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 991 records · Page 55Linked to original sources
PubMed1997

An empirical comparison of statistical tests for assessing the proportional hazards assumption of Cox's model.

In the analysis of survival data using the Cox proportional hazard (PH) model, it is important to verify that the explanatory variables analysed satisfy the proportional hazard assumption of the model. This paper presents results of a simulation study that compares five test statistics to check the proportional hazard assumption of Cox's model. The test statistics were evaluated under proportional hazards and the following types of departures from the proportional hazard assumption: increasing relative hazards; decreasing relative hazards; crossing hazards; diverging hazards, and non-monotonic hazards. The test statistics compared include those based on partitioning of failure time and those that do not require partitioning of failure time. The simulation results demonstrate that the time-dependent covariate test, the weighted residuals score test and the linear correlation test have equally good power for detection of non-proportionality in the varieties of non-proportional hazards studied. Using illustrative data from the literature, these test statistics performed similarly.

Bias↗
PubMed2000

On statistical analysis for placebo-challenging designs in clinical trials.

In clinical trials, appropriate designs are often chosen to address scientific/medical questions of particular interest to the investigator. For a chosen statistical design, however, standard statistical procedures may not be applicable owing to the nature of the design. In this paper we examine statistical methods for analysis of data collected from a placebo-challenging design which is often considered for assessment of the efficacy of drug products for indication of erectile dysfunction. An example concerning a clinical trial conducted with 120 male patients with erectile dysfunction is used to illustrate the derived statistical methods. Some recommendations to the randomization procedure for the study design of this kind are also made.

Cross-Over Studies↗
PubMed1999

A preliminary statistical examination of the effects of uncertainty and variability on environmental regulatory criteria for ozone.

Basing the quantitative expression of environmental regulatory standards and associated compliance criteria on statistical principles has recently received attention in Europe, most visibly in a study by the UK Royal Commission on Environmental Pollution. These issues are timely for consideration in the USA, where a recent periodic review of National Ambient Air Quality Standards (NAAQS) has led to revision of the regulatory standards for ambient ozone and particulate matter. Salient statistical issues include accounting for errors of the first and second kind due to sampling and measurement error. These issues appear routine statistically and also may seem absent from regulations, but neither is necessarily the case. This paper is directed towards developing a methodology for examining the problem of dealing with uncertainty and variation in environmental regulations and compliance criteria. Our approach is illustrated through statistical analysis of the (old) 1 hour and the (new) 8 hour standards for ambient ozone, based on intensive monitoring in California's San Joaquin Valley during summer 1990 performed under the SARMAP Project. This paper presents preliminary findings based on quantifying measurement error or precision in terms of small-scale spatial and temporal variability, laying the groundwork for future work.

Air Pollutants↗
PubMed1999

Better late than never? Injecting statistical know-how into legislation on water quality.

Many of the Directives we use to control the risk of water pollution do not distinguish the statistical concepts of population and sample. They also fail to acknowledge the uncertainty in estimates of summary statistics. This leads to faulty assessments of performance, some of which have serious consequences. This paper explains how calculations of compliance should be done in duplicate. First, we must follow the stipulations in the Directives. Second we must make a proper statistical estimation. The results from the second assessment show where failure is statistically significant, and where the Directives' rules preclude a good decision on the need to act.

Confidence Intervals↗
PubMed1997

Statistical and study design issues in assessing the quality and outcomes of care in rheumatic diseases.

As we have pointed out in this article, a health care study should have a well-defined intent that is matched to the study type, population of interest, and outcome. Care must be taken to collect samples in a meaningful way, so that results can be generalized to larger populations. Outcomes must be selected carefully to ensure that they will be sensitive to the types of care being considered, and only a few main outcomes should be selected so as to preserve the level of statistical significance of the research results. Sample sizes should be sufficient to detect an effect of reasonable size, after accounting for attrition in longitudinal studies and rates of occurrence with dichotomous outcomes. If the study purpose is to compare multiple institutions or health care providers, statistical adjustments for case mix will generally be required. Outcome research is an expanding area of development for rheumatology care and for medical care in general. It offers the promise of the use of administrative data bases to answer questions that are important both to arthritis researchers and to consumers of rheumatology care. As with all areas of clinical research, we must maintain appropriate levels of statistical rigor to protect the integrity of the results. Inadequate attention to the design and analysis of data can compromise research results before a study even gets started, and health care research studies have as many potential statistical pitfalls as other types of clinical research.

Epidemiologic Studies↗
PubMed2004

Challenges in constructing statistically based structure-activity relationship models for developmental toxicity.

Regulatory agencies are increasingly called upon to review large numbers of environmental contaminants that have not been characterized for their potential to pose a health risk. Additionally, there is special interest in protecting potentially sensitive subpopulations and identifying developmental toxicants that may be present in the environment. Thus, there is an urgent need for efficient methods to screen large numbers of chemicals for their potential to pose a developmental hazard. One potential screening method involves the use of statistically based structure-activity relationship (SAR) tools to predict activity of untested chemicals. Such systems rely on statistical analyses to discern relationships between structure and activity for a training set of substances. Predictions can then be made for an untested substance as long as its structural features are encompassed by chemicals of the training set. In theory, such systems could assist regulatory agencies in their screening efforts; however, to date, there has been little independent evaluation of these tools for this use. To contribute to such an evaluation, the International Life Sciences Institute Risk Science Institute (ILSI RSI) convened a Working Group to examine methodology used to construct statistically based SAR systems for developmental toxicity. This document reports on the deliberations of the Working Group, which concluded that an improved process is needed for utilizing developmental toxicity data in the construction of statistically based SAR models. The process must be objective, reproducible, rational and transparent. Moreover, it must be informed by the expertise of developmental toxicologists and biologists and must be subject to peer review.

Congenital Abnormalities↗
PubMed1986

Statistical analyses for in vitro cytogenetic assays using Chinese hamster ovary cells.

It is a widely held view that objective statistical criteria are needed for the evaluation of genetic toxicity assays. This paper presents statistical methods for the analysis of data from in vitro sister chromatid exchange (SCE) and chromosome aberration tests that use Chinese hamster ovary cells. For SCEs, an extensive study of solvent control results demonstrated that there is a substantial interday component of variability in the data, and that a Poisson sampling model is applicable to data generated via the protocol of Galloway et al [1985]. Consequently, a trend test for evidence of a dose response is proposed for such SCE data. As an illustration of this statistical method, analysis of data previously considered to be negative [Gulati et al, 1985] indicates that di(2-ethyl-hexyl) phthalate induces a weak, but reproducible, SCE dose response in CHO cells. Monte Carlo methods are used to show that the trend test is more sensitive than four other statistical procedures considered for the analysis of Poisson-distributed SCEs. A similar trend test for dose response in proportions is proposed for chromosome aberration data, where the percent of cells with chromosome aberrations is the response of interest. Sensitivity (or power) studies indicate that three doses and a control with 50 cells/dose point is a reasonable design for an in vitro SCE study that uses the Galloway et al protocol. For in vitro chromosome aberrations, however, three doses and a control with 100 cells/dose point appears to produce too insensitive an assay; an increase to 200 cells/dose point in the Galloway et al protocol seems worthy of serious consideration.

Animals↗
PubMed1993

Clinical importance, statistical significance and the assessment of economic and quality-of-life outcomes.

The assessment of economic and quality-of-life outcomes of health care interventions is moving into a new era, with such assessments increasingly being made within the context of controlled clinical trials. Traditionally the measurement of many variables in economic evaluations, particularly costs, has been deterministic. In the context of clinical trials the measurement of variables is stochastic, with the standard principles of statistical inference being applied to analyse differences between treatments in terms of effectiveness. Economists participating in clinical research are therefore being called upon to specify the sample size for the economic component of the evaluation and to undertake statistical tests for differences in cost or cost-effectiveness. This paper discusses the current methodological issues surrounding stochastic measurement in clinical trials, discusses the additional issues raised by the assessment of economic and quality-of-life outcomes and specifies the challenges facing economists if they are to answer the questions now being posed about economic analysis by statisticians and clinical researchers. It is concluded that application of the standard principles of statistical inference to economic data is not straightforward and will require value judgements to be made about statistical significance and economic importance, which may differ from those already made in purely clinical studies.

Confidence Intervals↗
PubMed2002

A statistical method for the determination of absorption rate constant estimated using the rat single pass intestinal perfusion model and multiple linear regression.

The guide "Waiver of In Vivo Bioavailability and Bioequivalence Studies for Immediate Release Solid Dosage Forms Containing Certain Active Moieties/Active Ingredients Based on a Biopharmaceutical Classification System" (Rockville, MD: CDER, 2000) outlined non-in vivo tests of permeability that may satisfy the classification of a compound in the biopharmaceutical classification system. However, absent from that document were specific statistical methods to legitimatize the non-in vivo tests. This report describes the appropriate statistical treatment of absorption data, and recommends its adoption in the estimation of absorption and/or permeability measurements. The calculation of the absorption rate constants (k(a)) of ten compounds by a new multiple linear regression (MLR) method was completed after the separate perfusion of each compound through the rat single pass intestinal perfusion model (n = 3 rats per compound). Studentized residuals were evaluated to determine whether any statistically significant outliers were present in the data. The standard error of k(a) was estimated using variance components from the random effects model. The results were compared with the "traditional method" for k(a) calculations. Although both methods produced similar values of k(a), the MLR method's error estimate included multiple components of variability, which was largely ignored by the traditional method. The MLR method provided objective tests for outliers and achievement of steady-state. A preferred method for the statistical analysis of absorption data was demonstrated. These methods should be applied to all forms of permeability measurements, especially the non-in vivo measurements that classify a compound in the biopharmaceutical classification system.

Absorption↗
PubMed2004

Reproducibility of statistical motor unit number estimates in amyotrophic lateral sclerosis: comparisons between size- and number-weighted modifications.

Motor unit number estimations (MUNEs) can directly assess motor unit populations in muscle and quantify the degree of physiological or pathological motor unit degeneration. A high degree of reproducibility and reliability is required of any effective quantitative tool. MUNE is being increasingly applied clinically, and statistical MUNE has several advantages over alternative techniques. Nevertheless, the optimal method of applying statistical MUNE with respect to its reproducibility has not been established. We performed statistical MUNE by selecting the most compensated compound muscle action potential (CMAP) area as a test area and modified the results obtained by using the weighted mean surface-recorded motor unit potential (SMUP). MUNE measurements made in patients with amyotrophic lateral sclerosis (ALS) showed better reproducibility after incorporating the size-weighted modification. Therefore, we suggest that the size-weighted MUNE in combination with the selection of testing "neurogenically compensated" CMAP areas is a more reliable method of statistical MUNE analysis in ALS patients.

Action Potentials↗
PubMed2007

Revised statistical motor unit number estimation in the Celecoxib/ALS trial.

Techniques to estimate motor unit number (MUNE) measure the number of functioning motor units in a muscle. As amyotrophic lateral sclerosis (ALS) is characterized by progressive motor unit loss, this disease offers an ideal setting for the use of MUNE. Statistical MUNE was employed in a recent multicenter trial of creatine in ALS, and was shown to be reliable, reproducible, and to decline with disease progression. However, motor unit amplitude stayed constant over 7 months, a finding believed to reflect an artifact of the method. The statistical method was revised to reflect more accurately the presence of larger motor units and employed in a 12-month study of Celecoxib in ALS. MUNE declined by 49% in 12 months; however, motor unit amplitude again stayed constant over the same period. Statistical MUNE estimates motor unit number based on the variability of response to a repeated stimulus of constant strength, with an underlying assumption that this variability is due solely to the number of motor units responding in an intermittent manner. Based on studies showing that single motor units in ALS display excessive amplitude variability when stimulated repeatedly, we show that response variability in ALS patients is in large part due to single unit changes. Thus, we conclude that the statistical method is not an appropriate measure of motor unit number in any disease associated with motor unit instability.

Action Potentials↗
PubMed2003

Statistical models of shape for the analysis of protein spots in two-dimensional electrophoresis gel images.

In image analysis of two-dimensional electrophoresis gels, individual spots need to be identified and quantified. Two classes of algorithms are commonly applied to this task. Parametric methods rely on a model, making strong assumptions about spot appearance, but are often insufficiently flexible to adequately represent all spots that may be present in a gel. Nonparametric methods make no assumptions about spot appearance and consequently impose few constraints on spot detection, allowing more flexibility but reducing robustness when image data is complex. We describe a parametric representation of spot shape that is both general enough to represent unusual spots, and specific enough to introduce constraints on the interpretation of complex images. Our method uses a model of shape based on the statistics of an annotated training set. The model allows new spot shapes, belonging to the same statistical distribution as the training set, to be generated. To represent spot appearance we use the statistically derived shape convolved with a Gaussian kernel, simulating the diffusion process in spot formation. We show that the statistical model of spot appearance and shape is able to fit to image data more closely than the commonly used spot parameterizations based solely on Gaussian and diffusion models. We show that improvements in model fitting are gained without degrading the specificity of the representation.

Computer Simulation↗
PubMed1993

Contrasting clinical and statistical significance within the research setting.

When designing a clinical trial or study, the value of the following interrelated parameters should be determined prior to collecting data: clinical significance, statistical significance, power, and sample size. Too often, clinical importance and the other design issues are ignored and only statistical significance dictates the conclusions of the study. In order to evaluate the frequency that each of these design parameters is addressed in the published literature, the topic of pulmonary function tests (specifically forced vital capacity) was chosen, and all relevant articles for one year (1990) were identified using Minnesota MEDLINE. A total of 121 articles met the selection criteria and were reviewed. Of all the articles, 13.2% discussed clinical significance, 21.5% discussed sample size, and only 5.0% addressed statistical power. As expected, the majority of the articles (92.6%) discussed statistical significance (P values). None of the articles mentioned all four factors. When choosing the level of clinical significance several methods may be used. Such might be well established in certain clinical areas or available from previous publications and references or they may be attainable from pilot study data and, in the absence of any prior information, a clinician may use personal experience. To minimize subjectivity, the clinical effect-size can be based on the population distribution of the measurement of interest.

Clinical Trials as Topic↗
PubMed2005

General and targeted statistical potentials for protein-ligand interactions.

We present a novel atom-atom potential derived from a database of protein-ligand complexes. First, we clarify the similarities and differences between two statistical potentials described in the literature, PMF and Drugscore. We highlight shortcomings caused by an important factor unaccounted for in their reference states, and describe a new potential, which we name the Astex Statistical Potential (ASP). ASP's reference state considers the difference in exposure of protein atom types towards ligand binding sites. We show that this new potential predicts binding affinities with an accuracy similar to that of Goldscore and Chemscore. We investigate the influence of the choice of reference state by constructing two additional statistical potentials that differ from ASP only in this respect. The reference states in these two potentials are defined along the lines of Drugscore and PMF. In docking experiments, the potential using the new reference state proposed for ASP gives better success rates than when these literature reference states were used; a success rate similar to the established scoring functions Goldscore and Chemscore is achieved with ASP. This is the case both for a large, general validation set of protein-ligand structures and for small test sets of actives against four pharmaceutically relevant targets. Virtual screening experiments for these targets show less discrimination between the different reference states in terms of enrichment. In addition, we describe how statistical potentials can be used in the construction of targeted scoring functions. Examples are given for cdk2, using four different targeted scoring functions, biased towards increasingly large target-specific databases. Using these targeted scoring functions, docking success rates as well as enrichments are significantly better than for the general ASP scoring function. Results improve with the number of structures used in the construction of the target scoring functions, thus illustrating that these targeted ASP potentials can be continuously improved as new structural data become available.

Binding Sites↗
PubMed2006

RT-PSM, a real-time program for peptide-spectrum matching with statistical significance.

The analysis of complex biological peptide mixtures by tandem mass spectrometry (MS/MS) produces a huge body of collision-induced dissociation (CID) MS/MS spectra. Several methods have been developed for identifying peptide-spectrum matches (PSMs) by assigning MS/MS spectra to peptides in a database. However, most of these methods either do not give the statistical significance of PSMs (e.g., SEQUEST) or employ time-consuming computational methods to estimate the statistical significance (e.g., PeptideProphet). In this paper, we describe a new algorithm, RT-PSM, which can be used to identify PSMs and estimate their accuracy statistically in real time. RT-PSM first computes PSM scores between an MS/MS spectrum and a set of candidate peptides whose masses are within a preset tolerance of the MS/MS precursor ion mass. Then the computed PSM scores of all candidate peptides are employed to fit the expectation value distribution of the scores into a second-degree polynomial function in PSM score. The statistical significance of the best PSM is estimated by extrapolating the fitting polynomial function to the best PSM score. RT-PSM was tested on two pairs of MS/MS spectrum datasets and protein databases to investigate its performance. The MS/MS spectra were acquired using an ion trap mass spectrometer equipped with a nano-electrospray ionization source. The results show that RT-PSM has good sensitivity and specificity. Using a 55,577-entry protein database and running on a standard Pentium-4, 2.8-GHz CPU personal computer, RT-PSM can process peptide spectra on a sequential, one-by-one basis in 0.047 s on average, compared to more than 7 s per spectrum on average for Sequest and X!Tandem, in their current batch-mode processing implementations. RT-PSM is clearly shown to be fast enough for real-time PSM assignment of MS/MS spectra generated every 3 s or so by a 3D ion trap or by a QqTOF instrument.

Algorithms↗
PubMed2005

Two-part statistics with paired data.

In epidemiology, we often study data from a mixed distribution, i.e. with a clump of observations at zero and positive continuous data. Lachenbruch developed statistics for this kind of data for independent samples. These tests are the sum of one test for equality of proportions of zero values and one conditional test for the continuous distribution. This paper concerns the adaptation of these tests to paired samples. Like Lachenbruch, we developed two statistics, which tend to a two-degree-of-freedom chi2 distribution. These two-part statistics are the sum of McNemar's test for testing the equality of proportions of zero values, and the Wilcoxon signed-rank test or the paired Student's test for testing the equality of the distribution of positive values. We studied the behaviour of these tests for various proportions of zeros, and mean values of the continuous distribution. All tests are efficient when the smaller proportion of zero values corresponds to the population with the larger mean. In all other situations, the two-part statistics are superior to the others. These methods are applied to a matched case-control study of lower limb venous insufficiency.

Adult↗
PubMed2005

Statistical consideration of the strategy for demonstrating clinical evidence of effectiveness--one larger vs two smaller pivotal studies.

As a regulatory strategy, it is nowadays not uncommon to conduct one confirmatory pivotal clinical trial, instead of two, to demonstrate efficacy and safety in drug development. This paper is intended to investigate the statistical foundation of such an approach. The one-study approach is compared with the conventional two-study approach in terms of power, type-I error, and fundamental statistical assumptions. Necessary requirements for a single-study model is provided in order to maintain equivalent evidence as that from a two-study model. In general, one-study model is valid only under a 'one population' assumption. In addition, higher data quality and more convincing and robust results need to be demonstrated in such cases. However, when 'one-population' assumption is valid and appropriate methods are selected, a one-study model can have a better power using the same sample size. The paper also investigates statistical assumptions and methods for making an overall inference when a two-study model has been used. The methods for integrated analysis are evaluated. It is important for statisticians to select correct pooling strategy based on the project objective and statistical hypothesis.

Clinical Trials, Phase III as Topic↗
PubMed2006

Properties of R(2) statistics for logistic regression.

Various R(2) statistics have been proposed for logistic regression to quantify the extent to which the binary response can be predicted by a given logistic regression model and covariates. We study the asymptotic properties of three popular variance-based R(2) statistics. We find that two variance-based R(2) statistics, the sum of squares and the squared Pearson correlation, have identical asymptotic distribution whereas the third one, Gini's concentration measure, has a different asymptotic behaviour and may overstate the predictivity of the model and covariates when the model is mis-specified. Our result not only provides a theoretical basis for the findings in previous empirical and numerical work, but also leads to asymptotic confidence intervals. Statistical variability can then be taken into account when assessing the predictive value of a logistic regression model.

Biometry↗