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 1,099 records · Page 61Linked to original sources
PubMed2005

Increased sensitivity in neuroimaging analyses using robust regression.

Robust regression techniques are a class of estimators that are relatively insensitive to the presence of one or more outliers in the data. They are especially well suited to data that require large numbers of statistical tests and may contain outliers due to factors not of experimental interest. Both these issues apply particularly to neuroimaging data analysis. We use simulations to compare several robust techniques against ordinary least squares (OLS) regression, and we apply robust regression to second-level (group "random effects") analyses in three fMRI datasets. Our results show that robust iteratively reweighted least squares (IRLS) at the 2nd level is a computationally efficient technique that both increases statistical power and decreases false positive rates in the presence of outliers. The benefits of IRLS are apparent with small samples (n = 10) and increase with larger sample sizes (n = 40) in the typical range of group neuroimaging experiments. When no true effects are present, IRLS controls false positive rates at an appropriate level. We show that IRLS can have substantial benefits in analysis of group data and in estimating hemodynamic response shapes from time series data. We provide software to implement IRLS in group neuroimaging analyses.

Algorithms↗
PubMed2001

Have randomized controlled trials of neuroprotective drugs been underpowered? An illustration of three statistical principles.

BACKGROUND AND PURPOSE: The results of phase III trials of neuroprotective drugs for acute ischemic stroke have been disappointing. We examine the question of whether these trials may have been underpowered. METHODS: Computer simulations were based on the binomial distribution. RESULTS: We illustrate that even small overestimates of the efficacy of an intervention can lead to a serious reduction in statistical power, that the use of data from phase II studies tends to lead to such overestimation, and that a minimum clinically important difference derived with cost-effectiveness modeling techniques is considerably smaller than might be suggested by intuition. CONCLUSIONS: We recommend placing more emphasis on minimum clinically important differences when planning stroke trials, with these differences being derived from an assessment of the public health impact obtained in conjunction with the use of epidemiological and cost-effectiveness models. Even small benefits, when averaged over a sufficiently large number of cases, will, in total, accrue to a large positive impact on the public health.

Brain Ischemia↗
PubMed2005

1H magnetic resonance spectroscopy investigation of the dorsolateral prefrontal cortex in bipolar disorder patients.

BACKGROUND: Magnetic resonance spectroscopy studies (MRS) reported abnormally low levels of N-acetylaspartate (NAA, a marker of neuronal integrity) in dorsolateral prefrontal cortex (DLPFC) of adult bipolar patients, suggesting possible neuronal dysfunction. Furthermore, recent MRS reports suggested possible lithium-induced increase in NAA levels in bipolar patients. We examined with in vivo (1)H MRS NAA levels in the DLPFC of adult bipolar patients. METHODS: Ten DSM-IV bipolar disorder patients (6 lithium-treated, 4 drug-free) and 32 healthy controls underwent a short echo-time 1H MRS session, which localized an 8 cm3 single-voxel in the left DLPFC using a STEAM sequence. RESULTS: No significant differences between the two groups were found for NAA, choline-containing molecules (GPC+PC), or phosphocreatine plus creatine (PCr+Cr) (Student t-test, p > 0.05). Nonetheless, NAA/PCr+Cr ratios were significantly increased in lithium-treated bipolar subjects compared to unmedicated patients and healthy controls (Mann-Whitney U-test, p < 0.05). LIMITATIONS: Relatively small sample size may have reduced the statistical power of our analyses and the utilization of a single-voxel approach did not allow for the examination of other cortical brain areas. CONCLUSIONS: This study did not find abnormally reduced levels of NAA in left DLPFC of adult bipolar patients, in a sample of patients who were mostly on medications. However, elevated NAA/PCr+Cr ratios were shown in lithium-treated bipolar patients. Longitudinal 1H MRS studies should further examine NAA levels in prefrontal cortex regions in untreated bipolar patients before and after mood stabilizing treatment.

Adult↗
PubMed1994

Sample size calculation for clinical trials in which entry criteria and outcomes are counts of events. ACIP Investigators. Asymptomatic Cardiac Ischemia Pilot.

In many chronic diseases, therapy aims to prevent or reduce the frequency of episodes of a disease manifestation, for example cardiac ischaemic episodes or epileptic seizures. Entry criteria for clinical trials typically include a minimum number of episodes within a baseline period, and regression to the mean should be anticipated. The distribution of the number of episodes at follow-up, the statistical power for treatment comparisons, and the difficulty of recruitment will depend on the entry criterion chosen. A gamma-Poisson mixture model is employed to describe the regression to the mean when the entry criterion and outcome measure are counts of discrete events. Sample size formulae which take account of the entry criterion are derived for comparison of mean number of events at follow-up and the proportion of patients with zero events at follow-up. Application of these formulae to screening data from the Asymptomatic Cardiac Ischemia Pilot (ACIP) Study is presented as an example.

Clinical Trials as Topic↗
PubMed2005

Clinical trials in ALS: a review of the role of clinical and neurophysiological measurements.

We have reviewed all the published clinical trials of ALS and, from those considered sufficiently large, and containing a control group, we have evaluated their methodology with regard to statistical power. This implies a critical analysis of the endpoint measurements. We have concluded that clinical endpoints used in clinical trials of ALS have frequently been insufficiently sensitive, non-linear, or even not intuitively highly relevant to the disease. We suggest that the ALS-FRS, perhaps also MUNE and the Neurophysiological Index, may be the best measures currently available. These techniques have complementary characteristics that allow them to be used to address different aspects of the disease and its treatment in various trials designs. In the past some trials may have failed to demonstrate a treatment effect because the chosen endpoint measures and the trial design were inappropriate.

Amyotrophic Lateral Sclerosis↗
PubMed1985

Improving the use of epidemiologic data in health risk assessment.

Epidemiologic data with quantitative exposure measures is infrequently available for specific environmental agents. This lack of exposure measures creates confusion in interpreting epidemiologic data and therefore has impeded its efficient use in health risk analysis. This paper discusses screening and evaluating epidemiologic studies for use in assessing health risk. It also describes the larger role of epidemiology in reducing uncertainties in risk analysis. The approach recognizes that the various designs used to increase statistical power and to control for covariables have different functions in contemporary risk assessment as practiced by regulatory agencies. Each of these study designs is categorized for its role in risk analysis as useful for hazard identification or for dose-response assessment. Studies presenting geographic correlations are construed to be not directly useful in health risk assessment. The numerical level of the exposure data is a deciding factor in using valid epidemiologic studies. However, data measured on an ordinal scale can be used in qualitative assessments and can demonstrate the strength of the relationship. The application of this procedure is illustrated using epidemiologic studies on the carcinogenicity of chemicals contaminated with dioxins.

Dioxins↗
PubMed2004

Discovery of meaningful associations in genomic data using partial correlation coefficients.

MOTIVATION: A major challenge of systems biology is to infer biochemical interactions from large-scale observations, such as transcriptomics, proteomics and metabolomics. We propose to use a partial correlation analysis to construct approximate Undirected Dependency Graphs from such large-scale biochemical data. This approach enables a distinction between direct and indirect interactions of biochemical compounds, thereby inferring the underlying network topology. RESULTS: The method is first thoroughly evaluated with a large set of simulated data. Results indicate that the approach has good statistical power and a low False Discovery Rate even in the presence of noise in the data. We then applied the method to an existing data set of yeast gene expression. Several small gene networks were inferred and found to contain genes known to be collectively involved in particular biochemical processes. In some of these networks there are also uncharacterized ORFs present, which lead to hypotheses about their functions. AVAILABILITY: Programs running in MS-Windows and Linux for applying zeroth, first, second and third order partial correlation analysis can be downloaded at: http://mendes.vbi.vt.edu/tiki-index.php?page=Software. SUPPLEMENTARY INFORMATION: Supplementary information can be found at: URL to be decided.

Algorithms↗
PubMedDate not supplied

Interpretation and comparison of treatment studies for uncomplicated urinary tract infections in women.

Recent treatment trials for uncomplicated urinary tract infection were reviewed to determine most commonly used methodologic approaches and to assess whether methodologic problems often compromise internal validity and comparability of these studies. The 62 studies surveyed fulfilled an average of 56% (standard deviation, 19.2%) of 12 standards necessary for accurate interpretation and comparability. Standards most often met were: reporting the incidence of adverse-effects (90%); recording presence or absence of pretreatment symptoms (81%); and describing clinical response to therapy (71%). Standards least often met were: having adequate statistical power to detect a meaningful difference between therapies (21%); double-blinded assigning of treatment regimens (37%); and clearly defining criteria for diagnosing cure and failure (35%). Two recurring characteristics that impaired both interpretation of individual studies and comparisons between studies were failure to separately randomize or stratify patients with risk factors known to adversely affect therapeutic response and inadequate description of outcome measures. Consideration of these factors and use of a proposed system for classifying therapeutic outcome in future treatment trials would improve the basis for clinical decision making.

Anti-Bacterial Agents↗
PubMed2005

An expectation-maximization-likelihood-ratio test for handling missing data: application in experimental crosses.

The mapping of quantitative trait loci (QTL) is an important research question in animal and human studies. Missing data are common in such study settings, and ignoring such missing data may result in biased estimates of the genotypic effect and thus may eventually lead to errant results and incorrect inferences. In this article, we developed an expectation-maximization (EM)-likelihood-ratio test (LRT) in QTL mapping. Simulation studies based on two different types of phylogenetic models revealed that the EM-LRT, a statistical technique that uses EM-based parameter estimates in the presence of missing data, offers a greater statistical power compared with the ordinary analysis-of-variance (ANOVA)-based test, which discards incomplete data. We applied both the EM-LRT and the ANOVA-based test in a real data set collected from F2 intercross studies of inbred mouse strains. It was found that the EM-LRT makes an optimal use of the observed data and its advantages over the ANOVA F-test are more pronounced when more missing data are present. The EM-LRT method may have important implications in QTL mapping in experimental crosses.

Algorithms↗
PubMed1995

A critical review of studies of newborn discharge timing.

The duration of hospitalization for newborns has declined dramatically, driven by efforts to control health-care costs as well as by efforts to demedicalize childbirth. In order to determine the clinical basis for this practice, the quality of the published literature on discharge timing was analyzed. Thirteen experimental or quasi-experimental studies were retrieved through a computer search. Seven characteristics that influenced the quality of these studies were reviewed: research design; measures of effect; sample descriptions; statistical methods; reliability measures; sample size; and the definition of early discharge, including the use of any related interventions. Although all 13 studies suggest that there are no differences between infants discharged early and their compeers, these studies have three limitations. First, with one exception, these reports are from hospitals where well-defined assessment and follow-up protocols have been established, potentially limiting their wide applicability. Second, these studies lack statistical power to assess the likelihood of rare events such as readmission. Third, few studies report outcomes other than readmission and medical conditions diagnosed within 1 to 6 weeks. Early discharge as the standard of care for well newborns has not been well established by empirical studies. Pediatricians and local public health officials have a responsibility to assure that the health objectives of hospitalization are met whether this occurs in the hospital or through other mechanisms, such as routine home visiting.

Adult↗
PubMed2006

Affective disorders, anxiety disorders and psychological distress in non-drinkers.

BACKGROUND: Non-drinkers have elevated levels of psychological distress but a recent study reported no elevation in prevalence of diagnosed disorders. We aimed to determine the prevalence of affective and anxiety disorders (from the CIDI-A) in current abstainers and contrast results with findings for psychological distress (K10) in the same sample. METHODS: Cross-sectional, representative household survey of adult Australians. RESULTS: Non-/occasional drinkers had higher levels of psychological distress than light drinkers, and distress in heavy drinkers was even higher. Heavy drinkers also had the highest rates of most disorders. Non-/occasional drinkers showed significantly elevated prevalence only of dysthymia, agoraphobia and posttraumatic stress disorder compared with light drinkers. LIMITATIONS: Statistical power was limited for investigating low prevalence disorders. History of alcohol consumption was not collected. The CIDI-A and K10 have finite validity. CONCLUSIONS: This study confirmed J-shaped relationships between psychological distress and alcohol consumption. Although affective and anxiety disorders also showed non-linear relationships with alcohol consumption, non-/occasional drinkers are not at increased risk for all disorders compared to light drinkers. The pattern of symptomatology in non-/occasional drinkers may be of a different character to that in heavy drinkers, as well as being less severe.

Adolescent↗
PubMedDate not supplied

Use of sequential medical trials in rehabilitation research.

Although randomized, clinical trial designs represent the pinnacle of research excellence, they are often difficult to implement in natural or clinical settings. An alternative approach to analysis of data collected during rehabilitation research is the sequential medical trial design. In a sequential medical trial, subjects are serially recruited and the results are continuously analyzed. The use of preconstructed sequential charts remove the need for clinicians to perform complicated statistical analyses and allow an immediate visual indication of the direction the results are taking. Benefits of sequential medical trials are as follows: as soon as enough data are collected to show treatment preference, the trial is stopped; a trial that is showing no beneficial or even harmful effects can be quickly terminated; fewer subjects are needed (without loss of statistical power); features such as randomization, blinding, and cross-over designs can be included; the design suits the serial nature of rehabilitation practice. Limitations to sequential medical trials are that in simple designs extraneous factors cannot be controlled for and that multiple dependent variables are difficult to assess. This article discusses the theory behind the sequential medical trial design and outlines how to plan a sequential medical trial, recruit subjects, construct and use a simple sequential chart, and analyze and interpret the results. Examples of sequential medical trials used in rehabilitation research are presented. It is hoped that this article will increase the exposure of this useful design and result in greater use of sequential medical trials by rehabilitation professionals.

Algorithms↗
PubMed1994

Quantitative analysis of CD34+ stem cells using RT-PCR on whole cells.

We have employed RT-PCR of whole cells to develop a quantitative method for estimating the number of rare cells expressing a unique mRNA in a large, mixed population of cells. We have demonstrated that RT-PCR can be done on whole cells without the need for extraction of the RNA. This allows for a great saving of time and effort, as well as allowing quantitative analysis to be based on the total number of cells analyzed in a given aliquot and the presence or absence of the specific RT-PCR product. We have employed a limiting dilution series on whole cells, with multiple aliquots at each cell concentration to achieve more statistical power in the analysis of a rare cell type. We have used a nested amplification of the CD34 mRNA to be able to detect a single cell expressing the CD34 mRNA in a larger population of non-CD34-expressing cells. We demonstrate that by using this technique, cells from blood and bone marrow containing the CD34 mRNA can be followed quantitatively during a multistep purification involving immunoadsorption followed by fluorescence-activated cell sorting. We also demonstrate that many cells that express the CD34 protein on their surface no longer contain detectable levels of CD34 mRNA, a phenomenon that appears to be developmentally regulated.

Antigens, CD↗
PubMed2005

"Harshlighting" small blemishes on microarrays.

BACKGROUND: Microscopists are familiar with many blemishes that fluorescence images can have due to dust and debris, glass flaws, uneven distribution of fluids or surface coatings, etc. Microarray scans show similar artefacts, which affect the analysis, particularly when one tries to detect subtle changes. However, most blemishes are hard to find by the unaided eye, particularly in high-density oligonucleotide arrays (HDONAs). RESULTS: We present a method that harnesses the statistical power provided by having several HDONAs available, which are obtained under similar conditions except for the experimental factor. This method "harshlights" blemishes and renders them evident. We find empirically that about 25% of our chips are blemished, and we analyze the impact of masking them on screening for differentially expressed genes. CONCLUSION: Experiments attempting to assess subtle expression changes should be carefully screened for blemishes on the chips. The proposed method provides investigators with a novel robust approach to improve the sensitivity of microarray analyses. By utilizing topological information to identify and mask blemishes prior to model based analyses, the method prevents artefacts from confounding the process of background correction, normalization, and summarization.

Algorithms↗
PubMed2005

Replication of putative candidate-gene associations with rheumatoid arthritis in >4,000 samples from North America and Sweden: association of susceptibility with PTPN22, CTLA4, and PADI4.

Candidate-gene association studies in rheumatoid arthritis (RA) have lead to encouraging yet apparently inconsistent results. One explanation for the inconsistency is insufficient power to detect modest effects in the context of a low prior probability of a true effect. To overcome this limitation, we selected alleles with an increased probability of a disease association, on the basis of a review of the literature on RA and other autoimmune diseases, and tested them for association with RA susceptibility in a sample collection powered to detect modest genetic effects. We tested 17 alleles from 14 genes in 2,370 RA cases and 1,757 controls from the North American Rheumatoid Arthritis Consortium (NARAC) and the Swedish Epidemiological Investigation of Rheumatoid Arthritis (EIRA) collections. We found strong evidence of an association of PTPN22 with the development of anti-citrulline antibody-positive RA (odds ratio [OR] 1.49; P=.00002), using previously untested EIRA samples. We provide support for an association of CTLA4 (CT60 allele, OR 1.23; P=.001) and PADI4 (PADI4_94, OR 1.24; P=.001) with the development of RA, but only in the NARAC cohort. The CTLA4 association is stronger in patients with RA from both cohorts who are seropositive for anti-citrulline antibodies (P=.0006). Exploration of our data set with clinically relevant subsets of RA reveals that PTPN22 is associated with an earlier age at disease onset (P=.004) and that PTPN22 has a stronger effect in males than in females (P=.03). A meta-analysis failed to demonstrate an association of the remaining alleles with RA susceptibility, suggesting that the previously published associations may represent false-positive results. Given the strong statistical power to replicate a true-positive association in this study, our results provide support for PTPN22, CTLA4, and PADI4 as RA susceptibility genes and demonstrate novel associations with clinically relevant subsets of RA.

Adult↗
PubMed2004

Randomized study of high-dose pulse calcitriol or placebo prior to radical prostatectomy.

BACKGROUND: Cancer chemoprevention trials require enormous resources due to the large numbers of patients and the years of follow-up needed to achieve sufficient statistical power. Examination of candidate prevention agents using biomarkers as surrogate end points has been proposed as a method to rapidly identify promising agents for prevention trials. Treatment of patients with candidate agents prior to scheduled biopsy or surgical resection of malignancy allows for direct examination of the treatment effects on tumor tissue. In this study, we selected this approach to test several hypotheses about the effect of calcitriol (1,25-dihydroxycholecalciferol), the active form of vitamin D, on early-stage human prostate cancer. METHODS: After selection of surgical treatment for histologically confirmed adenocarcinoma of the prostate, patients were randomized to either calcitriol 0.5 mug/kg or placebo weekly for 4 weeks. The expression levels of the vitamin D receptor (VDR), proliferating cell nuclear antigen, PTEN (MMAC1/TEP1), c-Myc, transforming growth factor (TGF) beta receptor type II (TGFbeta RII), and Bcl-2 were quantified using immunohistochemistry in the patients' prostate specimens post surgery. RESULTS: Thirty-seven of 39 prostate tumors were evaluable for molecular end points. VDR expression was reduced in patients treated with calcitriol (mean, 75.3% of cells) compared with those that received placebo (mean, 98.6%; P = 0.005). Calcitriol treatment did not result in a statistically significant change in the fraction of cells expressing TGFbeta RII, PTEN, or proliferating cell nuclear antigen. Bcl-2 and c-Myc expression was at the lower limits of detection in both the calcitriol group and the placebo group; therefore, we were unable to determine whether drug treatment induced a significant change in these biomarkers. CONCLUSIONS: High-dose calcitriol down-regulates VDR expression in human prostate cancer. Further study is needed to determine the biological consequences of VDR down-regulation in prostate cancer. This study shows that the use of the preprostatectomy model is feasible and can be used to test the effect of candidate chemopreventive agents on prostate cancer.

Adenocarcinoma↗
PubMed1987

Scaled subprofile model: a statistical approach to the analysis of functional patterns in positron emission tomographic data.

The data obtained from measurements of regional rCMRglu using [18F]fluorodeoxyglucose (FDG)/positron emission tomographic (PET) data contain more structure than can be identified with group mean rCMRglu profiles or regional correlation coefficients. This additional structure is revealed by a novel mathematical-statistical model of regional metabolic interactions that explicitly represents rCMRglu profiles as a combination of region-independent global effects, a group mean pattern and a mosaic of interacting networks. In its application to FDG/PET data, this model removes global subject effects [global scaling factors (GSFs)] and a group mean pattern (profile) so as to maximize statistical power for the detection and simultaneous discovery of all networks of two or more regions that form a significant and consistent linearly covarying pattern. The model approach presented here was applied to the combined rCMRglu data from 12 demented AIDS patients and 18 normal controls: Two significant metabolic covariance pattern descriptors that together accounted for 71 to 96% of the rCMRglu/GSF variation across subjects for 22/28 regions in the AIDS group were extracted. Each descriptor was found to be highly correlated with performance on several neuropsychological tests, providing independent validation of the analysis technique as a means of discovering and describing behaviorally related components of group rCMRglu profiles.

Acquired Immunodeficiency Syndrome↗
PubMed2004

A comparison of methods for characterizing the event-related BOLD timeseries in rapid fMRI.

Information about the shape and temporal duration of the blood oxygenation level dependent (BOLD) response can inform both functional neuroanatomy and psychological theory. However, the BOLD response evolves over 20 s or more, making it difficult to distinguish the unique characteristics of the response evoked by temporally adjacent stimuli. Fortunately, event-related BOLD signals can be extracted given that there is adequate variance in the distribution of inter-stimulus intervals (ISI). Unfortunately, the ISI distribution that yields the highest statistical efficiency is not always optimal from a psychological perspective; variability in the stimulus timing may complicate the interpretation of neuroimaging data in terms of underlying cognitive operations. In the present paper, Monte Carlo simulations are used to evaluate two techniques for estimating the event-related BOLD timeseries-event-related averaging and deconvolution using the Ordinary Least Squares estimate -with respect to maintaining acceptable levels of statistical power and experimental validity. While the unbiased deconvolution technique more robustly estimates the shape of the BOLD response functions, both methods succeed in accurately re-producing known differences between evoked BOLD responses when the stimulus ordering is randomized. However, the deconvolution method is more effective at preserving differences when there are sequential dependencies in the stimulus presentation order and restricted ISI distributions are used; particularly if the second of two sequentially dependent stimuli is omitted on some portion of the trials. Importantly, the successful re-production of the evoked BOLD response using restricted ISI distributions often maximizes the ability to make psychologically valid experimental conclusions.

Brain↗