PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “Factor Analysis, Statistical”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 37 records · Page 2Linked to original sources
PubMed1991

The analysis of categorical data from cross-over trials using a latent variable model.

A latent variable model for categorical data is described. The nuisance parameters in the model can be eliminated from the analysis through the use of a conditional likelihood and this leads to an analysis based on a log-linear model. It is shown how the structure of a cross-over trial allows considerable simplification in the formulation of this model and routine application of well-known statistical packages. The method is illustrated by data from a three-period cross-over trial on the relief of primary dysmenorrhea using the GLIM and SAS packages.

Analysis of Variance↗
PubMed1991

[Risk factors and survival prognosis in lymphogranulomatosis: a statistical analysis].

To examine risk factors and to design prognostic models for the total, relapse-free and after-relapse survival, use was made of the Cox model for unifactorial and multidimensional analysis based on the retrospective estimation of the results of the treatment of 235 patients with lymphogranulomatosis, stages II-III. Radiotherapy according to the radical program was carried out in conjunction with polychemotherapy according to the COPP scheme. The number of areas of damage, massiveness of injury to the mediastinum, involvement of the spleen into the process appeared the most informative risk factors for predicting both total and relapse-free survival. The prediction of the after-relapse survival was considerably influenced by the age combined with carrying out of not less than 3 cycles of polychemotherapy given to the patients during initial treatment or with the presence of intoxication symptoms and increased ESR. The tables of the survival probability were calculated.

Antineoplastic Combined Chemotherapy Protocols↗
PubMed1991

[A statistical investigation of the influence of allergic factors on intractable asthma by multiple factor analysis].

The relationships of the development of intractability in bronchial asthma with 11 factors, namely 1) sex, 2) age of onset, 3) duration of the disease, 4) severity of the disease, 5) disease type, 6) family history within the third degree of consanguinity, 7) history of smoking, 8) history of atopic dermatitis, 9) history of allergic rhinitis, 10) history of chronic sinusitis and 11) history of nasal polyp were studied by multiple factor analysis in 95 patients. The severity of the disease was shown to be the most important factor in whether the disease becomes intractable or not, followed by the age of onset. The history of atopic dermatitis had the least influence, and the influences of the other factors were not markedly different from one another. Evaluation of each factor according to the category score suggested that severe or moderate non-atopic bronchial asthma in males with a positive family history and positive histories of smoking, chronic sinusitis, nasal polyp and atopic dermatitis within a short period after the onset in the second decade or fifth decade or later tend to become intractable.

Adolescent↗
PubMed1991

Measuring contraceptive values: an alternative approach.

Previous assessments of individuals' values for various contraceptive consequences have employed one of four methodologies: free elicitation, direct ratings, multiple regression, or factor analysis. All four methodologies are flawed because they produce group rather than individual values, relying on rating scales, and fail to incorporate information regarding consequence trade-offs. Axiomatic conjoint measurement is proposed as an alternative methodology and used to determine individuals' values for a selected set of contraceptive consequences at two stages of the family-planning career.

Attitude↗
PubMed1976

[Genetico-statistical analysis of multiple birth factors in man. III. Component analysis of multiple birth factors].

Based on the component analysis of correlation matrices for five indicies (parents' age by the moment of twin birth, the number of proband's pregnancy, the beginning of mother's coitus and menarche) characterizing four specified family groups (MZ and DZ load multiple birth and MZ and DZ sporadic multiple birth) at least 6 independent trends of multiple birth factors influence have been revealed. Gases of burdened and sporadic multiple birth turned out to differ in some of the specific trends. At the same time while cases of MZ and DZ load multiple birth differ in some of the trends, they appear to be much more similar to one another than those of MZ and DZ sporadic multiple birth. The data obtained make it possible to assert that first, according to the mechanism of the appearance, there exist at least four main multiple birth groups, specified above, and second, there are multiple birth factors being both common to all groups and specific to each of them separately. At this some of the defined factors influence mainly as implementing hereditary determined trend to multiple burth, while the others in the absence of predisposing genotype perform apparently as casual. The present study data confirm the concept, formed on the basis of genealogical analysis, that genetical factors involved in the determination of MZ and DZ multiple birth are of definitely common character. In addition to that the results of the study make it possible to conclude that multiple aproach in the statistical analysis of quantitative characteristics (multiple birth factors in the present case) is extremely sensitive and results in non-routine conclusions the effects of which can be directly changed.

Age Factors↗
PubMed1992

Potential language and attentional networks revealed through factor analysis of rCBF data measured with SPECT.

We used changes in regional cerebral blood flow (rCBF) to disclose regions involved in central auditory and language processing in the normal brain. rCBF was quantified with a fast-rotating, single-photon emission computerized tomograph (SPECT) and inhalation of 133Xe. rCBF data were obtained simultaneously from parallel, transverse slices of the brain. The lower slice was positioned to include both Broca's and Wernicke's areas. The upper slice included regions generally regarded by neurobehaviorists as less related to primary auditory or linguistic functions. We presented three types of auditory stimuli to ten healthy, young volunteers: (a) diotically presented Danish speech, (b) dichotic word stimulation, and (c) white noise. Wilcoxon's signed ranks sum test revealed increased rCBF in language-related areas of cortex, viz., Wernicke's area and its right-sided homologous area as well as in Broca's area (left hemisphere), when subjects listened to narrative speech, compared to white noise (baseline). No significant rCBF differences were detected with this test during dichotic stimulation vs. white noise. A more sophisticated statistical method (factor analysis) disclosed patterns of functionally intercorrelated regions. The factor analysis reduced the highly intercorrelated rCBF measures from 28 regions of interest to a set of three independent factors. These factors accounted for 77% of the total variation in rCBF values. These three factors appeared to represent statistical analogues of independent brain networks involved in (I) auditory/linguistic, (II) attentional, and (III) visual imaging activity.

Acoustic Stimulation↗
PubMed1975

[Application of correspondance factor analysis to function testing of the thyroid].

Biological data concerning 249 thyroid patients have been treated by using correspondance factor analysis. This statistical treatment has enabled us to compare 2 types of associated tests (in vitro, in vivo) and to evaluate their respective contribution to the diagnosis of different thyroid diseases. There appeared also certain possibilities for the use of such a method in the future: determination of a biological profile for thyroid disease and for Grave's disease patients in course of treatment.

Binding, Competitive↗
PubMed1976

Statistical analysis of the different factors which could affect postextrasystolic potentiation in the human heart.

The effect of different coupling indices and intervals that could theoretically affect postextrasystolic potentiation has been investigated. A total of 150 ventricular premature beats corresponding to 20 patients submitted to routine cardiac catheterization were studied. Only single ventricular premature contractions following at least four regular sinus beats were considered. Percentage changes in left ventricular systolic pressure, end-diastolic pressure, and max dp/dt were correlated against seven indices and intervals. Index 2 (coupling interval/coupling interval + postextrasystolic pause) gave the better correlations. Besides, this Index includes two intervals that were demonstrated to have statistical significance when individually considered. It has been proved that in the first postextrasystolic beat the highest values of max dt/dt, or left ventricular systolic pressure occurred in early ventricular premature beats, giving a negative regression with Index 2, while in the second postextrasystolic beat the highest values of max dp/dt and left ventricular systolic pressure corresponded to late prematuring beats, giving therefore positive regressions with Index 2 (slope inversion phenomenon). The third and fourth postextrasystolic beats had similar positive regressions but with progressively smaller slopes. Correlations between left ventricular end-diastolic pressure and Index 2 were very poor. It is suggested that variations in baroreceptor activity could account for the different forms of potentiation observed in early and late extrasystoles. In five cases, there were no consistent differences in potentiation when premature beats were elicited from either right or left ventricles.

Adult↗
PubMed1979

Multivariate analysis of cancer mortalities for selected sites in 24 countries.

In order to analyze the pattern in the geographical distribution of cancer death in 24 countries of the world, correlation coefficients were calculated between pairs of mortality rates of different cancer sites, using the data for 13 sites in males and 14 sites in females over 18 years from 1950 to 1967. Then factor analysis by means of varimax method was performed on 13 x 13 correlation matrix for males, 14 x 14 correlation matrix for females and 27 x 27 correlation matrix for males and females combined. As a result of factor analysis, three factors are extracted, which are commonly recognized in both males and females. The first factor has high positive factor loadings on pancreas, prostate (for males), skin, and intestine cancers, and negative loadings on stomach and liver cancers. The second factor has high positive factor loading on rectum, intestine, and lung cancers, and the third factor on larynx, oral, and esophagus cancers. Factor analysis based on 27 x 27 correlation matrix revealed that the third factor of both sexes are heterogeneous with regards to the distributions of the factor score. In order that we may find some clues to develop an etiological hypothesis for each site of cancer, we obtained the correlation coefficient between the scores of the extracted factors and the variables on food and environmental agent, and performed stepwise regression methods as well. One of the most striking results we obtained was that excessive drinking of alcohol and the lack of appropriate intake of fruit are suspected as etiological promoters in the pathogenesis of oral, esophagus, and larynx cancers in males.

Adult↗
PubMedDate not supplied

A genetic factor model for the statistical analysis of multilocus DNA fingerprints.

A novel concept is described for the statistical analysis of multilocus DNA fingerprints. Utilizing this method, it is shown by simulation that the application of multilocus DNA fingerprints to paternity testing is robust against deviations from idealistic assumptions made about underlying models and parameters. Partial homozygosity, allelism and linkage at the DNA loci involved, as well as variations in estimates of band-sharing probabilities were studied for effects on the resulting paternity probabilities. None of the above-mentioned phenomena appear to change these values to an extent relevant for decision making in paternity cases.

Alleles↗
PubMed1992

Statistical analysis of clinical risk factors for coronary artery spasm: identification of the most important determinant.

Coronary artery spasm plays an important role in acute ischemic events, and it has a close relationship with coronary atherosclerosis. Thus we attempted to determine the most significant risk factor for coronary artery spasm. Among 3000 consecutive patients who underwent coronary cineangiography with ergonovine maleate testing, 330 with typical angina pectoris (group 1) and 294 with old myocardial infarction (group 2) were studied. We divided each group into three or four subgroups according to the presence of fixed organic stenosis (FOS+) or a positive reaction to ergonovine maleate (coronary artery spasm [CAS]+). We examined the relationship between coronary artery spasm and eight coronary risk factors: age, sex, hypertension, diabetes mellitus, smoking, and serum cholesterol, uric acid, and high-density lipoprotein cholesterol levels. The proportion of smokers in the subgroups with CAS(+) was significantly higher than in the subgroups with CAS(-)(p less than 0.01). There was no correlation between smoking and fixed organic stenosis. According to the results of multiple regression analysis, there was a positive correlation between smoking and CAS(+) and between serum high-density lipoprotein cholesterol levels and CAS(+)(p less than 0.01). Thus we concluded that smoking is the most significant risk factor in discriminating between patients with and without coronary artery spasm.

Cholesterol, HDL↗
PubMed1992

Structural equation models in medical research.

Structural equation modelling (SEM) is a modern statistical method that allows one to evaluate causal hypotheses on a set of intercorrelated nonexperimental data. The sample variances and covariances, and possibly the means, are compared to those predicted by a theory-based hypothetical model after optimal estimation of the parameters of the model. The goodness-of-fit of the empirical data to the hypothesized model is evaluated statistically. This review describes the underlying statistical theory and rationale of SEM. Both confirmatory factor analysis and latent variable path models are discussed. The applicability of SEM to assessment of reliability and validity is noted. A detailed example is provided, and several examples from the medical literature are briefly reviewed. Cautions regarding the possible misuse or misinterpretation of the technique are also mentioned. Possible future directions for the use of SEM in medical research are suggested. Two appendices provide more technical details.

Factor Analysis, Statistical↗
PubMedDate not supplied

[A clinico-statistical analysis of the risk factors for the occurrence of maxillodental anomalies in children].

The informative value of maxillodental abnormality risk factors was assessed in 421 children with deciduous occlusion in bits (information units). Sixteen real risk factors were distinguished, whose informative value depended on the type of abnormality. The factors that rank first are as follows: finger sucking, untimely loss of two or more molars and premolars, oral respiration, dummy sucking for more than one year, labor complications. The author claims that study of the informative value of risk factors is conducive to a differentiated approach to elimination of these factors with due consideration for the type of the abnormality.

Child, Preschool↗