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 289 records · Page 16Linked to original sources
PubMed1990

Lack of useful clinical predictors of response to splenectomy in patients with chronic idiopathic thrombocytopenic purpura.

We set out to identify clinical or analytical variables that might predict the response to splenectomy in patients with chronic idiopathic thrombocytopenic purpura (ITP). We retrospectively examined 138 mostly adult patients with chronic ITP, treated with splenectomy. Information was compiled from five Public Health Hospitals from a questionnaire and chart review. Sixty-one potentially prognostic variables were analysed by univariate and multivariate statistical methods. After multivariate analysis, age (relative risk = 1.02; CI 1-1.03) and a severity of the bleeding diathesis (relative risk = 1.6; CI 1.13-2.22) were independent prognostic factors for a sustained response to splenectomy. An independent analysis of the postsplenectomy counts showed that an early (days 1-3) thrombocyte count exceeding 156 x 10(9)/l cells increases the likelihood of a permanent unmaintained response. Our data indicate that the response to splenectomy in patients with chronic ITP cannot be adequately predicted on the basis of pre-splenectomy clinical or analytical variables.

Adult↗
PubMed2003

Hepatology and the Canadian gastroenterologist: interest, attitudes and patterns of practice: results of a national survey from the Canadian Association of Gastroenterology.

BACKGROUND: Hepatology has emerged as a subspecialty distinct from gastroenterology. Despite this, there is no formal certification examination or accredited training program, and training remains combined with gastroenterology. AIM: To determine attitudes, perceptions and patterns of practice with respect to liver disease among Canadian gastroenterologists. METHODS: A survey questionnaire was distributed to clinician gastroenterologists who are members of the Canadian Association of Gastroenterology. The responses of subgroups of respondents were compared by univariate and multivariate statistical techniques. RESULTS: Hepatologists constituted 20 of 201 respondents, the rest identifying themselves as gastroenterologists. Among gastroenterologists, liver disease constituted 10% of in- and out-patient practice. Despite this, 85% of gastroenterologists maintain an interest in hepatology, 49% perform liver biopsies, 60% treat hepatitis C, and 54% treat hepatitis B. In all of these areas, university-based gastroenterologists were consistently less likely than community-based gastroenterologists to maintain an interest and practice in hepatology, a finding that remained statistically significant on multivariate analysis. With regard to hepatology training, 90% of hepatologists and 94% of gastroenterologists felt that hepatology training should remain combined with gastroenterology, although 55% of hepatologists felt that current training was adequate compared with 79% of gastroenterologists, who were satisfied with the status quo. CONCLUSIONS: Hepatology remains relevant and important to Canadian gastroenterologists, especially those who have community-based practices. Canadian gastroenterologists and hepatologists are not in favour of separating hepatology training from existing gastroenterology training programs, although hepatologists feel that the current level of training is suboptimal.

Attitude of Health Personnel↗
PubMed1988

Simultaneous cleft lip and palate repair: an experimental study in beagles.

This study was designed to test the hypothesis that simultaneous lip and palate repair results in more severe craniofacial growth aberrations than lip repair or palate repair performed separately. Seventy-six purebred beagles were divided into five groups. Two of these groups were controls (unoperated and unrepaired animals); the three remaining groups were experimental (in one group only the lip was repaired, in another only the palate was repaired, and in the last the lip and palate were repaired simultaneously). Cephalometric measurements were analyzed using univariate and multivariate statistical techniques. In multivariate analysis, stepwise multiple regression and discrimination were applied to precisely assess the effects of the various surgical procedures. The results of this study indicate that simultaneous lip and palate repair results in more severe craniofacial growth aberrations than lip repair or palate repair performed separately.

Animals↗
PubMed1999

Multivariate quantitative structure-pharmacokinetic relationships (QSPKR) analysis of adenosine A1 receptor agonists in rat.

The aim of this study was to investigate the feasibility of a quantitative structure-pharmacokinetic relationships (QSPKR) method based on contemporary three-dimensional (3D) molecular characterization and multivariate statistical analysis. For this purpose, the programs SYBYL/CoMFA, GRID, and Pallas, in combination with the multivariate statistical technique principal component analysis were employed to generate a total of 16 descriptor variables for a series of 12 structurally related adenosine A1 receptor agonists. Subsequently, the multivariate regression method, partial least squares, was used to predict clearance (CL), volume of distribution (VdSS) and protein binding (fraction unbound, fU). The QSPKR models obtained could account for most of the variation in CL, VdSS, and fU (R2 = 0.82, 0.61 and 0.78, respectively). Cross-validation confirmed the predictive ability of the models (Q2 = 0.59, 0.41 and 0.62 for CL, VdSS, and fU, respectively). In conclusion, we have developed a multivariate 3D QSPKR model that could adequately predict overall pharmacokinetic behavior of adenosine A1 receptor agonists in rat. This methodology can also be used for other classes of compounds and may facilitate the further integration of QSPKR in drug discovery and preclinical development.

Animals↗
PubMed2002

Environmetric approaches to estimate pollution impacts on a coastal area by sediment and river water studies.

This paper represents an effort to demonstrate the opportunities of some environmetric methods like regression analysis, cluster analysis and principal components analysis. Their role for data modeling is stressed and the basic theoretical principles are given. The application of the multivariate statistical methods is illustrated by two major examples: Assessment of metal pollution based on multivariate statistical modeling of "hot spot" sediments from the Black Sea; and a trend study of Kamchia River water quality. In the first part of the study the environmetric approach makes it possible to separate three zones of the marine environment with different levels of pollution (Bourgas gulf, Varna gulf and lake buffer zone). Further, the extraction of four latent factors offers a specific interpretation of the possible pollution sources and separates the natural factors from the anthropogenic ones, the latter originating from contamination by chemical and steel-works and an oil refinery. In the second part of the study nine sampling sites along Kamchia River were considered as sources for water quality monitoring data. Trends for all parameters are calculated by the use of linear regression analysis and special attention is paid to a specific coastal site. Then five latent factors were extracted from the monitoring data set in order to gain information about some structural characteristics of the set.

Cluster Analysis↗
PubMed2002

A new statistical method for testing hypotheses of neuropsychological/MRI relationships in schizophrenia: partial least squares analysis.

We applied partial least squares (PLS) as a novel multivariate statistical technique to examine neuropsychological correlates of magnetic resonance imaging (MRI) measures of brain volumes in a well studied sample of 15 male patients with chronic schizophrenia. In the current study, because the total number of measures far surpassed the total number of subjects, extant multivariate techniques such as canonical correlation could not be used to examine relationships among simultaneous measures of MRI and neuropsychology. Moreover, because MRI measures were expected to be highly inter-correlated, as would be neuropsychological test scores, extant multivariate statistical techniques would be substantially limited because they typically assume statistical independence among sets of measures. PLS, on the other hand, proved to be especially well suited to examining the relationships among function and anatomy measures in this sample, where statistically significant relationships were demonstrated that were entirely consistent with prior studies using univariate correlation techniques. In particular, statistically significant relationships emerged among sets of MRI temporal lobe measures and neuropsychological tests of verbal memory and categorization as well as among MRI frontal measures and neuropsychological tests of working memory.

Adult↗
PubMed1997

[Pathological analysis and clinical importance of invasion and neck nodal metastasis in squamous cell carcinoma of the tongue].

To determine the absolute and relative of histopathological parameters and conventional indicators in correlation with invasion and nodal metastasis, we performed both univariate and multivariate statistical analysis in 92 patients with squamous cell carcinoma of the tongue by spss/pc+ statistical system. Mode of invasion (MI) and peritumoral lymphocytic infiltrating density (PLID) were significant and independent factors by multivariate analysis (P = 0.000, P = 0.009 respectively), although univariate analysis showed that MI, PLID, T-stage and Grading were significantly associated with nodal metastasis. A correlation exists only between T-stage and MI by multivariate statistical test (P = 0.045), while, PLID, T-stage and Grading were significant parameters by univariate analysis. A highly dense PLI around cancer nest was observed which blocked nest from tissue and capillary lymph vessels.

Adult↗
PubMed1996

Prognostic value of psychological testing in patients undergoing spinal cord stimulation: a prospective study.

OBJECTIVE: Associations between psychological and physical states are understood to exist, and the development of standardized psychological tests has allowed quantitative evaluation of this relationship. We tested whether associations exist between psychological test instruments and patients selected for therapeutic trials of spinal cord stimulation (SCS) for chronic, intractable pain. METHODS: Fifty-eight patients selected for SCS were tested prospectively with a battery of standardized psychological tests: Minnesota Multiphasic Personality Inventory with Wiggins content scales, Symptom Check List-90, and Derogatis Affects Balance Scale. Associations between treatment outcomes and preoperative test scores and clinical variables were tested by univariate and multivariate statistical analyses, in which the dependent variables were as follows: 1) the outcome of a therapeutic trial of stimulation (whether the patient derived sufficient reported pain relief with a temporary electrode to proceed with a permanent implant), and 2) long-term outcome of treatment with the permanent implant, as determined by disinterested third-party interview. RESULTS: Significant associations (P < or = 0.01) were observed between the outcome of the therapeutic trial of stimulation and psychological test results; patients with low "anxiety" scores on the Derogatis Affects Balance Scale and with high "organic symptoms" scores on the Wiggins test were significantly more likely to proceed to permanent implants, as determined by multivariate statistical models. There was an elevation in the Minnesota Multiphasic Personality Inventory hypochondriasis scale in these patients by univariate (P = 0.02), but not by multivariate, models. The multivariate model also identified young age, reproduction of leg pain by straight leg raising, and bilateral leg pain as favorable prognostic factors. The only association with favorable long-term outcome of implantation of a permanent device, by univariate analysis, was an elevated "joy" score on the Derogatis Affects Balance Scale. Multivariate analysis revealed no statistically significant predictors of long-term outcome. CONCLUSION: Because our study population was selected on the basis of recognized prognostic factors and long clinical experience, it may not be possible to generalize our findings to the overall pain clinic referral population. In the subpopulation we have chosen for SCS trials, psychological testing is of modest value and explains little of the observed variance in outcome. We find little evidence for selecting patients for SCS on the basis of psychological testing. Because self-reported outcome measures may themselves reflect the patient's psychological state, these findings should be considered carefully, in overall clinical context. A prospective study with additional objective outcome measures is underway, which will address some of these issues.

Adaptation, Psychological↗
PubMedDate not supplied

Estimating the accident potential of an Ontario driver.

To run a "demerit point" program, one uses routinely available information about drivers to identify those who are most likely to have an accident in the near future. On the basis of a four-year record for a large sample of Ontario drivers, we have examined several tools for the identification of such drivers and investigated how they perform. Each driver is thought to have an expected number of accidents, m. In a group of drivers with common traits (such as age, gender, record of convictions and accidents) the ms have a mean E(m) and a variance VAR(m). Estimates of E(m) and VAR(m) for all combinations of traits can be obtained within the framework of a multivariate statistical model. The same estimates can then be used to judge how well a model identifies drivers who have a large m. In such a multivariate model it is important to use data about previous accidents and convictions. However, the accuracy with which the m of a driver can be estimated is not improved much by distinguishing between offence type or between accidents as being "at fault" or "not at fault". Without much loss in estimation accuracy, one may attach a weight 1 to a conviction and 2 to an accident. Model performance is described in tangible terms: how many accidents are recorded by the drivers identified by a model, what proportion of identified drivers are "false positives," how many drivers with high m remain unidentified. We conclude that by using a multivariate statistical model one can do substantially better than by using a demerit point scheme in which points are assigned to offenses on the basis of their perceived seriousness. However, even when the best model is used to identify a large group of drivers, many will be false positives.

Accidents, Traffic↗
PubMedDate not supplied

Soluble metals in the atmosphere and their biological implications. A study to identify important aerosol components by statistical analysis of PIXE data.

Multivariate statistical analysis has been applied to time series measurements of aerosol elemental composition from PIXE analysis of filter samples, and principal components have been resolved that represent distinct particle types in an external mixture in the atmosphere. In this study, it is argued that a combination of chemical and statistical analyses of the data may be more powerful in determining chemical species in atmospheric aerosols than studied that employ mainly direct chemical analysis of chemical species in unresolved mixtures of aerosol particle samples. Sulfur is generally associated with mineral dust elements. It is reasoned that the association may represent sulfuric acid coatings on particles that can lead to mineral dissolution and solubilization of significant amounts of aluminum, iron, and other metals. Upon wet or dry deposition to the surface, the fluxes of these metals in biologically-available form may be sufficient to affect primary productivity in the world ocean and cause ecological damage in lakes. As a consequence, the fluxes of biogenic trace gases to the atmosphere may be changed, possibly leading to changes in the tropospheric concentration of ozone. The inputs to lakes of soluble aluminum, which is toxic to fish, may be partly by deposition directly from the atmosphere, thus not limited to leaching of soils by acid deposition. Human inhalation of soluble aluminum and other solubilized mineral metals may account, in part, for the observed geographic pattern of deaths attributed to chronic obstructive pulmonary disease (COPD) that show high rates in cities of the Western US and the southeast region, but low in most of the midwest and northeast.

Aerosols↗
PubMed1997

An electrophysiological technique for assessment of the development of spatial vision.

An objective visual evoked potential (VEP) technique was developed to evaluate spatial processing in humans over a wide range of ages. The constellation of stimulus conditions and analysis methods constitute a novel tool for the assessment of neural development. The key points that delineate this VEP technique are: (1) A brief, 6-s, swept-parameter stimulus with spatial frequency of square-wave grating patterns varied in octave steps, which facilitates correct accommodation and increases the likelihood of collecting uninterrupted, useable data; (2) data collection synchronous with stimulus presentation, which prevents contamination of the relevant frequency component from other frequency components in the response, thereby increasing the signal-to-noise ratio; (3) noise estimation at the response frequency of interest (second harmonic), based on a multivariate statistic (Tcirc2), which yields a realistic measure of signal-to-noise; (4) estimation of grating acuity from linear interpolation of the signal-to-noise measure; (5) monocular testing followed by multivariate statistical comparison of fellow eye data for each spatial frequency condition, which enables the determination of asymmetries within monocular neural pathways; (6) evaluation of maturation of the visual system based on vector-averaged amplitude and phase measures. Preliminary results indicate that reliable response functions are obtained from infants, children, and adults. Acuity estimates increased as a function of age. Phase values decreased consistently with increases in spatial frequency greater than 4 c/deg. Infants produced larger peak amplitude responses than did older observers, consistent with known developmental changes in cortical synaptic density. Phase data for the 2 c/deg condition provided additional evidence for the lack of maturity in the infant visual system as compared with that of older observers.

Adolescent↗
PubMed2001

Agricultural produces: synopsis of employed quality control methods for the authentication of foods and application of chemometrics for the classification of foods according to their variety or geographical origin.

A review of quality control methods and applications of multivariate statistical techniques on the authentication and classification of agricultural products is presented. The products reported within the frame of this article were vegetables, fruits, juices, jams, wines, cereals, bakery products, oils, tea, coffee, honey, sugar-syrups, salad dressings, and gums. The perspective of multivariate statistics as a promising tool to authenticate and classify these food products according to their geographical origin or variety was demonstrated. Several representative figures and informative synoptical tables for agricultural food products were provided both for the quality control methods employed and the multivariate analyses implemented.

Cluster Analysis↗
PubMed2001

Treating missing data in a clinical neuropsychological dataset--data imputation.

Missing data frequently reduce the applicability of clinically collected data in research requiring multivariate statistics. In data imputation, missing values are replaced by predicted values obtained from models based on auxiliary information. Our aim was to complete a clinical child neuropsychological data set containing 5.2% of missing observations. This was to be used in research requiring multivariate statistics. We compared four data imputation methods by artificially deleting some data. A real-donor imputation method which preserved the parameter estimates and which predicted the observed values with acceptable accuracy was used to complete the data set. In addressing the lack of studies with regard to treatment of missing data in neuropsychological data sets, this study presents information on the outcomes of applying data imputation methods to such data. The imputation modeling described can be applied to a variety of clinical neuropsychological data sets.

Child↗
PubMedDate not supplied

A multivariate strategy for prediction of psychoacoustic performance from electroacoustic characteristics of hearing aids.

Sentence length psychoacoustic material was processed through 16 hearing aids and presented to 220 listeners. Twenty-nine indices of electroacoustic performance were abstracted from a testing program under laboratory conditions. Through a multivariate statistical strategy, psychoacoustic achievement was shown to be positively correlated to and could be predicted by a knowledge of white noise gain, bandwidth below and above 1 kHz, frequency response regularity, and transiet distortion (multiple R = 0.77; F = 2.04; df = 5,10). It was concluded that a multivariate statistical tactic, when used with discretion, can be a useful and powerful tool to predict psychoacoustic performance from a knowledge of hearing aid electroacoustic characteristics.

Amplifiers, Electronic↗
PubMed1994

Combined survivorship and multivariate analyses of revisions in 799 hip prostheses. A 10- to 20-year review of mechanical loosening.

From 1970 to 1980 cemented metal-on-plastic total hip replacement was performed on 799 hips with primary osteoarthritis using one surgical technique. At the 10- to 20-year follow-up there had been 97 revisions for mechanical loosening. Univariate survivorship analysis showed that an increased risk of revision was associated with male gender, young age at primary THR, the Brunswik and Lubinus snap-fit prostheses with large femoral heads (as compared with the Charnley prosthesis), and varying experience of the surgeon. Multivariate statistical analysis showed a three-fold increased risk of revision for men (p < 0.0001), an increase in relative risk of 1.8 per 10 years younger at surgery (p < 0.0001), a fivefold increase in risk for the Brunswik prosthesis (p < 0.0001) and a twofold increase for the Lubinus prosthesis (p = 0.0067). Inexperience of the surgeon, however, was not validated as a risk factor. The study shows that the true risk factors for revision can be identified accurately by combining univariate survivorship and multivariate statistical analyses.

Aged↗
PubMed2002

An exploratory analysis of multiple mutation spectra.

The last decade has witnessed a remarkable increase in the number of mutations identified both in human disease-related genes and mutation reporter genes including those in mammalian cells and transgenic animals. This has led to the curation of a number of computerised databases, which make mutation data freely available for analysis. A primary interest of both the clinical researcher and the genetic toxicologist is determination of location and types of mutation within a gene of interest. Collections of mutation data observed for a disease-related gene or, for a gene exposed to a particular chemical, permits discovery of regions of sequence along the gene prone to mutagenesis and may provide clues to the origin of a mutation. The principal tool for visualising the distribution pattern of mutant data along a gene is the mutation spectrum: the distribution and frequency of mutations along a nucleotide sequence. In genetic toxicology, the current wealth of mutation data available allows us to construct many mutation spectra of interest to investigate the mutagenic mechanisms and mutational sites for one or a group of mutagens. Using the multivariate statistical methods principal components analysis (PCA) and cluster analysis (CA) we have tested the ability of these methods to establish the underlying patterns within and between 60 UV-induced, mitomycin C-induced and spontaneous mutations in the supF gene. The spectra were derived from human, monkey and mouse cells including both repair efficient and repair deficient cell lines. We demonstrate and support the successful application of multivariate statistical methods for exploring large sets of mutation spectra to reveal underlying patterns, groupings and similarities. The methods clearly demonstrate how different patterns of spontaneous and UV-induced supF mutation spectra can result from variation in plasmid, culture medium, species origin of cell line and whether mutations arose in vivo or in vitro.

Animals↗
PubMed1992

Crossover design in pharmacy research.

OBJECTIVE: Reports of pharmacy research using crossover designs were reviewed to determine if the studies adequately consider interaction effects and use appropriate statistical analyses. DATA SOURCES: All crossover studies published in DICP, The Annals of Pharmacotherapy during 1988 and 1989 were analyzed. STUDY SELECTION: Reports of crossover studies were included only if at least two treatments were applied in a different order to two or more groups of subjects. DATA EXTRACTION: The principal characteristics of crossover studies and the critical design variables were listed and each study analyzed according to these variables. The critical design variables included consideration of period, sequence, and carryover effects as well as the presentation of data by groups and the use of multivariate statistical analysis. The analysis was conducted independently by each author and conflicts were discussed until consensus was obtained. RESULTS: A total of 11 crossover studies were identified: 6 were bioavailability trials, 3 were treatment comparisons, and 2 had multiple objectives. The possibility of period, sequence, or carryover effects was less with bioavailability studies than with treatment comparisons. Only 1 study presented data by group and only 4 studies used multivariate analysis. CONCLUSIONS: The crossover design appears more appropriate for bioavailability trials than for treatment trials in pharmacy research. Analysis of data from crossover designs could be improved by presenting the data for each treatment group and using multivariate statistical analysis.

Analysis of Variance↗
PubMedDate not supplied

[Principles and methods of geometric morphometrics].

The basic concepts, notions and methods of geometric morphometrics (GM) are considered. This approach implies multivariate analysis of landmark coordinates located following certain rules on the surface of a morphological object. The aim of GM is to reveal differences between morphological objects by their shapes as such, the "size factor" being excluded. The GM is based on the concept of Kendall's space (KS) defined as a hypersphere with points distributed on its surface. These points are the shapes defined as aligned landmark configurations. KS is a non-Euclidian space, its metrics called Procrustes is defined by landmark configuration of a reference shape relative to which other shapes are aligned and compared. The differences among shapes are measured as Procrustes distances between respective points. For the linear methods of multivariate statistics to be applied to comparison of shapes, the respective points are projected onto the tangent plane (tangent space), the tangent point being defined by the reference. There are two principal methods of shape comparisons in GM: the Procrustes superimposition (a version of the least squares analysis) and thin-plate spline analysis. In the first case, Procrustes residuals are the outcome shape variables which remain after isometric alignment of the shapes being compared. Their summation over all landmarks yields Procrustes distances among these shapes. The Procrustes distances can be used in multivariate analyses just as the Euclidian distances. In the second case, the shapes are fitted to the references by stretching/compressing and shearing until complete identity of their landmark configurations. Eigenvectors of resulting bending energy matrix are defined as new shape variables, principal warps which yield another shape space with the origin defined by the reference. Projections of the shapes being compared onto principal warps yield partial warps, and their covariance matrix decomposition into eigenvectors yields relative warps which are similar to principal components (in particular, they are mutually orthogonal). Both partial and relative warps can be used in many multivariate statistic analyses as quantitative shape variables. Results of thin-plate spline analysis can be represented graphically by transformation grid which displays type, amount and localization of the shape differences. Basis rules of sample composition and landmark positioning to be used in GM are considered. At present, rigid (with minimal degrees of freedom) 2D morphological objects are most suitable for GM applications. It is important to recognize three type of real landmarks, and additionally semi-landmarks and "virtual" landmarks. Some procedures of thin-plate spline analysis are considered exemplified by some study cases, as well as applications of some standard multivariate methods to GM results. They make it possible to evaluate correlation between different shapes, as well as between a shape and some non-shape variables (linear measurements etc); to evaluate the differences among organisms by shape of a morphological structure; to identify landmarks which most accounted for both correlation and differences between the shapes. An annotated list of most popular softwares for GM is provided.

Animals↗