Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Multivariate Analysis”

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 19 recordsLinked to original sources
PubMed1993

Prognostic factors in lung cancer based on multivariate analysis.

Multivariate analysis was performed on 1,336 patients with lung cancer to determine the prognostic significance of stage, race, gender, age, and treatment in each histologic subtype. The study was designed to establish a subgroup of patients whose survival outcome might be better, based on these factors. On univariate analysis, stage and surgery were significant factors in each histologic subtype. The presence of liver metastases, was an important prognostic factor in all subtypes except large cell carcinoma. However, 131 of 140 patients with large cell carcinoma had liver metastases, and this factor may account for the observation that liver metastases was not a significant prognostic factor. In the multivariate analysis, good prognosis was associated with early stage disease and surgical treatment in all cell types. For a given stage, the improvements in relative risk due to surgery represent both the effect of treatment and the effects of other unmeasured patient characteristics, such as performance status and physiological status, that make the patient a suitable candidate for surgery.

Adult↗
PubMed1986

Studies on the evaluation of joint functional impairment in the lower extremities--examination by multivariate analysis.

Multivariate analysis was applied to develop systems which afford more objective and universal evaluation of joint function in the lower extremities such as the hip, knee, and ankle joint. Conventional methods have various drawbacks, such as selection of parameters, allocation of points to selected parameters, selection of indices to represent impairment of joint function as a whole, and functional evaluation in polyarticular impairment. In order to solve these problems, causative factors (pain, mobility, and stability) of the joint impairment and secondary factors such as activities of daily living (ADL) representing the overall effects of the causative factors were strictly discriminated, and only causative factors were selected to serve as functional parameters. Then, taking ADL as criterion variables and causative factors as predictor variables, multiple regression analysis was performed using 207 hips, 250 knees, and 50 ankle joints. Based on this analysis, an original system for evaluating joint function was developed.

Activities of Daily Living↗
PubMed1977

Changes of P50 in hypoxaemia, hypercapnia and polycythaemia: multivariate analysis.

Multivariate analysis of P50 changes in hypoxia, hypercapnia and polycythaemia was performed in an heterogeneous group of forty three patients: hypoxic subjects with or without hypercapnia, with or without polycythaemia and polycythaemic subjects without hypoxia. A statistical analysis was undertaken using comparison of the means, study of the correlations, principal component analysis, multiple regression and correspondence analysis. In the patients studied, P50 changes were not wholly explained by those of 2-3 DPG and pH; PaCO2, per se, did not play an important part. Haemoglobin concentration and P50 value would represent an adaptative mechanism to hypoxia: when hypoxia is moderate (80 greater than PaO2 greater than or equal to 65 torr) and isolated, oxygen haemoglobin affinity decreases (P50 increases); when hypoxia is severe (PaO2 less than 65 torr) and combined with hypercapnia and disturbed acid-base equilibrium, P50 comes back to normal range but haemoglobin increases, restoring thus, the normal blood oxygen content.

Adult↗
PubMed1983

Langur monkey mother loss: profile analysis with multivariate analysis of variance for separation subjects and controls.

We present an analysis of observational data on the behavior profiles for two groups of infant langur monkeys (Presbytis entellus): one that had experienced mother loss, and the other, a control group, that had not undergone mother loss. Observations were analyzed in three 2-week time periods: 2 preseparation, 2 separation, and 2 postseparation periods for both groups. The profile analysis application of the multivariate analysis of variance (MANOVA) was employed to test the hypothesis that the two groups had similar patterns of change across time. For each group a vector of 5 difference scores was computed; then MANOVA was used to test the similarity of the mean vectors of difference scores for the two groups. Profiles for the separation and control infants were significantly different for some behaviors, and changes between some time periods were also significant.

Analysis of Variance↗
PubMed1994

Age and sex modulate effects of stress on the immune system: a multivariate analysis.

Multivariate analyses indicate that both age and sex can modulate examination stress effects on immune factors. Stress lowered eosinophil counts in females but raised them in males. Age modulated stress effects on CD4 and CD8 cells, hemoglobin, and erythrocytes. CD4 decreased more in older subjects; CD8 increased in older and decreased in younger subjects; hemoglobin decreased in younger but not older subjects; erythrocytes increased in older and decreased in younger subjects. Initial age and/or sex differences in levels of neutrophils and lymphocytes were not statistically altered by stress. Stress effects not modulated by age or sex increased serum IgA and IgM, CD19, and stimulated phagocytic activity but decreased serum IgG, CD3, basophils, and unstimulated phagocytic activity. The immunological effects of stress are multiple and are influenced by variations in age and sex of the person.

Adult↗
PubMed1991

Morphometry and classification of the cat hypothalamic capillaries using multivariate analysis.

Multivariate statistical methods were used to do morphological determination of the amount and the relation of the astrocytic foot processes (AF), astrocytic lamina (AL) and dense zones (DZ) on hypothalamic capillaries. We corroborated that 100% of the perivascular membrane is surrounded by glial astrocytic cytoplasm. Principal component analysis shows that the original variables are independent of one another. Cluster analysis confirms the individual situation obtained through the principal component analysis. Discriminant analysis shows that the discriminant capacity of the 3 variables is very significant. For the potential interrelations of astrocytic foot processes, astrocytic lamina and dense zones we have clearly differentiated four groups of hypothalamic capillaries.

Animals↗
PubMed1996

Regression analysis and multivariate analysis.

Proper evaluation of data does not necessarily require the use of advanced statistical methods; however, such advanced tools offer the researcher the freedom to evaluate more complex hypotheses. This overview of regression analysis and multivariate statistics describes general concepts. Basic definitions and conventions are reviewed. The types of regression analysis are then discussed, including simple regression, multiple regression, multivariate multiple regression, and logistic regression. The various steps required to perform these analyses are described, and the advantages and disadvantages of each is detailed.

Analysis of Variance↗
PubMed1985

Prognostic factors in survival after portasystemic shunts. Multivariate analysis.

Multivariate analyses correlated short-term survival and long-term survival with clinical data from 141 patients with portasystemic shunts for bleeding esophageal varices over the 8 years from 1974 through 1981. By logistic regression analysis, the elements with independent prognostic significance for operative death were an emergency operation, serum albumin and bilirubin levels, age, and sex. A cutpoint probability value of 0.75 marked the single level above which 84% of patients could be expected to survive, but below which 77% are likely to die. By a Cox regression model, elements with independent prognostic significance were identical. Male sex (p = 0.02) and prolonged partial thromboplastin time (p = 0.04) indicated a poor prognosis after an emergency operation; after an elective operation only the serum albumin level was prognostic (p = 0.02). Normal blood clotting is the major determinant of survival after an emergency shunt, and the serum albumin level chiefly determines survival after elective portasystemic decompression.

Adult↗
PubMed1993

Long-term prognosis after myocardial infarction: univariate and multivariate analysis of clinical characteristics in 1,000 patients. Kyoto and Shiga Myocardial Infarction (KYSMI) Study Group.

To examine the clinical outcome for patients with myocardial infarction and to analyze clinical predictors for long-term prognosis, 1,000 patients were studied retrospectively. Between January 1983 and December 1987, 1,002 consecutive patients with myocardial infarction, who resided in the Kyoto and Shiga districts, were reviewed after coronary arteriography, but in two patients medical records were not located. During 3.3 +/- 2.0 years, 75 patients died of cardiac causes and 301 experienced cardiac events (death, reinfarction, and revascularization). Overall 5-year cardiac mortality was 8% and cardiac event rate was 35%. Among determinants of age, smoking, hypertension, diabetes mellitus, heart failure, postinfarction angina, serious arrhythmia, mitral regurgitation, digitalis and diuretics administration, ejection fraction (EF), left main trunk disease, and number of diseased coronary arteries selected by univariate analysis, multivariate analysis revealed that heart failure, EF, number of coronary vessel disease, diabetes, and mitral regurgitation were the important predictors of cardiac death. For cardiac events, multivariate analysis demonstrated that the number of diseased coronaries, postinfarction angina, and left main trunk disease were significant predictors. Therefore, impaired left ventricular function and myocardial ischemia appear to be important markers for cardiac death, but impaired left ventricular function does not appear to be a predictor for cardiac events. The data obtained in this study will be useful in the assessment of patients with myocardial infarction and will be of clinical significance in predicting mortality and cardiac events.

Adult↗
PubMed2001

[A confirmatory positioning analysis: a multivariate analysis for finding change of image and evaluation of cognitive map].

A confirmatory approach to the positioning method (Toyoda, 2001) that can be used for analyzing three-mode multivariate data measured by the semantic differential (SD) method is considered. Two examples of the confirmatory method are given to demonstrate the applicability of the model. The first is a method to test the difference of image, between an experimental group and a control group. The second is a method to evaluate cognitive maps that are drawn subjectively. The conventional standard methods are not capable of solving these problems.

Cognition↗
PubMed1986

Seemingly complete removal of histologically benign intracranial meningioma: late recurrence rate and factors predicting recurrence in 657 patients. A multivariate analysis.

From 1953 to 1980, 657 patients survived a seemingly complete removal of a histologically benign intracranial meningioma. Recurrence of the tumor was found in 59 of these 657 patients, but some cases of recurrence may have been missed inasmuch as only 69 patients were followed for 20 years. With ordinary life-table analysis, the overall recurrence rate was estimated to be 19% at 20 years. The reliability of this result was evaluated with a modified life-table analysis. Multivariate analysis showed that coagulation of the dural insertion, invasion of bone, and soft consistency of tumor were strong risk factors (p less than 0.01) for recurrence. The recurrence rate at 20 years was estimated to be 11% for the 249 patients with none of these risk factors, 15-24% for those with one factor, and 34-56% for those with two.

Brain Neoplasms↗
PubMed1990

Empirical considerations in orthopaedic research design and data analysis. Part III: Multivariable analysis.

To make a contribution to a base of knowledge, research design must be sound and the data must be analyzed by the most efficient, powerful method available. In conducting orthopaedic research as well as when interpreting the orthopaedic literature, knowledge of multivariable analysis is essential in evaluating the effects of variables that may confound or influence the relationship of interest. An understanding of the basic tenets, underlying assumptions, and purposes of frequently used types of multivariable analysis is necessary for an accurate, critical evaluation of a study's results.

Humans↗
PubMed1983

Smoking and the occurrence of congenital malformations and spontaneous abortions: multivariate analysis.

A multivariate analysis was carried out with retrospective data on the effects of tobacco smoking on congenital malformations and spontaneous abortions. Congenital malformations were collected from the Finnish Register of Congenital Malformation. Tens of possible confounding variables, including maternal and family characteristics, obstetric history, medicine taking, and diseases during the pregnancy, were taken into consideration. In the final analysis 13 confounders were controlled. The odds ratios for the smoker's child to be born with central nervous system, oral cleft, or musculoskeletal malformations were 1.25, 1.25, and 0.75, respectively. All the differences were statistically nonsignificant. The effect of smoking on spontaneous abortions was analyzed by means of a questionnaire study on hospital personnel. The smokers of over 10 cigarettes per day had more spontaneous abortions than the nonsmokers, but the differences were not significantly statistically. The data were controlled for age, parity, and coffee and alcohol drinking.

Abortion, Spontaneous↗
PubMed1990

[The analysis of prognostic factors for prostate cancer by multivariate analysis].

Multivariable analysis (quantification method I and II) was used to investigate the relationship between survival rates and prognostic factors. The items studied were age, stage, pathological differentiation, gait disturbance, acid phosphatase and erythrocyte sedimentation rate (ESR). The data were available for 353 patients for all items. The significance of each item was evaluated in all patients-group, subgroup stratified by their cause of death, and subgroup stratified by their stage. The significance of age and gait disturbance which represent the general condition was proven in all patients, non-cancer death group and stage ABC group. In the cancer death group, stage D group was the most significant as prognostic factor. ESR was the most significant in the stage D group and the significance was gradually increased with the observation time.

Humans↗
PubMedDate not supplied

Determination of the optimum dose level of lipiodol in transcatheter arterial embolization of primary hepatocellular carcinoma based on retrospective multivariate analysis.

PURPOSE: Multivariate analysis was used to study the effectiveness and optimum dose level of Lipiodol (LP) in transcatheter arterial embolization (TAE) of hepatocellular carcinoma (HCC). METHODS: A total of 219 cases of nodular type HCC, with a tumor diameter less than 7 cm, were studied. TAE was performed using both Gelfoam sponge (GS) and LP in 158 cases; in the remaining 61 cases only GS was used. RESULTS: Statistical stepwise variable selection revealed that only LP had a negative T-value, suggesting that LP is a useful factor for prognosis. The most favorable effect on patient prognosis was obtained with an LP dose level (expressed in mm) of 1-1.5 times the absolute value of the tumor diameter (expressed in cm). CONCLUSION: A significant difference (p < 0.01, log-rank test) in survival was found between the GS with LP group and the GS only group, using Cox's proportional hazard model.

Carcinoma, Hepatocellular↗
PubMed1998

Statistical methodology: IV. Analysis of variance, analysis of covariance, and multivariate analysis of variance.

Medical research frequently involves the statistical comparison of >2 groups, often using data obtained through the application of complex experimental designs. Fortunately, inferential statistical methodologies exist to address these situations. Analysis of variance (ANOVA) in its many forms is used to simultaneously test the equality of all groups in a study. One-way (with 1 independent variable), 2-way (with 2 independent variables), and repeated-measures (patients serve as their own controls) ANOVAs are forms of this technique. Each form has been developed to analyze data from a specific experimental design. Analysis of covariance (ANCOVA) allows the researcher to control for confounding variables that may influence the response of the dependent variable. Finally, multivariate analysis of variance (MANOVA) evaluates the simultaneous responses of multiple dependent variables to > or = 1 independent variable. Whereas ANOVA is the correct alternative to statistically inappropriate multiple t-tests, MANOVA is the correct alternative to statistically inappropriate multiple univariate ANOVA calculations. Use of each of these statistical methods requires an appropriate experimental design and data meeting a number of assumptions. When used properly, each of these methods provides a powerful statistical analysis technique.

Analysis of Variance↗
PubMed1993

The application of computerized analysis of nuclear images and multivariate analysis to the understanding of the effects of antineoplastic agents and their mechanism of action.

The present work involves the creation of a model that makes the rapid study of anticancer drugs possible. In order to make possible the multilevel study of antineoplastic derivatives (their effects on cell proliferation, their cell kinetics, their chromatin texture and the study of their potential operating mechanisms), the computerized analysis of nuclear images linked to multivariate analysis was chosen, and, in so doing, took into account 15 computer-calculated morphonuclear parameters for Feulgen-stained cell nuclei. In order to validate the method it was subjected to a series of well-known cancer drugs. The present results are in agreement with those obtained by conventional methods similar to those used for the study of the cytotoxicity of anticancer molecules or their effect on cell kinetics. Multivariate analysis applied to the parameters described by image analysis should make possible the rapid study of new anticancer drugs together with the rapid orientation of studies undertaken to ascertain their operating mechanisms.

Animals↗