Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Multivariate statistics”

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 847 records · Page 47Linked to original sources
PubMed1992

Effect of human cervical mucus on human sperm motion and hyperactivation in vitro.

The aim of our experiment was to examine the effect of exposure to human cervical mucus on quantitative sperm motility with specific reference to hyperactivated sperm motility. Human spermatozoa were allowed to penetrate cervical mucus for 20 min before swimming into Earle's balanced salt solution tissue culture medium for 25 min. The sperm motion characteristics were compared to those which had been obtained from a direct swim-up for 45 min. Spermatozoa treated with mucus were more 'active' than the control group. Multivariate statistical analysis indicated that cervical mucus promotes hyperactivated motility and that sperm sub-populations exposed to cervical mucus are very heterogeneous, as indicated by the numbers and motility characteristics of spermatozoa.

Cervix Mucus↗
PubMed1999

Serum transferrin receptor levels are increased in asymptomatic and mild Plasmodium falciparum-infection.

BACKGROUND AND OBJECTIVE: The serum transferrin receptor (sTfR) concentration in an individual reflects the extent of erythropoietic activity and is considered a useful marker of iron deficiency independent of concurrent inflammation or infection. However, data on the impact of malaria on this parameter are ambiguous. We have examined potential associations of asymptomatic and mild Plasmodium falciparum-infections and of several erythrocyte variants with sTfR values in South West Nigeria. DESIGN AND METHODS: In a cross-sectional study among 161 non-hospitalized children, sTfR concentrations and P. falciparum parasitemia were assessed. In addition, hemoglobin (Hb) and serum ferritin values, Hb-types, glucose-6-phosphate dehydrogenase (G6PD)deficiency and a-globin genotypes were determined and the effects of these factors on sTfR levels were analyzed by univariate and multivariate statistical methods. RESULTS: P. falciparum-infection was present in 77% of the children. Mean sTfR levels were higher in infected than in non-infected children (geometric mean, 3.68, 95% confidence interval [3.5-3.9] vs. 2.99 [2.7-3.3] mg/L; p = 0.0009). There was a significant trend for higher sTfR values with increasing parasite density. sTfR values decreased continuously with age. Hb-types, G6PD-, and a-globin genotypes did not correlate with sTfR levels. In the multivariate analysis, age, Hb and log ferritin values, and parasite density of P. falciparum were independently associated with log sTfR values. INTERPRETATION AND CONCLUSIONS: sTfR concentrations are increased in asymptomatic and mild P. falciparum-infections suggesting adequate bone marrow response in this condition. The diagnostic value of sTfR levels for iron deficiency may be impaired in areas where stable malaria occurs.

Age Factors↗
PubMed1994

Identification of the larvae of the Simulium damnosum complex from Nigeria by analysis of cuticular hydrocarbons.

The larvae of three sibling species of the Simulium damnosum complex were studied with a view to identification, using a Gas Liquid Chromatograph to analyze the hydrocarbons in the cuticular wax extracts. Multivariate statistics were applied to the chromatographic data to differentiate three sibling species: S. damnosum s.s., S. squamosum and S. yahense. Paired comparisons between the species gave greater than 96% correct classifications in all cases.

Animals↗
PubMed1984

Quantitative histological study of duodenitis in biopsies.

In 48 patients, suffering from an endoscopically verified ulcerative or nonulcerative duodenitis, biopsies of the bulb and of the lower duodenum have been studied with quantitative histological methods. The evaluation of the data with simple and multivariate statistical methods revealed: In healthy duodenal mucosa the height of the villous epithelium significantly decreases with rising age, while the cellular infiltration of the stroma increases. Conclusive, sex dependent differences are not present. Many histological variables are significantly correlated to the endoscopical degree of inflammation, if the endoscopically normal mucosa of the lower duodenum is used as normal control. In contrast, the endoscopically normal bulb mucosa histologically showed inflammatory alterations in the majority of cases. As shown by a multivariate discriminant analysis, a correct histological diagnosis of duodenitis will be reached in nearly all patients, if a biopsy of the endoscopically normal lower duodenum of the same patients is used as matched control. As shown by a multiple stepwise regression analysis, the same procedure results in a good correspondence of the endoscopically estimated degree of inflammation and the values, calculated from histology. In some patients, the lower duodenum slightly takes part in the inflammation of the bulb. But no significant correlation was found between the calculated degree of inflammation of the bulb and of the lower duodenum, neither in ulcerative duodenitis nor in nonulcerative duodenitis. From a practical point of view, our findings elucidate, that the simultaneous histological examination of mucosa of the bulb and of endoscopically healthy mucosa of the lower duodenum will be an essential help in the diagnosis and grading of duodenitis and should be introduced as routine method.

Adult↗
PubMed2002

Exercise testing scores as an example of better decisions through science.

INTRODUCTION: The application of common statistical techniques to clinical and exercise test data has the potential to become a useful tool for assisting in the diagnosis of coronary artery disease, assessing prognosis, and reducing the cost of evaluating patients with suspected coronary disease. Since general practitioners function as gatekeepers and decide which patients must be referred to the cardiologist, they need to optimally use the basic tools they have available (i.e., history, physical exam, and the exercise test). METHODS: Review of the literature with a focus on the scientific techniques for aiding the decision-making process. RESULTS: Scores derived from multivariable statistical techniques considering clinical and exercise data have demonstrated superior discriminating power when compared using receiver-operating-characteristic curves with the ST segment response. In addition, by stratifying patients as to probability of disease and prognosis, they provide a management strategy. While computers as part of information management systems can calculate complicated equations to provide scores, physicians are reluctant to trust them. Thus, these scores have been represented as nomograms or simple additive tables so physicians are comfortable with their application. Scores have also been compared with physician judgment and been found to estimate the presence of coronary disease and prognosis as well as expert cardiologists, and often better than nonspecialists. CONCLUSION: Multivariate scores can empower the clinician to assure the cardiac patient with access to appropriate and cost-effective cardiological care.

Adult↗
PubMed2005

Early predictive variables for upper canine impaction as derived from posteroanterior cephalograms.

The aim of this study was to identify a model of cephalometric variables for early diagnosis of the displacement of upper permanent canine and early prognosis of upper canine impaction. Postero-anterior (PA) cephalograms of 43 subjects (22 males and 21 females) with different types of malocclusions in the mixed dentition were analyzed at the time of first observation (mean age 8 years and 5 +/- 9 months). All subjects were reevaluated at the mean age of 14.3 years. At this time the sample was divided into 2 groups according to the presence of canine impaction: nonimpacted canine group (31 subjects) and impacted canine group (12 subjects). Stepwise variable selection on the measurements at the time of first observation identified 2 predictive variables on PA cephalograms, ie, the distance between the center of the canine crown and the midsagittal plane (A3cc to Cg Vertical) and the distance between the jugal process and the midsagittal plane (J to Cg Vertical). The closer the canine crowns to the midsagittal plane and the larger the posterior portion of the hemimaxilla, the higher the probability of canine impaction. Discriminant analysis assigned a classification power of 95.3% to the predictive model. On the basis of the equation generated by the multivariate statistical method, impaction of the maxillary canine for each new case at the age of 8 years can be predicted. The importance of PA cephalograms in the diagnosis and prediction of canine eruption disturbances is emphasized.

Cephalometry↗
PubMed1995

LDL subclass phenotypes and the risk factors of the insulin resistance syndrome.

The insulin resistance syndrome is characterized by a constellation of risk factors including obesity, central body fat distribution, hypertension, glucose intolerance, elevated plasma insulin levels, increased triglyceride and decreased high-density lipoprotein (HDL) cholesterol. Similarly, low-density lipoprotein (LDL) subclass phenotype B, characterized by a predominance of small, dense LDL particles, is associated with increased triglyceride and lower HDL cholesterol levels. Both the insulin resistance syndrome and phenotype B have also been related to increased risk of coronary heart disease. Using a sample of nearly 700 women who participated in the Kaiser Permanente Women Twins Study, we investigated the associations of LDL subclass phenotype B with the risk factors that characterize the insulin resistance syndrome. The results demonstrate that LDL subclass phenotype B was strongly associated with both age and diabetes status. Among nondiabetic women, phenotype B was more prevalent in those with higher body mass index and in those with higher waist-hip ratio. In addition to the expected associations with triglyceride and HDL cholesterol, mean values of both systolic and diastolic blood pressure were higher in women with phenotype B. Finally, fasting and post-load insulin levels and post-load glucose levels were higher in those with phenotype B. In general, these relationships persisted in multivariate statistical analyses. Therefore, a predominance of small, dense LDL particles appears to be an integral feature of the risk factors that characterize the insulin resistance syndrome.

Adult↗
PubMed1996

Criteria for generalized and focal osteoarthritis.

STUDY OBJECTIVE: to identify homogeneous, distinct subgroups of osteoarthritis patients based on distribution of osteoarthritis lesions. PATIENTS-METHODS STUDY DESIGN: cross-sectional, prospective, multicenter. Patients with osteoarthritis of the hip, knee, fingers, or spine. DATA COLLECTED: clinical and radiologic features at 41 joint sites. METHODS: multivariate statistical analysis including a) k-mean clustering analysis followed by ascending hierarchical classification b) and a tree-structured discriminant method to confirm and to characterize the subgroups obtained using the clustering method. RESULTS: The 1021 patients were first classified into five categories with an error rate (obtained by cross-validation) of 7.6%. The tree obtained by segmentation took into account manifestations of osteoarthritis at the hands, knees and spine. Irrespective of the reason for seeking medical advice, patients with bilateral involvement of the fingers or with involvement of the spine and both femorotibial joints were classified as having generalized osteoarthritis; in contrast, focal osteoarthritis was defined as the absence of involvement of the fingers and of either the spine or the knees. CONCLUSION: The statistical analysis provided a classification system that would be easy to use in everyday clinical practice. Prospective studies are needed to evaluate the potential clinical relevance of this system.

Cluster Analysis↗
PubMed2002

Statistical process control for large scale microarray experiments.

MOTIVATION: Maintaining and controlling data quality is a key problem in large scale microarray studies. In particular systematic changes in experimental conditions across multiple chips can seriously affect quality and even lead to false biological conclusions. Traditionally the influence of these effects can be minimized only by expensive repeated measurements, because a detailed understanding of all process relevant parameters seems impossible. RESULTS: We introduce a novel method for microarray process control that estimates quality based solely on the distribution of the actual measurements without requiring repeated experiments. A robust version of principle component analysis detects single outlier microarrays and thereby enables the use of techniques from multivariate statistical process control. In particular, the T(2) control chart reliably tracks undesired changes in process relevant parameters. This can be used to improve the microarray process itself, limits necessary repetitions to only affected samples and therefore maintains quality in a cost effective way. We prove the power of the approach on 3 large sets of DNA methylation microarray data.

Algorithms↗
PubMed2004

ROC analysis of an erythroblast morphologic scoring system to improve identification of fetal cells in maternal blood.

OBJECTIVES: Nucleated red blood cells are used in research settings as a target cell type for investigations of noninvasive prenatal diagnosis, and these cells have a characteristic nuclear morphology and hemoglobin staining pattern that makes them distinguishable from maternal cells. Recently, we developed an erythroblast scoring system based on these characteristics. Here, we employ statistical analyses to further characterize the utility of this scoring system. METHODS: A total of 170 nucleated red blood cells isolated from peripheral blood of four women undergoing elective termination of a trisomy 21 male fetus were analyzed. Each of the four scoring system parameters, whose values range from 0 to 3 points, served as an independent variable to determine its significance for the correct identification of a cell as fetal or maternal. A logistic regression was used as a multivariable statistical tool. RESULTS: Forty-four patterns of the four parameters were found in the overall series. Some patterns were exclusively associated with maternal cells (e.g. 1-1-1-1 and 1-1-2-2), and others were exclusively associated with fetal cells (e.g. 3-2-2-2 and 3-2-3-2). A detection rate of 73.9% at a false-positive rate of 5% resulted from a random simulation model performed with a 1:5 case:control matched set. The variables most predictive of a cell being fetal were gamma hemoglobin staining intensity of cytoplasm and nuclear roundness. CONCLUSIONS: The modified scoring system presented here improves upon the previously reported, unmodified erythroblast scoring system. These statistical analyses suggest that the scoring system is a promising method that aids in distinguishing fetal and maternal NRBCs for prenatal diagnostic applications and that it may be amenable to automated microscopy by applying the discrete morphological parameters as computational classifiers.

Cell Nucleus↗
PubMed2003

Multivariate analyses of invertebrate community responses to a C12-15 AE-3S anionic surfactant in stream mesocosms.

The ecological effects of a C(12-15)AE-3S linear alkylethoxysulfate anionic surfactant on aquatic organisms were studied in stream mesocosms. Analyses of the experimental data have yielded multiple population-level no-observed effect concentrations (NOECs) but did not provide a measure of community-level response to surfactant exposure. In the present study, the invertebrate data from the mesocosm study were further examined for community-level effects using multivariate techniques. Two multivariate statistical methods were selected as representative of recommended methods and used to analyze invertebrate abundance. One method was similarity analysis using the Bray-Curtis index, and the second method was the canonical ordination technique principal response curves recently developed for analysis of mesocosm data. Both sets of techniques detected statistically significant changes in invertebrate community structure following surfactant exposure. Ordinations constructed from Bray-Curtis similarity matrices provided clearer two-dimensional representations of these changes than ordinations from redundancy analysis. Re-displaying treatment and temporal effects on community structure on separate axes as response curves provided a clearer method for interpreting community response to surfactant exposure. Both approaches detected treatment effects in low and high abundance taxa. The highest rank abundance taxa in the pretreatment streams contributed very little to the differences observed between the control and treated streams. The same set of taxa was identified as the dominant drivers for structural differences between control and treated communities despite underlying mathematical differences in these two methods. Invertebrate community-level NOECs estimated from these different multivariate methods (2.5-5.0 mg/l) were comparable and these NOECs were similar to invertebrate population NOECs determined for this surfactant.

Alkanesulfonic Acids↗
PubMed2002

Risk factors for the formation of a steinstrasse after extracorporeal shock wave lithotripsy: a statistical model.

PURPOSE: We studied the various stone, renal and therapy factors that could affect steinstrasse formation after extracorporeal shock wave lithotripsy (ESWL), Dornier Medical Systems Inc., Marietta, Georgia to define the predictive factors for its formation. Thus, steinstrasse could be anticipated and prophylactically avoided. MATERIALS AND METHODS: Between February 1989 and May 1999, 4,634 patients were treated with a Dornier MFL 5000 lithotriptor (Dornier Med Tech, GmbH, Germany). Renal stones were encountered in 3,403 patients and ureteral stones in 1,231. Steinstrasse were recorded in 184 patients, of whom 74 required intervention and formed the "complicated group." All patient data, stone and renal characteristics, and data of ESWL were reviewed. Univariate and multivariate statistical analyses of patients, stones and therapy characteristics in correlation with the incidence of steinstrasse formation were performed to assign the factors that had a significant impact on steinstrasse formation. RESULTS: The overall incidence of steinstrasse was 3.97%. The steinstrasse was in the pelvic ureter in 74% of the cases, lumbar ureter in 21.7% and iliac ureter in 4.3%. Steinstrasse incidence significantly correlated with stone size and site, the power level (kV.) used during therapy and radiological renal features. Steinstrasse was more common with renal stones more than 2 cm. in diameter in a dilated system, especially with the use of high power (greater than 22 kV.) for disintegration. A statistical model was constructed to estimate the risk of steinstrasse formation accurately. CONCLUSIONS: Stone size and site, renal morphology and shock wave energy are the significant predictive factors controlling steinstrasse formation. If a patient has a high probability of steinstrasse formation, close followup with early intervention or prophylactic pre-ESWL ureteral stenting is indicated.

Adolescent↗
PubMed1999

Novel indications for BRCA1 screening using individual clinical and morphological features.

Since there is a lack of common family profile among BRCA1-gene carriers, and since the risk of being a mutation carrier is not limited to women with a family history of breast or ovarian cancer, multivariate statistical analysis using the logistic-regression model was carried out, to discriminate between sporadic cases and BRCA1-breast cancers (BRCA1-BCs), especially when information about the family history of breast/ovarian cancer and ethnicity are irrelevant or unavailable, in order to offer specific medical treatment to this population. We examined 32 BRCA1-BCs selected at cancer genetic clinics and 200 consecutive controls without family history of breast cancer for age at onset and current morphological parameters. Following the multivariate analysis, 3 parameters only, namely, early age at cancer onset [odds ratio (OR) for each year = 1.16; p < 0.0001], estrogen-receptor negativity (OR = 5.7; p = 0.01) and poor differentiation (OR = 5; p = 0.03) were found significant factors for predicting BRCA1-carrier status. The expected impact in BRCA1 screening of our model was estimated using data on 5700 breast-cancer cases from a hospital-based registry. Only 50 and 15% of tumours with early age at onset below 35 years present one or the other 2 discriminant parameters respectively. Consequently, whereas the probability of finding a BRCA1 mutation is rated low (6.2%) when the sole criterion of early onset up to the age of 35 years is used, based on our model, in the sub-group of women with a tumor that is both estrogen-receptor-negative and poorly differentiated the mutation-detection rate is predicted to be above the 10% chance level recommended by the ASCO guidelines. This sub-group of women, representing about 1% of all breast-cancer cases in Western countries, consequently deserves to be tested.

Adult↗
PubMedDate not supplied

A multivariate analysis of HIV-1 protease inhibitors and resistance induced by mutation.

This paper describes the use of the multivariate statistical procedure principal component analysis as a tool to explore the inhibitory activity of classes of protease inhibitors (PIs) against HIV-1 viruses (wild type and more-frequent single mutants, V82A, V82F, and I84V) and against protease enzymes. The analysis of correlations between biological activity and molecular descriptors or similarity indexes allowed a reliable classification of the 51 derivatives considered in this study. The best results were obtained in the case of the I84V mutant for which a high number of predictions was achieved. On this basis, this statistical approach is proposed as a reliable method for the prediction of the activity of PIs, for which the data against mutant strains have not been reported.

Drug Resistance, Viral↗
PubMedDate not supplied

Risk factors associated with early respiratory disease complex in broiler chickens.

Early respiratory disease complex (ERDC) is a term coined to describe an acute disease characterized by depression, respiratory distress, and increased mortality in 2-to-3-wk-old broiler chickens. Postmortem lesions include airsacculitis, fibrinous pericarditis, and perihepatitis. Colisepticemia is the primary cause of death. In order to investigate the association between ERDC and farm management factors, a retrospective case-control study was conducted by collecting data covering a 6-mo period (January-June 1997) from four broiler integrators on the Delmarva peninsula. Univariate and multivariate statistical analyses revealed that flock size was positively associated (P = 0.02) and layout time was negatively associated (P = 0.05) with ERDC.

Acute Disease↗
PubMedDate not supplied

Identifying the determinants of viable microorganisms in the air and bulk metalworking fluids.

Exposure assessment was conducted for an epidemiologic study of the respiratory effects of exposure to metalworking fluids (MWF). As part of the study, airborne microorganisms were collected with a two-stage microbial impactor, and a sample of the bulk soluble MWF was collected from each machine sump, as well as information about the work environment. These data were then used to develop multivariate statistical models of the determinants bulk MWF and airborne microbial levels. Microbial concentrations in the bulk MWF ranged from 5 x 10(4) to 5 x 10(10) colony-forming units (CFU)/mL, with a geometric mean of 3.4 x 10(7) CFU/mL. The geometric mean airborne microbial level was 182 CFU/m3 (for particles size <8 microm) with a range of 1 to 8,308 CFU/m3. In modeling the determinants of bulk microorganisms, fluid-related factors were the most important characteristics associated with microbial levels, followed by process-related and environmental factors. The final full multivariate model predicted a significant reduction in bulk microbial levels by increasing pH of the fluid and reducing the amount of tramp oil leaking into the fluid. For the airborne microbial models, process-related factors were the major characteristics associated with microbial levels, followed by factors related to worker activities and environmental factors. The final full multivariate model predicted a significant control of airborne microorganisms by increasing worker distance from the machine, reducing the number of machines within 10 feet of the worker, decreasing the bulk microbial levels, and adding machine enclosures. These models can be used to prioritize nonbiocidal interventions to control microbial contamination of the bulk MWF and the air.

Air Microbiology↗
PubMedDate not supplied

Disease activity in systemic lupus erythematosus: report of the Consensus Study Group of the European Workshop for Rheumatology Research. II. Identification of the variables indicative of disease activity and their use in the development of an activity score. The European Consensus Study Group for Disease Activity in SLE.

A European Consensus Group study, involving 29 centres from 14 countries, was performed in order to reach agreement on the definition of disease activity in systemic lupus erythematosus (SLE) and to construct a new disease index. Data on 704 lupus patients were collected and analysed, using univariate and multivariate statistical procedures, to select those clinical and laboratory features of the disorder which best correlate with the global assessment of disease activity assigned to the patients by the physician of each participating centre. A combination of 15 clinical and laboratory variables was shown to be the best predictor of disease activity in SLE. A European Consensus Lupus Activity Measurement (ECLAM) was then formulated. This index included the 15 selected variables, weighted (with some adjustments) according to their respective regression coefficients in the multivariate model. ECLAM appears to be an effective instrument for scoring patients with different degrees of disease activity. This is the first SLE disease activity index based on data from a very large number of lupus patients followed at a large number of lupus centres in different countries. It might therefore very well serve as a standardised measure for future European clinical studies. Final assessment of the validity, reliability and sensitivity of this index is now underway.

Europe↗
PubMed1993

Immunohistochemical analysis of proliferating cell nuclear antigen (PCNA) in infiltrating ductal carcinomas: comparison with clinical and pathologic variables.

PC10, a monoclonal antibody directed against proliferating cell nuclear antigen (PCNA), was used to study the percentage of cells and range of immunopositivity in formalin-fixed paraffin-embedded tissues of 39 infiltrating ductal carcinomas. The percentage of immunostained nuclei was compared to the results of seven clinical and pathologic variables and also to each of the three components of tumor grade. The range of percent positivity was 3% to 85% (median: 36%). There was no statistically significant correlation between percent PCNA positivity and patient age, axillary lymph node status, estrogen or progesterone receptor levels, ploidy, tubule formation, or nuclear pleomorphism. Significant correlations were seen between percent PCNA positivity and tumor size (p = 0.03), grade (p = 0.0005), and mitotic rate (p = 0.0001). These data suggest that, in studies examining PCNA immunoreactivity as a potentially important prognostic parameter for patients with infiltrating ductal carcinoma, tumor size and either grade or mitotic count must be considered as variables in the multivariate statistical analysis.

Antibodies, Monoclonal↗