Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Multivariate 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 451 records · Page 25Linked to original sources
PubMed2006

Testing statistical significance of multivariate time series analysis techniques for epileptic seizure prediction.

Nonlinear time series analysis techniques have been proposed to detect changes in the electroencephalography dynamics prior to epileptic seizures. Their applicability in practice to predict seizure onsets is hampered by the present lack of generally accepted standards to assess their performance. We propose an analytic approach to judge the prediction performance of multivariate seizure prediction methods. Statistical tests are introduced to assess patient individual results, taking into account that prediction methods are applied to multiple time series and several seizures. Their performance is illustrated utilizing a bivariate seizure prediction method based on synchronization theory.

Algorithms↗
PubMed1976

On the classification of antidepressant drugs.

Clinical, pharmacologic, and biochemical profiles of antidepressant drugs have been analyzed statistically. A method derived from multivariate statistics separates mere drug potency from the spectral information contained in the profiles. The dimensionality of the spectra is also reduced and this results in a diagram of associations and dissociations between the antidepressants and their scales of observation. The spectra of antidepressants show three poles which have been labeled as D (desipraminelike), A (amitriptylinelike), and M (MAOI). Clinical spectra agree better between one another than various pharmacologic spectra do. Monoamine oxidase inhibitors are readily differentiated from the tricyclic compounds. The latter appear in the sequence: desipramine, nortriptyline, imipramine, and amitriptyline. This sequence can also be reproduced from biochemical spectra of central and peripheral blockade of norepinephrine and serotonin reuptake.

Affect↗
PubMedDate not supplied

[Non-Q infarct. Clinical and prognostic aspects].

Prognosis of on-Q wave myocardial infarction (nQMI) has been the subject of considerable controversy over the last few years and a systematically aggressive approach with PTCA or coronary by-pass surgery (CBS) has been advocated as a means to reduce subsequent coronary events rate. To investigate clinical outcome and possibly identify high-risk subgroups 131 consecutive patients (pts) meeting diagnostic criteria for nQMI, admitted to our CCU for their first myocardial infarction between January 1980 and June 1985, were followed-up for a mean period of 34 months (range 6-72). Mean age of pts was 59 +/- 7 yrs; 101 (76%) were males, 30 (24%) females. No pt was lost to follow-up. During the same period 684 pts were admitted for Q wave myocardial infarction. Major coronary events such as angina, reinfarction, CBS, cardiac death as well as overall early and late mortality were considered for statistical evaluation with uni-multivariate analysis taking into account multiple data from pts history and acute clinical presentation. Angina appeared or recurred in 71 pts (54%), reinfarction occurred in 14 (10.7%); 25 pts underwent CBC (19%). Early cardiac deaths were 7 (5.3%), late cardiac deaths 12 (9.2%); overall mortality rate 18.1%. Uni- and multivariate statistical analysis did not disclose significant criteria predicting major coronary events and no high-risk subgroup could be identified, confirming uncertainties and doubts from literature.

Aged↗
PubMed1998

Prognostic significance of p53, angiogenesis, and other conventional features in operable breast cancer: subanalysis in node-positive and node-negative patients.

To validate the prognostic value of the determination of p53 expression, intratumoral microvessel density (IMD) (a measure of angiogenesis), and the conventional features, we studied 531 patients operated of breast cancer (271 node-positive and 260 node-negative), with a median follow-up exceeding 6 years. IMD was assessed by using the anti-CD31 antibody to identify the microvessels. p53, estrogen receptor (ER) and progesterone receptor (PgR) were determined by immunocytochemistry using the antibodies PAb1801, H-222 Sp2y and KD-68, respectively. The prognostic value of the markers was analyzed by univariate and multivariate statistical analyses. In the overall series p53 expression, IMD, nodal status, ER and PgR were statistically significant prognostic indicators for both relapse-free survival (RFS) and overall survival (OS) in the final multivariate model. Likewise, tumor size and menopausal status were significant prognostic indicators for RFS and OS, respectively. In the subgroup of node-negative patients who did not receive adjuvant therapy only p53, IMD, and tumor size were statistically significant in multivariate analysis. In the subgroup of node-positive patients treated with adjuvant chemotherapy, IMD, the number of involved nodes and PgR were statistically significant in multivariate analysis. In the subgroup of node-positive patients treated with adjuvant tamoxifen, IMD and ER (and the number of involved nodes, only for OS) were statistically significant for both RFS and OS in the final multivariate model. Different markers played a diverse prognostic role in the diverse subgroups studied. Angiogenesis was the sole marker which retained prognostic value in all the sub-groups analyzed. p53 retained significance only in the subgroup of node-negative patients, whilst ER and PgR were statistically significant in the subgroups of node-positive patients treated with adjuvant hormone therapy or chemotherapy, respectively.

Adult↗
PubMed1997

Theoretical calculation and prediction of Caco-2 cell permeability using MolSurf parametrization and PLS statistics.

PURPOSE: To statistically model the permeability across Caco-2 cell monolayers using theoretically computed molecular descriptors and multivariate statistics. METHODS: Seventeen structurally diverse compounds were investigated. The program MolSurf was used to compute theoretical molecular descriptors related to physico-chemical properties such as lipophilicity, polarity, polarizability and hydrogen bonding. The multivariate Partial Least Squares Projections to Latent Structures (PLS) method was used to delineate the relationship between the permeability across Caco-2 cell monolayers and the theoretically computed molecular descriptors. RESULTS: Excellent statistical models were derived. Properties associated with hydrogen bonding had the largest impact on diffusion through the monolayers and should be kept at a minimum to promote high permeability. High lipophilicity and the presence of surface electrons, i.e. valence electrons, which are not tightly bonded to the molecule, were also found to have a favorable influence to achieve high permeability. CONCLUSIONS: The results indicate that theoretically computed molecular MolSurf descriptors in conjunction with multivariate statistics of PLS type can be used to successfully model permeability across Caco-2 cell monolayers and, thus, differentiate drugs with poor permeability from those with acceptable permeability at an early stage of the preclinical drug discovery process.

Caco-2 Cells↗
PubMed1998

Theoretical calculation and prediction of brain-blood partitioning of organic solutes using MolSurf parametrization and PLS statistics.

Sixty-three structurally diverse compounds were investigated to statistically model the brain-blood partitioning of organic solutes using theoretically computed molecular descriptors and multivariate statistics. The program MolSurf was used to compute theoretical molecular descriptors related to physicochemical properties such as lipophilicity, polarity, polarizability, and hydrogen bonding. The multivariate Partial Least Squares Projections to Latent Structures (PLS) method was used to delineate the relationship between the brain-blood partitioning of organic solutes and the theoretically computed molecular descriptors. Good statistical models were derived. Properties associated with polarity and Lewis base strength had the largest impact on the blood-brain partitioning and should be kept to a minimum to promote high partitioning. The absence of atoms capable of hydrogen bonding interactions as well as high lipophilicity and the presence of polarizable surface electrons, i.e., valence electrons, were also found to promote high brain-blood partitioning. The results indicate that theoretically computed molecular MolSurf descriptors in conjunction with multivariate statistics of PLS type can be used to successfully model the brain-blood partitioning of organic solutes and hence differentiate drugs with poor partitioning from those with acceptable partitioning at an early stage of the preclinical drug-discovery process.

Blood-Brain Barrier↗
PubMed1993

The contribution of individual variables to Hotelling's T2, Wilks' lambda, and R2.

We examine the effect of each variable on the following statistics: the one-sample and two-sample Hotelling's T2, Wilks' lambda for multivariate analysis of variance, and R2 in multiple regression. For T2, the net effect of each variable is an increase in the multivariate statistic, and the particular factors determining the amount of increase are (i) the multiple correlation of the variable with all other variables, and (ii) how well the variable's contribution to falsifying the hypothesis can be linearly predicted from the other variables. The effect of each predictor variable on R2 is similar to the effect of each variable on T2. For Wilks' lambda, each variable induces a decrease, due to (i) the F for that variable alone, and (ii) the change in multiple correlation from within-sample to total-sample.

Analysis of Variance↗
PubMed1997

Identification of procaryotic developmental stages by statistical analyses of two-dimensional gel patterns.

Multivariate statistical comparisons of two-dimensional protein (2-D) gel patterns were used for the first time to define stages of a biological developmental system. The differentiating procaryote, Streptomyces coelicolor, was radiolabeled in liquid cultures at 16 intervals during development, and radioactive proteins were separated and quantified on 2-D gels. Cluster, principal component, and correlation analyses classified these gel patterns into four distinct groups, each reflecting a pattern of gene expression specific for a stage of development. These studies focused our attention on a phase of arrested growth as a key regulatory transition leading to secondary metabolism and a phase of renewed growth. Proteins whose synthesis was switched on or off during the "transitional" phase (some 21 and 18, respectively) were identified and will be the focus of future studies designed to identify their physiological or regulatory function.

Algorithms↗
PubMed2001

Statistical detection of between-group differences in event-related potentials.

OBJECTIVE: Many event-related potential (ERP) situations do not fulfill the multivariate statistics requirement of more cases than measurements. Five simulation studies were carried to select a sensitive test for comparing two independent groups of ERPs. METHODS: Simulated signal and noise waveforms were generated using different combinations of parameters: cases or replications per group, correlation between data points, number of points, signal-to-noise ratio (S/N), etc. The false alarm (FA) rate of each method was assessed and their sensitivity compared over sets of 2000 simulated experiments per condition. RESULTS: Study 1 identified the 'Projection onto Centroids Difference Vectors' (PCDV) method of Haig and Gordon (Brain Topogr 1995;8:67) as very good, but its FA rate was erratic under several conditions. The following studies served to shape the implementation parameters of a version of PCDV, termed PCDVp, that assesses significance through random permutations of the case labels. The final form is very sensitive. For instance, with two groups of 48 trials of 30-point EEG-like waveforms, its power for alpha=0.05 is about 50% at S/N 1.0 and 90% at S/N 1.5 (amplitude). CONCLUSION: PCDVp requires no a priori knowledge and is sensitive to detect differences between independent sets of waveforms, topographies or spatio-temporal data.

Event-Related Potentials, P300↗
PubMed1994

Multiple regression analysis of cytogenetic human data.

Biomonitoring studies on cytogenetic outcomes in humans should be considered as epidemiological studies, rather than randomized trials. Under this light the emphasis given to the achievement of a significant p-value should be reduced, since this measure suffers from major limitations. The use of a point estimate (and its corresponding confidence interval) to measure the association between exposure and effect offers several advantages, including the adjustment for confounding, and the evaluation of possible interaction between factors. In most instances the use of multivariate statistical methods allows an efficient analysis of these studies, even in presence of a small sample size and several covariates. In this paper we re-analyzed four biomonitoring studies by using multivariate methods to estimate relative risks through statistical modeling. The use of multiple regression techniques allowed the computation of point estimates of association and their confidence intervals for each covariate evaluated by the studies considered; the estimate of the effect of confounding variables such as smoking habits, age and gender; and the presence of interaction between covariates. Measures of association estimated through univariate and multivariate statistical approaches are compared. The advantages of the latter technique are discussed.

Adult↗
PubMed1998

The path analysis approach for the multivariate analysis of infant mortality data.

PURPOSE: This paper reviews the use of the Path Analysis (PA) methodology in health determinants modeling, with special reference to infant mortality modeling. METHODS: A review of the literature on PA applications in the modeling of infant mortality and similar problems is presented, together with a discussion of the conceptual basis of PA and its relation to other multivariate statistical techniques. Important aspects of the technique are discussed: 1) criteria for path formulation; 2) parameter estimation methods; 3) direct, indirect, spurious, and joint effects; and 4) goodness-of-fit and modification indices. RESULTS AND CONCLUSION: The review of the literature suggests that PA represents a methodological improvement regarding multivariate techniques used in modeling some health-related issues. PA allows investigation of more complex models, providing information that could have been previously overlooked, such as how the interrelations among independent variables in a model affect the dependent ones.

Humans↗
PubMed1987

An analysis of the arrangement of neurons in the cingulate cortex of schizophrenic patients.

A series of computer-assisted stereomorphometric analyses of the spatial arrangements of neurons and glia in postmortem cerebral cortex specimens has been developed and applied to both control subjects and schizophrenic patients. The data suggest that the anterior cingulate cortex of schizophrenic patients may contain domains or aggregates of neurons, particularly in layer II, which are smaller in size and separated by wider distances than those observed in the control group. Verification of the inferences made from the computer-generated data has been obtained by direct microscopic visualization and measurement of neuronal aggregates of layer II in Nissl-stained cingulate specimens from the control and schizophrenic groups. Statistical correction of the data, using multivariate statistics, for effects of age, hypoxia, postmortem interval, fixation, and neuroleptic exposure does not eliminate the differences in size and separation of neuronal aggregates in layer II of schizophrenic patients. The possible relevance of these findings to our understanding of schizophrenic symptomatology is discussed.

Aged↗
PubMed2001

Addition of radiation therapy to androgen ablation improves outcome for subclinically node-positive prostate cancer.

OBJECTIVES: To determine the outcome for node-positive prostate cancer treated by early androgen ablation with or without prostatic radiation. METHODS: Two hundred fifty-five men with lymphadenectomy-proven pelvic nodal metastases treated with early androgen ablation alone (n = 183) or with combined ablation and radiation (n = 72) between 1984 and 1998 were retrospectively reviewed for disease outcome and survival. Post-treatment disease status was based on the prostate-specific antigen levels or on the clinical and radiographic status for patients treated before 1987. Univariate and multivariate statistics were used to determine the prognostic factors and assess the influence of radiation treatment. RESULTS: With a median follow-up of 9.4 years, the 5, 10, and 13-year overall survival rate for those treated with early ablation alone was 83%, 46%, and 21%, respectively. The freedom from relapse or rising prostate-specific antigen rate for these patients was 41%, 25%, and 19% at 5, 10, and 13 years, respectively. Distant metastasis and local recurrence occurred with a 10-year actuarial incidence of 44% and 51%, respectively. With a median follow-up of 6.2 years, the 5 and 10-year overall survival rate for those treated with radiation and ablation was 92% and 67%, respectively. The freedom from relapse or rising prostate-specific antigen rate in these men was 91% and 80% at 5 and 10 years, respectively. The superior outcome for combined ablation and radiation was substantial and statistically significant in the univariate and multivariate analyses. CONCLUSIONS: Early androgen ablation alone has little curative potential for node-positive prostate cancer. The addition of prostatic radiation to ablation resulted in substantial and significant improvement in disease control and patient survival.

Adult↗
PubMed1998

Latent structure of EEG sleep variables in depressed and control subjects: descriptions and clinical correlates.

In this study, we aimed to determine the latent structure of multiple EEG sleep variables in patients with major depressive disorder (MDD) and in healthy control subjects and to examine associations between sleep factors and clinical variables. Subjects included 109 women with MDD and 54 healthy control women. EEG sleep data were collected prior to any treatment. Principal components analysis (PCA) was conducted on a set of 24 sleep variables. Separate PCAs were run for patients with MDD, control subjects, and a matched group of patients and controls. Other analyses included correlations, t-tests and MANOVA. Each PCA identified four sleep factors that explained 70% of the total variance in individual sleep variables: slow wave sleep, REM sleep, sleep continuity and REM latency/delta sleep ratio (RL/DSR). Patients with MDD and healthy controls differed on the mean value of the sleep continuity factor, and a multivariate analysis of variance based on the PCA identified MDD-control differences in REM sleep and sleep continuity. In the MDD group, slow wave sleep correlated inversely with age and personality disorder symptoms; sleep continuity correlated with subjective sleep quality and anxiety; and RL/DSR correlated inversely with age. The mean value of the REM factor was higher among treatment non-responders than responders. EEG sleep variables have a similar latent structure in women with MDD and in healthy controls. These sleep factors are supported conceptually and empirically, and correlate with clinical measures in women with MDD. Multivariate statistical techniques decrease the risk of Type I and Type II errors when using a large number of collinear sleep measures, and can clarify conceptual issues related to sleep and depression.

Adult↗
PubMed2007

A tale of two matrices: multivariate approaches in evolutionary biology.

Two symmetric matrices underlie our understanding of microevolutionary change. The first is the matrix of nonlinear selection gradients (gamma) which describes the individual fitness surface. The second is the genetic variance-covariance matrix (G) that influences the multivariate response to selection. A common approach to the empirical analysis of these matrices is the element-by-element testing of significance, and subsequent biological interpretation of pattern based on these univariate and bivariate parameters. Here, I show why this approach is likely to misrepresent the genetic basis of quantitative traits, and the selection acting on them in many cases. Diagonalization of square matrices is a fundamental aspect of many of the multivariate statistical techniques used by biologists. Applying this, and other related approaches, to the analysis of the structure of gamma and G matrices, gives greater insight into the form and strength of nonlinear selection, and the availability of genetic variance for multiple traits.

Biological Evolution↗
PubMedDate not supplied

An empirical investigation of pressure ulcer risk factors.

Despite many improvements in practice, pressure ulcers continue to be a source of concern to nurses and distress to patients. Clearly, there are no simple solutions to this problem and it is necessary to improve our understanding of pressure ulcer epidemiology and the effectiveness of preventive methods to continue to progress. Risk assessment is an important aspect of such work and forms the topic of this article. The authors first consider some of the difficulties associated with risk assessment and suggest that progress requires the use of multivariate statistical methods. They then describe the difficulty of comparing existing studies, which gives rise to the need for further work. A description of the study currently being undertaken follows, together with a presentation of the preliminary findings and a discussion of their implications for practice.

Aged↗
PubMed1994

[The short-term results and angiographic predictors of subacute thrombosis in the coronary implantation of the Palmaz-Schatz endoprosthesis].

INTRODUCTION AND OBJECTIVES: Intracoronary stenting has been proposed as an adjunct to balloon angioplasty in order to improve the immediate and long-term results. The purpose of this study was evaluate the short-term results, subacute closure rate and to try to identify angiographic predictors of subacute thrombotic after Palmaz-Schatz stent implantation. METHODS: Through a prospective registry, we have evaluated in 500 patients the safety and efficacy of Palmaz-Schatz stent implantation (580 in total) in coronary arteries and saphenous vein grafts. The identification of clinic and angiographic predictors of subacute closure have been evaluated with the assistance of a BMDP statistical software using an univariate and multivariate statistical analysis (logistic regression). The determination of diameter and stenosis has been achieved by electronic caliper. RESULTS: The stent was implanted successfully in 98.6% of the patients. There was no abrupt closure (< or = 1 day), however 36 patients (7.2%) developed subacute thrombotic closure (among 2nd-21st day after stenting). The major complications were: death 9 patients (1.8%), bypass surgery 7 patients (1.4%) and myocardial infarction 21 patients (4.2%). The predictors of subacute thrombotic closure through univariate statistical were: stenting for bail-out (S.T.: 27%; p < or = 0.0001), multiple stenting (S.T.: 24.1%; p < or = 0.0001), final diameter stent < or = 3.25 mm (S.T.: 12.6%; p < or = 0.013), and left ventricular ejection fraction < or = 45% (S.T.: 15.7%; p < or = 0.022). We showed with logistic regression that final diameter stent < or = 3.25 mm; p < or = 0.0030, left ventricular ejection fraction < or = 45%; p < or = 0.0012, stenting for bail-out; p < or = 0.0195 and multiple stenting; p < or = 0.0252, were predictors of subacute thrombotic closure. CONCLUSIONS: The Palmaz-Schatz coronary stenting will preferably realize in those arteries bigger than 3.25 mm and left ventricular ejection fraction > 45%, showing multiple stenting and stenting for bail-out greater subacute thrombotic closure rate.

Acute Disease↗
PubMed2005

Statistical power for analyses of changes in randomized controlled trials.

Randomized controlled trials (RCTs) are widely recommended as the most useful study design to generate reliable evidence and guidance to daily practices in medicine and dentistry. However, it is not well-known in dental research that different statistical methods of data analysis can yield substantial differences in study power. In this study, computer simulations are used to explore how using different univariate and multivariate statistical methods of analyzing change in continuous outcome variables affects study power, and the sample size required for RCTs. Results show that, in general, analysis of covariance (ANCOVA) yields greater power than other statistical methods in testing the superiority of one treatment over another, or in testing the equivalence between two treatments. Therefore, ANCOVA should be used in preference to change score or percentage change score to reduce type II error rates.

Analysis of Variance↗