PubMed2002

Phylogeny shape and the phylogenetic comparative method.

We explored the impact of phylogeny shape on the results of interspecific statistical analyses incorporating phylogenetic information. In most phylogenetic comparative methods (PCMs), the phylogeny can be represented as a relationship matrix, and the hierarchical nature of interspecific phylogenies translates into a distinctive blocklike matrix that can be described by its eigenvectors (topology) and eigenvalues (branch lengths). Thus, differences in the eigenvectors and eigenvalues of different relationship matrices can be used to gauge the impact of possible phylogeny errors by comparing the actual phylogeny used in a PCM analysis with a second phylogenetic hypothesis that may be more accurate. For example, we can use the sum of inverse eigenvalues as a rough index to compare the impact of phylogenies with different branch lengths. Topological differences are better described by the eigenvectors. In general, phylogeny errors that involve deep splits in the phylogeny (e.g., moving a taxon across the base of the phylogeny) are likely to have much greater impact than will those involving small perturbations in the fine structure near the tips. Small perturbations, however, may have more of an impact if the phylogeny structure is highly dependent (with many recent splits near the tips of the tree). Unfortunately, the impact of any phylogeny difference on the results of a PCM depends on the details of the data being considered. Recommendations regarding the choice, design, and statistical power of interspecific analyses are also made.

Biological Evolution↗
PubMed2026

Contribution of a combined approach using refined enterotyping and non-negative matrix factorization (NMF) to the characterization of the gut microbiota in Tunisia, North Africa.

This pilot study aimed to assess the enhanced capabilities of a combined approach using refined enterotyping and non-negative matrix factorization (NMF) for identifying specific microbiota patterns in healthy adults in Tunisia. Shotgun metagenomic sequencing was performed on 21 stool samples. Taxonomic classification was carried out using Kraken2, followed by Bracken analysis. Enterotype (ET) assignment was performed using a publicly available, reference-based classification tool involving Fuzzy-k-means (FKM) clustering. Next, NMF was applied to identify 'enterosignatures' (ESs). The FKM approach revealed a co-dominance of Prevotella-ET (P-ET, 57%) and Firmicutes-ET (F-ET, 38%) with 41% of P-ET samples exhibiting a significant deviation from the reference enterotype center. These latter had a lower proportion of Prevotella-ES and a higher proportion of Bacteroides/Phocaeicola-, Firmicutes- and/or Bifidobacterium-enriched ESs. The F-ET samples were differentially enriched by Blautia (P = 0.007) and Vescimonas (P = 0.007). NMF revealed within this group, a candidate Firmicutes-associated ES driven by Blautia and encompassing Vescimonas, Akkermansia, and Methanobrevibacter. These findings demonstrate the combined power of refined enterotyping and NMF in characterizing gut microbiota, providing a key methodology for future large-scale research. However, our relatively small sample size limits statistical power and biological interpretation, making this study exploratory in nature. Candidate ES requires validation in larger independent datasets.

Humans↗
PubMed1999

Statistical limitations in functional neuroimaging. II. Signal detection and statistical inference.

The field of functional neuroimaging (FNI) methodology has developed into a mature but evolving area of knowledge and its applications have been extensive. A general problem in the analysis of FNI data is finding a signal embedded in noise. This is sometimes called signal detection. Signal detection theory focuses in general on issues relating to the optimization of conditions for separating the signal from noise. When methods from probability theory and mathematical statistics are directly applied in this procedure it is also called statistical inference. In this paper we briefly discuss some aspects of signal detection theory relevant to FNI and, in addition, some common approaches to statistical inference used in FNI. Low-pass filtering in relation to functional-anatomical variability and some effects of filtering on signal detection of interest to FNI are discussed. Also, some general aspects of hypothesis testing and statistical inference are discussed. This includes the need for characterizing the signal in data when the null hypothesis is rejected, the problem of multiple comparisons that is central to FNI data analysis, omnibus tests and some issues related to statistical power in the context of FNI. In turn, random field, scale space, non-parametric and Monte Carlo approaches are reviewed, representing the most common approaches to statistical inference used in FNI. Complementary to these issues an overview and discussion of non-inferential descriptive methods, common statistical models and the problem of model selection is given in a companion paper. In general, model selection is an important prelude to subsequent statistical inference. The emphasis in both papers is on the assumptions and inherent limitations of the methods presented. Most of the methods described here generally serve their purposes well when the inherent assumptions and limitations are taken into account. Significant differences in results between different methods are most apparent in extreme parameter ranges, for example at low effective degrees of freedom or at small spatial autocorrelation. In such situations or in situations when assumptions and approximations are seriously violated it is of central importance to choose the most suitable method in order to obtain valid results.

Biometry↗
PubMed2004

Association between the TPH gene A218C polymorphism and suicidal behavior: a meta-analysis.

Genes encoding proteins involved in serotonergic metabolism are major candidates in association studies of suicidal behavior. The tryptophan hydroxylase (TPH) gene, which codes for the rate-limiting enzyme of serotonin biosynthesis, is a major candidate gene and has been extensively studied in association studies of suicidal behavior, providing conflicting results. It is difficult to interpret these conflicting results due to lack of power, ethnic heterogeneity, and variations in the sampling strategies (in particular for controls) and in the polymorphism of the TPH gene studied. Meta-analyses can improve the statistical power for the analysis of the effects of candidate vulnerability factors. The analysis of the sources of heterogeneity that contribute to these conflicting results is an important step in the interpretation of these conflicting association results and in the interpretation of the results of a meta-analysis. We selected all of the published association studies between the TPH gene polymorphism and suicidal behavior. Nine association studies between the A218C TPH polymorphism and suicidal behavior fulfilled the inclusion criteria. A significant association was observed between the A218C polymorphism and suicidal behavior using the fixed effect method (odds ratio (OR) = 1.62; 95% confidence interval (CI) = [1.26; 2.07]) and the random effect method (OR = 1.61; 95% CI = [1.11; 2.35]). The analysis of the sources of heterogeneity showed that two studies (one positive and one negative) significantly deviated from the calculated global effect. The meta-analysis performed after removing those two studies also revealed a significant association between the TPH A218C polymorphism and suicidal behavior. Both analyses suggested that the A allele has a dose-dependent effect on the risk of suicidal behavior.

Alleles↗
PubMed1992

Gamma distribution and house 222Rn measurements.

The statistical distribution of 222Rn measurements from basements and first floors of homes in northeastern Pennsylvania was investigated. The gamma distribution was statistically significantly superior to the normal distribution (p less than 0.005) in describing the frequency distribution of the logarithm of observed 222Rn levels. The fit to the data was closer both in the central portion and in the upper tail. The gamma distribution has certain characteristics that make it generally useful in the study of environmental toxic agents where several different exposures over a lifetime occur and must be combined, as for risk assessment or for statistical power calculations for epidemiologic studies.

Air Pollution, Indoor↗
PubMed1993

Measuring male reproductive hormones for occupational field studies.

As part of our longitudinal study of unexposed workers, we drew blood samples and analyzed the individual endocrine profiles for 45 men. The blood collection was between 8 AM and 8 PM, and three blood samples were drawn 20 minutes apart on three occasions during the course of the study (June, October, and February). Serum concentrations of follicle-stimulating hormone, luteinizing hormone, testosterone, and prolactin were determined. A component of variance model was used to estimate variability between the 20-minute blood draws. Statistical power analysis using this component showed that three blood draws provide a marginal improvement over a single blood draw in detecting population shifts. Also, if the prospect of three blood draws reduces subject participation by 10 to 20%, the increase in power would be negated.

Adult↗
PubMed1994

Low power? Use two-dimensional confidence regions as a graphical method for depicting uncertainty.

Prospective studies of rare outcomes, such as HIV seroconversion or obsessive-compulsive disorder, can often result in small sample sizes with limited power for detecting associations. For this reason, it is useful to develop graphical procedures that enable researchers to depict uncertainty around parameter estimates and examine the direction of association when statistical power is low. Classical procedures include the reporting of confidence intervals, which typically are derived from asymptotic normality of parameters estimated using large samples. In this paper, we present a likelihood-based procedure for the estimation of the confidence region of two parameters from a conditional logistic regression of a nested case-control study with a relatively small number of cases. Graphical depiction of the confidence regions provides an easily comprehensible procedure to quantify the uncertainty of the estimation based on small samples.

Adult↗
PubMedDate not supplied

Neurobehavioral Evaluation System (NES): comparative performance of 2nd-, 4th-, and 8th-grade Czech children.

The Neurobehavioral Evaluation System was designed for field studies of workers, but many NES tests can be performed satisfactorily by children as young as 7 or 8 years old and a few tests, such as simple reaction time, can be performed by preschool children. However, little comparative data from children of different ages or grade levels are available. Studies of school children in the Czech Republic indicate that 2nd-grade children could perform the following NES tests satisfactorily: Finger Tapping, Visual Digit Span. Continuous Performance, Symbol-Digit Substitution, Pattern Comparison, and simpler conditions of Switching Attention. Comparative scores of boys and girls from the 2nd, 4th, and 8th grades and power analyses to estimate appropriate sample size were presented. Performance varied systematically with grade level and gender. Larger samples were needed with younger children to achieve comparable levels of statistical power. Gender comparisons indicated that boys responded faster, but made more errors than girls.

Adolescent↗
PubMed2005

Magnetic resonance criteria for future trials of cardiac resynchronization therapy.

Current patient selection criteria for Cardiac Resynchronization Therapy (CRT), an efficacious treatment for heart failure, include no measure of disconjugate cardiac contractility other than prolonged QRS on electrocardiogram. Using cardiac magnetic resonance imaging, we examined the roles of cardiac asymmetry, asynchrony, and circumferential strain in DCC with the principal aim of generating a robust numerical index for use in future trials of CRT. Standard cardiac magnetic resonance imaging was done on a GE 1.5 Tesla Signa LX MRI clinical scanner (GE Healthcare, Milwaukee, WI, USA) and analyzed by MASS Analysis (MEDIS, Leiden, The Netherlands). The methods were evaluated in eleven patients with advanced heart failure due to ischemic and non-ischemic cardiomyopathy, who did not qualify under current criteria for CRT, five CRT candidates pre-op and eleven normal subjects. Using t-test and standardized differences (SD = sd/diff, Power (N) = number of patients to reach p < .05) we determined efficacy. Indices of asymmetry and asynchrony (Ism and Isn, respectively) could be measured with accuracy and provided excellent statistical power when used as surrogate markers to delineate heart failure and CRT patients from control subjects. Asymmetry and asynchrony in heart contraction are both critical components of dilated cardiomyopathy that can be improved by CRT. Magnetic resonance asynchrony is efficacious in screening patients and should now be compared with recently published echocardiography data to improve outcome for this costly but valuable therapy.

Adult↗
PubMed2004

Prospects for admixture mapping of complex traits.

Admixture mapping extends to human populations the principles that underlie linkage analysis of an experimental cross. For detecting genes that contribute to ethnic variation in disease risk, admixture mapping has greater statistical power than family-linkage studies. In comparison with association studies, admixture mapping requires far fewer markers to search the genome and is less affected by allelic heterogeneity. Statistical-analysis programs for admixture mapping are now available, and a genomewide panel of markers for admixture mapping in populations formed by West African-European admixture has been assembled. Some of the remaining technical challenges include the ability to ensure that the statistical methods are robust and to develop marker panels for other admixed populations. Where admixed populations and panels of markers informative for ancestry are available, admixture mapping can be applied to localize genes that contribute to ethnic variation in any measurable trait.

Black People↗
PubMed2001

Estimates of appropriate number of rats: interaction with housing environment.

An extensive list of physiological parameters from previous experiments was re-analysed in order to evaluate the effects of enrichment, cage type and group size on the within-group variation and hence on the number of animals needed in studies using Wistar rats. The independent factors studied in these experiments included the provision of aspen gnawing blocks for enrichment, solid bottom cages (SBCs) and grid floor cages (GFCs) and animal number per cage (varied from 1-4). SOLO power analysis was used to calculate the smallest number of animals (n) needed to detect an arbitrarily chosen 20% effect size, when significance was set at P = 0.05 and statistical power at 0.90. N ratios (nlarger/nsmaller) were calculated for the effect of enrichment, cage type and group size to compare the 'treatment group' with the 'control group'. The n values of adrenal gland, interscapular brown adipose tissue (BAT) and epididymal adipose tissue (EAT) weights seemed to vary most, whereas final body weight (FBWJ and growth seemed to be the least variable ones. According to one-sample t-test, the N ratios of most physiological parameters differed significantly from zero (except the ones of FBW) indicating that n values in 'treatment' and 'control' groups were unequal. The results indicate that some of the physiological parameters are susceptible to variability attributable to environmental modifications in general whereas some are not. Furthermore, they suggest that the variation of different parameters may vary from one experiment to another and between different environments thus hindering the estimations of appropriate number of animals.

Animal Husbandry↗