Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Statistics”

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 1,117 records · Page 62Linked to original sources
PubMed2005

Viewpoint: pancakes and medical statistics.

Everyone is willing to expend considerable effort seeking authenticity; for example, the author sometimes travels over 100 miles to eat authentic and delicious homemade pancakes in a small-town restaurant. In the same way, physicians seek authenticity of care, which is another way of saying they seek the truest knowledge available when treating their patients. Because there are so many possible causes of many patients' complaints, physicians invoke the statistical theory behind the chief complaint (e.g., Which test for the complaint has the highest value?). In a world where medicine must be practiced with attention to resources and cost and where the answer must be reached in the fewest steps, evidence-based medicine (EBM) has risen to prominence. But how do clinicians ensure that the EBM literature is giving them the truest available knowledge? The author observes that clinicians and others have trusted the peer-review system to safeguard them against errors in the clinical literature. But he contends that errors are getting through at an increasing rate, and that physicians cannot automatically trust the peer-review process. Instead, they must become judges of experimental design, statistics, and analysis and assume the responsibilities that they had hoped the peer-review system would bear for them. He speculates about the reasons that peer review is no longer sufficiently ensuring authenticity in the literature. And to alleviate this problem, he recommends that current approaches to educating physicians about experimental design and statistics be augmented by making a beginning knowledge of statistics a requirement for entry into medical school.

Evidence-Based Medicine↗
PubMed2002

Selection of inferential statistics: an overview.

This column serves as an introduction to selection of appropriate statistical methods. In the next five columns we will discuss conceptually, and in more depth, these statistical methods. We will use clinical examples and discuss why the author(s) selected a particular statistical method and how the results of the statistical method were interpreted.

Humans↗
PubMed2000

Statistical modeling of positron emission tomography images in wavelet space.

A new method is introduced for the analysis of multiple studies measured with emission tomography. Traditional models of statistical analysis (ANOVA, ANCOVA and other linear models) are applied not directly on images but on their correspondent wavelet transforms. Maps of model effects estimated from these models are filtered using a thresholding procedure based on a simple Bonferroni correction and then reconstructed. This procedure inherently represents a complete modeling approach and therefore obtains estimates of the effects of interest (condition effect, difference between conditions, covariate of interest, and so on) under the specified statistical risk. By performing the statistical modeling step in wavelet space. the procedure allows the direct estimation of the error for each wavelet coefficient; hence, the local noise characteristics are accounted for in the subsequent filtering. The method was validated by use of a null dataset and then applied to typical examples of neuroimaging studies to highlight conceptual and practical differences from existing statistical parametric mapping approaches.

Artifacts↗
PubMed1976

23. The long-term care component of the Massachusetts Cooperative Health Statistics Program.

The Office of Health Planning and Statistics of the Massachusetts Department of Public Health, with the support of the National Center for Health Statistics, has over the past two years established two data programs in long-term care. The first of these involves experimentation with large-scale collection of patient-specific data from long-term care facilities, with the goal of developing statistically valid algorithms to predict the most appropirate level of care for individual patients. The second program has been the development of a data base for home health care agencies, the elements of which are an agency-specific annual statistical report and a patient-specific discharge abstract program. These two programs mark an effort to document home health care activities so as to provide a base of information for program development and evaluation.

Home Care Services↗
PubMed1991

Problem of between-eye correlation for statistical hypothesis testing: rabbit corneal thickness.

The two eyes of a subject often yield correlated data. Statistical analysis which treats correlated data as if it were independent is most likely to be biased toward statistical significance; that is, the probability of a type I error is likely to be inflated. To illustrate the importance of lack of independence to the inferential process, data from an experimental design commonly used in optometric research are used to demonstrate (1) the potential magnitude of between-eye correlation, (2) the statistical bias toward a significant outcome when the between-eye correlation is ignored via inappropriate analysis, and (3) simple ways by which the bias can be avoided. The researcher must be aware of the between-eye correlation which exists for the particular effect under study, and the statistical bias that ensues from the correlation when the data are not handled correctly.

Animals↗
PubMed2002

The prevalence of negative studies with inadequate statistical power: an analysis of the plastic surgery literature.

Studies published in the medical literature often neglect to consider the statistical power needed to detect a meaningful difference between study groups. Small sample sizes tend to produce negative results because of low statistical power. Studies that cannot make conclusive statements about their hypotheses can waste resources, deter further research, and impede advances in clinical treatment. The current study reviewed three of the most frequently read plastic surgery journals from 1976 to 1996 to determine the prevalence of inadequately (<80 percent) powered clinical trials and experimental studies that found no difference (negative studies) in the response variable of interest between comparison groups. The statistical power of 54 negative studies using continuous response variables was calculated to detect a difference of 1 SD (+/-1 SD) in means between the comparative groups. The power of another 57 negative studies with dichotomous response (yes/no) variables was calculated to detect a relative change in proportions of 25 percent and 50 percent from the experimental to the control group. It was found that 85 percent of the studies with continuous response variables had inadequate power to detect the desired mean difference of +/-1 SD. In studies with dichotomous response variables, 98 percent had inadequate power to detect a desired 25 percent relative change in proportions, and 74 percent had inadequate power to detect a desired 50 percent relative change in proportions. These results indicate that many of the studies in the plastic surgery literature lack adequate power to detect a moderate-to-large difference between groups. The lack of power makes the interpretation of the studies with negative findings inconclusive. Proper study design dictates that investigators consider a priori the difference between groups that is of clinical interest, and the sample size per group that is needed to provide adequate statistical power to detect the desired difference.

Animals↗
PubMedDate not supplied

Risk factors, confounding, and the illusion of statistical control.

When experimental designs are premature, impractical, or impossible, researchers must rely on statistical methods to adjust for potentially confounding effects. Such procedures, however, are quite fallible. We examine several errors that often follow the use of statistical adjustment. The first is inferring a factor is causal because it predicts an outcome even after "statistical control" for other factors. This inference is fallacious when (as usual) such control involves removing the linear contribution of imperfectly measured variables, or when some confounders remain unmeasured. The converse fallacy is inferring a factor is not causally important because its association with the outcome is attenuated or eliminated by the inclusion of covariates in the adjustment process. This attenuation may only reflect that the covariates treated as confounders are actually mediators (intermediates) and critical to the causal chain from the study factor to the study outcome. Other problems arise due to mismeasurement of the study factor or outcome, or because these study variables are only proxies for underlying constructs. Statistical adjustment serves a useful function, but it cannot transform observational studies into natural experiments, and involves far more subjective judgment than many users realize.

Bias↗
PubMed2001

An efficient and robust statistical modeling approach to discover differentially expressed genes using genomic expression profiles.

We have developed a statistical regression modeling approach to discover genes that are differentially expressed between two predefined sample groups in DNA microarray experiments. Our model is based on well-defined assumptions, uses rigorous and well-characterized statistical measures, and accounts for the heterogeneity and genomic complexity of the data. In contrast to cluster analysis, which attempts to define groups of genes and/or samples that share common overall expression profiles, our modeling approach uses known sample group membership to focus on expression profiles of individual genes in a sensitive and robust manner. Further, this approach can be used to test statistical hypotheses about gene expression. To demonstrate this methodology, we compared the expression profiles of 11 acute myeloid leukemia (AML) and 27 acute lymphoblastic leukemia (ALL) samples from a previous study (Golub et al. 1999) and found 141 genes differentially expressed between AML and ALL with a 1% significance at the genomic level. Using this modeling approach to compare different sample groups within the AML samples, we identified a group of genes whose expression profiles correlated with that of thrombopoietin and found that genes whose expression associated with AML treatment outcome lie in recurrent chromosomal locations. Our results are compared with those obtained using t-tests or Wilcoxon rank sum statistics.

Acute Disease↗
PubMed2001

Bayesian analysis of systems with random chemical composition: renormalization-group approach to Dirichlet distributions and the statistical theory of dilution.

We investigate the statistical properties of systems with random chemical composition and try to obtain a theoretical derivation of the self-similar Dirichlet distribution, which is used empirically in molecular biology, environmental chemistry, and geochemistry. We consider a system made up of many chemical species and assume that the statistical distribution of the abundance of each chemical species in the system is the result of a succession of a variable number of random dilution events, which can be described by using the renormalization-group theory. A Bayesian approach is used for evaluating the probability density of the chemical composition of the system in terms of the probability densities of the abundances of the different chemical species. We show that for large cascades of dilution events, the probability density of the composition vector of the system is given by a self-similar probability density of the Dirichlet type. We also give an alternative formal derivation for the Dirichlet law based on the maximum entropy approach, by assuming that the average values of the chemical potentials of different species, expressed in terms of molar fractions, are constant. Although the maximum entropy approach leads formally to the Dirichlet distribution, it does not clarify the physical origin of the Dirichlet statistics and has serious limitations. The random theory of dilution provides a physical picture for the emergence of Dirichlet statistics and makes it possible to investigate its validity range. We discuss the implications of our theory in molecular biology, geochemistry, and environmental science.

Algorithms↗
PubMed2001

Statistics of the integrated backscatter estimate from a blood-mimicking fluid.

This work evaluates the variance of the integrated backscatter (IBS) from moving blood [or blood-mimicking fluid (bmf)] as a way of determining the quality of the mean IBS estimate. The main motivation for this work comes from the fact that absolute IBS values from tissues adjacent to arterial blood can be found by normalizing the measured backscatter energy with the IBS of moving, deaggregated blood. The paper describes the parameters that control the statistics of the IBS estimate, which is calculated for the stochastic ultrasound backscatter signals from flowing blood. It further formulates how the measurement parameters should be specified so that an appropriately low blood IBS variance is ensured or, alternatively, a specified accuracy of the tissue IBS estimate is obtained. First, the paper provides an analytic formulation of the statistics of the IBS, based on a sequence of sampled echoes from a nonstationary Gaussian scattering medium. The analysis incorporates the correlation between the sample values as well as the correlation between the IBS of the individual echoes. The estimate of the mean IBS has been shown to be chi-squared distributed with a determinable order. With the degree of correlation between the samples and between the IBS of individual echoes specified, the number of measurements required to obtain an IBS estimate with a specified variance is readily calculated. Next, a sequence of synthetic echoes is produced and arranged as columns in a data matrix. The echoes are generated such that the second-order statistics along the rows and columns of the matrix match that of actually observed echoes. The actual variance of the mean IBS estimate for the synthetic echoes is calculated and compared with the variance determined from the analytic model, and a good agreement has been found. Finally, sequences of actual backscattered echoes from circulating blood-mimicking fluid are acquired and analyzed to determine the variance of their mean IBS estimate. Based on the measured second-order statistics of the rows and columns of the data matrix for the actual echoes, the observed variance of the mean IBS estimate was compared with the analytically determined variance and with good agreement. Thus, the paper has shown through modeling, simulations, and experiments how the variance of the IBS estimate of the blood backscatter signal can be quantified and reduced to a specified tolerable level.

Blood↗
PubMed2004

Thresholding in edge detection: a statistical approach.

Many edge detectors are available in image processing literature where the choices of input parameters are to be made by the user. Most of the time, such choices are made on an ad-hoc basis. In this article, an edge detector is proposed where thresholding is performed using statistical principles. Local standardization of thresholds for each individual pixel (local thresholding), which depends upon the statistical variability of the gradient vector at that pixel, is done. Such a standardized statistic based on the gradient vector at each pixel is used to determine the eligibility of the pixel to be an edge pixel. The results obtained from the proposed method are found to be comparable to those from many well-known edge detectors. However, the values of the input parameters providing the appreciable results in the proposed detector are found to be more stable than other edge detectors and possess statistical interpretation.

Algorithms↗
PubMed2003

Automatic construction of 3-D statistical deformation models of the brain using nonrigid registration.

In this paper, we show how the concept of statistical deformation models (SDMs) can be used for the construction of average models of the anatomy and their variability. SDMs are built by performing a statistical analysis of the deformations required to map anatomical features in one subject into the corresponding features in another subject. The concept of SDMs is similar to statistical shape models (SSMs) which capture statistical information about shapes across a population, but offers several advantages over SSMs. First, SDMs can be constructed directly from images such as three-dimensional (3-D) magnetic resonance (MR) or computer tomography volumes without the need for segmentation which is usually a prerequisite for the construction of SSMs. Instead, a nonrigid registration algorithm based on free-form deformations and normalized mutual information is used to compute the deformations required to establish dense correspondences between the reference subject and the subjects in the population class under investigation. Second, SDMs allow the construction of an atlas of the average anatomy as well as its variability across a population of subjects. Finally, SDMs take the 3-D nature of the underlying anatomy into account by analysing dense 3-D deformation fields rather than only information about the surface shape of anatomical structures. We show results for the construction of anatomical models of the brain from the MR images of 25 different subjects. The correspondences obtained by the nonrigid registration are evaluated using anatomical landmark locations and show an average error of 1.40 mm at these anatomical landmark positions. We also demonstrate that SDMs can be constructed so as to minimize the bias toward the chosen reference subject.

Algorithms↗
PubMed2004

A statistical model for contours in images.

In this paper, we describe a statistical model for the gradient vector field of the gray level in images validated by different experiments. Moreover, we present a global constrained Markov model for contours in images that uses this statistical model for the likelihood. Our model is amenable to an Iterative Conditional Estimation (ICE) procedure for the estimation of the parameters; our model also allows segmentation by means of the Simulated Annealing (SA) algorithm, the Iterated Conditional Modes (ICM) algorithm, or the Modes of Posterior Marginals (MPM) Monte Carlo (MC) algorithm. This yields an original unsupervised statistical method for edge-detection, with three variants. The estimation and the segmentation procedures have been tested on a total of 160 images. Those tests indicate that the model and its estimation are valid for applications that require an energy term based on the log-likelihood ratio. Besides edge-detection, our model can be used for semiautomatic extraction of contours, localization of shapes, non-photo-realistic rendering; more generally, it might be useful in various problems that require a statistical likelihood for contours.

Algorithms↗
PubMed2002

Interval estimation for a difference between intraclass kappa statistics.

Model-based inference procedures for the kappa statistic have developed rapidly over the last decade. However, no method has yet been developed for constructing a confidence interval about a difference between independent kappa statistics that is valid in samples of small to moderate size. In this article, we propose and evaluate two such methods based on an idea proposed by Newcombe (1998, Statistics in Medicine, 17, 873-890) for constructing a confidence interval for a difference between independent proportions. The methods are shown to provide very satisfactory results in sample sizes as small as 25 subjects per group. Sample size requirements that achieve a prespecified expected width for a confidence interval about a difference of kappa statistic are also presented.

Computer Simulation↗
PubMed2004

Editors can lead researchers to confidence intervals, but can't make them think: statistical reform lessons from medicine.

Since the mid-1980s, confidence intervals (CIs) have been standard in medical journals. We sought lessons for psychology from medicine's experience with statistical reform by investigating two attempts by Kenneth Rothman to change statistical practices. We examined 594 American Journal of Public Health (AJPH) articles published between 1982 and 2000 and 110 Epidemiology articles published in 1990 and 2000. Rothman's editorial instruction to report CIs and not p values was largely effective: In AJPH, sole reliance on p values dropped from 63% to 5%, and CI reporting rose from 10% to 54%; Epidemiology showed even stronger compliance. However, compliance was superficial: Very few authors referred to CIs when discussing results. The results of our survey support what other research has indicated: Editorial policy alone is not a sufficient mechanism for statistical reform. Achieving substantial, desirable change will require further guidance regarding use and interpretation of CIs and appropriate effect size measures. Necessary steps will include studying researchers' understanding of CIs, improving education, and developing empirically justified recommendations for improved statistical practice.

Biomedical Research↗
PubMedDate not supplied

Use of statistics to assess the global burden of breast cancer.

A variety of statistics are used to quantify the burden (occurrence and outcome) of cancer generally and of breast cancer specifically. When undertaking any cancer control program, understanding these statistics, their source, and their quality is important for assessing the current situation, allocating resources to different control strategies, and evaluating progress. Two core statistics are the cancer incidence rate and the cancer mortality rate, which provide estimates of the average risk of acquiring and of dying from the disease, respectively. About 16% of the world's population is covered by registration systems that produce cancer incidence statistics, while mortality data are available for about 29%. Breast cancer incidence and mortality vary considerably by world region. In general, the incidence is high (greater than 80 per 100,000) in developed regions of the world and low (less than 30 per 100,000), though increasing, in developing regions; the range of mortality rates is much less (approximately 6-23 per 100,000) because of the more favorable survival of breast cancer in (high-incidence) developed regions. The incidence of breast cancer is increasing almost everywhere. This unfavorable trend is due in part to increases in risk factors (decreased childbearing and breast-feeding, increased exogenous hormone exposure, and detrimental dietary and lifestyle changes, including obesity and less physical activity). On the other hand, mortality is now decreasing in many high-risk countries due to a combination of intensified early detection efforts and the introduction of mammographic screening, resulting in the diagnosis of more small, early stage tumors, and advances in treatment.

Breast Neoplasms↗
PubMed1996

Drug related involvement of specific sites in fixed eruptions: a statistical evaluation.

Different drugs produce fixed eruptions over different parts of the body. However, the significance of preferential site involvement in fixed eruptions due to specific drugs has not been statistically evaluated. One hundred and twenty five patients of fixed drug eruption (FDE) were studied to examine this question. Different sites affected by individual drugs were classified as lips alone, genitalia alone, lips and genitalia together, trunk alone, trunk and limbs together, and generalized. Statistical analysis was carried out for 7 common drugs causing FDE in 5 or more patients. Cotrimoxazole, a combination of sulfamethoxazole and trimethoprim, was the most common offender for FDE (32.8%), followed by analgin (12%), tetracycline (8%), pyrazolones (8%), metronidazole (6.4%), ciprofloxacin (5.6%), and phenytoin sodium (4%). Major sites involved by FDE were trunk and limbs (24%), lips alone (20.8%), genitalia alone (20%), generalised (14.4%), lips and genitalia together (11.2%), and trunk alone (8.8%). Seventy-two (57.6%) patients had multiple lesions; 33 (42.4%) had single lesions. Involvement of mucocutaneous junctions of lips and genitalia by FDE was frequently caused by cotrimoxazole and tetracycline; trunk and limbs, by analgin and pyrazolones. Generalised eruptions were commonly caused by cotrimoxazole and phenytoin. The present statistical analysis confirms the clinical impression that genitalia alone are frequently involved in FDE due to tetracyclines (p < 0.001). FDE over trunk and limbs was significantly associated with analgin (p < 0.001), and generalised eruptions, with phenytoin sodium (p < 0.001). The association of FDE over lips with cotrimoxazole was statistically not significant.

Adolescent↗
PubMed1993

Endodontic leakage studies reconsidered. Part II. Statistical aspects.

The aim of many endodontic studies is to compare two or more treatment methods, techniques or materials, for example, to detect differences in mean leakage scores. As it is not feasible to study large populations, samples are taken. The important question then arises as to how large the sample sizes have to be in order to establish the 'true' (= populations') mean scores. First, it must be determined which magnitude of the difference (= v) between the mean scores is of endodontic interest. Based upon v and a few related statistical parameters, one may calculate how large the samples must be in order that a statistical test yields a significant result for a difference that is of endodontic importance. In other words, the 'power' of a test, depending on the sample size among other factors, must be large enough to detect the 'true' a priori determined difference between the populations. The use of small sample sizes may imply that a rather large difference between two mean leakage scores is not found to be significant, thereby leading to incorrect conclusions. This article describes the power and the statistical related factors that determine the adequate size of samples. Examples of power calculation are presented. Next, the power of publicized endodontic leakage studies was evaluated. Almost two-thirds of the sample sizes were 10 or less, and about 90% were 20 or less. Less than one-half of the tests had an adequate power (conventionally > or = 0.80). It is necessary to be cautious when extrapolating the results of such studies, because of the limited power of the statistical tests. The power may be increased by using larger sample sizes or, alternatively, by enlarging the 'effect size', by either taking an interest in a larger difference between the mean scores, or by minimizing the variability of the data.

Analysis of Variance↗