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 667 records · Page 37Linked to original sources
PubMed1991

[Influence of the risk factors of atherosclerosis on the onset of early peri-operative complications in peripheral arterial surgery].

The influence of classical atherosclerosis risk factors (male sex, smoking, hypercholesterolemia, diabetes, overweight and hypertension on peri-hospital mortality and morbidity were studied in 83 patients (68 men and 15 women) aged 63 +/- 13 years and undergoing peripheral arterial surgery. None of these factors was statistically correlated with peri-operative complications with the exception of hypertension but the correlation was slight (0.5 less than p less than 0.10). In contrast, there was a powerful statistical (p less than 0.0003) link between age and morbidity and mortality associated with this type of surgery.

Aged↗
PubMed1996

[Statistics in the clinical research on drugs. A study of original articles emanating from Spanish centers].

BACKGROUND: The aim of this study was to evaluate statistical analysis reported in clinical investigation articles of Spanish drugs. METHODS: Original articles provided by Spanish centers and indexed in EMBASE in 1975, 1980, 1985 and 1990 were studied. The type of statistics used, their description and different aspects of results presentation are reported. RESULTS: Two hundred eighty-eight articles were studied of which 73.3% used inferential statistics with 57.3% presenting an adequate description of the same. The most frequently used statistical tests were bivariant techniques, mainly the Student's t test (33.7%) and chi square test (28.8%). The complexity of the statistical tests increased progressively in the years reviewed being greater in the articles published in foreign journals. The results were presented adequately and in a comprehensive form in 40% of the cases. Confidence intervals were used in 22.2% in the presentation of the results. In 49.1% of the articles statistical significance favored the group receiving the therapy studied. Of the 46.6% which did not present differences only 9.3% the statistical power was calculated. CONCLUSIONS: Although an improvement was observed in the years evaluated, the articles on clinical drug investigation carried out in Spanish centers still present insufficient information on statistical methodology, particularly in those published in Spanish journals.

Clinical Trials as Topic↗
PubMed2000

Gene mapping in the 20th and 21st centuries: statistical methods, data analysis, and experimental design.

In the 20th century geneticists began to unravel some of the simpler aspects of the etiology of inherited diseases in humans. The theory of linkage analysis was developed and applied long before the advent of molecular biology, but only the technological advances of the second half of the 20th century made large-scale gene mapping with a dense genome-spanning set of markers a reality. More recently, the primary topic of interest has shifted from simple Mendelian diseases, for which genotypes of some gene are the cause of disease, to more complex diseases, for which genotypes of some set of genes together with environmental factors merely alter the probability that an individual gets the disease, although individual factors are typically insufficient to cause the disease outright. To this end, a great deal of dogma has evolved about the best way to skin this cat, although to date success has been minimal with any approach. We postulate that the main reason for this is a lack of attention to experimental design. Once the data have been ascertained, the most powerful statistical methods will not be able to salvage an inappropriately designed study (Andersen 1990). Each phenotype and/or population mandates its own individually tailored study design to maximize the chances of successful gene mapping. We suggest that careful consideration of the available data from real genotype-phenotype correlation studies (as opposed to oversimplified theoretically tractable models), and the practical feasibility of different ascertainment schemes dictate how one should proceed. In this review we review the theory and practice of gene mapping at the close of the 20th century, showing that most methods of linkage and linkage disequilibrium analysis are similar in a fundamental sense, with the differences being related more to study design and ascertainment than to technical details of the underlying statistical analysis. To this end, we propose a new focus in the field of statistical genetics that more explicitly highlights the primacy of study design as the means to increase power for gene mapping.

Algorithms↗
PubMed1981

Multivariate data analysis for outcome studies.

The use of multivariate statistical techniques for analyzing the complex data often gathered in outcome studies is discussed. The multivariate analysis of variance (MANOVA) is suggested for multiple group studies common to outcome studies. This technique can be utilized for a large number of specific research designs whenever multiple outcome measures are collected. MANOVA offers two specific advantages over more familiar univariate approaches: it presents better control over Type 1 error rates while preserving statistical power, and it allows more thorough analysis of complex data.

Humans↗
PubMed2001

Statistical test to assess rank-order imaging studies.

RATIONALE AND OBJECTIVES: Rank-order experiments often provide a reasonable method of determining whether a large-scale receiver operating characteristic study can be justified. The authors' purpose was to formalize a proposed method for analyzing rank-order imaging experiments and provide methods that can be used in determining sample sizes for both cases and raters. MATERIALS AND METHODS: Simulations were conducted to determine the adequacy of the normal approximation of a statistic used to test the null hypothesis of random ordering. For a multireader experiment, formulas are presented and guidelines are provided to enable investigators to determine the number of required readers (raters) and cases for a specific study. RESULTS: When there are at least five ordered images per case, 10 cases are sufficient to test a random rank order. When there are only three or four images for a case, 20 cases are required. The authors constructed tables of statistical power for selected numbers of ordered images, numbers of cases, and degrees of trend, and they also provide an approximation for use in situations that are not tabled. CONCLUSION: The statistical methods for analyzing rank-order experiments and estimating sample sizes for study planning are relatively simple to implement. The derived formulas for sample size estimation, when applied to typical imaging experiments, indicate that modest numbers of cases and readers are required for rank-order studies.

Humans↗
PubMed2000

Bayesian statistics in medical research: an intuitive alternative to conventional data analysis.

Statistical analysis of both experimental and observational data is central to medical research. Unfortunately, the process of conventional statistical analysis is poorly understood by many medical scientists. This is due, in part, to the counter-intuitive nature of the basic tools of traditional (frequency-based) statistical inference. For example, the proper definition of a conventional 95% confidence interval is quite confusing. It is based upon the imaginary results of a series of hypothetical repetitions of the data generation process and subsequent analysis. Not surprisingly, this formal definition is often ignored and a 95% confidence interval is widely taken to represent a range of values that is associated with a 95% probability of containing the true value of the parameter being estimated. Working within the traditional framework of frequency-based statistics, this interpretation is fundamentally incorrect. It is perfectly valid, however, if one works within the framework of Bayesian statistics and assumes a 'prior distribution' that is uniform on the scale of the main outcome variable. This reflects a limited equivalence between conventional and Bayesian statistics that can be used to facilitate a simple Bayesian interpretation based on the results of a standard analysis. Such inferences provide direct and understandable answers to many important types of question in medical research. For example, they can be used to assist decision making based upon studies with unavoidably low statistical power, where non-significant results are all too often, and wrongly, interpreted as implying 'no effect'. They can also be used to overcome the confusion that can result when statistically significant effects are too small to be clinically relevant. This paper describes the theoretical basis of the Bayesian-based approach and illustrates its application with a practical example that investigates the prevalence of major cardiac defects in a cohort of children born using the assisted reproduction technique known as ICSI (intracytoplasmic sperm injection).

Bayes Theorem↗
PubMed2002

Giant-cell tumor of the appendicular skeleton.

The common objective of all surgical procedures in the treatment of giant-cell tumor of bone is to minimize the incidence of local recurrence. The purpose of this study was to determine what, if any, patient factors, tumor characteristics, or surgical practices correlate with local recurrence. Seventy-five patients treated for a giant-cell tumor of the appendicular skeleton were followed up for at least 2 years. The mean duration of followup was 62 months (range, 24-224 months). The highest proportion of patients had intralesional curettage, high-speed burring, and adjuvant treatment. Ten patients (13%) had a local recurrence. Bivariate analysis revealed that, with the numbers available, none of the patient variables, tumor variables, or surgical approaches correlated with local recurrence. Post hoc power analysis revealed the power of the study to be 33% to detect a clinically significant difference between treatment groups. The data presented here potentially could contribute to a metaanalysis, which would have the statistical power to determine which tumor-related factors and surgical techniques are most important in predicting recurrence in giant-cell tumor of bone.

Adolescent↗
PubMed2001

On sample size and inference for two-stage adaptive designs.

Proschan and Hunsberger (1995, Biometrics 51, 1315-1324) proposed a two-stage adaptive design that maintains the Type I error rate. For practical applications, a two-stage adaptive design is also required to achieve a desired statistical power while limiting the maximum overall sample size. In our proposal, a two-stage adaptive design is comprised of a main stage and an extension stage, where the main stage has sufficient power to reject the null under the anticipated effect size and the extension stage allows increasing the sample size in case the true effect size is smaller than anticipated. For statistical inference, methods for obtaining the overall adjusted p-value, point estimate and confidence intervals are developed. An exact two-stage test procedure is also outlined for robust inference.

Biometry↗
PubMed1995

Linkage detection under heterogeneity and the mixture problem.

Linkage analysis has contributed to the localization of many human disease genes. The presence of locus heterogeneity reduces statistical power and can prejudice the detection of linkage if the analysis assumes homogeneity. Nevertheless, mixed genetic models are not routinely used in gene searches. The null distribution of the test statistic is not uniquely defined. In this paper, a transformation is used to determine an approximate asymptotic distribution of the test statistic under a mixture model. The equivalent critical values of the test are computed and the performance of the test under various levels of heterogeneity and family size is investigated. For gene searches, we recommend the routine use of an admixture model with a critical lod score of 3.44.

Algorithms↗
PubMed2005

A call for reporting the relevant exposure term in air pollution case-crossover studies.

The exposure term in the case-crossover design consists in the difference between the ambient concentration on the event day and the concentration(s) on some control day(s). So far, all air pollution case-crossover studies presented the distribution of the daily ambient pollutant concentrations but do not publish the distributional properties of the relevant exposure term--that is, the concentration difference. This article shows that this difference can be very small for a large fraction of event days, therefore, seriously limiting the statistical power to refute the null hypothesis. Publishing the distribution of the relevant differences will improve the interpretation and discussion of findings from case-crossover studies, particularly in cases with statistically non-significant associations.

Air Pollutants↗
PubMed1995

Design issues in studies of radon and lung cancer: implications of the joint effect of smoking and radon.

Many case-control studies have been undertaken to assess whether and to what extent residential radon exposure is a risk factor for lung cancer. Nearly all these studies have been conducted in populations including smokers and nonsmokers. In this paper, we show that, depending on the nature of the joint effect of radon and tobacco on lung cancer risk, it may be very difficult to detect a main effect due to radon in mixed smoking and nonsmoking populations. If the joint effect is closer to additive than multiplicative, the most cost-effective way to achieve adequate statistical power may be to conduct a study among never-smokers. Because the underlying joint effect is unknown, and because many studies have been carried out among mixed smoker and nonsmoker populations, it would be desirable to conduct some studies with adequate power among never-smokers only.

Case-Control Studies↗
PubMed2006

Short communication: effectiveness of sample duplication to control error in ruminant digestion studies.

Eight ruminally cannulated lactating dairy cows from a study on the effect of dietary rumen-degraded protein on production and digestion of nutrients were used to assess using sample duplication to control day-to-day variation within animals and errors associated with sampling and laboratory analyses. Two consecutive pooled omasal samples, each representing a feeding cycle, were obtained from each cow in each period. The effectiveness of sample duplication in error control was tested by comparing the variance of the difference in treatment means when taking 2 samples from each cow in each period to the variance when taking only one sample. Compared with no duplication, sample duplication improved precision by reducing variance by 50, 40, 31, 23, 23, and 9% for, respectively, rumen-undegraded protein flows, ruminal neutral detergent fiber digestibility, microbial nonammonia N flow, microbial efficiency, organic matter flow, and organic matter truly digested in the rumen. For these same variables, reductions in the standard errors of the difference between treatment means due to sample duplication represented 100, 87, 73, 59, 58, and 27% of the predicted reductions resulting from doubling the number of experimental units without sample duplication. Sample duplication can substantially reduce experimental error originating from day-to-day variation within cows, sample collection, and laboratory analyses, thus improving statistical power in ruminant digestion studies.

Animals↗
PubMed2005

Methodology for estimation of days dry effects.

The primary objective of this research was to determine if, with appropriate methodology, unbiased estimates of days dry (DD) effects on subsequent lactation milk yield can be obtained from field data, particularly when DD is correlated with cow effects. Another objective was to ascertain relevant sampling properties of designed trials for estimation of DD effects. Simulated records were used to assess methodology. Along with a model with no adjustments for cow effects, alternative models including 1) previous lactation milk yield, 2) a prior adjustment for cow effects estimated from an animal model, and 3) a combination of 1 and 2, were tested. Estimates from the unadjusted model were biased downward; however, the 3 alternative analyses provided estimates of DD effects that were essentially unbiased, with a prior adjustment for cow effects and previous milk yield in the model providing the best results in terms of elimination of bias. Therefore, DD effects can be estimated from field data without bias from cow effects.A designed trial with 2 groups and 10 or fewer cows/ group is noninformative and has an unacceptably high probability of leading to invalid conclusions. A minimum of 30 cows/group is considerably better and should be used whenever possible. Even with 30 cows/group, however, the power is low unless the difference between DD groups for yield is at least 1130 kg. Prior correction of 305-d, mature equivalent records for cow effects, using predicted producing abilities, could be done in designed trials to improve the statistical power of tests and accuracy of estimates.

Animals↗
PubMed2006

Case-control studies in the genomic era: a clinician's guide.

The goal of case-control association studies is to find genetic variants in the human genome that influence common traits. The Human Genome and HapMap projects have added fresh impetus to this goal by cataloguing the raw genetic data behind human DNA variation. Studies that associate these genetic variants with phenotype improve both molecular diagnostics and drug discovery and offer clinicians important opportunities to improve care of patients. In this review I focus on case-control studies, which are the most widely used design and expected to be the most powerful. I also address the problem of case-control non-replication, which is widespread despite enormous effort and use of resources. Important causes of non-replication include inadequate statistical power to detect small and moderate effects, phenotype heterogeneity, population stratification, publication bias, and multiple comparison testing.

Case-Control Studies↗
PubMed2006

Comparative phylogeographic summary statistics for testing simultaneous vicariance.

Testing for simultaneous vicariance across comparative phylogeographic data sets is a notoriously difficult problem hindered by mutational variance, the coalescent variance, and variability across pairs of sister taxa in parameters that affect genetic divergence. We simulate vicariance to characterize the behaviour of several commonly used summary statistics across a range of divergence times, and to characterize this behaviour in comparative phylogeographic datasets having multiple taxon-pairs. We found Tajima's D to be relatively uncorrelated with other summary statistics across divergence times, and using simple hypothesis testing of simultaneous vicariance given variable population sizes, we counter-intuitively found that the variance across taxon pairs in Nei and Li's net nucleotide divergence (pi(net)), a common measure of population divergence, is often inferior to using the variance in Tajima's D across taxon pairs as a test statistic to distinguish ancient simultaneous vicariance from variable vicariance histories. The opposite and more intuitive pattern is found for testing more recent simultaneous vicariance, and overall we found that depending on the timing of vicariance, one of these two test statistics can achieve high statistical power for rejecting simultaneous vicariance, given a reasonable number of intron loci (> 5 loci, 400 bp) and a range of conditions. These results suggest that components of these two composite summary statistics should be used in future simulation-based methods which can simultaneously use a pool of summary statistics to test comparative the phylogeographic hypotheses we consider here.

Classification↗
PubMed2000

Power to detect homework effects in psychotherapy outcome research.

Studies of homework effects in psychotherapy outcome have produced inconsistent results. Although these findings may reflect the comparability of psychotherapy with and without homework assignments, many of these studies may not have been sensitive enough to detect the effects sizes (ESs) likely to be found when examining homework effects. The present study evaluated the power of homework research and showed that, on average, current power levels are relatively weak in controlled studies ranging from 0.58 for large ESs to 0.09 for small ESs. Thus, inconsistent findings between studies may very well be due to low statistical power.

Bias↗
PubMed2004

Statistical problems encountered in trapping studies of scolytids and associated insects.

Traps baited with semiochemicals are often used to investigate the chemical ecology of scolytids and associated insects. One statistical problem frequently encountered in these studies are treatments that catch no insects and, thus, have zero mean and variance, such as blank or control traps. A second problem is the use of multiple comparison procedures that do not control the experimentwise error rate. We conducted a literature survey to determine the frequency of these two statistical problems in Journal of Chemical Ecology for 1990-2002. Simulations were then used to examine the effects of these problems on the validity of multiple comparison procedures. Our results indicate that both statistical problems are common in the literature, and when combined can significantly inflate both the experimentwise and per comparison error rate for multiple comparison procedures. A possible solution to this problem is presented that involves confidence intervals for the treatment means. Options to increase the statistical power of trapping studies are also discussed.

Animals↗
PubMed2003

Nonreplication in genetic association studies of obesity and diabetes research.

The objective of this article is to provide an overview of the existing literature concerning the identification of genetic markers associated with obesity and diabetes. Specifically, this article will review recent association studies of diabetes and obesity with an emphasis on the need for the replication of findings. Unfortunately, a substantial number of the published associations between genetic markers and phenotypes, including diabetes and obesity, have not been replicated. Literature that addresses the potential reasons for the nonreplication of association studies (population stratification, publication bias, effect heterogeneity, Type I errors and lack of statistical power) is summarized. Recommendations to improve future association studies are presented.

Diabetes Mellitus↗