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 1,027 records · Page 57Linked to original sources
PubMed1992

The Lancet's statistical review process: areas for improvement by authors.

The Lancet now incorporates statistical review of submitted papers which remain candidates for publication after conventional review. We summarise here criticisms noted by the statistical reviewers for 191 such papers received between November, 1990, and June, 1991. Only 54% of papers were deemed acceptable or acceptable after revision; the others were either recommended for rejection (14%) or for more substantial revision and re-review (32%). Descriptions of methods and of results were found inadequate in about half of the papers; about one-quarter of papers had inadequate abstracts and conclusions. Major errors of inference were made in 48 papers and went hand in hand with major criticisms of analysis or design in those papers. The natural focus of statistical review is whether conclusions drawn are justified by study design and statistical analysis. In this, there is room for improvement by authors.

Bias↗
PubMed1986

On the use of some multivariate statistical methods in pharmacological research.

Using an interaction experiment with apomorphine and scopolamine effects on exploratory behavior as an illustrative example, four multivariate statistical methods are described and compared with univariate statistical methods, with respect to their utility in pharmacological research. The utility of multivariate analysis of variance and Hotelling's T2 test is compared with univariate analysis of variance and Student's t-test. A novel use of principal component analysis is reported. This latter method transfers knowledge about the dose-response pattern of an agonist to subsequent interaction experiments involving the agonist and putative antagonists. The procedure considerably increases the sensitivity of the statistical analysis and reduces the risk for spurious results (statistical type I errors). Finally, some possibilities of a recently developed method for modeling with latent variables, the partial least squares method, are explored. It is demonstrated by the example how the interaction between apomorphine and scopolamine can be decomposed into one apomorphine-related pattern unaffected by scopolamine, one scopolamine-related pattern sensitive to apomorphine, and one interaction pattern. The similarities and differences between principal components and partial least squares analyses are also discussed.

Animals↗
PubMed1986

Brain asymmetry: facts and fallacies in statistical inference.

In a recent article, C. M. Clark, R. Kessler, and R. Margolin (1985, Brain and Cognition, 4, 7-12) argue that standard statistical tests should not be applied to measures of regional cerebral metabolism. They offer no evidence that such measures uniquely escape the mathematical foundations of statistical inference but, instead, merely note the definitional relations among various descriptive and inferential statistics. They falsely claim that these relations invalidate conclusions drawn from statistical tests, and in making this claim, they commit several quite serious logical fallacies. These fallacies are examined.

Brain Mapping↗
PubMed1996

Trough-to-peak ratios: some statistical considerations.

The trough-to-peak ratio has recently come into prominence as a statistic to evaluate the clinical duration of effect for an antihypertensive medication. The statistic is unusual in that it is defined as the ratio of mean drug effect (compared to placebo) at trough compared to that at peak. There is no corresponding measure available on individual patients. This paper will examine the criteria needed to make the trough-to-peak ratio a useful, interpretable statistic. In addition, other possible trough-to-peak measures and statistics are considered for their potential usefulness in describing diurnal blood pressure (BP) data.

Antihypertensive Agents↗
PubMed2005

Toward an accurate statistics of gapped alignments.

Sequence alignment has been an invaluable tool for finding homologous sequences. The significance of the homology found is often quantified statistically by p-values. Theory for computing p-values exists for gapless alignments [Karlin, S., Altschul, S.F., 1990. Methods for assessing the statistical significance of molecular sequence features by using general scoring schemes. Proc. Natl. Acad. Sci. USA 87, 2264-2268; Karlin, S., Dembo A., 1992. Limit distributions of maximal segmental score among Markov-dependent partial sums. Adv. Appl. Probab. 24, 13-140], but a full generalization to alignments with gaps is not yet complete. We present a unified statistical analysis of two common sequence comparison algorithms: maximum-score (Smith-Waterman) alignments and their generalized probabilistic counterparts, including maximum-likelihood alignments and hidden Markov models. The most important statistical characteristic of these algorithms is the distribution function of the maximum score S(max), resp. the maximum free energy F(max), for mutually uncorrelated random sequences. This distribution is known empirically to be of the Gumbel form with an exponential tail P(S(max)>x) approximately exp(-lambdax) for maximum-score alignment and P(F(max)>x) approximately exp(-lambdax) for some classes of probabilistic alignment. We derive an exact expression for lambda for particular probabilistic alignments. This result is then used to obtain accurate lambda values for generic probabilistic and maximum-score alignments. Although the result demonstrated uses a simple match-mismatch scoring system, it is expected to be a good starting point for more general scoring functions.

Algorithms↗
PubMed2006

Statistical models in assessing fold change of gene expression in real-time RT-PCR experiments.

Real-time RT-PCR has been frequently used in quantitative research in molecular biology and bioinformatics. It provides remarkably useful technology to assess expression of genes. Although mathematical models for gene amplification process have been studied, statistical models and methods for data analysis in real-time RT-PCR have received little attention. In this paper, we briefly introduce current mathematical models, and study statistical models for real-time RT-PCR data. We propose a generalized estimation equations (GEE) model that properly reflects the structure of repeated data in RT-PCR experiments for both cross-sectional and longitudinal data. The GEE model takes the correlation between observations within the same subjects into consideration, and prevents from producing false positives or false negatives. We further demonstrate with a set of actual real-time RT-PCR data that different statistical models yield different estimations of fold change and confidence interval. The SAS program for data analysis using the GEE model is provided to facilitate easy computation for non-statistical professionals.

Adipose Tissue↗
PubMed2006

Some statistical issues related to multiple linear regression modeling of beach bacteria concentrations.

As a fast and effective technique, the multiple linear regression (MLR) method has been widely used in modeling and prediction of beach bacteria concentrations. Among previous works on this subject, however, several issues were insufficiently or inconsistently addressed. Those issues include the value and use of interaction terms, the serial correlation, the criteria for model selection, and model assessment. The present work shows that serial correlations, as often present in sequentially observed data records, deserve full attention from the modeler. The testing and adjustment for the time-series effect should be implemented in a statistically rigorous framework. The R(2) and Cp-statistic as joint criteria are recommended for the model selection process, while using the t-statistics associated with the full model is erroneous. During model selection, using interaction terms can often help to decrease the bias in reduced models, although the resulting improvement in the numerical performance may be limited. For the assessment of the model predictive capacity, which is different from testing the goodness of fit, a comprehensive set of statistics are advocated to allow for an objective evaluation of different models. Results obtained from the data at Huntington Beach, OH, show that erroneous conclusions could be drawn if only the model R(2) and the count of type I and type II errors are considered. In this sense, several previous works deserve further investigation.

Bathing Beaches↗
PubMed2006

The use of a registry database in clinical trial design: assessing the influence of entry criteria on statistical power and number of eligible patients.

Randomized clinical trials (RCTs) are prospective empirical studies used to investigate the effect of a particular medical intervention. The design of a clinical trial is a delicate decision process, as each of the decisions that are taken in this process influences the eventual result of the clinical trial. Despite the efforts that are put into trial design, many trials fail to show an effect of the intervention. In some of these situations the intervention may be truly ineffective, however, more often this is caused by problems with the inclusion of patients and a resulting lack of statistical power to show the effect. To avoid this problem, in the design of a trial, the statistical power that can be achieved with the current design choices is calculated and balanced with economic considerations. In the choice of the entry criteria however, an important step in the design process, the influence of the chosen criteria on statistical power and number of eligible patients is not quantified. As these criteria influence the characteristics of the study population and the number of patients that will be eligible for the trial, and thereby the chances of finding an effect of the intervention, we believe that also in the choice of entry criteria explicit estimates of the number of eligible patients should be made. This paper presents a method to arrive at precise, objective estimates of statistical power and the number of eligible patients, using a registry database. Furthermore, we describe how this method is incorporated in the process of choosing entry criteria for a clinical trial. We illustrate the method with an example in the area of severe sepsis.

Humans↗
PubMed2004

Mathematical coupling can undermine the statistical assessment of clinical research: illustration from the treatment of guided tissue regeneration.

OBJECTIVES: Previous periodontal literature has shown that there is a strong relationship between treatment effects, such as guided tissue regeneration (GTR), and baseline disease severity. However, relating change to baseline values using correlation or regression is methodologically flawed due to mathematical coupling, where the statistical procedure of testing the null hypothesis-that the coefficient of correlation or slope of regression is equal to zero-becomes erroneous. The aim of this study is to investigate if baseline disease severity is genuinely associated with the treatment outcome of intrabony defects using GTR after adjustment for mathematical coupling. In particular, we seek to demonstrate the potential effect that mathematical coupling has in distorting the results from the statistical analyses of trials of dental treatment, using data from the periodontal literature on GTR. The erroneous results arising from the use of simple correlation and regression techniques to analyse this association will be demonstrated, also the methodological flaw where the statistical procedure tests the null hypothesis-that the coefficient of correlation or the slope of regression is equal to zero. METHODS: Three main periodontal journals were electronically and manually searched to extract the data for the clinical outcomes of pocket probing depth (PPD) and lifetime cumulative attachment loss (LCAL) in the studies using GTR. The relationship between clinical outcomes and baseline measurements were reanalysed using Oldham's method and the variance ratio test. RESULTS: The results of these analyses were compared with those from the papers where the authors used the standard approach of correlation or regression. This shows that mathematical coupling caused spurious correlations between baseline disease severity and treatment effect. Ten out of 12 studies for PPD and nine out of 14 for LCAL initially claimed a significant positive relationship; after using either of the more appropriate statistical methods of adjustment, only three correlations in each group of studies remained significant. CONCLUSIONS: Previous evidence suggesting an association between baseline disease severity and treatment effect for GTR is challenged and therefore needs to be critically reviewed. All future clinical research should avoid using mathematically coupled data in correlation or regression analysis. In seeking to examine the bivariate association between baseline and subsequent change, Oldham's method is recommended.

Alveolar Bone Loss↗
PubMed2005

Basic concepts of statistical analysis for surgical research.

Appropriate statistical analyses are an integral part of surgical research. The purpose of this work is to assist surgeons and clinicians with the interpretation of statistics by providing a general understanding of the basic concepts that lead to choosing an appropriate statistical test for common study designs. It is extremely important to understand the nature of the data before embarking on a statistical analysis. A researcher must design an appropriate study around the research hypothesis. Initially, data should be inspected using frequency distributions and graphical techniques. If the data are continuous, the normality of the distribution must be assessed. In addition, the data must be defined as independent or dependent. For normally distributed and independent samples, a two-sample t test is appropriate. A paired t test should be used for dependent data. The nonparametric counterpart to the t test is the Mann-Whitney U and the paired counterpart is the Wilcoxon signed rank. For binary data, contingency table methods such as a chi2 test apply unless the expected value is < 5; then, use the Fisher's exact test. The McNemar test applies to paired binary data. Correlation coefficients assess the association between two continuous distributions. Linear regression assesses trend. Multiple regression analysis is appropriate for multivariate analyses with a continuous outcome variable. Logistic regression methods would apply for binary outcomes. The quality of the analysis and subsequent results of any research project depend on an appropriate study design, data collection, and analysis to make meaningful conclusions.

Biomedical Research↗
PubMed2005

Statistical mechanics of myosin molecular motors in skeletal muscles.

Statistical mechanics provides the link between microscopic properties of matter and its bulk properties. The grand canonical ensemble formalism was applied to contracting rat skeletal muscles, the soleus (SOL, n = 30) and the extensor digitalis longus (EDL, n = 30). Huxley's equations were used to calculate force (pi) per single crossbridge (CB), probabilities of six steps of the CB cycle, and peak muscle efficiency (Eff(max)). SOL and EDL were shown to be in near-equilibrium (CB cycle affinity 2.5 kJ/mol) and stationary state (linearity between CB cycle affinity and myosin ATPase rate). The molecular partition function (z) was higher in EDL (1.126+/-0.005) than in SOL (1.050+/-0.003). Both pi and Eff(max) were lower in EDL (8.3+/-0.1 pN and 38.1+/-0.2%, respectively) than in SOL (9.2+/-0.1 pN and 42.3+/-0.2%, respectively). The most populated step of the CB cycle was the last detached state (D3) (probability P(D3): 0.890+/-0.004 in EDL and 0.953+/-0.002 in SOL). In each muscle group, both pi and Eff(max) linearly decreased with z and statistical entropy and increased with P(D3). We concluded that statistical mechanics and Huxley's formalism provided a powerful combination for establishing an analytical link between chemomechanical properties of CBs, molecular partition function and statistical entropy.

Adenosine Triphosphatases↗
PubMed2004

Generalized image models and their application as statistical models of images.

A generalized image model (GIM) is presented. Images are represented as sets of four-dimensional (4D) sites combining position and intensity information, as well as their associated uncertainty and joint variation. This model seamlessly allows for the representation of both images and statistical models (such as those used for classification of normal/abnormal anatomy and for inter-patient registration), as well as other representations such as landmarks or meshes. A GIM-based registration method aimed at the construction and application of statistical models of images is proposed. A procedure based on the iterative closest point (ICP) algorithm is modified to deal with features other than position and to integrate statistical information. Furthermore, we modify the ICP framework by using a Kalman filter to efficiently compute the transformation. The initialization and update of the statistical model are also described. Preliminary results show the feasibility of the approach and its potentialities.

Algorithms↗
PubMed2005

Statistical analysis of fractal-based brain tumor detection algorithms.

Fractals are geometric objects that have a noninteger fractal dimension (FD). The FD has been exploited for various biomedical recognition applications such as breast tumor and lung tumor detection. Our previous work shows that the FD is useful in the detection of brain tumors when a reference nontumor image is available. In this work, we extend our previous work by statistically validating the results of FD analysis on a set of 80 real MR and CT images. Our half-image technique requires that the tumor is located in one half of the brain whereas our whole-image technique does not. Furthermore, we alleviate the need for a reference (control) nontumor image to compute the tumor FD, which was necessary in our previous work. We also compare the brain tumor detection performance of our algorithms with other fractal-based algorithms in the literature and statistically validate our results against manually segmented tumor images. We find that the tumor region offers a statistically significant lower FD compared with that of the nontumor area for most of the FD algorithms studied in this work. Thus, our statistical analysis suggests that these FD algorithms may be exploited successfully to determine the possible presence and location of brain tumors in MR and CT images.

Algorithms↗
PubMed2004

Variation of BOLD hemodynamic responses across subjects and brain regions and their effects on statistical analyses.

Estimates of hemodynamic response functions (HRF) are often integral parts of event-related fMRI analyses. Although HRFs vary across individuals and brain regions, few studies have investigated how variations affect the results of statistical analyses using the general linear model (GLM). In this study, we empirically estimated HRFs from primary motor and visual cortices and frontal and supplementary eye fields (SEF) in 20 subjects. We observed more variability across subjects than regions and correlated variation of time-to-peak values across several pairs of regions. Simulations examined the effects of observed variability on statistical results and ways different experimental designs and statistical models can limit these effects. Widely spaced and rapid event-related experimental designs with two sampling rates were tested. Statistical models compared an empirically derived HRF to a canonical HRF and included the first derivative of the HRF in the GLM. Small differences between the estimated and true HRFs did not cause false negatives, but larger differences within an observed range of variation, such as a 2.5-s time-to-onset misestimate, led to false negatives. Although small errors minimally affected detection of activity, time-to-onset misestimates as small as 1 s influenced model parameter estimation and therefore random effects analyses across subjects. Experiment and analysis design methods such as decreasing the sampling rate or including the HRF's temporal derivative in the GLM improved results, but did not eliminate errors caused by HRF misestimates. These results highlight the benefits of determining the best possible HRF estimate and potential negative consequences of assuming HRF consistency across subjects or brain regions.

Adolescent↗
PubMed2005

Use of the scan statistic on disaggregated province-based data: foot-and-mouth disease in Iran.

The spatial scan statistic was applied to density-smoothed data that approximated the spatial distribution within the area and reduced the potential bias produced when location data have been aggregated for large areas. The method is illustrated, using data on the location of foot-and-mouth disease (FMD) outbreaks in Iran. Data examined were 4477 FMD outbreaks reported on a per province basis between June 1996 and September 2003. A kernel density of the outbreak locations was estimated, using a fixed radius and the centroid of each province as the designated location of all cases reported for the province. The radius that produced a density map with the highest correlation with expert opinion was 4 degrees (latitude/longitude). Livestock density was used as a proxy for the underlying population at risk of acquiring FMD. Livestock and outbreak density maps were overlain to obtain the number of outbreaks and livestock in each of 15,599 cells covering the mapped surface of the country. A spatial scan statistic was applied to the density-smoothed data assuming that the outbreaks had a Poisson distribution. Results were compared with those obtained using a spatial scan statistic on provincially aggregated data. Application of the spatial scan statistic on the density-smoothed data allowed identification of clusters (P<0.01) related more to the actual geographic distribution of cases (expert opinion) and of animals at risk, than to the distribution of the provinces. Significant clusters of FMD were identified that coincided with roads, neighboring countries, and high-density population areas, suggesting that the region may represent a route for cross-continent transmission of FMD.

Animals↗
PubMed2004

Waiting time information services: an evaluation of how well clearance time statistics can forecast a patient's wait.

Governments in some countries have created web-based information services so that patients requiring elective surgery can compare the waiting times of surgical units. This study investigated how accurately the waiting times of patients about to join a waiting list can be forecast by various clearance time statistics. It used 3 years of elective surgical activity data that covered 46 surgeons in 10 specialties within a public hospital. Six clearance time functions were tested, and the best function was compared with average waiting time statistics derived from census and throughput data. The forecast accuracy of the clearance time functions was found to be greatly affected by the characteristics and behaviour of a surgeon's waiting list. Although there was less difference in performance among the six functions, systematic differences between them were also found. The best of these performed better than the statistics derived from waiting time data, especially where waiting times exceeded 6 months. Yet, its accuracy was still poor. For each surgeon with an average waiting time of more than 6 months, at least 20% of patients waited more than 90 days beyond the clearance time forecast. Consequently, while waiting time information services should consider adopting the clearance time approach, they need to be explicit about its statistical limitations.

Elective Surgical Procedures↗
PubMedDate not supplied

Statistical power analysis for hemodynamic cardiovascular safety pharmacology studies in beagle dogs.

INTRODUCTION: We studied the statistical power of a replicated Latin square design where eight animals each receive a vehicle control and three dose levels of a drug on four separate dosing days. Cardiovascular parameters evaluated in the study were systolic arterial pressure, diastolic arterial pressure, left ventricular heart rate, and dP/dt(max). METHODS: Observations were simulated based on historical data and drug response profiles from cardiovascular safety pharmacology studies conducted at Lilly Research Laboratories. Statistical analysis for treatment effects was performed using a linear mixed model. Monotonicity of dose response was examined using sequential linear trend tests based on ordinal spacing of dose levels. RESULTS: The replicated Latin square design for cardiovascular safety pharmacology studies is shown to have at least an 80% power of detecting changes from control of at least a 10% increment in systolic and diastolic pressure and a 15% increment in heart rate and dP/dt(max). The power is not sensitive to the shape of dose response profile over time. DISCUSSION: Several unique features of our statistical power evaluation include the comparison of different covariance structures and drug response profiles. The procedure can also be applied to future power evaluations of other cardiovascular parameters, such as the QT interval, and the loss of statistical power due to missing observations.

Animals↗
PubMed2002

Bayesian analysis: a new statistical paradigm for new technology.

Full Bayesian analysis is an alternative statistical paradigm, as opposed to traditionally used methods, usually called frequentist statistics. Bayesian analysis is controversial because it requires assuming a prior distribution, which can be arbitrarily chosen; thus there is a subjective element, which is considered to be a major weakness. However, this could also be considered a strength since it provides a formal way of incorporating prior knowledge. Since it is flexible and permits repeated looks at evolving data, Bayesian analysis is particularly well suited to the evaluation of new medical technology. Bayesian analysis can refer to a range of things: from a simple, noncontroversial formula for inverting probabilities to an alternative approach to the philosophy of science. Its advantages include: (1) providing direct probability statements--which are what most people wrongly assume they are getting from conventional statistics; (2) formally incorporating previous information in statistical inference of a data set, a natural approach which we follow in everyday reasoning; and (3) flexible, adaptive research designs allowing multiple looks at accumulating study data. Its primary disadvantage is the element of subjectivity which some think is not scientific. We discuss and compare frequentist and Bayesian approaches and provide three examples of Bayesian analysis: (1) EKG interpretation, (2) a coin-tossing experiment, and (3) assessing the thromboembolic risk of a new mechanical heart valve.

Bayes Theorem↗