PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “multivariate analysis”

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 1,081 records · Page 60Linked to original sources
PubMed1994

Multivariate analysis of the insulin resistance syndrome in women.

The insulin resistance syndrome (IRS) is characterized by a constellation of interrelated coronary heart disease (CHD) risk factors, including dyslipidemia, obesity, central obesity, elevated systolic blood pressure, and hyperinsulinemia. Factor analysis was used to investigate the clustering of these risk factors in individuals by examining the correlational structure among these variables. Data from 281 genetically unrelated nondiabetic women who participated in exam 2 (1979 to 1980) of the Kaiser Permanente Women Twins Study were used. Factor analysis reduced 10 correlated risk factors to 3 uncorrelated factors, each reflecting a different aspect of the IRS: factor 1 (increased body weight, waist circumference, fasting insulin, and glucose), factor 2 (increased postload and fasting glucose and insulin and systolic blood pressure), and factor 3 (larger low-density lipoprotein particles, decreased plasma triglycerides, and increased high-density lipoprotein). Together, the factors explained nearly 66% of the total variance in the data. Thus, factor analysis defined three distinct aspects of the IRS in this sample of nondiabetic women. These factors may reflect separate underlying mechanisms of the syndrome, each of which may also be involved in CHD risk.

Blood Glucose↗
PubMed1998

Multivariate analysis ('forensiometrics')--a new tool in forensic medicine. Differentiation between sharp force homicide and suicide.

Multivariate projective statistical methods (PCA, PLS-DA) and logistic regression analysis were used to create models to make predictions regarding whether a certain fatality shows similarities to homicide or suicide. The 'model set' consisted of 174 deaths due to sharp force injuries that in previous medicolegal investigation had been judged as homicides and 105 as suicides. The models were then validated on a new set of 40 homicides and 27 suicides that had not been used to create the models (test set validation). The model based on the PLS-DA technique had regarding its ability to identify homicides a sensitivity of 40/40 = 100% and a specificity of 25/27 = 93%. The model's predictions agreed with previously performed medicolegal investigations except in two suicides which according to the model were likely to be homicides. The reliability of this model was somewhat better than predictions achieved by means of logistic regression analysis, where six otherwise proven homicides were wrongly classified as suicides and two actual suicides were misclassified as homicides. The technique not only identifies variables but also ranks their importance. Ranked according to falling positive correlation (falling 'importance' of a finding) to the dependent variable 'death caused by homicide', the predictors were: Injuries to clothing, blood alcohol level, presence of defence injuries, injuries due to other type of violence than sharp force, chest stabs with vertical axis of the entrance wound, sharp force injuries to the upper extremity (except wrist and crook of the arm), sharp force injuries to the head and back. Ranked in increasing positive correlation to 'death caused by suicide' were the predictors: sharp force injuries to the crook of the arm, venue being the victim's home, presence of farewell letter, victim's age, sharp force injuries to the wrist, known suicidal ideation and presence of tentative injuries.

Diagnosis, Differential↗
PubMedDate not supplied

Predictive factors for success in surgical treatment for partial epilepsy: a multivariate analysis.

We conducted a longitudinal self-controlled study of 131 patients aged 4-60 treated with resective surgery for medically uncontrolled partial epilepsy from 1949 to 1988. Using multivariate logistic regression, we showed that pre- and perioperative variables can be used to predict "success" or "failure" of surgical resective treatment in approximately 79% of cases. If the predicted probability is > 0.75 or < 0.25, the model predicts a correct result in 87% of cases. Eight predictive factors emerged with a backward multivariate logistic regression model with the likelihood-ratio (LR) test to exclude variables from the equation: (a) the influence of the surgical team and surgical procedure, (b) the presence of paresis preoperatively, (c) duration of disease, (d) age at treatment, (e) positive neuroradiologic findings in preoperative investigations, (f) preoperative complex partial seizures (CPS), (g) nonepileptic EEG abnormalities, and (h) generalized spike activity in EEG preoperatively. Sex, age at first seizure, area of resection, presence of simple or generalized seizures preoperatively, preoperative seizure frequency, tissue pathology, use of computed tomography/nuclear magnetic resonance (CT/NMR) in preoperative investigations, degree of preoperative neurologic deficit, perioperative electrocorticographic results, and bilateral EEG spikes did not have predictive value in the model.

Adolescent↗
PubMed2001

Prediction of Caco-2 cell permeability using partial least squares multivariate analysis.

The permeability across Caco-2 cell monolayers of structurally diverse compounds were predicted using computed molecular descriptors and multivariate. Partial Least Squares (PLS). The molecular descriptors including log polarization, log solvent accessible surface area, hydration energy, heat of formation, and dipole moment were calculated with Hyperchem and ChemPlus QSAR programs for Windows. Other physiochemical properties such as hydrogen acceptor for oxygen atoms, hydrogen acceptor for nitrogen atoms, hydrogen bond donors, hydrogen bond forming ability, molecular weight, and log distribution coefficient were also used as descriptors. Cross validation with internal test set and prediction with external test set indicated the usefulness of the derived model for Caco-2 cell permeability. Hydrogen bonding is one of the important factors associated with permeability. While increased logD and hydration energy facilitate permeability, an increased dipole moment of molecules has a negative effect on permeability.

Algorithms↗
PubMedDate not supplied

Determinants of no-show appointment behavior: the utility of multivariate analysis.

A multitude of determinants have been identified as predictive of broken appointments. The majority of prior studies have been limited to univariate analysis of the relationship between predictors and appointment keeping behavior. The present report studied 25 independent predictors of no-show behavior using both univariate and multivariate analyses. A total of 579 kept appointments and 84 failed appointments were analyzed. Results of univariate analysis indicated the following significant relationships with appointment behavior: age, ethnicity, marital status, mode of payment, chronic illness, telephone in house, type of care, prior visits to center, cost of care, transportation to center, physician ethnicity, and linguistic capability. However, multiple logistic function analysis revealed only six significant associations: type of care, chronic illness, linguistic capability, mode of payment, physician-patient sex differences, and marital status of the patient. Multivariate analysis may yield a more accurate and clinically useful model of no-show behavior. For example, language barrier may be more of a problem than the race of the patient. Prospective studies might benefit from these considerations.

Appointments and Schedules↗
PubMed1991

Multivariate analysis of factors predicting local relapse after radical prostatectomy--possible indications for postoperative radiotherapy.

Between 1970 and 1983, 273 patients underwent radical surgery (radical prostatectomy--261, radical cystoprostatectomy--12) for newly diagnosed adenocarcinoma of the prostate at Duke University Medical Center and received no adjuvant radiotherapy. A total of 46 patients developed local recurrence. Forty developed local relapse only and six developed simultaneous local and distant failure. The crude local relapse rate was 17% (46/273). The actuarial local failure rate at 5, 10, and 15 years was 12%, 32%, and 35%, respectively. Univariate and multivariate analyses were performed to identify factors predictive of local relapse after radical surgery. Possible prognostic factors analyzed were: age, type of biopsy, use of adjuvant hormonal therapy, histologic grade, histologic involvement of seminal vesicles, positive surgical margins, clinical stage, and elevated acid phosphatase. Factors identified as significant predictors of local relapse by univariate analysis were: poorly differentiated histology (p = 0.0001), seminal vesicle involvement (p = 0.0009), and positive surgical margins (p = 0.0001). An elevated preoperative acid phosphatase was of borderline significance (p = 0.06). On multivariate analysis, poorly differentiated histology (p = 0.0007), positive margins (p = 0.0015), and elevated acid phosphatase (p = 0.0273) were significant predictors of local failure. Seminal vesicle involvement was no longer a significant predictor of local failure. However, on subsequent univariate and multivariate analyses, seminal vesicle involvement was the only significant predictor for the development of distant metastases (p = 0.0019, multivariate). Thus, patients with poorly differentiated tumors, positive surgical margins, or elevated preoperative acid phosphatase are at high risk for local relapse after radical prostatectomy. These patients should be included in future clinical trials studying the role of adjuvant radiotherapy after radical prostatectomy, or offered adjuvant radiotherapy if they cannot or will not participate in such trials.

Adenocarcinoma↗
PubMed2007

A multivariate analysis of neuroanatomic relationships in a genetically informative pediatric sample.

An important component of brain mapping is an understanding of the relationships between neuroanatomic structures, as well as the nature of shared causal factors. Prior twin studies have demonstrated that much of individual differences in human anatomy are caused by genetic differences, but information is limited on whether different structures share common genetic factors. We performed a multivariate statistical genetic analysis on volumetric MRI measures (cerebrum, cerebellum, lateral ventricles, corpus callosum, thalamus, and basal ganglia) from a pediatric sample of 326 twins and 158 singletons. Our results suggest that the great majority of variability in cerebrum, cerebellum, thalamus and basal ganglia is determined by a single genetic factor. Though most (75%) of the variability in corpus callosum was explained by additive genetic effects these were largely independent of other structures. We also observed relatively small but significant environmental effects common to multiple neuroanatomic regions, particularly between thalamus, basal ganglia, and lateral ventricles. These findings are concordant with prior volumetric twin studies and support radial models of brain evolution.

Adolescent↗
PubMed2005

Factors affecting assisted reproductive technology (ART) pregnancy rates: a multivariate analysis.

PURPOSE: To determine factors instrumental in achieving a clinical pregnancy in assisted reproductive technology (ART) patients. METHODS: This study included 205 women undergoing their first ART cycle. Univariate and multivariate analyses were performed for patient demographics, in vitro production data, and factors associated with embryo transfer. Odds ratios (OR) were performed where appropriate. RESULTS: Our analyses indicated that age (OR: 0.879), specific year in which the cycle was performed (OR: 2.959), and use of intracytoplasmic sperm injection (OR: 2.867) altered potential pregnancy rate. In addition, percent fertilization (OR: 1.028), number of embryos transferred (OR: 1.842), type of catheter used to transfer the embryos (OR: 0.377), presence of blood on the catheter (OR .414), and embryologist (OR: 2.338) also altered pregnancy rate. CONCLUSIONS: Our data indicate patients' age, use of the Cook catheter, and presence of blood on the catheter reduce pregnancy rates. Performing ART in 1999, using ICSI, increasing fertilization rate, increasing number of embryos transferred (albeit less desirable when the chance of multiple gestation may occur), and transferring embryos via a particular embryologist, improve pregnancy rates.

Adult↗
PubMedDate not supplied

DNA ploidy, proliferative index and EGF-R status in 130 cases of resected gastric cancer--a multivariate analysis.

BACKGROUND/AIMS: The purpose of this study was to define the prognostic role of DNA ploidy, proliferative index and EGF-R status in resected gastric cancer. MATERIALS AND METHODS: Ten clinico-pathological parameters and three biological factors obtained from flow cytometry and immunohisto-chemistry were evaluated in a series of 130 gastric cancer patients who received surgical treatment, including 28 stage IV cases (21.6%), using paraffin-embedded and fresh specimens in 77.7% and 22.3% of the cases, respectively. These variables were first analyzed and tested for correlation within the whole series and then weighted against survival in 117 applicable cases through univariate and multivariate analyses. RESULTS: Aneuploidy was significantly related to higher proliferative activity, EGF-R expression and deeper stomach wall infiltration. Higher proliferative activity was significantly related to deeper stomach wall infiltration and larger tumor diameter. The latter showed a significant relationship to EGF-R expression. Univariate analysis showed the significant variables for survival to be DNA ploidy, pT, pN, M, stage, histological type according to Lauren and tumor diameter. Multivariate analysis calculated on these significant variables using the Cox multiple stepwise regression model detected three factors which independently influence survival: pathological stage (p < 0.00001), histological type according to Lauren (p < 0.002) and DNA ploidy (p < 0.03). CONCLUSIONS: DNA ploidy was shown to be a significant prognostic parameter in resected gastric cancer after pathological stage and histological type according to Lauren. The prognostic roles of proliferative activity and EGF-R status require further investigation.

Adenocarcinoma↗
PubMed1998

Predictors of psychiatric hospitalization: a multivariate analysis.

Inpatient treatment continues to be the most expensive form of mental health service. This study sought to improve the methodological weaknesses, e.g., poor statistical controls, in the literature by using multivariate statistics to predict hospitalization. Results revealed that aftercare, i.e., outpatient treatment, is an important factor in reducing the utilization of inpatient resources, even when controlling for demographic and psychiatric history variables. Further, background characteristics, while easily measured, are not important predictors of hospitalization.

Adult↗
PubMed2006

Comparison of the performance of linear multivariate analysis methods for normal and dyplasia tissues differentiation using autofluorescence spectroscopy.

We compared the performance of three widely used linear multivariate methods for autofluorescence spectroscopic tissues differentiation. Principal component analysis (PCA), partial least squares (PLS), and multivariate linear regression (MVLR) were compared for differentiating at normal, tubular adenoma/epithelial dysplasia and cancer in colorectal and oral tissues. The methods' performances were evaluated by cross-validation analysis. The group-averaged predictive diagnostic accuracies were 85% (PCA), 90% (PLS), and 89% (MVLR) for colorectal tissues; 89% (PCA), 90% (PLS), and 90% (MVLR) for oral tissues. This study found that both PLS and MVLR achieved higher diagnostic results than did PCA.

Algorithms↗
PubMed1993

[Differences in prediction of mortality and non-fatal complications after acute myocardial infarction: a multivariate analysis].

The global prognostic after acute myocardial infarction depends of the left ventricular function and the residual ischemia. However, there is controversy about the capacity of some variables reflecting this prognostic factors to predict futures complications. The main objective of this study is to know the influence of the kind of complications in the prognostic value of the variables more frequently used in the clinical practice. We studied 121 consecutive patients with acute myocardial infarction discharged from hospital. The previous and during acute infarction clinical variables were analyzed. The regional and global left ventricular function was studied by 2D echocardiography. An exercise test symptoms-limited was performed at 3rd week after discharge in 94 patients selected. The patients were followed during selected one year. The first year cardiac mortality rate was 8.6%. In the multivariate, the independent predictors of mortality were: age > 60 years (p = 0.02) and the impossibility to perform exercise test (p = 0.0002). In the univariate analysis the echocardiographic motility score > or = 17, eyection fraction < 35% and Killip class > or = II were also death predictors. Non fatal complications (angina, reinfarction o heart failure) were present in 47 patients (42.3%) of 111 survivors. The ischemic complications were predicted in multivariate analysis only by postinfarction angina (p = 0.0007), and heart failure by eyection fraction < 35% (p = 0.006), previous infarct (p = 0.002) and Killip class (p = 0.04).(ABSTRACT TRUNCATED AT 250 WORDS)

Adult↗
PubMed1993

A multivariate analysis of choice criteria for hospitals.

This study determines the choice criteria used by consumers in selecting a hospital and provides information useful to hospital administrators in planning and implementing marketing efforts directed toward potential consumers. The findings suggest that physical plot, previous experience with the hospital, location of hospital, overall cost, and reputation of the hospital were important factors in selecting a hospital.

Adult↗