Regression analysis.
Explore the source record for details and available documents.
Regression Analysis↗
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.
[Risk factor of liver disorders caused by flutamide--statistical analysis using multivariate logistic regression analysis].
The antiandrogenic drug, flutamide (Odyne), is widely used in the treatment of carcinoma of prostate. It is well known that flutamide has adverse effects of liver disorders. To ascertain the risk of liver disorders before administering this drug, past history and lifestyle preferences were resurveyed in 123 patients who had been treated with flutamide. The results obtained were assessed in relation to the occurrence of liver disorders by multivariate logistic regression analysis. The incidence of liver disorders was 26% (33/123), with 64% of the disorders occurring within 9 months. The chi-square test for dependent variables revealed that three variables, i.e., body mass index, past history of liver disorders and elevated glutamic-pyruvic transaminase levels were significantly related to the incidence of liver disorders (p > 0.05). Multivariate analysis indicated that a history of liver disorders and elevated alanine aminotransferase (ALT) levels were related to a higher incidence of liver disorders. Elevated ALT levels were associated with a higher incidence of liver disorders and smoking was related to a lower incidence of the liver disorders.
Alanine Transaminase↗
Determination of sex by discriminant function analysis and stature by regression analysis: a lateral cephalometric study.
Determination of sex and estimation of stature from the skeleton is vital to medicolegal investigations. In the present study, an attempt is made to determine sex and estimate stature of an individual using data derived from lateral cephalogram in Central Indian population. Skull is composed of hard tissue and is the best preserved part of skeleton after death, hence, in many cases it is the only available part for forensic examination. Lateral cephalogram is ideal for the skull examination as it gives details of various anatomical points in a single radiograph, also it easily provides architectural and morphological details of skull superstructures and intra-cranial details for comparisons. A discriminant function derived from 10 cephalometric variables provided 99% reliability in sex determination. The formulae obtained from regression analysis using the maximum length of skull showed very high degree of reliability for estimation of stature in males as well as females.
Adult↗
Diagnosing coeliac disease by rectal gluten challenge: a prospective study based on immunopathology, computerized image analysis and logistic regression analysis.
The purpose of this study was to evaluate the use of rectal gluten challenge in the diagnosis of coeliac disease. A total of 103 patients with features suggestive of this diagnosis were prospectively enrolled into the study; a diagnosis of coeliac disease was based on strictly defined criteria used in judging the proximal jejunal biopsy. On that basis, 45 out of the 103 patients were deemed to have coeliac disease. A slurry of gluten powder in physiological saline was introduced into the rectum, and biopsies taken before and at 2 h or 4 h after the challenge were examined immunohistochemically by computerized image analysis. Cell counts were analysed by logistic regression, and the best equations were obtained for each challenge group. The 2 h challenge yielded diagnostic sensitivity and specificity of 69.6% and 78.6% respectively. The 4 h challenge provided sensitivity and specificity of 100% and 100% respectively. These results were compared with other clinical diagnostic predictors,including anti-endomysial antibodies, which yielded diagnostic sensitivity and specificity of 70% and 98% respectively. It is concluded that a 4 h rectal challenge is a highly sensitive means of identifying gluten-sensitized individuals, and would be of particular value in cases showing negative antibody screening or equivocal biopsy appearances.
Adult↗
ROC curve regression analysis: the use of ordinal regression models for diagnostic test assessment.
Diagnostic tests commonly are characterized by their true positive (sensitivity) and true negative (specificity) classification rates, which rely on a single decision threshold to classify a test result as positive. A more complete description of test accuracy is given by the receiver operating characteristic (ROC) curve, a graph of the false positive and true positive rates obtained as the decision threshold is varied. A generalized regression methodology, which uses a class of ordinal regression models to estimate smoothed ROC curves has been described. Data from a multi-institutional study comparing the accuracy of magnetic resonance (MR) imaging with computed tomography (CT) in detecting liver metastases, which are ideally suited for ROC regression analysis, are described. The general regression model is introduced and an estimate for the area under the ROC curve and its standard error using parameters of the ordinal regression model is given. An analysis of the liver data that highlights the utility of the methodology in parsimoniously adjusting comparisons for covariates is presented.
Algorithms↗
Problems due to small samples and sparse data in conditional logistic regression analysis.
Conditional logistic regression was developed to avoid "sparse-data" biases that can arise in ordinary logistic regression analysis. Nonetheless, it is a large-sample method that can exhibit considerable bias when certain types of matched sets are infrequent or when the model contains too many parameters. Sparse-data bias can cause misleading inferences about confounding, effect modification, dose response, and induction periods, and can interact with other biases. In this paper, the authors describe these problems in the context of matched case-control analysis and provide examples from a study of electrical wiring and childhood leukemia and a study of diet and glioma. The same problems can arise in any likelihood-based analysis, including ordinary logistic regression. The problems can be detected by careful inspection of data and by examining the sensitivity of estimates to category boundaries, variables in the model, and transformations of those variables. One can also apply various bias corrections or turn to methods less sensitive to sparse data than conditional likelihood, such as Bayesian and empirical-Bayes (hierarchical regression) methods.
Bias↗
Responses to phenelzine and amitriptyline absence of differential predictors by multiple regression analysis.
Multiple regression analyses using initial symptoms to predict outcome were carried out on data from an outpatient controlled comparison of amitriptyline, phenelzine and placebo. Separate analyses were carried out in the three treatment groups and regression equations were compared. Significant prediction was obtained only for phenelzine. On only one of three outcome measures analysed, however were there significant differences between treatment groups in regressions. Individual predictors were not easily interpretable. There was little evidence for different clinical predictors or response to phenelzine and amitriptyline in this sample.
Adult↗
Biological age and its estimation. II. Assessment of biological age of albino rats by multiple regression analysis.
Multiple regression model of biological age (BA) theoretically gives agreement with the main concept of BA. When assessment of BA is based on the model, the age being in regression center, the method provides satisfactory results, whereas BA estimates of individuals in extreme age groups are erroneous. Investigation of male and female Wistar rats of age 5-29 months showed the BA estimates calculated from 4-10 physiological indices in young (5-7 mo) animals are overestimated, and in old (24-28 mo) animals are underestimated. Coincidence of average BA in one-age group of animals with its chronological age served as a criterion for the correspondence of the estimate to "real" BA. The paper also examines the following questions: the necessary and sufficient number of physiological indices; the sample size from the intact animal population to establish normal aging standard; the relationship between BA and animal weight.
Aging↗
Prognostic factors in osteosarcomas. A regression analysis.
A multivariate regression analysis of survival data, using the Cox proportional hazards model (PHM), was performed on the retrospective material of 184 osteosarcoma patients treated at the Aarhus and Copenhagen oncology centers, Denmark, from 1963 to 1984. All patients were previously untreated. Radical surgery, in general ablative when possible, was the primary treatment goal throughout this period. A number of clinical and pathologic variables were tested in the model to elucidate their prognostic importance. Tumors localized to the trunk, pelvis, or femur, and symptom duration of less than 6 months were poor prognostic signs. Tumors dominated by fibroblastic cells and a patient age of approximately 25 to 30 years were associated with an especially good prognosis. The prognosis worsened with advancing age. Children, adolescents, and adults ages 5 to 25 years had significantly poorer prognosis than young adults 25 to 30 years of age. Sex, radiologic appearance, and year of referral had no significant prognostic value in this series. Based on the regression model, a prognostic index is derived and survival is calculated for a good and a poor prognostic case. The overall 10-year survival with one standard deviation was 28.6 +/- 3.5%. Cancer deaths continue to occur 10 years after initial treatment, and the estimated hazard rate is still four times greater than that of a sex- and age-matched group of healthy individuals.
Adolescent↗
[Linear regression analysis of values of the calciuria index, N-telopetide and bone mineral density in patients with osteoporosis].
UNLABELLED: The authors examined 49 patients (n=49) with primary and secndary osteoporosis incl. 37 women aged 41-58 years and 12 men aged 38-56 years. In all they assessed the 24-hour calciuria index (Ca/creatinine), the aminoterminal N-telopetide of collagen I of the organic matrix (NTx24) and the bone mineral density (BMD) in the first three lumbar vertebrae using quantitative computed tomography (QC). OBJECTIVE: To find an answer to the questions:1. Is the calciuria index in the 24-hour diuresis a parameter which provides information on osteresorption? 2. Is there a significant negative correlation between values NTx24 and BMD? 3. Is NTx a sensitive parameter of osteoresorption in men and women with primary or secondary osteoporosis? STATISTICAL ANALYSIS: Linear regression analysis. RESULTS: Correlation coefficients for the correlated groups NTx24 - Ca/creatinine (r = 0.023, p>0.05) and NTx24 - BMD (r = -0.238, p>0.05) were not statistically significant. Statistical significance was recorded in the correlation coefficient for pathologically elevated values of NTx24(zNTx24), correlated with BMD (r = -0.3115, p<0.048). CONCLUSION: 1. The calciuria index in the 24-hour diuresis does not provide information on osteoresorption. 2. Between pathologically elevated values of NTx24 and BMD in L1-L3 is a statistically significant negative correlation. 3. NTx is a sensitive parameter of osteresorptioon in men and women with primary and secondary osteoporosis, it reflects the influence of the bone turnover on the density of the bone mineral.
Adult↗
Importance of events per independent variable in proportional hazards regression analysis. II. Accuracy and precision of regression estimates.
The analytical effect of the number of events per variable (EPV) in a proportional hazards regression analysis was evaluated using Monte Carlo simulation techniques for data from a randomized trial containing 673 patients and 252 deaths, in which seven predictor variables had an original significance level of p < 0.10. The 252 deaths and 7 variables correspond to 36 events per variable analyzed in the full data set. Five hundred simulated analyses were conducted for these seven variables at EPVs of 2, 5, 10, 15, 20, and 25. For each simulation, a random exponential survival time was generated for each of the 673 patients, and the simulated results were compared with their original counterparts. As EPV decreased, the regression coefficients became more biased relative to the true value; the 90% confidence limits about the simulated values did not have a coverage of 90% for the original value; large sample properties did not hold for variance estimates from the proportional hazards model, and the Z statistics used to test the significance of the regression coefficients lost validity under the null hypothesis. Although a single boundary level for avoiding problems is not easy to choose, the value of EPV = 10 seems most prudent. Below this value for EPV, the results of proportional hazards regression analyses should be interpreted with caution because the statistical model may not be valid.
Computer Simulation↗
Using the correct regression analysis technique in method comparison studies.
Regression analysis frequently is used to evaluate a new clinical laboratory method. Results from the new method are compared to results from an existing more established method. If measurement error exists in the established method, then least squares may not be an appropriate statistical method to use for the regression analysis. An errors variable regression analysis model was used to evaluate data from five method comparison studies. Results were compared to least squares analyses performed on the same data. When significant measurement error existed in the "reference" method, the errors variable analysis provided a less biased estimate of the regression statistics, which differed markedly from the least squares results. Inappropriate use of least squares in method comparison studies can lead to erroneous conclusions about the relationship of a new analytical method to an existing method.
Journal Article↗
The logistic regression analysis of psychiatric data.
Logistic regression is presented as the statistical method of choice for analyzing the effects of independent variables on a binary dependent variable in terms of the probability of being in one of its two categories vs the other. The method, which must be applied by computer, is illustrated on data from the DSM-III field trials. The dependent variable is treatment with behaviourally-oriented psychotherapy vs treatment with psychoanalytically-oriented psychotherapy, and the independent variables are several patient and clinician characteristics. Like ordinary multiple regression, the method is shown capable of analyzing categorical as well as continuous independent variables. Unlike ordinary multiple regression when applied to binary data, logistic regression analysis necessarily yields estimated probabilities that lie between 0 and 1. The measure of association derived from logistic regression analysis, the odds ratio, is defined. Methods for making inferences about it are presented and illustrated.
Attitude of Health Personnel↗
The use of radionuclide angiography in the diagnosis of coronary artery disease--a logistic regression analysis.
We applied logistic regression analysis to a group of 736 patients with chest pain to determine which radionuclide angiographic (RNA) parameters were most useful in the diagnosis of significant coronary artery disease. The most useful parameters were exercise ejection fraction, exercise heart rate, "ischemia score," and the presence of a regional wall motion abnormality at exercise. Ten clinical variables were used in one logistic regression model to estimate each patient's pretest probability of disease. A second logistic regression model considered these clinical variables and the four important RNA parameters to estimate each patient's posttest probability. These models were applied prospectively to a group of 76 patients with chest pain who did not have a high pretest probability of disease. Twenty-four patients (32%) could be diagnosed with 90% probability; 32 patients (42%) could be diagnosed with 85% probability. RNA testing is therefore helpful in the noninvasive diagnosis of coronary artery disease. However, a majority of patients who do have a low or intermediate pretest probability of disease will require additional testing for a definitive diagnosis.
Cardiac Catheterization↗
Fusidic acid disk diffusion testing of clostridium difficile can be calibrated using single-strain regression analysis.
Single-strain regression analysis (SRA) was employed to calibrate the disk diffusion antibiotic susceptibility test for fusidic acid and Clostridium difficile. MIC determinations of 40 clinical isolates of C. difficile were performed with the E-test. The disk diffusion test was standardized according to the Swedish Reference Group for Antibiotics (SRGA). Disks used for SRA contained 1.5, 5, 15, 50 and 150 microg fusidic acid and the routine disk contained 50 microg fusidic acid. A control strain, ATCC 9689, was also tested. SRA constants A and B of the regression lines were calculated. This permitted the determination of zone breakpoints for C. difficile. When applying the pharmacological MIC S and R limits set by SRGA to the E-test results I strain of C. difficile was interpreted as resistant. Zone breakpoints corresponding to the pharmacological MIC limits and calculated using the mean SRA constants for the 40 clinical isolates lead to all strains being interpreted as susceptible. SRA calculations enable laboratories to set up calibrated disk tests with species-related and laboratory-specific interpretations.
Anti-Bacterial Agents↗
Numerical analysis of kinetic data: collapsed regression analysis for first order reactions.
Explore the source record for details and available documents.
Enzymes↗
[Multivariate analysis in computer diagnosis. 4. Regression analysis and some remarks on its application].
Explore the source record for details and available documents.
Diagnosis, Computer-Assisted↗
An application of a microcomputer compiler program to multiple logistic regression analysis.
Microcomputer programs for multiple logistic regression analysis were written in BASIC language to determine the usefulness of microcomputers for multivariate analysis, which is an important method in epidemiological studies. The program, carried out by an interpreter system, required a comparatively long computing time for a small amount of data. For example, it took approximately thirty minutes to compute the data of 6 independent variables and 63 matched sets of case and controls (1:4). The majority of the calculation time was spent computing a matrix. The matrix computation time increased cumulatively in proportion to additions in the number of subjects, and increased exponentially with the number of variables. A BASIC compiler was utilized for the program of multiple logistic regression analysis. The compiled program carried out the same computations as above, but within 4 minutes. Therefore, it is evident that a compiler can be an extremely convenient tool for computing multivariate analysis. The two programs produced here were also easily linked with spreadsheet packages to enter data.
Computers↗