PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “Regression 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 253 records · Page 14Linked to original sources
PubMed1997

Characterization of sonographically indeterminate ovarian tumors with MR imaging. A logistic regression analysis.

PURPOSE: The goal of this study was to maximize the discrimination between benign and malignant masses in patients with sonographically indeterminate ovarian lesions by means of unenhanced and contrast-enhanced MR imaging, and to develop a computer-assisted diagnosis system. MATERIAL AND METHODS: Findings in precontrast and Gd-DTPA contrast-enhanced MR images of 104 patients with 115 sonographically indeterminate ovarian masses were analyzed, and the results were correlated with histopathological findings. Of 115 lesions, 65 were benign (23 cystadenomas, 13 complex cysts, 11 teratomas, 6 fibrothecomas, 12 others) and 50 were malignant (32 ovarian carcinomas, 7 metastatic tumors of the ovary, 4 carcinomas of the fallopian tubes, 7 others). A logistic regression analysis was performed to discriminate between benign and malignant lesions, and a model of a computer-assisted diagnosis was developed. This model was prospectively tested in 75 cases of ovarian tumors found at other institutions. RESULTS: From the univariate analysis, the following parameters were selected as significant for predicting malignancy (p< or =0.05): a solid or cystic mass with a large solid component or wall thickness greater than 3 mm; complex internal architecture; ascites; and bilaterality. Based on these parameters, a model of a computer-assisted diagnosis system was developed with the logistic regression analysis. To distinguish benign from malignant lesions, the maximum cut-off point was obtained between 0.47 and 0.51. In a prospective application of this model, 87% of the lesions were accurately identified as benign or malignant. CONCLUSION: Benign and malignant ovarian lesions can be distinguished in most sonographically indeterminate lesions by means of parameters obtained from contrast-enhanced MR imaging.

Contrast Media↗
PubMed1993

A multiple logistic regression analysis of the risk and relative odds of temporomandibular disorders as a function of common occlusal features.

A multiple logistic regression analysis was used to compute the odds ratios for 11 common occlusal features for asymptomatic controls (n = 147) vs. five temporomandibular disorder groups: Disc Displacement with Reduction (n = 81), Disc Displacement without Reduction (n = 48), Osteoarthrosis with Disc Displacement History (n = 75), Primary Osteoarthrosis (n = 85), and Myalgia Only (n = 124). Features that did not contribute included: retruded contact position (RCP) to intercuspal position (ICP) occlusal slides < or = 2 mm, slide asymmetry, unilateral RCP contacts, deep overbite, minimal overjet, dental midline discrepancies, < or = 4 missing teeth, and maxillo-mandibular first molar relationship or cross-arch asymmetry. Groupings of a minimum of two to at most five occlusal variables contributed to the TMD patient groups. Significant increases in risk occurred selectively with anterior open bite (p < 0.01), unilateral maxillary lingual crossbite (p < 0.05 to p < 0.01), overjets > 6-7 mm (p < 0.05 to p < 0.01), > or > 5-6 missing posterior teeth (p < 0.05 to p < 0.01), and RCP-ICP slides > 2 mm (p < 0.05 to p < 0.01). While the contribution of occlusion to the disease groups was not zero, most of the variation in each disease population was not explained by occlusal parameters. Thus, occlusion cannot be considered the unique or dominant factor in defining TMD populations. Certain features such as anterior open bite in osteoarthrosis patients were considered to be a consequence of rather than etiological factors for the disorder.

Age Factors↗
PubMed1999

Multivariate regression analysis of the influence of aortic pressure, end-diastolic pressure, and heart rate on left ventricular relaxation in isolated ejecting rat and guinea pig hearts.

The effect of moderate changes of peak aortic and end-diastolic pressure and of heart rate on the left ventricular relaxation of isolated working rat and guinea pig hearts was investigated by multivariate regression analysis. Each of these three independent variables was set to three different levels, yielding 27 sets of data by combination in each experiment. Relaxation was quantified by the maximum pressure fall velocity (min LVdP/dt) and by the time constant tau of left ventricular isovolumic pressure fall. tau was obtained by fitting that pressure curve to the three-parametric exponential regression model p(t) = P infinity + (P0-P infinity) exp (-t/tau) and to an extended four-parametric model p(t) = P infinity + (P0-P infinity) exp [(-t/(tau 0 + r tau t)]. The influence of the three independent variables on min LVdP/dt, tau, and r tau was checked by analysis of variance and quantified by standardized regression coefficients obtained by trivariate regression analysis. A positive dependence of min LVdP/dt on precedent maximum pressure and of the three-dimensionally (but not four-dimensionally) estimated tau on beat interval length are the only unequivocally significant effects of the investigated hemodynamic changes on the lusitropic parameters min LVdP/dt and tau. Principal species differences do not occur. It is concluded that considerable lusitropic effects seen especially in pharmacologic studies must be attributed to intrinsic effects of the substance rather than to hemodynamic changes caused by the substance if the former remain moderate. The four-dimensionally calculated tau is slightly more often found to be independent of the investigated hemodynamic parameters than the usual three-dimensional estimate. This may indicate higher reliability of the four-dimensional regression model of isovolumic pressure fall.

Animals↗
PubMed1999

[Regression analysis in the assessment of the state of the immune system].

Optimal assessment of the immune status is one of the key problems in modern clinical immunology. Mathematical methods of diagnosis help select the minimal number of the most informative tests. Regression analysis of linear relationship permitted determining the regression coefficients and correlation pairs of equations which may be used for predicting, with few reference monoclonal sera (CD3 and CD5), the lacking cellular phenotype in immunogram. The resultant correlation pairs of equations were tried in a large number of examinees at laboratory of ecological immunology and are recommended for clinical diagnostical laboratories.

Antigens, CD↗
PubMed1990

Laparoscopic findings of fatty liver using multiple regression analysis. "Spotty" and "diffuse" yellow color.

Laparoscopic features of fatty liver including the cases with fibrotic change were studied by multiple regression analysis. The coloration of the liver surface under laparoscopy was classified into three types: "spotty" yellow color (YC) type, "diffuse" YC type, and without YC type. The presence of YC on the liver surface depended on the amount of deposited lipid (standardized estimate = 0.606, p = 0.0001). However, the presence of YC and whether it appeared "spotty" or "diffuse" was independent of the intralobular distribution and size of the deposited lipid droplets. When YC was observed in cases without fibrotic change, 19 of the 23 with more than 30% lipid droplets in the biopsy specimens had "spotty" YC. When YC was found in cases with fibrotic change, all of the 14 cases with more than 30% lipid droplets were "diffuse" YC. A linear regression model showed that the yellow colored liver with fibrotic change had a "diffuse" appearance (yes = 1, no = 0) standardized estimate = 0.569, p = 0.0009). These data suggested that the amount of lipid deposited was responsible for the presence of YC on the liver surface both with and without fibrotic change, and that the co-existence of fibrotic change was a major factor for giving the yellow colored liver a "diffuse" appearance.

Biopsy↗
PubMed1993

Histological features of prognostic significance in CML--an immunohistochemical and morphometric study (multivariate regression analysis) on trephine biopsies of the bone marrow.

To evaluate the prognostic significance of clinical as well as histological disease features at the time of diagnosis, an immunohistochemical and morphometric study was performed on bone marrow trephine biopsies in 130 patients with Ph(1+)-CML. For identification of all cell elements of the megakaryocytopoiesis we used the monoclonal antibody CD61 (Y2/51) and for the macrophages, the recently characterized antibody PG-M1. Density of argyrophilic fibers was determined per fat cell-free marrow area. Based on a multivariate analysis-derived risk model, the reproducibility of the prognostic score described by Sokal and co-workers was tested, particularly with regard to histological variables. Additionally, we calculated the disease-specific loss in life expectancy. Our prognostic model (Cox model) consisted of the variables: age, spleen size, peripheral erythro-normoblasts, pseudo-Gaucher cells, and fiber density. To assess the validity of this new CML score, a receiver-operating curve (ROC) of sensitivity and specificity was constructed. The improved prognostic efficiency of this newly developed risk model in predicting death within 3 years after diagnosis of CML was demonstrated in comparison with generally accepted staging systems. Immunohistochemistry revealed that not the total number of macrophages, but only the subfraction of pseudo-Gaucher cells exerted a significant impact on survival. Furthermore, it was feasible to calculate the number of atypical micromegakaryocytes and pro- and megakaryoblasts. This abnormal and immature cell population showed a significant correlation with fiber density and prognosis. Finally, the practical value of the Hannover classification was tested. This histological classification enabled a discrimination between two groups with different survival patterns, i.e., granulocyte and/or megakaryocyte-rich subtypes versus subtypes with increase in reticulin and collagen fibers.

Adolescent↗
PubMed1992

Nonconditional logistic regression analysis of risk factors in rheumatic heart disease.

The prognostic risk factors in 4323 patients with rheumatic heart disease, admitted from 1970 to 1990, were analysed. The overall mortality was 31.75% in this group of patients. Single factor analysis indicated that cardiac functional classification, time of death, month of death, cardiothoracic ratio, valvular lesions, cardiogenic shock, digitalis-induced arrhythmias, intercurrent pneumonia, pleurorrhea, and hypotension were related to the overall mortality and cardiac death. Multiple factor logistic analysis indicated that for the overall mortality, the independent prognostic factors included presence of cardiac functional classification, cardiothoracic ratio and cardiogenic shock; for cardiac death, the independent factors included cardiac functional classification, cardiothoracic ratio, cardiogenic shock, digitalis-induced arrhythmias and valvular lesions. The data analysis showed that these five factors were contributory to rheumatic heart disease with synergism.

Adolescent↗
PubMed1989

[Multiple stepwise regression analysis of etiological factors of esophageal cancer in Cixian county].

Cixian county, one of the high-risk counties of esophageal cancer in the world, has a standardized mortality of 142.19/10(5) population, 1969-1971. The incidence of esophageal cancer had dropped year by year from 1974 to 1982. The significance of the incidence tendency was studied. The results are highly significant (P less than 0.001). The causative factors of esophageal cancer including five independent variables: X1 (number of people taking sanitized water), X2 (number of people on pickled Chinese cabbage), X3 (annual output of fruit), X4 (annual output of fresh vegetable) and X5 (annual output of sweet potato) and one dependent variable Y (morbidity of esophageal cancer) were studied by correlative analysis and multiple stepwise regression. Three correlative factors (X1, X2, and X5) with significant effect on the esophageal cancer were selected from the five suspected factors. The result indicated that taking sanitized water, reducing the number of people on pickled Chinese cabbage, changing the structure of food and keeping the nutrient balance, might decrease the incidence of esophageal cancer.

China↗
PubMed1986

[Regression analysis of clinical data for purposes of predicting the efficacy of psychopharmacotherapy of episodes of schizophrenia].

Validity of a multidimensional statistics procedure for the prediction of the psychotropic drugs efficacy (PDE) was tested in a study on 60 schizophrenics showing the signs of an acute condition before treatment. Pretreatment ratings of 78 psychopathological symptoms underwent a computer analysis. A multiple stepwise regression equation was obtained which predicted PDE using ratings of 15 symptoms. A high correlation between the predicted and actual PDE was shown at cross-validation (r = +0.87), as well as a higher proportion of the PDE variance explained by the "regression" predictor rather than traditional "syndromal" one or the former's combination with the diagnosed type of the course (42%, 57.4% and 71.9%, respectively). Some methodological problems (homogeneity of patients and treatment conditions, the adequacy of the mathematical model applied) are discussed.

Computers↗
PubMed2003

Detection of glaucomatous visual field changes using the Moorfields regression analysis of the Heidelberg retina tomograph.

PURPOSE: To evaluate the sensitivity and specificity of optic disk examinations performed using the Moorfields regression analysis (MRA) of the Heidelberg Retina Tomograph (HRT) in differentiating normal from glaucomatous eyes. DESIGN: Retrospective cross-sectional study. METHODS: Five hundred and nineteen patients were included in the study for a total of 193 normal eyes, 213 with suspected glaucoma (primary open-angle glaucoma [POAG]), and 113 with POAG. The intervention consisted of optic disk imaging by means of HRT I. A mean of three repeated images was analyzed using version 2.01 software. The optic disk was classified as "normal/outside normal limits" on the basis of the MRA. The visual field was examined using the DS 24 II program (Humphrey perimeter), with a glaucomatous visual field being defined on the basis of an abnormal Glaucoma Hemifield Test and a statistically significant corrected pattern standard deviation greater than 2 dB. The results obtained with MRA were compared with those obtained using the multivariate discriminant analysis (MDA) provided in the HRT I. The main outcome measures were sensitivity, specificity, and positive and negative predictive values of HRT examination. RESULTS: The sensitivity and specificity of the HRT-MRA examination were, respectively, 74% and 94% (83% and 75% with MDA) when the patients with suspected POAG were excluded from the analysis; the figures were 74% and 85%, and 41% and 94% (83% and 64%, and 60% and 75% with MDA) when the same patients were included as being normal or having POAG. CONCLUSIONS: In a broad clinical setting including normal subjects, patients with suspected POAG, and POAG patients, the HRT-MRA showed a high degree of diagnostic accuracy. The MRA was less sensitive but more specific than MDA. Greater diagnostic ability may be added by HRT-MRA examinations than by HRT-MDA to standard POAG diagnostic studies.

Adult↗
PubMed1999

Economic transition and changing relation between income inequality and mortality in Taiwan: regression analysis.

OBJECTIVE: To examine the changing relation between income inequality and mortality through different stages of economic development in Taiwan. DESIGN: Regression analysis of mortality on income inequality for three index years: 1976, 1985, and 1995. SETTING: 21 counties and cities in Taiwan. MAIN OUTCOME MEASURES: All age mortality and age specific mortality in children under age 5. RESULTS: When median household disposable income was controlled for, the association between income inequality and mortality became stronger in 1995 than in 1976. Especially, the association between income inequality and mortality in children aged under 5, with adjustment for differences in median household disposable income, changed from non-significant in 1976 to highly significant in 1995. In 1995, the level of household income after adjustment for income distribution no longer had a bearing on mortality in children under 5. CONCLUSION: The health of the population is affected more by relative income than by absolute income after a country has changed from a developing to a developed economy.

Adolescent↗
PubMed1995

Age estimation on the female sternum by quantification theory I and stepwise regression analysis.

One hundred and nine sternums from Chinese females aged 18-50 years were examined in this study. Six morphological characteristics and their changing degree in the female sternum were observed and scored. The data were statistically processed by quantification theory I and stepwise regression analysis on a computer and the following regression equation for age estimation of the female sternum was calculated: y = 19.28 + 1.83x1 + 1.66x2 + 3.02x3 + 1.57x4 + 3.02x5 + 7.75x6 + 1.25x7 + 3.45x8 + 4.88x9 + 0.82x10 + 2.76x11 + 2.48x12 + 7.84x13 + 1.26x14 + 3.80x15 (the correlation coefficient R = 0.9774, the standard deviation S = 2.20, F > 0.01). For convenience, the changing degree of the morphological characters and the equation were also converted into two tables. This method of age estimation on the female sternum is simple and accurate and is of importance to forensic medicine and anthropology.

Adult↗
PubMed1987

Linear regression analysis of ultrasound follicular growth series: statistical relationship of growth rate and calculated date of growth onset to total growth period.

Linear regression analysis was used to assess the serial growth of carefully measured individual mean follicular diameters as a function of menstrual age in a group of normal volunteers in natural, unstimulated cycles (n = 18). Follicle diameter (FD) growth, as measured with serial pelvic ultrasonography, can be shown to be essentially linear (mean R2, 0.92) with respect to time up to the point of ovulation. Growth rate (k) and the calculated date of growth onset (Do) are obtainable from the individual regression equations. Because an approximation of the date of ovulation (Dov, measured by menstrual age at the date of peak mean follicular size) is an available parameter, total growth period (TGP) is calculated with the equation (TGP = Dov - Do). Do, k, and TGP are highly correlated with one another. The mean parameters (+/- standard deviation) are Do = 4.29 +/- 2.77 day, k = 0.24 +/- 0.07 cm/day, TGP = 10.32 +/- 2.35 day, peak FD = 2.34 +/- 0.28 cm, and Dov = 14.61 +/- 2.00 day. The internal correlation of these parameters predicts that information from the first several scans in an individual series may be used to forecast date of ovulation. This may represent an improvement over the use of follicle size or menstrual age alone in planning for human chorionic gonadotropin administration, insemination timing, follicular aspiration and other interventions.

Adult↗
PubMed1996

A simulation study of the number of events per variable in logistic regression analysis.

We performed a Monte Carlo study to evaluate the effect of the number of events per variable (EPV) analyzed in logistic regression analysis. The simulations were based on data from a cardiac trial of 673 patients in which 252 deaths occurred and seven variables were cogent predictors of mortality; the number of events per predictive variable was (252/7 =) 36 for the full sample. For the simulations, at values of EPV = 2, 5, 10, 15, 20, and 25, we randomly generated 500 samples of the 673 patients, chosen with replacement, according to a logistic model derived from the full sample. Simulation results for the regression coefficients for each variable in each group of 500 samples were compared for bias, precision, and significance testing against the results of the model fitted to the original sample. For EPV values of 10 or greater, no major problems occurred. For EPV values less than 10, however, the regression coefficients were biased in both positive and negative directions; the large sample variance estimates from the logistic model both overestimated and underestimated the sample variance of the regression coefficients; the 90% confidence limits about the estimated values did not have proper coverage; the Wald statistic was conservative under the null hypothesis; and paradoxical associations (significance in the wrong direction) were increased. Although other factors (such as the total number of events, or sample size) may influence the validity of the logistic model, our findings indicate that low EPV can lead to major problems.

Bias↗
PubMed1975

The prediction of glaucoma from ocular biometric data. Part 1. An application of multiple regression analysis.

Two sets of multiple regression equations (prediction systems) were derived from the analysis of ocular biometric data obtained from glaucoma patients (16 open angle; 16 angle-closure) and 75 normal subjects. Discriminant scores were established for both sets of equations which minimised the number of false negatives. One set, the 'Glaucoma Equations' was applied to the data to segregate the glaucoma from the normal subjects. The other prediction system, the 'Classification Equations,' was then applied to the group defined as glaucomatous to discriminate between patients in the angle-closure and open angle categories. The performance of these equations, obtained by comparing the predicted and actual classifications for this sample, was such that between 9 and 12% of false positives and 0 and 3% false negatives were found on the 'Glaucoma Equations' and between 6 and 12% of false positives with no false negatives on the 'Classification Equations.'

Age Factors↗
PubMed2000

Derivation of a needs based capitation formula for allocating prescribing budgets to health authorities and primary care groups in England: regression analysis.

OBJECTIVE: To develop a weighted capitation formula for setting target allocations for prescribing expenditures for health authorities and primary care groups in England. DESIGN: Regression analysis relating prescribing costs to the demographic, morbidity, and mortality composition of practice lists. SETTING: 8500 general practices in England. SUBJECTS: Data from the 1991 census were attributed to practice lists on the basis of the place of residence of the practice population. MAIN OUTCOME MEASURES: Variation in age, sex, and temporary resident originated prescribing units (ASTRO(97)-PUs) adjusted net ingredient cost of general practices in England for 1997-8 modelled for the impact of health and social needs after controlling for differences in supply. RESULTS: A needs gradient based on the four variables: permanent sickness, percentage of dependants in no carer households, percentage of students, and percentage of births on practice lists. These, together with supply characteristics, explained 41% of variation in prescribing costs per ASTRO(97)-PU adjusted capita across practices. The latter alone explained about 35% of variation in total costs per head across practices. CONCLUSIONS: The model has good statistical specification and contains intuitively plausible needs drivers of prescribing expenditure. Together with adjustments made for differences in ASTRO(97)-PUs the model is capable of explaining 62% (35%+0.65% (41%)) of variation in prescribing expenditure at practice level. The results of the study have formed the basis for setting target budgets for 1999-2000 allocations for prescribing expenditure for health authorities and primary care groups.

Budgets↗
PubMed2003

Comparable results between standardization methods and regression analysis in predicting mortality rate among samples with burns.

This report suggests an alternative statistical model for estimating the mortality rate of samples of patients with burns. Three direct standardization methods were applied to nine samples (48,379 patients) selected randomly from the medical literature, and compared with the technique of three regression models, The results showed the similar mortality rates in these samples using the three different methods for standardization in comparison to the rates of LA(50) using three regression models. Of the different standardization methods used to measure the of mortality rate, standardization of death rate for age and area burned may be the best method to objectively evaluate the results of burns care, if combined with LA(50) using regression analysis. Furthermore, the values of LA(50) and 95% Cl in a sample are changed with different regression models. It is necessary to specify the regression model, 95% Cl or correlation coefficient of LA(50) in probit analysis or linear regression, respectively, when LA(50) of one sample is compared with that of another.

Adolescent↗