Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Statistical power analysis”

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 775 records · Page 43Linked to original sources
PubMed2000

First measured plasma concentration value as C(max); impact on the C(max) confidence interval in bioequivalence studies.

In bioequivalence studies, the first blood or plasma sample taken after dosing sometimes yields a higher assayed drug concentration than any samples drawn thereafter. This circumstance ('first C(max)' or 'FCM'), is usually considered undesirable, since a 'true C(max)' requires that the sampled concentrations immediately preceding and immediately after the 'true C(max)' concentration should be lower than the 'true C(max)' concentration. Therefore, a question arises whether the presence of FCM in a bioequivalence study affects the power and accuracy of the computed statistical confidence interval (CI) for C(max). This study examines what effect, if any, the inclusion or exclusion of FCM data has on the statistical power and accuracy of the 90% CI computed for C(max) in the analysis of results for in vivo bioequivalence studies. Actual experimental study data as well as data from simulated studies were evaluated. In the simulated studies, up to half of the study subjects exhibited FCM, and various levels of intrasubject variability were incorporated into the absorption rate constant. The two one-sided tests procedure was used to assess equivalence of C(max) for the test versus reference products when either a complete set of C(max) data was analysed (designated 'CC(max)'), which included subjects with FCM profiles; or a truncated set of data was analysed (designated 'TC(max)'), that excluded all subjects with FCM. The results showed that the CC(max) metric had greater statistical power and comparable or greater statistical accuracy compared to TC(max) for both bioequivalent and non-bioequivalent drug product formulations. Even when up to 50% of the study subjects had FCM, the power and accuracy of the 90% CI for rate of absorption (i.e. C(max)) was not significantly affected. Consequently, this study shows that, in the analysis of data from conventional in vivo bioequivalence studies, the inclusion of 50% of the subjects exhibiting FCM does not greatly impact the statistical results obtained for C(max). Published in 2000 by John Wiley & Sons, Ltd.

Absorption↗
PubMed2003

Radiation dose is reduced with a single-pass whole-body multi-detector row CT trauma protocol compared with a conventional segmented method: initial experience.

Radiation dose data were collected from a calibrated multi-detector row computed tomographic (CT) scanner during trauma CT. One protocol (used with 10 case subjects) involved a single-pass continuous whole-body acquisition from cranial vertex to symphysis pubis, while the other, conventional protocol (used with 10 control subjects) involved scouting and scanning body segments (head, cervical spine, chest, abdomen, and pelvis) individually. Technical factors were kept constant within each body segment for the single-pass and the segmented protocols. Statistics included univariate analysis, two-tailed t testing to evaluate statistical significance of the summary statistic, and power and subject population contingency tables. The mean dose length product (DLP) with the single-pass protocol was 17% lower than the sum of the DLPs of each of the individual body segment scans (P <.001). Analysis of power and subject population by using a difference in mean of 500 mGy. cm and an alpha of.05 revealed a (1-beta) of higher than 0.90 for a sample of 10 patients. Thus, a whole-body single-pass trauma protocol, compared with a typical segmented acquisition protocol matched for imaging technique, resulted in reduced total radiation dose. The reduction in radiation dose is thought to represent a reduction in redundant imaging at overlap zones between body segments scanned in the segmental protocol but not in the continuous acquisition.

Humans↗
PubMed2003

Within-sample variation of fecal glucocorticoid measurements.

Fecal glucocorticoid metabolite analysis is a useful tool for monitoring adrenocortical activity in captive and free-ranging wildlife. Glucocorticoid metabolites may not be evenly distributed within fecal samples and this variability could affect the interpretation of glucocorticoid metabolite measurements. Furthermore, the precision (i.e., repeatability of measurements) of fecal glucocorticoid measurements from well-mixed samples is unknown. We collected fresh white-tailed deer (Odocoileus virginianus) feces at various times pre- and post-adrenocorticotropin injection to provide samples with low (<75 ng/g), medium (75-90 ng/g), and high (>90 ng/g) glucocorticoid concentrations in case variability differs in samples of dissimilar hormone metabolite concentrations. We compared two sampling methods (selection of three pellet groups [one from each end of the fecal mass and one from the center] versus sampling three small portions of the thoroughly mixed fecal mass) to estimate within-sample variation of glucocorticoid metabolites. Glucocorticoid metabolite measures from pellet groups were higher than fecal glucocorticoid measures from mixed samples in the low (F=3.10; df = 1,56; P=0.08) and medium concentration (F=7.28; df = 1,50; P=0.01) groups. Fecal glucocorticoid metabolite estimates from mixed samples were less variable than glucocorticoid metabolite measures using pellet groups from the same fecal mass, although these differences were not statistically significant (low group: F=0.59; df = 1,38; P=0.45; medium group: F=0.13; df = 1,34; P=0.72; high group: F=2.30; df = 1,28; P=0.14). The mean coefficient of variation was <10% across all treatment groups and sampling methods. However, a power analysis indicated the mixed sub-sample technique requires fewer samples to detect statistically significant differences than pellet groups. Our results suggest that glucocorticoid metabolites may not be evenly distributed in white-tailed deer feces. Consequently, using only a few pellets from a fecal mass may bias assay interpretation. We suggest researchers homogenize the entire fecal mass before removing a sub-sample of fecal material for analysis. Also, other sources of variation must be considered when interpreting results of fecal glucocorticoid studies.

Animals↗
PubMed2006

A comparison of two dietary instruments for evaluating the fat-breast cancer relationship.

BACKGROUND: Previous research suggests food diaries may be more efficient than food frequency questionnaires (FFQ) in detecting a dietary fat-breast cancer relationship. We assessed this further using 4 day food records (FRs) and FFQs in a large sample. METHODS: Participants were from the non-intervention group of the dietary modification component of the Women's Health Initiative Clinical Trial: 603 breast cancer cases and 1206 controls matched on age, clinic, and length of follow-up. Relative risks (RRs) were estimated using unconditional logistic regression, adjusted for confounders and for the selection into the trial of women with an FFQ report exceeding 32% calories from fat. Direct comparison of the statistical power of the two instruments used the standardized log RR. An alternative analysis after removing subjects with missing covariate data was also conducted. RESULTS: The RR estimate for breast cancer in the top quintile of total fat intake, adjusted for confounders and total energy, was 1.82 (P for trend 0.02) for the FR but 0.67 for the FFQ (P for trend 0.24). Following adjustment for selection, estimates were 2.09 (P for trend 0.008) for the FR (alternative: 2.54, P for trend 0.006) and 1.71 (P for trend 0.18) for the FFQ (alternative: 1.24, P for trend 0.41). Similar results were seen for fat subtypes, particularly unsaturated fats. Comparisons showed higher statistical power for the FR than the FFQ (e.g. total fat, P = 0.08: alternative P = 0.01). CONCLUSIONS: Alternative instruments, such as FRs, may be preferable to FFQs for evaluating diet-disease relationships in cohort studies. The results support a positive association between dietary fat and breast cancer.

Aged↗
PubMed2004

Resolution of large and small differences in gene expression using models for the Bayesian analysis of gene expression levels and spotted DNA microarrays.

BACKGROUND: The detection of small yet statistically significant differences in gene expression in spotted DNA microarray studies is an ongoing challenge. Meeting this challenge requires careful examination of the performance of a range of statistical models, as well as empirical examination of the effect of replication on the power to resolve these differences. RESULTS: New models are derived and software is developed for the analysis of microarray ratio data. These models incorporate multiplicative small error terms, and error standard deviations that are proportional to expression level. The fastest and most powerful method incorporates additive small error terms and error standard deviations proportional to expression level. Data from four studies are profiled for the degree to which they reveal statistically significant differences in gene expression. The gene expression level at which there is an empirical 50% probability of a significant call is presented as a summary statistic for the power to detect small differences in gene expression. CONCLUSIONS: Understanding the resolution of difference in gene expression that is detectable as significant is a vital component of experimental design and evaluation. These small differences in gene expression level are readily detected with a Bayesian analysis of gene expression level that has additive error terms and constrains samples to have a common error coefficient of variation. The power to detect small differences in a study may then be determined by logistic regression.

Algorithms↗
PubMed1994

Statistical power to detect occupationally related respiratory cancer risk in a cohort of female employees in the US man-made vitreous fiber industry.

The current update of the US man-made vitreous fiber production worker cohort includes women for the first time. Preliminary comparisons of 3,820 female and 27,767 male workers from 11 participating fibrous glass plants show different hiring patterns during World War II. The gender-specific person-year distributions are similar with respect to duration of employment and time since first employment. The current follow-up of 118,559 person-years for women provides an estimated 80% power to detect a threefold relative risk of respiratory cancer for women who worked more than 14 years in these plants, based on a Poisson regression analysis of the cohort rates. When women comprise a small fraction of the cohort, the statistical power may be inadequate to detect risks of the magnitude typically of interest in studies of men.

Aged↗
PubMed1996

A comparison of t test, F test, and coherence methods of detecting steady-state auditory-evoked potentials, distortion-product otoacoustic emissions, or other sinusoids.

Sinusoids in background noise can conveniently be detected using unsegmented power spectra, comparing power at the signal frequency to average power at several neighbor frequencies. In this case, the F test is preferable to t tests based on rms or dB values, because of the skewed distributions of rms and dB when signal-to-noise ratio (SNR) = 0. F-test performance improves as the number of frequencies increases, to about 15, but can be degraded if the background noise is not white, with a slope exceeding about 10 dB for the range of frequencies sampled. Segment analysis, using magnitude-squared coherence (MSC) or related statistics, has equivalent statistical power; MSC and F each yield unbiased SNR estimates that have identical distributions when SNR = 0. Selection of F or MSC for detection of sinusoids will usually be a matter of convenience.

Evoked Potentials, Auditory↗
PubMed1988

[Evaluation of the methodological quality of randomized therapeutic trials].

A grid devised to evaluate the methodological quality of randomized therapeutic trials is presented. Fourteen items are analysed: description of the principal criterion of assessment, criteria of inclusion, number of subjects seen and excluded, number of subjects randomized to each group, number of subjects excluded who are alive or lost to follow-up and reasons for the fall-outs, blind character of the doctors, patients and persons responsible for the assessment criteria, calculation of the number of subjects required before starting the trial, method of randomization and its blind character, analysis and discussion of covariables, statistical tests used, taking into account of the fall-outs and power of the trials with non-significant results. This grid, used for the analysis of 527 randomized trials, showed that about one-half of them were of mediocre quality, the most frequent defects encountered being the multiplicity of assessment criteria, the lack of description of the subjects excluded, the absence of calculation of the number of subjects required before starting the trial and the small number of subjects in trials with non-significant results.

Clinical Trials as Topic↗
PubMed1997

Cephalometric and demographic characteristics of obstructive sleep apnea: an evaluation with partial least squares analysis.

Obstructive sleep apnea (OSA) is caused by repeated obstruction of the upper airway during sleep. The purpose of this study was to test the relative contributions of specific demographic and cephalometric measurements to OSA severity. Demographic, cephalometric, and overnight polysomnographic records of 291 male OSA patients and 49 male nonapneic snorers were evaluated. A partial least squares (PLS) analysis was used for statistical evaluation. The results revealed that the predictive powers of obesity and neck size variables for OSA severity were higher than the cephalometric variables used in this study. Compared with other cephalometric characteristics, an extended and forward natural head posture, lower hyoid bone position, increased soft palate and tongue dimensions, and decreased nasopharyngeal and velopharyngeal airway dimensions had relatively higher associations with OSA severity. The respiratory disturbance index (RDI) was the OSA outcome variable that was best explained by the demographic and cephalometric predictor variables. We conclude that the PLS analysis can successfully summarize the correlations between a large number of variables, and that obesity, neck size, and certain cephalometric measurements may be used together to evaluate OSA severity.

Adolescent↗
PubMed1994

Analysis of factorial survival experiments.

Several new methodological issues that arise within two-way factorial designs for survival experiments are discussed within the framework of asymptotic theory for the proportional hazards model with two binary treatment covariates. These issues include: the proper formulation of null hypotheses and alternatives, the choice among log-rank and adjusted or stratified log-rank statistics, the asymptotic correlation between test statistics for the separate main effects, the asymptotic power (under the various possible methods of analysis) of tests to detect main effects and interactions, the comparison of power to detect main effects within a 2 x 2 factorial design with power in a three-group trial where no patients are randomized simultaneously to both treatments, and the problems of analysis arising when accrual or exposure to one of the treatments is terminated early for ethical reasons.

Biometry↗
PubMed1999

A relative power table for nested matched case-control studies.

OBJECTIVE: To calculate relative powers for nested case-control studies for different values of both relative risk and numbers of controls per case, given a fixed number of cases available for analysis. METHODS: Algebraic and numerical methods. RESULTS: In nested case-control studies, statistical power is a function of relative risk, rarity of exposure, number of case-control sets, and the number of controls per case. CONCLUSION: The dictum that sufficient power will be obtained in a nested case-control study by selecting only four controls per case cannot be sustained. Appropriate numbers need to be calculated for specific studies.

Case-Control Studies↗
PubMed2006

Meta-analysis based on control of false discovery rate: combining yeast ChIP-chip datasets.

MOTIVATION: High-throughput microarray technology can be used to examine thousands of features, such as all the genes of an organism, and measure their expression. Two important issues of microarray bioinformatics are first, how to combine the significance values for each feature across experiments with high statistical power, and second, how to control the proportion of false positives. Existing methods address these issues separately, in spite of their linked usage. RESULTS: We present a novel method (ESP) to address the two requirements in an interdependent way. It generalizes the truncated product method of Zaykin et al. to combine only those significance values which clear their respective experiment-specific false discovery restrictive thresholds, thus allowing us to control the false discovery rate (FDR) for the final combined result. Further, we introduce several concepts that together offer FDR control, high power, quality control and speed-up in meta-analysis as done by our algorithm. Computational and statistical methods of research synthesis like the one described here will be increasingly important as additional genome-wide datasets accumulate in databases. We apply our method to combine three well-known ChIP-chip transcription factor binding datasets for budding yeast to identify significant intergenic regulatory sequences for nine cell cycle regulating transcription factors, both with high power and controlled FDR.

Algorithms↗
PubMed2001

A robust multipoint linkage statistic (tlod) for mapping complex trait loci.

Classical parametric two-point linkage analysis is a powerful analysis tool, however there are clear disadvantages too, including the sensitivity to allele frequency mis-specification. Conversely, multipoint linkage analysis is not sensitive to allele frequency mis-specification, but it is sensitive to genetic model mis-specification. Göring and Terwilliger [Am J Hum Genet 66:1095-106, 2000] proposed a new robust multipoint statistic that increased the robustness of multipoint analyses. In this paper we have referred to this new statistic as the tlod. We applied this new statistic to the Genetic Analysis Workshop (GAW) 12 data using affected status (AFF) as the phenotype of interest. The heterogeneity tlod and two-point hlod scores correlated highly across the genome (p < 0.0001), as expected, but the het-tlod had a lower number false positives. In addition, the tlod analysis handled missing data better, as would be expected for a multipoint method. When one-third of the genotype data was removed (dead people) the tlod analysis was less affected than the two-point analysis. When tlod scores were compared with multipoint lod scores in true gene locations, the robustness of the tlod to model mis-specification was clearly evident. When the "best" replicate from the general population was analyzed, a borderline genome-wide significant two-point hlod result (3.6) was found 4 cM from MG6 and MG7 on chromosome 6. The heterogeneity tlod score was lower than the two-point hlod score (1.8), but greater than the heterogeneity multipoint lod score (0.4). However, when replicate 1 of the isolated population was analyzed none of the true gene locations were identified with either statistic.

Chromosome Mapping↗
PubMed2000

Selection of an adaptive test statistic for use with multiple comparison analyses of neuroimaging data.

Statistical analysis of neuroimages is commonly approached with intergroup comparisons made by repeated application of univariate or multivariate tests performed on the set of the regions of interest sampled in the acquired images. The use of such large numbers of tests requires application of techniques for correction for multiple comparisons. Standard multiple comparison adjustments (such as the Bonferroni) may be overly conservative when data are correlated and/or not normally distributed. Resampling-based step-down procedures that successfully account for unknown correlation structures in the data have recently been introduced. We combined resampling step-down procedures with the Minimum Variance Adaptive method, which allows selection of an optimal test statistic from a predefined class of statistics for the data under analysis. As shown in simulation studies and analysis of autoradiographic data, the combined technique exhibits a significant increase in statistical power, even for small sample sizes (n = 8, 9, 10).

Algorithms↗
PubMed1994

Updates of bioequivalence programs (including statistical power approximated by Student's t)

Updates are given of the earlier published MS-DOS programs BIOEQV37, BIOPAR37 and BIOEQNEW for the statistical analysis of bioequivalence studies. The updates are named BIOEQV52, BIOPAR40 and BIOEQNEW, respectively. Modifications and improvements in the updated programs are discussed. Calculations of statistical power for two-period crossover studies are now based on Schuirmann's two one-sided tests procedure. It is demonstrated that good approximations of power can be obtained by Student's t-statistic. This not only applies to the equations published elsewhere for the 'additive model', but also to hitherto unpublished equations for the 'multiplicative model'.

Software↗
PubMed1993

Leadership instability in hospitals: the influence of Board-CEO relations and organizational growth and decline.

This study tested whether leadership instability--a systemic pattern of frequent succession in the top management position of an organization--was associated with sociopolitical structures that define the relationship between the board and chief executive officer (CEO), controlling for temporal patterns of the organizational life-cycle stage. In organizations that are not profit maximizing and subject to considerable uncertainty, such governance properties were hypothesized to affect leadership instability independent of organizational growth or decline. Results of regression analyses demonstrate strong main effects of board-CEO relations, net of the impact of organizational life cycle, on leadership instability.

Chi-Square Distribution↗
PubMed2001

Statistical analysis of nuclear genome size of plants with flow cytometer data.

BACKGROUND: There is a mismatch between the sophistication of the cytometer and the resulting data, made possible by the computing power of today, and the traditional statistical methods based on the computing power of the early 1930s. The purpose here is to apply modern statistical techniques that similarly take advantage of this computer power. METHODS: Likelihood functions and their graphs are introduced as direct measures of plausibility of the parameters of interest. These methods are valid for samples of any size. They are exemplified on an experimental plant flow cytometer data set with n = 2 replications. RESULTS: The likelihood functions revealed important features of the data that would have been missed by the traditional methods, and in fact would invalidate them. CONCLUSIONS: The likelihood function produced highly informative graphs that allow quantitative comparisons of different aspects of 2C DNA nuclear contents among different groups or varieties of plants.

Analysis of Variance↗
PubMed2000

Stratified tests, stratified slopes, and random effects models for clinical trials with missing data.

Because missing observations may affect the size and power of statistical tests of equality, various analytical techniques explicitly or implicitly condition the analysis on the amount of information available per person. We illustrate the difference between stratifying a slope estimate and stratifying a test statistic based on slopes. We compare a nonparametric version of the latter approach with the parametric tests available from SAS Proc Mixed. Power and size of these two approaches are considered under different parametric settings, distributions, and missing data mechanisms.

Algorithms↗