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 1,297 records · Page 72Linked to original sources
PubMed1992

Beyond Campbell and Fiske: assessment of convergent and discriminant validity.

While it is not an innovation in the more established social sciences, the multitrait multimethod (MTMM) technique (Campbell & Fiske, 1959) and additional, more powerful statistical methods of interpreting a MTMM matrix of correlations, have not been used extensively in nursing to assess construct validity of new instruments. This article reviews the Campbell and Fiske method for the assessment of convergent and discriminant validity, reviews the application of the method in the nursing literature, and describes the limitations of the original method. In addition, the procedures, strengths, and limitations of the analysis of variance and confirmatory factor analytic approaches to the quantification of Campbell and Fiske's criteria are described.

Data Interpretation, Statistical↗
PubMedDate not supplied

The risk of liver disorders in women prescribed cyproterone acetate in combination with ethinyloestradiol (Dianette): a nested case-control study using the GPRD.

PURPOSE: To explore the risk of liver disorders associated with cyproterone acetate combined with ethinyloestradiol (CPA/EE). CPA/EE is licensed in the UK for the treatment of women with acne and hirsutism and is a treatment option for polycystic ovary syndrome (PCOS). It acts as a contraceptive also. METHODS: Using the General Practice Research Database, we conducted a cohort analysis and case-control study in women aged 15-39 with acne, hirsutism or PCOS to estimate the risk of liver disorders associated with CPA/EE. RESULTS: Compared with cases exposed to conventional combined oral contraceptives (COCs), the age-adjusted incidence rate ratio for liver disorders in women using CPA/EE was 1.7 (95% CI: 0.9, 3.4) and compared with no use it was 1.5 (95% CI: 0.8, 2.8). In the case-control study, the adjusted odds ratio (OR) for liver disorders in women exposed to CPA/EE was 1.6 (95% CI: 0.7, 3.5) and 0.8 (95% CI: 0.5, 1.3) for exposure to conventional COCs, compared with no use. The risk of liver disorders in women prescribed CPA/EE was not significantly greater than that in women prescribed conventional COCs (OR: 2.1 [95% CI: 0.9, 4.8]). CONCLUSION: Our results do not indicate an increased risk for liver disorders associated with CPA/EE use in women with acne, hirsutism or PCOS after adjusting for potential confounding. This may be due to lack of statistical power.

Adolescent↗
PubMed2002

Hormones and prostate cancer: current perspectives and future directions.

Prostate cancer is the most commonly diagnosed non-skin cancer in men in most western countries. Despite the high morbidity and mortality from prostate cancer, its etiology remains obscure. Although compelling laboratory data suggest a role for androgens in prostate carcinogenesis, most epidemiologic data on humans are inconclusive. To provide insights and directions for future epidemiologic research on hormones and prostate cancer, this review focuses on current perspectives of serum-based studies and polymorphisms in relevant hormone-related genes. We highlight the importance of methodologic studies and investigations of hormone levels in the prostatic tissue to help clarify the often-contradictory data on serologic studies. We recommend careful analysis and cautious interpretation of studies of genetic markers, including repeats and single nucleotide polymorphisms (SNPs), as false positive and negative results may arise in many current and future studies with limited statistical power and non-representative samples from the population. The review also highlights the reasons to perform functional analyses of SNPs, a critical and often under-appreciated component of molecular epidemiologic investigations. The time is ripe for large-scale multidisciplinary investigations that incorporate molecular genetics, biochemistry, histopathology, and endocrinology into traditional epidemiologic studies. Such collaboration will lead to a deeper understanding of the etiologic pathways of prostate cancer, ultimately yielding better preventive, diagnostic, and therapeutic strategies.

Androgens↗
PubMed2004

Comparing completely and stratified randomized designs in cluster randomized trials when the stratifying factor is cluster size: a simulation study.

Stratified randomized designs are popular in cluster randomized trials (CRTs) because they increase the chance of the intervention groups being well balanced in terms of identified prognostic factors at baseline and may increase statistical power. The objective of this paper is to assess the gains in power obtained by stratifying randomization by cluster size, when cluster size is associated with an important cluster level factor which is otherwise unaccounted for in data analysis. A simulation study was carried out using a CRT where UK general practices were the randomized units as a template. The results show that when cluster size is strongly associated with a cluster level factor which is predictive of outcome, the stratified randomized design has superior power results to the completely randomized design and that the superiority is related to the number of clusters.

Cluster Analysis↗
PubMed2004

A simple significance test for quantile regression.

Where OLS regression seeks to model the mean of a random variable as a function of observed variables, quantile regression seeks to model the quantiles of a random variable as functions of observed variables. Tests for the dependence of the quantiles of a random variable upon observed variables have only been developed through the use of computer resampling or based on asymptotic approximations resting on distributional assumptions. We propose an exceedingly simple but heretofore undocumented likelihood ratio test within a logistic regression framework to test the dependence of a quantile of a random variable upon observed variables. Simulated data sets are used to illustrate the rationale, ease, and utility of the hypothesis test. Simulations have been performed over a variety of situations to estimate the type I error rates and statistical power of the procedure. Results from this procedure are compared to (1) previously proposed asymptotic tests for quantile regression and (2) bootstrap techniques commonly used for quantile regression inference. Results show that this less computationally intense method has appropriate type I error control, which is not true for all competing procedures, comparable power to both previously proposed asymptotic tests and bootstrap techniques, and greater computational ease. We illustrate the approach using data from 779 adolescent boys age 12-18 from the Third National Health and Nutrition Examination Survey (NHANES III) to test hypotheses regarding age, ethnicity, and their interaction upon quantiles of waist circumference.

Adolescent↗
PubMed2005

FDR-controlling testing procedures and sample size determination for microarrays.

Microarrays are used increasingly to identify genes that are truly differentially expressed in tissues under different conditions. Planning such studies requires establishing a sample size that will ensure adequate statistical power. For microarray analyses, false discovery rate (FDR) is considered to be an appropriate error measure. Several FDR-controlling procedures have been developed. How these procedures perform for such analyses has not been evaluated thoroughly under realistic assumptions. In order to develop a method of determining sample sizes for these procedures, it needs to be established whether these procedures really control the FDR below the pre-specified level so that the determined sample size indeed provides adequate power. To answer this question, we first conducted simulation studies. Our simulation results showed that these procedures do control the FDR at most situations but under-control the FDR when the proportion of positive genes is small, the most likely scenarios. Thus, these existing procedures can overestimate the power and underestimate the sample size. Accordingly, we developed a simulation-based method to provide more accurate estimates for power and sample size.

Algorithms↗
PubMed2007

Bias reduction in effectiveness analyses of longitudinal ordinal doses with a mixed-effects propensity adjustment.

A mixed-effects propensity adjustment is described that can reduce bias in longitudinal studies involving non-equivalent comparison groups. There are two stages in this data analytic strategy. First, a model of propensity for treatment intensity examines variables that distinguish among subjects who receive various ordered doses of treatment across time using mixed-effects ordinal logistic regression. Second, the effectiveness model examines multiple times until recurrence to compare the ordered doses using a mixed-effects grouped-time survival model. Effectiveness analyses are initially stratified by propensity quintile. Then the quintile-specific results are pooled, assuming that there is not a propensity x treatment interaction. A Monte Carlo simulation study compares bias reduction in fully specified propensity model relative to misspecified models. In addition, type I error rate and statistical power are examined. The approach is illustrated by applying it to a longitudinal, observational study of maintenance treatment of major depression.

Adult↗
PubMed2007

Challenge of multiple co-primary endpoints: a new approach.

There are many disorders where regulatory agencies have required a new treatment to demonstrate efficacy on multiple co-primary endpoints, all significant at the one-sided 2.5 per cent level, before accepting the treatment's effect for the disorder. This requirement, rooted in the intersection-union (IU) test, has led many researchers to increase the study sample size to make up for the reduction in the statistical power at the study level. Unfortunately, the increase in sample size could be substantial when the endpoints are minimally correlated and the treatment effects on the multiple endpoints are comparable. In this paper, we demonstrate that the frequentist concept of controlling the maximum false positive rate, even when applied to a restricted null space, has only limited success in keeping the sample size increase at a reasonable level. We therefore propose an approach that is based on the notion of controlling an average type I error rate. By employing an upper bound for the average type I error rate, the new approach provides an adjustment to the significance level that depends only on the correlation among the endpoints. For the most common case of two or three co-primary endpoints, the adjusted significance level is at most 5 per cent (one-sided) when the endpoints are moderately correlated. We show how sample size could be calculated under the proposed approach and contrast the needed sample size with that required under the IU test. We provide additional comments and discuss why the new approach is consistent with the principle requiring evidence of significance in the drug development and approval process.

Algorithms↗
PubMed1992

Impact of measurement error and temporal variability on the estimation of event probabilities for risk factor intervention trials.

The impact of measurement error and temporal variability of risk factors on estimates of disease probabilities based on the logistic function is discussed. Monte Carlo results and empirical findings from the Multiple Risk Factor Intervention Trial indicate that the degree of attenuation of logistic parameter estimates is well approximated by the reliability coefficient when the errors are assumed to be normal random variates and event probabilities are small. In the design of intervention studies, measurement error and temporal variability of risk factors do not usually influence estimates of the probability of developing the disease in the control group, but can result in mis-estimation of the probability of developing the disease in the experimental group, substantially reducing the statistical power of the clinical trial.

Bias↗
PubMed2001

A three-stage clinical trial design for rare disorders.

Many clinical trials of uncommon diseases are underpowered because of the difficulty of recruiting adequate numbers of subjects. We propose a clinical trial design with improved statistical power compared to the traditional randomized trial for use in clinical trials of rare diseases. The three-stage clinical trial design consists of an initial randomized placebo-controlled stage, a randomized withdrawal stage for subjects who responded, and a third randomized stage for placebo non-responders who subsequently respond to treatment. Test level and power were assessed by computer-intensive exact calculations. The three-stage clinical trial design was found to be consistently superior to the traditional randomized trial design in all cases examined, with sample sizes typically reduced by 20 per cent to 30 per cent while maintaining comparable power. When a treatment clearly superior to placebo was considered, our design reached a power of 75 per cent with a sample of 21 patients compared with the 52 needed to attain this power when only a randomized controlled trial was used. In situations where patient numbers are limited, a three-stage clinical trial design may be a more powerful design than the traditional randomized trial for detecting clinical benefits.

Clinical Trials as Topic↗
PubMed1986

Exogenous sex hormone exposure and the risk for VACTERL association.

In several studies investigators have suggested that maternal use of exogenous sex hormones during early pregnancy may be associated with various congenital malformations. A group of malformations, the VACTERL (vertebral, anal, cardiac, trachea, esophageal, renal, limb-acronym) association, has been statistically associated with maternal exposure to exogenous sex hormones during the first trimester of pregnancy. The VACTERL association is a nonrandom group of major malformations that occur together more often than would be expected on the basis of chance. To assess this association, we conducted a case-control study of first-trimester exposure to sex hormones among mothers of 34 infants with the VACTERL association and of 1,024 comparison infants with one or more of ten major malformations or Down syndrome. The study subjects were malformed infants born between July 1970 and June 1979 and registered in a population-based birth defects registry. Information concerning the use of exogenous sex hormones during pregnancy was obtained by systematically interviewing the mothers of the malformed infants. Most of the mothers were interviewed within 6 months of their children's births. Each mother was interviewed within a year of her child's birth. We found an odds ratio of 0.98 (90% confidence limits 0.40, 2.38) for the relationship between VACTERL association and use of any sex hormone in the first trimester of pregnancy. Our study had adequate statistical power to detect a true relative risk of 2.8 or greater.

Abnormalities, Multiple↗
PubMed1992

The use of dysmorphology in birth defects epidemiology.

Most human teratogens have been identified by the clinical recognition of characteristic patterns of congenital anomalies among children whose mothers were exposed to a particular agent during pregnancy. Although this dysmorphologic method has been valuable, it is criticized because it is not easily amenable to statistical evaluation. Conventional birth defects epidemiological studies are designed to permit rigorous statistical assessment, but such investigations usually classify congenital anomalies without adequate consideration of their known etiological heterogeneity. It is possible to combine the best aspects of these two approaches to identifying human teratogens in a "dysmorphologic case/control study." Instead of including all available cases with a given defect, only individuals having the anomaly in the context of a multiple congenital anomaly pattern without a recognizable cause would be selected for inclusion among the case group. The frequency of exposure to the putative teratogen would be determined among these selected cases and among appropriately chosen controls; conventional statistical analysis would then be performed. Although this design reduces the size of the case group compared to a conventional case/control study, the statistical power is unchanged or increased. In addition, biological plausibility is increased by concentrating upon a group of cases that is more likely to have a teratogenic cause.

Abnormalities, Drug-Induced↗
PubMed1995

Detection of genotype-environment interaction in case-control studies of birth defects: how big a sample size?

Detecting interactions between risk factors in case-control studies of birth defects and other conditions usually requires increasing the sample size beyond that needed to detect marginal effects. A special case of such interaction is genotype-environment interaction in which the effects of an exposure on disease risk are modified by genetic susceptibility. When case-control studies are designed to detect marginal effects of an exposure (i.e., in the whole population), under many plausible interaction schemes, no additional case and control subjects are needed to detect genotype-environment interaction. On the contrary, inclusion of genotypic information generally can improve the statistical power of the original study. Using the example of oral clefts, maternal cigarette smoking, and genetic variation at the transforming growth factor alpha gene, we illustrate sample size and power issues in designing case-control studies when prior information is available on both the marginal effects of the exposure and the genetic factor.

Case-Control Studies↗
PubMed1995

The effect of body weight on tumor incidence and carcinogenicity testing in B6C3F1 mice and F344 rats.

Associations between animal body weights and tumor incidence were examined using individual control animal data from 55 mouse and 53 rat studies conducted by the National Toxicology Program. Several statistically significant associations were found, the strongest of which were positive relationships between body weight and risk of liver tumors in both sexes of mice, pituitary gland tumors in both sexes of rats, and mammary gland tumors in female rats. The most compelling evidence that these relationships were causal in nature was the replication of the correlations across individual experiments. In addition, significant correlations between tumor occurrence and body weights occurred in animals as young as 9 weeks of age. Logistic regression models relating 12-month body weight to tumor risk were developed for the strongest relationships, and utilized in the reanalysis of tumor data from two National Toxicology Program studies with treatment-related decreases in body weight. A simulation study based on the logistic regression models indicated that statistical power to detect a treatment-related increase in tumor incidence can be seriously diminished when mean body weight in treated animals is depressed by as little as 10%. Conversely, when mean body weight in control animals is 10% lower than that of treated animals, false positive rates can increase to 20-30%. The severity of the effects of such commonly observed treatment-related disparities in body weight suggests that alternative data analysis methods or experimental designs should be considered to address this potential problem in carcinogenicity testing.

Animals↗
PubMed2002

Gender differences in the validity of testosterone measured in saliva by immunoassay.

We rigorously evaluated gender differences in the measurement validity of salivary testosterone. Matched serum, saliva, and finger stick blood spot specimens were collected from 40 (20 males) young adults (aged 18-27 years). Saliva was assayed for testosterone by two independent (isotopic and non-isotopic) immunoassay methods. Serum was assayed by commercially available immunoassay kits for free and total testosterone. An immunoassay was developed for the measurement of testosterone in dried blood spots and is presented in detail so as to be reproducible from this report. Regardless of assay method, salivary testosterone levels are modestly correlated with serum levels for males but not necessarily for females. Blood spot assay results were highly correlated with serum total and free testosterone for both males and females. Substitution of saliva assay results for serum values substantially underestimates known testosterone-behavior associations, and this effect is much more pronounced for females than for males. The findings have important implications for the use and potential misuse of noninvasive measures of testosterone, and with respect to statistical power, the probability of observing significant testosterone-behavior relationships.

Adult↗
PubMed1998

Line Mixing and Broadening in the Raman Q Branch of HD at 304.6 K.

The Q-branch lines of pure HD were measured at densities ranging from 1 to 7 Amagat at 304.6 K. Each profile was fitted to the well-known Rosenkranz expression to extract the size of the asymmetry due to line mixing as well as to the linewidth. Line mixing and broadening coefficients were obtained by fitting the asymmetries and widths to a straight line as a function of gas density. Apart from a single existing measurement for the Q(0) line, our mixing coefficients are the first direct measurements of the asymmetry due to line mixing in HD. Our broadening coefficients are consistent with the best earlier measurements but are an order of magnitude more precise. Agreement is found with some existing semiclassical calculations of broadening. We have fitted our HD broadening coefficients to a variety of empirical energy gap laws. Our conclusions are that none of the exponential gap law (EGL), the modified exponential gap (MEG) law, and the statistical power gap (SPG) law successfully models our broadening coefficients. We present a modified version of the EGL and the MEG laws, which are successful in reproducing the experimental results. Using the fitted parameters of the new gap law, we have calculated the relaxation matrix of HD at room temperature. With this relaxation matrix, we have simulated the Q-branch spectrum at a number of densities between 49.1 and 490 Amagat and compared the results with previous high-density measurements. At all densities and frequencies, the simulated spectral intensity was found to agree with the measured strength within about 5% of the peak of the spectrum. In addition, the comparison provides evidence of a nonlinear vibrational dephasing shift in HD. Copyright 1998 Academic Press.

Journal Article↗
PubMed1998

A standardized model of partial thickness scald burns in mice.

BACKGROUND: Large mammal partial thickness wound models were developed primarily for their anatomical similarity to human wounds, yet lack the economy, ease of handling, and statistical power afforded with rodent models. Previous small mammal models of partial thickness burn injury have failed to demonstrate complete reepithelialization in less than 3 weeks. We present a murine partial thickness scald model with a reepithelialization rate comparable to that of porcine wound models. METHODS: Thirty-eight adult male mice were secured in a burn template allowing exposure of a 2 x 3-cm area of the shaved dorsum to 60 degreesC water for 45 s, followed by 4 degreesC water for 45 s. Four wounds were harvested daily on Postburn Days 1-7, 10, and 14 for histologic evaluation. RESULTS: Histologic evidence of partial thickness dermal injury with sparing of dermal appendage epithelial cells was seen in all wounds. Of 134 wound sections evaluated 26 contained some areas of full thickness dermal injury, with only 8 of these 26 sections showing full thickness injury in 50% or more of the cross-sectional area of the wounds. Complete wound reepithelialization was seen between Postburn Days 10 and 14. The viable dermal thickness in all burn cross sections was at least 40-80 micrometers, and up to 1400 micrometers in reepithelialized wounds. CONCLUSIONS: This murine model of partial thickness scald injury provides a standardized thermal wound with consistent depth of injury, low mortality, and a reepithelialization rate between 10 and 14 days. A simple protocol allows easy production of 30-50 wounds daily with one technician.

Animals↗
PubMed2001

Effects of changes in experimental design on PET studies of isometric force.

Based on single-cell recordings in primates, the relationship between neuronal activity and force magnitude is thought to be monotonic, at least for a subset of pyramidal cells in the motor cortex. Functional neuroimaging studies have also suggested a monotonic relationship between cerebral activation and force magnitude. In order to more precisely define this relationship and to characterize the activation pattern(s) associated with the modulation of static force, we studied 40 normal subjects using [(15)O]water PET and a simple visuomotor task-application of static force on a micro force sensor with the thumb and index finger of the right hand. When our experimental design did not produce the expected result (evidence of a relationship between cerebral activation and force magnitude in ten subjects), we made serial changes in the experimental protocol, including the addition of control (baseline) trials, and increased the number of subjects in an effort to increase our sensitivity to variations in force magnitude. We compared univariate and multivariate data-analytic strategies, but we relied on our multivariate results to elucidate the interaction of attentional and motor networks. We found that increasing the number of subjects from 10 to 20 resulted in an increase in statistical power and a more stable (i.e., more replicable) but qualitatively similar result, and that the inclusion of control trials in a 10-subject group did not enhance our ability to discern significant brain-behavior relationships. Our results suggest that sample sizes greater than 20 may be required to detect parametric variation in some instances and that failure to detect such variation may result from unanticipated neurobehavioral effects.

Adult↗