SEARCH · PubMed Health
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.
A new unified approach to solving and studying the factor analysis parameter estimation problem is proposed. The maximum likelihood and least squares formulations of factor analysis are considered. The approach leads to globally convergent procedures for simultaneous estimation of the factor analysis parameters. The method presented necessarily leads to proper factor analysis estimations.
AIM: The goals of this study were: (1) to analyse the underlying associations between coronary risk factors and the metabolic syndrome and (2) to evaluate the construct validity of the variables used to measure each factor. METHODS: The subjects were from a previously studied cohort of 284 middle-aged Caucasian males from Goteborg, Sweden, who were selected from the National Population Register. A confirmatory factor analysis was performed using EQS Multivariate Software Version 5.7b with maximum likelihood estimation. Hypertension, obesity, insulin resistance and hyperlipidaemia were the latent factors hypothesized. RESULTS: The final, four-factor model showed good fit, with significant intercorrelations noted between all factors. The highest correlations were noted between the insulin resistance factor and the obesity factor (r = 0.887) and the insulin resistance factor and the lipid factor (r = 0.835). All factors exhibited good values for construct reliability and variance extracted except for the insulin resistance factor, which was measured with the variables of fasting insulin and fasting glucose levels. CONCLUSIONS: A four-factor model of metabolic syndrome including the coronary heart disease risk factors of hypertension, obesity, insulin resistance and hyperlipidaemia was developed using this sample of European, middle-aged Caucasian males. Insulin resistance was not well defined using the variables of fasting insulin and fasting glucose levels. Other possible variables to include in the measurement of this factor are discussed.
This paper (the first in a series) applies new, empirical factor analysis methods to the problem of "banding" EEG power spectra. A measure is introduced for the comparison of factor analysis results (factor loading matrices). The measure, Ambient Matrix Coherence (AC) is "geometrically unbiased", and invariant of so-called "oblique rotations." AC is used in a "stability computation" to find the dimension of the stable factor analysis solution common to several subsets of a given dataset. If the factor analysis model is appropriate, then the correct number of factors is empirically determined in this way. Stability computations were first performed on various simulated datasets to establish the robustness and efficacy of this method (for various noise levels). These techniques were then applied to EEG power spectra datasets for each of 8 leads. Comparison of these results indicated 3 stable factors in common to all 8 leads, an additional less stable factor in common to 5 leads, and weak stability for the six-dimensional solution for one lead.
We investigated possible influence of 17 allergy-associated factors on atopic dermatitis and allergic rhinitis using Multiple factor analysis in 150 asthmatic children. Atopic dermatitis was complicated in ninety-seven cases and allergic rhinitis in ninety-seven cases. 17 allergy-associated factors were as follows: 1) sex, 2) age, 3) onset age of asthma, 4) family history of allergy, 5) peripheral eosinophil counts, 6) IgE RIST, 7) IgE RAST score to egg white, 8) IgE RAST score to milk, 9) IgE RAST score to soybean, 10) IgG4 antibody titers to egg white, 11) IgG4 antibody titers to milk, 12) IgG4 antibody titers to soybean, 13) IgE RAST score to house dust, 14) IgE RAST score to Dermatophagoides farinae, 15) severity of asthma, 16) exercise-induced asthma, 17) atopic dermatitis or allergic rhinitis. We concluded as follows: 1) Factors which more strongly influenced both atopic dermatitis and allergic rhinitis were IgE RAST score to D.f., positive family history of allergy, IgE RIST and eosinophil counts. 2) Combination with high levels of IgG4 antibody to 3 food allergens such as egg-white, milk and soybean and IgE RAST to egg-white has a strong influence on atopic dermatitis, but high levels of IgG4 antibody to 3 food allergens except high level of IgG4 antibody to soybean have a weak influence on allergic rhinitis.
The purpose of the present study was to examine sequentially the evaluations of the self and others in psychotherapy using the Rating Grid Method, one of the Repertory Grid Techniques. A construct system of Subject A was composed of "pleasant-painful", "steady-hopeless", "stable-anxious", "comfortable-irritable", "feeling of freedom-feeling of pressure", "specific-vague", "strong-week", and "refreshing-gloomy". That of Subject B consisted of "fearful-not fearful", "composed-restless", "pleasant-painful", "dignified-undignified", "eager-lazy", "realistic-mysterious", "perfectionism-slipshod", and "decisive-irresolute". The evaluations of the two subjects in the Rating Grid Method were analyzed using three-mode factor analysis. These results were then compared with diagnoses obtained through psychotherapy.
The variations of lymphocyte subsets have been studied both in neoplastic and chronic inflammatory elderly patients compared to a control group. The interpretation of the Multivariate Analysis of Variance (MANOVA) and of the Factorial Analysis has demonstrated the opposite role of TCD4 subset, mainly involved in the inflammatory process, and of TCD8 subset in regard of the neoplastic ones. We report the slight and yet significant increase of NK related to age. Independently from the disease pattern, factorial analysis permitted a comparison between the variations of lymphocyte subsets and the different grades of immunoresponse.
The well-known problem of fitting the exploratory factor analysis model is reconsidered where the usual least squares goodness-of-fit function is replaced by a more resistant discrepancy measure, based on a smooth approximation of the lI norm. Fitting the factor analysis model to the sample correlation matrix is a complex matrix optimization problem which requires the structure preservation of the unknown parameters (e.g. positive definiteness). The projected gradient approach is a natural way of solving such data matching problems as especially designed to follow the geometry of the model parameters. Two reparameterizations of the factor analysis model are considered. The approach leads to globally convergent procedures for simultaneous estimation of the factor analysis matrix parameters. Numerical examples illustrate the algorithms and factor analysis solutions.
OBJECTIVE: The aim of this study was to test the goodness-of-fit of all previously published five-factor models of the Positive and Negative Syndrome Scale (PANSS). METHODS: We used confirmatory factor analysis (CFA) with a large data set (N = 5769). RESULTS: The different subsamples were tested for heterogeneity and were found to be homogeneous. This indicates that despite variability in age, sex, duration of illness, admission status, etc., in the different subsamples, the structure of symptoms is the same for all patients with schizophrenia. Although previous research has shown that a five-factor model fits the data better than models with three or four factors, no satisfactory fit for any of the 25 published five-factor models was found with CFA. CONCLUSIONS: Variability in age, sex, admission status and duration of illness has no substantial effect on the structure of symptoms in schizophrenia. The lack of fit can be caused by ill-defined items that aim to measure several properties in a single rating. Another explanation is that well-defined symptoms can have two or more causes. Then a double or triple loading item should not be discarded, but included because the complexity of symptoms in schizophrenia is represented by these multiple loadings. Such a complex model not only needs confirmation by CFA, but also has to be proven stable. A 10-fold cross-validation is suggested to develop a complex and stable model.
A factor analysis was performed on Philips & Hunter's (1981) Pain Behaviour Checklist for headache sufferers. Three intuitively meaningful factors emerged. All were similarly associated with overall intensity; pain severity does not determine type of pain behaviour. Differences in pain behaviour emerged between migraine and tension headache groups.
OBJECTIVE: To explore the significance of the combination of factor analysis and systematic cluster analysis in classification of traditional Chinese medical syndromes in patients with posthepatitic cirrhosis, and to provide a scientific basis for the criterion of the classification. METHODS: We designed a clinical questionnaire according to the clinical characteristics and the demands of traditional Chinese medical information collection for patients with posthepatitic cirrhosis. By means of clinical epidemiological research, with the four diagnosis methods for clinical information collection of traditional Chinese medicine, symptoms, physical signs, tongue conditions and pulse conditions in 310 patients with posthepatitic cirrhosis were collected, and the characteristics of traditional Chinese medical syndromes in these patients were explored with statistical methods, such as factor analysis, varimax and systematic cluster analysis. RESULTS: Analyzed by factor analysis and systematic cluster analysis with SPSS 11.0, the traditional Chinese medical syndromes in 287 of the 310 cases (92.58%) of posthepatitic cirrhosis could be classified. The syndromes could be divided into 7 categories, which were internal accumulation of damp-heat (55 cases), insufficiency of the spleen with overabundance of dampness (74 cases), accumulation of blood stasis plus deficiency of liver-yin and kidney-yin (73 cases), accumulation of blood stasis plus deficiency of both blood and qi (40 cases), deficiency of both blood and qi (16 cases), deficiency of yin and blood heat (6 cases) and stagnation of the liver-qi and deficiency of the spleen (23 cases). The traditional Chinese medical syndromes in the other 23 cases could not be classified. CONCLUSIONS: The clinical information collected with the four diagnostic methods of traditional Chinese medicine can be classified into different categories with the factor analysis and systematic cluster analysis. The factor analysis and systematic cluster analysis can reveal the characteristics and regularity of traditional Chinese medical syndromes in patients with posthepatitic cirrhosis in a way, and have value in researching the syndromes of traditional Chinese medicine.
The multivariate genetic factor analysis is used, as first attempt, to study genetic bases of correlation variability of neurodynamic and psychodynamic levels of individual organization among isolates of Daghestan. Closer similarity between factors described in templates of phenotypic correlations is explained by lower heritability of the parameters under study. Interpopulation differences revealed by the multivariate genetic analysis are the result of differences in the genetic structure of the populations.
Factor analysis, a multivariate correlation technique, has been used to provide insight into the underlying structure of metabolic syndrome, which is characterized by physiological complexity and strong statistical intercorrelation among its key variables. The majority of previous factor analyses, however, have used only surrogate measures of insulin sensitivity. In addition, few have included members of multiple ethnic groups, and only one has presented results separately for subjects with impaired glucose tolerance. The objective of this study was to investigate, using factor analysis, the clustering of physiologic variables using data from 1,087 nondiabetic participants in the Insulin Resistance Atherosclerosis Study (IRAS). This study includes information on the directly measured insulin sensitivity index (S(I)) from intravenous glucose tolerance testing among African-American, Hispanic, and non-Hispanic white subjects aged 40-69 years at various stages of glucose tolerance. Principal factor analysis identified two factors that explained 28 and 9% of the variance in the dataset, respectively. These factors were interpreted as 1) a " metabolic" factor, with positive loadings of BMI, waist, fasting and 2-h glucose, and triglyceride and inverse loadings of log(S(I)+1) and HDL; and 2) a "blood pressure" factor, with positive loadings of systolic and diastolic blood pressure. The results were unchanged when surrogate measures of insulin resistance were used in place of log(S(I)+1). In addition, the results were similar within strata of sex, glucose tolerance status, and ethnicity. In conclusion, factor analysis identified two underlying factors among a group of metabolic syndrome variables in this dataset. Analyses using surrogate measures of insulin resistance suggested that these variables provide adequate information to explore the underlying intercorrelational structure of metabolic syndrome. Additional clarification of the physiologic characteristics of metabolic syndrome is required as individuals with this condition are increasingly being considered candidates for behavioral and pharmacologic intervention.
Factor analysis of dynamic scintigraphic studies has been proposed for a variety of clinical applications. This method also called FADS (Factor Analysis of Dynamic Structures) enables spatial separation of complex images into discrete factors according to their time/activity characteristics. FADS, which does not require a priori formulation of a compartmental model of tracer kinetics, is particularly suitable for the evaluation of new radiolabeled compounds as potential radiopharmaceuticals. In animals as well as in humans it is possible to obtain information on the spatial time-distribution of tracers by analyzing computer acquired scintigraphic studies. On the basis of data obtained and analyzed with this method using [123I]IMP in humans, dogs, rabbits and rats, with two 99mTc labeled monoclonal antibodies in dogs and with 99mTc DTPA in renal transplants, we recommend this method as an adjunct in radiopharmaceutical development and evaluation.
In previous published research, a factor analysis approach has been applied to airborne spectrometric data of Al-Awabed area, Northern Palmyrides, Syria. A model of four factors (F1, F2, F3 and F4) has proven to be sufficient to represent the acquired data, where 94% of the total data variance is explained. A powerful tool for direct differentiation of various rocks units is obtained through the mapping of these four factors, where a scored lithological map including 11 radiometric units is established. Ninety nine rock samples have been taken according to the four factors to be analyzed by the gamma-spectrometry technique in order to determine their content of eU, eTh and K%. The analysis of 65 samples according to F1 indicates that uranium concentration varies between 2.74 and 123.3 ppm with an average of 58.85 ppm and a standard deviation of 32.53 ppm. The analysis of 18 samples taken according to F2 indicates that K% concentration varies between 0.001 and 0.324 with an average of 0.145 and a standard deviation of 0.122. The analysis of 16 samples taken according to F3 indicates that K% concentration varies between 0.024 and 0.558 with an average of 0.227 and a standard deviation of 0.133. These gamma-results are expected and fit very well with the results obtained by the factor analysis approach. Therefore, the validity and efficacy of the factor analysis approach, to be widely used as a guide in exploration and smart sampling for mining programs, are well demonstrated. The established phosphate maps show a width extension and distribution, and clearly indicate the potential of the research area, and it merits to be followed by economic exploration.
Achievement of an unambiguous solution in factor analysis of dynamic radionuclide studies depends on constraints reflecting the known properties of factors. The constraints should be tight enough to prevent ambiguity but sufficiently general in order to ensure the data-based derivation of factors. In dynamic scintigraphy, the non-negativity of factors is their essential property which is implied by the physical nature of measured quantities. Considering factors as the images of compartments in the distribution space of a radiopharmaceutical (i.e. performing the factor analysis in the spatial domain), a powerful additional constraint can be applied. This constraint is based on the presence of segments in the image matrix where the subtotal number of compartments is projected. Using this constraint, the existence of physiologically related unique solution in factor analysis can be proved providing the number of factors is chosen properly.
In factor analysis, structures are separated on the basis of their temporal behaviour, even if there is a partial overlap. Usually, the temporal behaviour of the system under study cannot be sampled in its true form due to a total overlap by the background. In the ROI (region of interest) method, background subtraction is used as a means to correct the problem. In factor analysis, this problem has been ignored for a long time. We prove that the factor analysis method gives incorrect results when there is a total overlap of a structure. By assuming a local homogeneity of the overlapping structure, we can greatly improve the solution found. Compared with the classical ROI method, our method is operator independent and organ specific.
Factor analysis by the Jöreskog method was applied to data obtained from measurements of 19 skeletal measurements of human physique, carried out in 1971 on 166 men and 122 women students of the Warsaw Technical University. Separate analyses were performed for three factors and four factors in both sexes. The basic three factors could be interpreted as: (i) Length of long bones (limb length), (ii) Size of hands and feet, (iii) Body breadth (skeletal frame size). In the four-factor analysis the fourth factor consisted of trunk length in both sexes, linked with head and neck length and biacromial diameter in men, but not in women.
OBJECTIVE: This paper reviews research relating to the factor analysis of the GHQ-12. We explore the question of whether there is a consistent replicable structure to the GHQ-12 using: (i) a comparative analysis of fit between identified factor models; and (ii) a confirmatory factor analysis of GHQ-12 data from our own study. METHOD: The factor models proposed from the literature were reviewed. The published factor loadings were used to carry out a factor matching analysis to identify similarities between the various factor models that have been identified. In addition, 490 patients visiting their general practitioner completed the General Health Questionnaire (GHQ-12) in the first phase of a longitudinal study evaluating service delivery to rural Tasmania. Three different methods for scoring the GHQ-12 were utilized and each resultant data set was analysed using a Confirmatory Factor Analysis (CFA) to establish which of the various factor models provided the most consistent description of the data. RESULT: None of the complete factor models that have been proposed have been consistently replicated across studies. Isolated factors were replicated between some studies but no single factor structure was replicated across all studies. All of the models had adequate fit to the Tasmanian data when the usual scoring was used. However, only one model had a consistently high 'goodness of fit' across scoring methods. CONCLUSION: It was concluded that the 'best fit' was achieved by a model based on an early factor analytic study using an Australian sample. It was suggested that researchers wanting to extract scales from the GHQ-12 could use this model.