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 1,027 records · Page 57Linked to original sources
PubMed2004

[Multivariate analysis of prognostic factors in endometrial carcinoma].

BACKGROUND & OBJECTIVE: A number of prognostic factors have been evaluated in endometrial carcinoma. Among them, only a few factors have been recognized as major adverse factors affecting prognosis independently. This study was to explore the independent prognostic factors in endometrial carcinoma. METHODS: A total of 265 patients who accepted primary surgical treatment in our hospital from Jan. 1990 to Dec. 2000 were studied retrospectively. Prognostic factors were analyzed using univariate model and multivariate regression model. RESULTS: The 5-year recurrence-free survival rate of all patients was 83.3%, and 5-year overall survival rate was 84.3%. The univariate model revealed that clinical stage, pathological stage, pathologic grade, pathologic subtype, depth of muscularis invasion, cervical invasion, lymph node metastasis, peritoneal cytology, lymph-vascular invasion, and adnexa metastasis significantly associated with 5-year recurrence-free survival, and 5-year overall survival (P< 0.05), while age, and complications have no significant association with prognosis (P >0.05). However,on multivariate regression analysis, only pathohistological stage, pathologic grade, depth of muscularis invasion, and cervical invasion significantly correlated with 5-year recurrence-free survival, and 5-year overall survival (P< 0.05); clinical stage correlated with 5-year recurrence-free survival significantly (P< 0.001), but not with 5-year overall survival (P=0.074). Hematogenous dissemination rate of patients with muscularis invasion of >50% was significantly higher than that with muscularis invasion of <or=50% (12.9% vs 0.6%, P< 0.001). Lymphatic failure rate of patients with cervical invasion was significantly higher than that without cervical invasion (21.1% vs 3.6%, P< 0.001). CONCLUSIONS: FIGO stage, pathologic grade, depth of muscularis invasion, and cervical invasion are independent prognostic factors in endometrial carcinoma. Comparing with clinical stage, pathohistological stage could assess prognosis more objectively. Deep muscularis invasion is the strongest predictor of hematogenous dissemination; and cervical invasion increases the lymphatic failure rate.

Adult↗
PubMed1978

Multivariate analysis for matched case-control studies.

A multivariate method based on the linear logistic model is presented for the analysis of case-control studies with pairwise matching. This technique enables one to investigate the effect of several variables simultaneously in the analysis while allowing for the matched design. The odds ratio is used as the basic measure of risk. One is able to control for variables which are not matching variables while investigating the odds ratio for a particular factor, and to estimate the change in the odds ratio as the level of one or more interval variables changes. The computing methods used for obtaining maximum conditional likelihood estimates of the parameters of interest are modifications of standard programs for logit regression.

Epidemiologic Methods↗
PubMed1990

Multivariate analysis of mental and physical load components in sinus arrhythmia scores.

In many studies on the relationship between mental load and sinus arrhythmia (SA) scores the effect of minor motor actions inevitably accompanying light work has been disregarded. However, because the chemoreceptors for muscle cell activities are directly coupled with the heart rate (HR) control centre in the brain, even minor motor actions can influence SA. In this study mental and physical loads were scored respectively by information processing and finger tapping rates. Percentage changes in SA and HR above resting levels were scored from the heart beat intervals of the subjects. According to a multivariate regression analysis of SA scores on mental and physical load components, an increase in mental load decreased SA, but did not increase HR significantly. An increase in physical load decreased SA and increased HR significantly.

Adult↗
PubMed1990

Long-term results of resection for large hepatocellular carcinoma: a multivariate analysis of clinicopathological features.

Recurrent or metastatic disease is frequently encountered among patients who have had resection of their primary hepatocellular carcinoma. A retrospective study on 117 patients (104 men, 13 women; mean age +/- standard deviation: 53.8 +/- 12.4 yr) who had hepatectomy for large hepatocellular carcinoma (diameter greater than or equal to 5 cm) was conducted to identify an at-risk population for tumor recurrence. Disease-free survival was correlated with 22 clinical (n = 5), serological (n = 2), gross pathological (n = 3) and histological (n = 12) features of the resected specimens using Cox's multivariate regression analysis. Recurrent hepatocellular carcinoma was detected in 74 patients within a median follow-up period of 13.7 mo. Although 17 patients had extrahepatic disease alone, recurrence was confined to the hepatic remnant in 40 patients. Disease-free survival rates at 1, 3 and 5 yr were 40%, 19% and 12%, respectively. Two of the five histological parameters isolated, negative resection margin (p less than 0.01) and encapsulation (p less than 0.006), were identified as favorable independent prognostic predictors. When patients with positive margins were excluded from the analysis, repeated calculation showed that encapsulation was the only important determinant. From this analysis, detailed histological study of the resected tumor is seen to be the only satisfactory means for assessing long-term prognosis. An aggressive approach is warranted among patients with encapsulated tumors. Even with a clear resection margin, adjuvant treatment should be considered for those patients who have unencapsulated lesions.

Adult↗
PubMed2002

Adenocarcinoma of the stomach: univariate and multivariate analysis of factors associated with survival.

Gastric cancer is the most frequent tumor of the digestive tract in Mexico. Most patients are diagnosed at advanced stages, and fatal outcome is expected. One hundred fifty patient charts were retrospectively reviewed. Univariate and multivariate analyses were performed to evaluate the impact of clinicopathologic and treatment variables on survival. Most patients (75%) were at advanced stages, harboring poorly differentiated tumors. Surgery, mostly palliative, was performed on 114 patients. Chemotherapy was administered to 47 patients. On univariate analysis, significant prognostic factors were TNM stage, chemotherapy, surgical attempt, performance status, histology, and tumor site (p < 0.001). On multivariate analysis, independent prognostic factors were TNM stage, histology, tumor site, surgical attempt, and chemotherapy (p < 0.01). Median survival for patients with palliative or adjuvant chemotherapy was 11.4 and 10.4 months, respectively, compared with +/- 3 months for patients with no chemotherapy (p < 0.03). Nonsurgical patients receiving chemotherapy survived 5.4 months versus 1.1 months for those without chemotherapy. The favorable influence of chemotherapy persisted after a stratified analysis of subgroups eliminating potential biases. We identified prognostic factors for survival. Chemotherapy should be considered even for advanced-stage patients with either adjuvant or palliative attempts, because we consistently found a favorable impact on the median survival time. However, phase III prospective randomized trials are awaited.

Adenocarcinoma↗
PubMed1997

Monte Carlo methods in clinical research: applications in multivariable analysis.

BACKGROUND: Monte Carlo methods use "simulated" analyses with random numbers for solving problems, particularly those that defy solutions using mathematical theory alone. Research using Monte Carlo simulations is very popular in many branches of science and is sometimes done in clinical investigation. The origins and basic strategy of the technique, however, may not be well known to clinical researchers. The purpose of this paper is to describe the history and general principles of Monte Carlo methods and to demonstrate how Monte Carlo simulations were recently applied to examine a phenomenon in multivariable statistical analysis called the number of outcome events per independent variable (EPV). For example, in a cohort of 200 people, with 50 deaths and 5 independent (predictor) variables, EPV = 50/5 = 10. METHODS: The "real-world" data came from a clinical trial of 673 patients in which 7 variables were cogent predictors of 252 deaths, so that EPV = 252/7 = 36. For the Monte Carlo simulations, special models were used while allowing simulations of proportional hazards and logistic regression to maintain the basic relationship of variables and the same size of the original population, at EPV values of 2, 5, 10, 15, 20, and 25. RESULTS: The Monte Carlo simulations confirmed a previously undocumented "rule of thumb" stating that when the EPV is less than 10-20, the algebraic models used in logistic regression and proportional hazards regression may be unreliable, leading to imprecise or spurious results. CONCLUSION: Monte Carlo techniques offer attractive methods for clinical investigators to use in solving problems that are not amenable to customary mathematical approaches.

Humans↗
PubMed1999

A longitudinal study into the perceptions of caring among student nurses using multivariate analysis of the Caring Dimensions Inventory.

A longitudinal study involving student nurses' responses to the Caring Dimensions Inventory (CDI) was carried out to investigate the development of perceptions of caring. This study continues previously reported work in which the development of the CDI was described. The current study was designed to see if the CDI could detect changes in perceptions of caring among student nurses as they progressed through their training. Data were collected on the same subjects at three points, each a year apart, throughout the study. A cohort of 168 student nurses entering training was recruited into the present study and data from the CDI was analysed using exploratory factor analysis. A four factor structure for perceptions of caring throughout the study was supported, with some evidence for a five factor structure at 24 months into the programme of nurse education. A professional and technical factor increased in congruence with a similar factor, identified in a previous study, throughout the present study and was used to provide factor scores for individual students. There was a statistically significant and moderately sized correlation between the score on this factor and the age of the student nurses. The five factor solution yielded a factor of accountability. Professional and educational consequences of the study are discussed along with the problem of attrition in longitudinal studies.

Adult↗
PubMed1992

Multivariate analysis of craniomandibular disorders.

This article presents the clinical application of a special statistical method, the Multiple Correspondence Analysis (MCA), derived from data analysis and particularly suited to study a restricted population (here, 202 patients with temporomandibular joint [TMJ] disorders) whose characteristics are defined by a large number of modalities (140). Processing of a Burt-type data table, or Complete Disjunctive Table (CDT), provided a geometrical visualization of information, and permitted determination of the prevalent factor axes. Results showed a significant relationship between all the articular disorders, either peripheral or focal (TMJ), thus revealing the possibility of a yet, little known morphogenic influence on the genesis of TMJ disorders.

Adolescent↗
PubMed2004

Genetic relationships among Aster species by multivariate analysis and AFLP markers.

Fourteen species of Aster were characterized for morphological traits of commercial interest, chromosome number and genetic diversity by AFLP markers to exploit this material not only for breeding purposes but also as source of pharmaceuticals. The variation observed among the Aster species for five morphological traits was summarized by means of discriminant analysis. Three significant canonical variables, accounting for about 96% of total variance, were mainly correlated with capitulum diameter, number of ligulae and leaf length. The morphological variation appeared related to ploidy level of the species. A diploid chromosome number ranging from 10 to 18 was evident in seven species whereas the other species are polyploids with variable chromosome numbers up to 52. The pollen production and stainability were quite variable among Aster species. Furthermore, seven species produced big pollen grains besides normally-sized ones, at a frequency ranging from 1 to 12%, possibly due to 2n pollen production. The possibility that evolution of Aster genus could be driven by 2n gametes and sexual polyploidization is discussed. Implications of 2n gametes for Aster breeding are also mentioned. AFLP analysis of the fourteen Aster species based on six primer combinations revealed a total number of 421 polymorphic amplified DNA fragments. The genetic similarities estimated between the Aster species, based on both shared and unique amplification products ranged from 0.335 to 0.604 across all species and revealed a geographically based clustering within the genus. The highest similarity was detected between A. alpinus and A. amellus with Eurasian origin.

Aster Plant↗
PubMed1997

Identification and correlation of human footfall load parameters using multivariate analysis.

This research had two main objectives: to identify and quantify the multiple reaction parameters of human footfall load histories for 24 subjects; and to seek statistical correlations of the reaction parameters with two gaits: fitness walking and running, with two footfall surfaces: rigid and mat, and with two subject attributes: gender and arch index. These reaction parameters, measured with a force plate, include the subjects' foot reaction forces in the three orthogonal directions and the particular features of these forces such as their duration, average values, peak values, and rates of loading. An automated data retrieval-software system evaluated these reaction parameters. The statistical correlations were made using principal component analysis (PCA), a method that projected the 13 identified footfall reaction parameters onto subsets of three or four parameters called principal components that contained most of the variance of the original thirteen. The results, among others, show couplings between the vertical and the peak medial load, but an uncoupling of the posterior-anterior loads with the loads in the other two directions.

Adult↗
PubMed2001

The dimensionality of alcohol abuse and dependence: a multivariate analysis of DSM-IV symptom items in the National Longitudinal Survey of Youth.

OBJECTIVE: This article examines the factor structure of 22 symptom items used to configure the criteria of DSM-IV (Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition) alcohol abuse and dependence and relates the factor structure to background characteristics. METHOD: Data for this study were drawn from the National Longitudinal Study of Labor Market Experience in Youth (NLSY). The symptom items were related to the covariates using the statistical technique of structural equation modeling generalized to dichotomous outcomes. The present model is a special case of structural equation modeling, a multiple causes and multiple indicators (MIMIC) model, in which one or more latent variables (i.e., alcohol abuse and dependence) intervene between a set of observed background variables predicting a set of observed response variables (i.e., DSM-IV symptom items). RESULTS: The results of the structural equation analysis provide further support for two dimensions underlying the DSM-IV symptom items. Although the two-factor dimension bore a strong resemblance to the DSM-IV conceptions of abuse and dependence, there were notable differences in the item content of the symptom items for each dimension. The dependence dimension drew upon items related to the abuse criteria for continued drinking despite social problems and recurrent drinking resulting in failure to fulfill role obligations. The abuse dimension drew upon items related to the abuse criterion for hazardous drinking and the dependence criterion for larger amounts over time. The two factors were shown to have different relationships to the background variables. Alcohol dependence was related to family history of alcoholism and educational status. Age was not related to dependence and inversely related to alcohol abuse. CONCLUSIONS: Findings from this study replicate the two-dimensional model for DSM-IV criteria found in other studies and provide further support for the validity of alcohol dependence in general population samples. A major implication of the factor structure in the present study relates to the different classification of cases that would otherwise be obtained with DSM-IV criteria. These departures were shown to affect abuse, which retained only 40% of DSM-IV diagnoses, more strongly than dependence, which retained 91% of DSM-IV diagnoses.

Adolescent↗
PubMed1990

[Multivariate analysis of meteorological factors and evaluation of circadian rhythm: their relation to the occurrence of acute myocardial infarction].

We statistically analysed the occurrence of acute myocardial infarction (AMI) in relation to meteorology and circadian rhythm. The study subjects included 581 patients with AMI in Asahikawa observed for the last 10 years, and 177 in Yamagata observed for the past three years. Monthly fluctuations of AMI occurrence were not statistically significant. The expected frequency of AMI estimated by grading meteorological factors every 5 degrees C from maximum to minimum temperatures did not correlate with the real frequency. A relatively cold period was defined as a term during which mean atmospheric temperature was below 0 degrees C in Asahikawa and below 3 degrees C in Yamagata Canonical discriminatory analyses were made of 10 meterological factors between days with and without AMI occurrences during a 10-year term in Asahikawa; and between days with and without outdoor occurrences. To compare regional difference, the same analysis was performed during a three-year cold period in Yamagata. Correlations among the three analyses were negligible (0.03%, 1.53% and 0.97%, respectively); thus, the days on which AMI occurred could not be identified with the 10 meteorological factors. Circadian rhythm of two cycles/day was recognized concerning the time of occurrence by power spectral analysis in 562 patients in Asahikawa, in whom the time of onset of AMI was recorded. It was concluded that there are no external meteorological risk factors for the occurrence of AMI. However, the intrinsic biological rhythm supposedly participates in triggering the occurrence of AMI as an intrinsic transient risk factor.

Adult↗
PubMed1993

Multivariate analysis of progression in superficial bladder cancer.

A total of 217 patients with superficial bladder cancer (SBC) category Ta-T1 were retrospectively analysed for tumour progression and prognostic factors; 46% of patients presented with recurrences, 13% with increasing grades of anaplasia and 12% with progression in tumour category. The mean annual index of recurrences was 0.7 +/- 0.1; 4% of patients died from tumour progression and the overall 5-year survival rate was 88% +/- 5.3. Factors that influenced progression significantly were peritumoural dysplasia, inflammatory infiltrate, residual tumour following transurethral resections, increasing grade of anaplasia and index of recurrences > 0.7. Multivariate statistical analysis revealed that only tumour category and increasing grade of anaplasia had significant prognostic value. Increasing grade of anaplasia predicts the likelihood of tumour progression with a 91% global predictive capacity. Thirty percent of patients with tumour progression did not have an increasing grade of anaplasia. Among these 87% were Ta, 50% were GI and the mean index of recurrences was 0.9. All were conservatively treated and 87% are alive. Among the patients with an increasing grade of anaplasia and tumour progression the mean index of recurrences was 1.7; 67% developed muscle infiltrating cancer and 7% metastatic spread without local progression; 64% of these patients died from cancer. An increasing grade of anaplasia affects prognosis. It defines patients with a high risk of tumour progression and tumour-related death, especially those with T1 bladder cancer.

Adolescent↗
PubMed1992

Multivariate analysis of serum tumor markers for diagnosis of skeletal metastases.

Serum tumor markers, including carcinoembryonic antigen (CEA), alpha-fetoprotein (AFP), carbohydrate antigen 15-3 (CA 15-3), carbohydrate antigen 19-9 (CA 19-9), cancer antigen 125 (CA 125), and tissue polypeptide antigen (TPA), were measured in 26 patients with skeletal metastases and 11 patients with primary malignant bone tumors. TPA, which was elevated in 16 patients (61.5%), was the most sensitive marker for detection of skeletal metastases. Combined measurement of these markers was useful in detecting skeletal metastases from primary lesions, although tumor markers had little organ specificity. In addition, skeletal metastases could be completely differentiated from primary lesions by the use of multivariate discriminant analysis of markers. The most and least powerful discriminating factors were AFP and CA 19-9, respectively. On multidimensional scaling, the distance between AFP and CEA was longest, with the other markers scattered between them. Expression of individual markers can not be linked to that of other markers.

Antigens, Neoplasm↗
PubMedDate not supplied

Heritability of somatotype components from early adolescence into young adulthood: a multivariate analysis on a longitudinal twin study.

BACKGROUND: Several studies with different designs have attempted to estimate the heritability of somatotype components. However they often ignore the covariation between the three components as well as possible sex and age effects. Shared environmental factors are not always controlled for. AIM: This study explores the pattern of genetic and environmental determination of the variation in Heath-Carter somatotype components from early adolescence into young adulthood. SUBJECTS AND METHODS: Data from the Leuven Longitudinal Twin Study, a longitudinal sample of Belgian same-aged twins followed from 10 to 18 years (n = 105 pairs, equally divided over five zygosity groups), is entered into a multivariate path analysis. Thus the covariation between the somatotype components is taken into account, gender heterogeneity can be tested, common environmental influences can be distinguished from genetic effects and age effects are controlled for. RESULTS: Heritability estimates from 10 to 18 years range from 0.21 to 0.88, 0.46 to 0.76 and 0.16 to 0.73 for endomorphy, mesomorphy and ectomorphy in boys. In girls, heritability estimates range from 0.76 to 0.89, 0.36 to 0.57 and 0.57 to 0.76 for the respective somatotype components. Sex differences are significant from 14 years onwards. More than half of the variance in all somatotype components for both sexes at all time points is explained by factors the three components have in common. CONCLUSIONS: The finding of substantial genetic influence on the variability of somatotype components is further supported. The need to consider somatotype as a whole is stressed as well as the need for sex- and perhaps age-specific analyses. Further multivariate analyses are needed to confirm the present findings.

Adolescent↗
PubMed1995

[Surgical risk assessment using multivariate analysis].

INTRODUCTION: Over and above the risk of a disease itself, surgical treatment implies additional risk involving possible temporary or permanent deterioration in a patient's health. Most studies of anesthetic-surgical risk are based on mortality or serious morbidity (severe, permanent deterioration). They may therefore be of limited applicability in hospitals where such outcomes are rare due to the types of patients treated. In order to compare and classify patients as well as improve outcome in such health care settings other parameters must be evaluated. PATIENTS AND METHODS: Indicators presaging adverse situations related directly to surgery in our postoperative intensive care units involve blood transfusion and perioperative events. By applying multivariate (regression) analysis we can determine the factors that condition these unfavorable situations. RESULTS: In the setting we studied, the main determinants of unfavorable events after surgery are patient health as shown by ASA classification, duration of operation and type of surgery. CONCLUSIONS: 1) Peripheral surgery in patients in good health is very safe. 2) Easily obtained parameters can be used to obtain a fairly efficient prediction of risk. 3) Single-setting risk studies are easy to carry out and useful for providing information about a specific hospital center.

Adolescent↗
PubMed2006

What are the most powerful determinants of endoscopic vesicoureteral reflux correction? Multivariate analysis of a single institution experience during 6 years.

PURPOSE: As the indications for endoscopic correction of vesicoureteral reflux continue to expand, the emergence of potential predictive variables has been noted. We used univariate and multivariate statistical analyses to find the most significant predictors of correction to improve patient selection. MATERIALS AND METHODS: A consecutive series of patients treated at a single institution was reviewed. Between August 1998 and August 2004, 232 children endoscopically injected with polydimethylsiloxane were identified, representing 351 refluxing units. A total of 23 variables were subjected to statistical analysis to detect predictors of reflux correction after injection. All identified patients with complete data and followup evaluations were included irrespective of anatomical variations, previous interventions or comorbidities. RESULTS: The overall success rate by patient and renal unit was 65% and 72%, respectively. In patients with a single system low grade (1-3) vesicoureteral reflux who did not previously undergo injection this success rate increased to 80%. Univariate analysis demonstrated that higher physician experience, low preoperative vesicoureteral reflux grade, absent renal scars and no previous injections were statistically significant predictors of vesicoureteral reflux correction (p <0.05). A history of febrile urinary tract infections and a duplex system did not attain significance (p = 0.069 and 0.076, respectively). On multivariate statistical evaluation only physician experience, preoperative vesicoureteral reflux grade and the number of previous injections remained significant. CONCLUSIONS: Multivariate analysis of our data showed the most important determinants of vesicoureteral reflux correction after endoscopic injection. Prospective validation will allow us to generate nomograms to better select and counsel patients who would benefit from vesicoureteral reflux treatment.

Child↗