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 901 records · Page 50Linked to original sources
PubMed1985

Assessing the power and quality of epidemiologic studies of asbestos-exposed populations.

This paper briefly discusses criteria for evaluating epidemiologic studies for risk assessment purposes, using asbestos as an example. Asbestos is one of the few carcinogens for which substantial data exist on exposures to humans. However, there are major difficulties in using these data for conducting risk assessments. In particular, exposure data are often incomplete, and risk assessments usually involve extrapolating from the higher exposures of the occupational environments to the lower levels typically encountered in the nonoccupational environment. The term "asbestos" refers to the fibrous form of several minerals, and levels of exposures to these fibers are not easily assessed. Criteria for evaluating epidemiologic studies used in an Ontario Royal Commission report on asbestos are discussed. The importance of considering the statistical power of studies to detect an excess risk is examined using as examples major cohort studies of asbestos-exposed workers, as summarized in a report by the U.S. National Research Council.

Asbestos↗
PubMed2002

The Human Genome Project and the future of diagnostics, treatment and prevention.

The Human Genome Project, the mapping of our 30,000-50,000 genes and the sequencing of all of our DNA, will have major impact on biomedical research and the whole of therapeutic and preventive health care. The tracing of genetic diseases to their molecular causes is rapidly expanding diagnostic and preventive options. The increased insights into molecular pathways, gained from high-throughput 'functional genomics', using DNA-chip and protein-chip approaches and specially designed animal model systems, will open great prospects for pharmacological and genetic therapies. Powerful bioinformatics and biostatistics will further improve our pattern recognition and accelerate progress. A rapidly expanding area of high expectations is that of 'pharmacogenomics': the design of more effective drugs with lower toxicity through tailoring of drug treatment to individual, genetically determined differences in drug metabolism. Not only will this decrease the cost of health care through reduction of adverse drug reactions, but a better stratification of populations will also provide more statistical power farther upstream in drug trials. However, the optimal benefits from the current explosion of 'data mining' will only be realized when the basic data are made and kept publicly accessible, while at the same time safeguarding the protection of intellectual property arising from downstream inventions. This is one of the goals of HUGO, the international Human Genome Organization, established 13 years ago to assist coordination of data acquisition and exchange and societal implementation of the genome project. Additional points of attention in this historic endeavour are the prevention of stigmatization and discrimination and the safeguarding of a worldwide balance in the contribution by--and benefits to--different populations, while respecting the diversity in cultures and traditions.

Ethics, Medical↗
PubMed2003

Efficiency robust tests for mapping quantitative trait loci using extremely discordant sib pairs.

In 1972, Haseman and Elston proposed a pioneering regression method for mapping quantitative trait loci using randomly selected sib pairs. Recently, the statistical power of their method was shown to be increased when extremely discordant sib pairs are ascertained. While the precise genetic model may not be known, prior information that constrains IBD probabilities is often available. We investigate properties of tests that are robust against model uncertainty and show that the power gain from further constraining IBD probabilities is marginal. The additional linkage information contained in the trait values can be incorporated by combining the Haseman-Elston regression method and a robust allele sharing test.

Chromosome Mapping↗
PubMed1997

Effect size and power for clinical trials that measure years of healthy life.

Some clinical trials perform repeated measurements on patients over time, plot those measures against time, and summarize the results in terms of the area under the curve. If the measured variable is health status, the summary outcome is sometimes referred to as years of healthy life (YHL), or quality-adjusted life years (QALY). This paper investigates some theoretical and practical aspects of randomized trials designed to assess measures such as YHL. We first derived algebraic expressions for the effect size of YHL measures under several theoretical models of the treatment's effect on health. We used these expressions to examine how the length of the study, the number of measurements per person and the correlations among health measurements over time influence the effect size. We also explored the relative statistical power of analyses based on YHL versus analyses based on change-scores using the same data. We present an example. Findings suggest that: (i) the number of measurements per person need not be large; (ii) high correlation among measures over time tends to lower the power of a study using YHL; (iii) a longer study will not always provide more power than a shorter study, and (iv) analyses based on YHL may have less power than change-score analyses. Some of these findings depend on the model of change in health status caused by the treatment. Such models require further study.

Effect Modifier, Epidemiologic↗
PubMed2001

Can placebo controls reduce the number of nonresponders in clinical trials? A power-analytic perspective.

BACKGROUND: There is ongoing debate regarding the ethics of placebo-controlled clinical trials when a moderately effective standard treatment exists. One aspect of the debate--the number of nonresponders--tends to be overlooked. A larger between-group effect size is expected in placebo-controlled trials than in trials with an active comparator. For that reason, substantially fewer subjects need to be enrolled in placebo-controlled trials; consequently, there tend to be far fewer nonresponders in placebo-controlled trials. OBJECTIVE: This analysis was undertaken to illustrate that the use of placebo as a control can reduce the number of subjects who are unnecessarily exposed to delayed treatment. METHODS: Statistical power analyses were used to estimate the sample size required to detect various population treatment differences and the resulting number of nonresponders for 2-tailed chi-square tests. RESULTS: Empiric evidence of the phenomenon is provided for a wide range of rates of response to placebo, investigational, and comparator treatments. For example, 24 subjects (ie, 12 per group) are needed to detect differences between placebo (10% response rate) and an investigational drug (70% response); 15 of these would not respond. In contrast, if the investigational drug (70% response) is initially compared with a standard therapy (60% response), 752 subjects would be required, 263 of whom would not respond. CONCLUSIONS: This paper shows empirically that placebo controls can reduce the number of nonresponders in a randomized controlled trial. The number of subjects who are exposed to unproven, albeit promising, investigational drugs should be kept to a minimum until placebo-controlled trials support their use.

Clinical Trials as Topic↗
PubMed2007

Multivariate scan statistics for disease surveillance.

In disease surveillance, there are often many different data sets or data groupings for which we wish to do surveillance. If each data set is analysed separately rather than combined, the statistical power to detect an outbreak that is present in all data sets may suffer due to low numbers in each. On the other hand, if the data sets are added by taking the sum of the counts, then a signal that is primarily present in one data set may be hidden due to random noise in the other data sets. In this paper, we present an extension of the spatial and space-time scan statistic that simultaneously incorporates multiple data sets into a single likelihood function, so that a signal is generated whether it occurs in only one or in multiple data sets. This is done by defining the combined log likelihood as the sum of the individual log likelihoods for those data sets for which the observed case count is more than the expected. We also present another extension, where the concept of combining likelihoods from different data sets is used to adjust for covariates. Using data from the National Bioterrorism Syndromic Surveillance Demonstration Project, we illustrate the new method using physician telephone calls, regular physician visits and urgent care visits by Harvard Pilgrim Health Care members cared for by Harvard Vanguard Medical Associates, a large multi-specialty group practice in Massachusetts. For upper and lower gastrointestinal (GI) illness, there were on average 20 telephone calls, nine urgent care visits and 22 regular physician visits per day. The strongest signal was generated by a single data set and due to a familial outbreak of pinworm disease. The second and third strongest signals were generated by the combined strength of two of the three data sets.

Boston↗
PubMed2006

Informativeness of genetic markers for pairwise relationship and relatedness inference.

Measuring the information content of markers in relationship/relatedness inferences is important in selecting highly informative markers to attain a given statistical power with the minimal genotyping effort. Using information-theoretic principles, I introduce the informativeness for relationship (I(R)) and the informativeness for relatedness (I(r)) to measure the amount of information provided by markers in inferring pairwise relationships (R) and relatedness (r), respectively. I also propose a fast and accurate algorithm to calculate the power (PW(R)) of a set of markers in differentiating two candidate relationships, and the reciprocal of the mean squared deviations of relatedness estimates (RMSD) to measure the amount of information of markers actually used by an estimator in estimating relatedness. All of the four measurements (I(R), I(r), PW(R), RMSD) apply to dominant and codominant markers, haploid and diploid individuals, and take into account of mutations and typing errors in data. The statistical properties of the four measurements and their relationships are investigated analytically and are examined by applying these methods to simulated and empirical data.

Algorithms↗
PubMed2001

A whole genome scan for quantitative trait loci affecting milk protein percentage in Israeli-Holstein cattle, by means of selective milk DNA pooling in a daughter design, using an adjusted false discovery rate criterion.

Selective DNA pooling was employed in a daughter design to screen all bovine autosomes for quantitative trait loci (QTL) affecting estimated breeding value for milk protein percentage (EBVP%). Milk pools prepared from high and low daughters of each of seven sires were genotyped for 138 dinucleotide microsatellites. Shadow-corrected estimates of sire allele frequencies were compared between high and low pools. An adjusted false discovery rate (FDR) method was employed to calculate experimentwise significance levels and empirical power. Significant associations with milk protein percentage were found for 61 of the markers (adjusted FDR = 0.10; estimated power, 0.68). The significant markers appear to be linked to 19--28 QTL. Mean allele substitution effects of the putative QTL averaged 0.016 (0.009--0.028) in units of the within-sire family standard deviation of EBVP% and summed to 0.460 EBVP%. Overall QTL heterozygosity was 0.40. The identified QTL appear to account for all of the variation in EBVP% in the population. Through use of selective DNA pooling, 4400 pool data points provided the statistical power of 600,000 individual data points.

Alleles↗
PubMed1999

[Phase III trials in oncology].

The randomised trial is the best context for the assessment of any new molecule in oncology. The authors emphasize the methodological aspects of randomised trials and suggest strategies to be adopted in phase III trials, based on the newly orthodox concept of evidence-based medicine, while underscoring the need for context specific adaptations to increase relevance and specificity of sought endpoints, including other means of controlling adequately the clinical experiment without randomization (patient as his own control models). The authors also suggest certain strategic changes. For example, restrictive but relevant eligibility criteria may help decrease the number of patients needed without compromising statistical power as well as other strategies aimed at proving the efficacy of a new molecule without carrying out large trials, which are too long and costly.

Antineoplastic Agents↗
PubMed2002

Adapted physical activity in old age: effects of a low-intensity training program on isokinetic power and fatigability.

BACKGROUND AND AIMS: This pilot study measured the effects of an original low-intensity training program, called "posture-balancing-mobility" (PBM), on muscular function. METHODS: Nine non-disabled, elderly (74.3 +/- 6 years) subjects were trained with the PBM technique for 11 weeks (2 sessions per week). Mean power and fatigue index parameters were measured using an isokinetic dynamometer before and after training and compared with those recorded in another group of 9 elderly (71.1 +/- 4.3 years) subjects, who had performed aquatic exercises during the same period and with the same frequency. RESULTS: The mean power of the knee extension muscles increased slightly but significantly on the dominant (15.6%, p<0.05) and non-dominant sides (13.4%, p<0.05) in the PBM group, with no significant fatigue index variation. None of the parameters changed significantly for the aquatic group, and comparison of variations obtained in the two groups showed no significant difference between their respective effects. CONCLUSIONS: Although the results showed slightly enhanced strength production in the PBM group, the low statistical power does not allow conclusions about the impact of this type of training intervention in its current form. Nevertheless, this pilot study provides some indication of the benefits that can be obtained from such an individualized approach. Its efficiency would most likely be improved by further exploration of the minimal threshold of intensity required for strength exercises and by measurement of its effect on functions involving neuromuscular parameters, such as balance or gait.

Aged↗
PubMed2005

Experimental investigation of geologically produced antineutrinos with KamLAND.

The detection of electron antineutrinos produced by natural radioactivity in the Earth could yield important geophysical information. The Kamioka liquid scintillator antineutrino detector (KamLAND) has the sensitivity to detect electron antineutrinos produced by the decay of 238U and 232Th within the Earth. Earth composition models suggest that the radiogenic power from these isotope decays is 16 TW, approximately half of the total measured heat dissipation rate from the Earth. Here we present results from a search for geoneutrinos with KamLAND. Assuming a Th/U mass concentration ratio of 3.9, the 90 per cent confidence interval for the total number of geoneutrinos detected is 4.5 to 54.2. This result is consistent with the central value of 19 predicted by geophysical models. Although our present data have limited statistical power, they nevertheless provide by direct means an upper limit (60 TW) for the radiogenic power of U and Th in the Earth, a quantity that is currently poorly constrained.

Journal Article↗
PubMed1994

Effects of truncation on reaction time analysis.

Many reaction time (RT) researchers truncate their data sets, excluding as spurious all RTs falling outside a prespecified range. Such truncation can introduce bias because extreme but valid RTs may be excluded. This article examines biasing effects of truncation under various assumptions about the underlying distributions of valid and spurious RTs. For the mean, median, standard deviation, and skewness of RT, truncation bias is larger than some often-studied experimental effects. Truncation can also seriously distort linear relations between RT and an independent variable, additive RT patterns in factorial designs, and hazard functions, but it has little effect on statistical power. The authors report a promising maximum likelihood procedure for estimating properties of an untruncated distribution from a truncated sample and present in an appendix a set of procedures to control for truncation biases when testing hypotheses.

Bias↗
PubMed1990

Illustration of the effects of genotype misclassification on the measurement of familial aggregation in epidemiologic studies.

Exposure misclassification is a well-known problem in epidemiologic studies and can considerably dilute relative risk (RR) measures toward unity. A vivid illustration of such misclassification occurs in familial aggregation studies, where "exposure" is defined as a specified genetic relationship to an affected individual (case) or to an unaffected individual (control). Even in the simplest single-gene model, only a fraction of relatives of cases have the disease genotype. Also, if the trait is not fully penetrant, a proportion of relatives of controls can have the genotype. Thus the familial RR measures are subject to misclassification bias if they are interpreted as the relative effect of a susceptible genotype. The effects of this form of "exposure" (or more properly genotype) misclassification on familial RR measures were quantified and applied empirically to known disease-genetic trait associations. Expected RRs in first-, second-, and third-degree relatives were generated and plotted for different genotype RRs, disease, and allele frequencies. In general, familial RRs are substantially lower than genotypic RRs. Even in the case of strong associations such as HLA-B27 and ankylosing spondylitis (RR = 100), RR measures in first-, second-, and third-degree relatives are only 6.0, 3.5, and 2.2, respectively. Such strong misclassification effects may result in considerable reduction of statistical power in family studies. It is suggested that etiologic studies of disease explore directly the role of measurable genetic traits in epidemiologic studies in populations as well as in families.

Classification↗
PubMed2001

Lack of interaction between asbestos exposure and glutathione S-transferase M1 and T1 genotypes in lung carcinogenesis.

An interaction between occupational carcinogens and genetic susceptibility factors in determining individual lung cancer risk is biologically plausible, but the interpretation of available studies are limited by the small number of exposed subjects. We selected from the international database on Genetic Susceptibility and Environmental Carcinogens the studies of lung cancer that included information on metabolic polymorphisms and occupational exposures. Adequate data were available for asbestos exposure and GSTM1 (five studies) and GSTT1 (three studies) polymorphisms. For GSTM1, the pooled analysis included 651 cases and 983 controls. The odds ratio (OR) of lung cancer was 2.0 [95% confidence interval (CI) 1.4-2.7] for asbestos exposure and 1.1 (95% CI 0.9-1.4) for GSTM1-null genotype. The OR of interaction between asbestos and GSTM1 polymorphism was 1.1 (95% CI 0.6-2.1) based on 54 cases and 53 controls who were asbestos exposed and GSTM1 null. The case-only approach, which was based on 869 lung cancer cases and had an 80% power to detect an OR of interaction of 1.56, also provided lack of evidence of interaction. The analysis of possible interaction between GSTT1 polymorphism and asbestos exposure in relation to lung cancer was based on 619 cases. The prevalence OR of GSTT1-null genotype and asbestos exposure was 1.1 (95% CI 0.6-2.0). Our results do not support the hypothesis that the risk of lung cancer after asbestos exposure differs according to GSTM1 genotype. The low statistical power of the pooled analysis for GSTT1 genotypes hampered any firm conclusion. No adequate data were available to assess other interactions between occupational exposures and metabolic polymorphisms.

Adult↗
PubMed2000

A general test of association for complex diseases with variable age of onset.

Analysis of age of onset is a key factor in linkage and association studies of some complex genetic traits. Recent methodological developments are mainly concentrated on binary or quantitative traits, in which age of onset information is ignored. We propose a linkage disequilibrium-based Cox model and a robust score test for testing association between a marker and a disease with variable age of onset. The proposed model is semi-parametric, with an unspecified baseline hazard function. Simulation results indicate that the proposed methods have correct error rates and good statistical power, even in the presence of population admixture. This approach offers a solution to the problem of testing for marker associations when there is a variable age of onset.

Age of Onset↗
PubMedDate not supplied

Key concepts in biostatistics: using statistics to answer the question "is there a difference?".

Biostatistics seeks to answer the question "Is there a difference?" in the rate of a disease or characteristic among subgroups of patients. The goal of this article is to introduce and define measures used in epidemiology and discuss different types of analyses in clinical research with an emphasis on the concepts and implications of the analyses rather than the mathematics. The implications of the use of measures such as incidence and prevalence, as well as odds, risk, and hazards ratios may affect study conclusions. An understanding of the distinction between these summary measures is essential. The concepts of univariate and multivariate analyses, a discussion of what it means to control for potential confounders, and a description of statistical power and significance are also presented. These concepts are integral to the design and analysis of clinical studies. An understanding of their advantages and applications will enhance the reader's ability to understand and evaluate the literature.

Analysis of Variance↗
PubMed1999

Bioequivalence review for drug interchangeability.

To monitor the performance of the approved generic copies of a brand-name drug, we propose some methods in assessing bioequivalence among generic copies and the brand-name drug, and among generic copies themselves, using data from several bioequivalence studies adopting the standard 2 x 2 crossover design without carryover effects. We propose a meta-analysis method that increases statistical power when the between-subject variability is not large. A nonmeta-analysis is also considered. A numerical example of applying both methods is presented for illustration.

Area Under Curve↗
PubMed2007

Power matters in closing the phenotyping gap.

Much of our understanding of physiology and metabolism is derived from investigating mouse mutants and transgenic mice, and open-access platforms for standardized mouse phenotyping such as the German Mouse Clinic (GMC) are currently viewed as one powerful tool for identifying novel gene-function relationships. Phenotyping or phenotypic screening involves the comparison of wild-type control mice with their mutant or transgenic littermates. In our study, we explored the extent to which standardized phenotyping will succeed in detecting biologically relevant phenotypic differences in mice generated and provided by different collaborators. We analyzed quantitative metabolic data (body mass, energy intake, and energy metabolized) collected at the GMC under the current workflow, and used them for statistical power considerations. Our results demonstrate that there is substantial variability in these parameters among lines of wild-type C57BL/6 (B6) mice from different sources. Given this variable background noise in mice that serve as controls, subtle phenotypes in mutant or transgenic littermates may be overlooked. Furthermore, a phenotype observed in one cohort of a mutant line may not be reproducible (to the same extent) in mice coming from a different environment or supplier. In the light of these constraints, we encourage researchers to incorporate information on intrastrain variability into future study planning, or to perform advanced hierarchical analyses. Both will ultimately improve the detectability of novel phenotypes by phenotypic screening.

Animals↗