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 73 records · Page 4Linked to original sources

PubMed1985

Discriminant analysis for predicting dystocia in beef cattle. I. Comparison with regression analysis.

Data from 131 calvings of Chianina crossbred cows (2 to 5 yr old) bred to Chianina bulls were used to compare stepwise multiple regression analysis (RA) and stepwise, two-group discriminant analysis (DA) for predicting dystocia. Variables (21) studied in relation to dystocia included both prebreeding and precalving cow and calf effects. Calving was categorized as either unassisted or assisted without regard to the severity of dystocia. During this study, 30 (22.9%) assisted births occurred. All variables were standardized to a mean of zero and a variance of one before statistical analyses. Models were developed based on precalving variables and with both precalving and postcalving variables with both RA and DA. Average discriminant scores (centroids) were different (P less than .01) between assisted and unassisted cows. Significant precalving DA variables were cow age and precalving pelvic height. This model correctly predicted 26 of 30 (86.7%) of the occurrences of dystocia. Significant precalving RA variables were prebreeding pelvic width and precalving pelvic height. The amount of variation accounted for by these two factors was 31.5%. Calf birth weight, calf chest depth, calf height, precalving pelvic area, cow age and precalving cow weight were selected by DA for use in the combined precalving and postcalving prediction model. Calf birth weight was 58% more important than either pelvic size or cow age. Percentage correctly classified with this model was 87.4. Significant postcalving variables selected by RA in order of importance were prebreeding pelvic width, calf birth weight and calf shoulder width (R2 = .399).(ABSTRACT TRUNCATED AT 250 WORDS)

Animals↗

PubMed1984

Single-strain regression analysis for quality control of cephalothin-susceptibility testing and determination of interpretive breakpoints.

Histogram analysis of inhibition zone diameters around the 30 micrograms cephalothin disk for E. coli, P. mirabilis, and K. pneumoniae in samples from 1975 to 1982 showed a marked reproducibility of the disk-diffusion antibiotic-susceptibility test in the routine laboratory. A comparison of interpretive breakpoints with histograms for E. coli, P. mirabilis, K. pneumoniae, S. aureus, coagulase-negative staphylococci, and S. faecalis showed a higher proportion of possible misinterpretations using the breakpoints of the Swedish Reference Group, SRG, as compared to international (NCCLS) breakpoints. Further analysis using single-strain regression analysis revealed two major causes of interpretive errors. Firstly, the laboratory-related regression line for a bacterial species can be different from the general regression line of the reference laboratory. This difference has to be corrected by using species-related breakpoints. For E. coli, a species-specific breakpoint was determined to R = greater than 13 mm. Secondly, MIC limits recommended for the susceptibility categories of cephalothin by SRG are lower than the international limits and close to the true MIC values of many bacterial isolates, leading to misinterpretations due to the methodological variation. These studies suggest an adoption of international MIC limits for the susceptibility categories of cephalothin in Scandinavia. The "I" category should denote an indeterminate zone. A multi-laboratory quality control assessment using histogram analysis is recommended with optional single-strain regression analysis to determine breakpoints for problem combinations of bacterial species and antibiotics.

Bacteria↗

PubMed1993

Multivariate Cox regression analysis of covariates for patency rates after femorodistal vein bypass grafting.

Multivariate Cox regression analysis of patency rates for 750 consecutive femorodistal autogenous vein graftings for chronic lower limb ischemia showed that significant independent prognostic covariates were the type of graft (long saphenous or arm vein), presence of diabetes, and absence of a past history of myocardial ischemia. Analysis assumes that patients withdrawn with patent grafts due to death or loss to follow-up would have followed the same course as those who remain, and the degree to which this could distort results was studied. Patients who died with patent grafts were more likely to have had past myocardial ischemia and critical lower limb ischemia. Cox regression analysis for 600 operations after excluding patients who died with patent grafts then showed that significant independent covariates were the type of graft (long saphenous or arm vein) and indication (claudication or critical ischemia); then age, sex, hypertension, diabetes, myocardial ischemia, date of operation, surgeon, site of distal anastomosis, or first compared to repeat operations had no significant influence. Cox regression analysis helps determine which covariates influence graft patency rates, but results are affected by censored data, particularly from patients who die with patent grafts.

Adult↗

PubMed1983

Risk of operative mortality in surgery for coronary heart disease (a multiple regression analysis of perioperative hemodynamic and electrocardiographic data).

Using multiple regression analysis, we examined perioperative hemodynamic and electrocardiographic variables as predictors of operative mortality in surgery for coronary heart disease. Data were first analyzed as univariates and, if significantly related to mortality, they were subjected to stepwise logistic multivariate regression analysis. The preoperative predictor variables were: heart rate, ventricular arrhythmia and ST-T shift. The intraoperative predictor variables were: ventricular arrhythmia and ST-T shifts; and the postoperative predictor variables were: left ventricular stroke work index, blood pressure, mixed venous oxygen content and intrapulmonary shunt. When only electrocardiographic data were analyzed, the preoperative variables were: ventricular arrhythmia, ST-T shift and anterior wall infarction. The intraoperative variables were: ventricular arrhythmia and ST-T shift. The postoperative electrocardiogram did not give additional information. The common denominator of the relevant hemodynamic and electrocardiographic variables appears to be an accumulation of pre- and perioperative myocardial damage, which leads to operative mortality.

Cardiac Surgical Procedures↗

PubMed1996

Regression analysis with missing covariate data using estimating equations.

In regression analysis, missing covariate data has been among the most common problems. Frequently, practitioners adopt the so-called complete-case analysis, i.e., performing the analysis on only a complete dataset after excluding records with missing covariates. Performing a complete-case analysis is convenient with existing statistical packages, but it may be inefficient since the observed outcomes and covariates on those records with missing covariates are not used. It can even give misleading statistical inference if missing is not completely at random. This paper introduces a joint estimating equation (JEE) for regression analysis in the presence of missing observations on one covariate, which may be thought of as a method in a general framework for the missing covariate data problem proposed by Robins, Rotnitzky, and Zhao (1994, Journal of the American Statistical Association 89, 846-866). A generalization of JEE to more than one such covariate is discussed. The JEE is generally applicable to estimating regression coefficients from a regression model, including linear and logistic regression. Provided that the missing covariate data is either missing completely at random or missing at random (in addition to mild regularity conditions), estimates of regression coefficients from the JEE are consistent and have an asymptotic normal distribution. Simulation results show that the asymptotic distribution of estimated coefficients performs well in finite samples. Also shown through the simulation study is that the validity of JEE estimates depends on the correct specification of the probability function that characterizes the missing mechanism, suggesting a need for further research on how to robustify the estimation from making this nuisance assumption. Finally, the JEE is illustrated with an application from a case-control study of diet and thyroid cancer.

Biometry↗

PubMed2004

Reliability of calculating the cepstral peak without linear regression analysis.

Measures of cepstral peak prominence, using the smoothing algorithm and linear regression analysis software developed by Hillenbrand, have been shown to be reliable predictors of dysphonia in voice samples.(1-4) Recently, the Computerized Speech Laboratory [(CSL) Kay Elemetrics, Pinebrook, New Jersey] has introduced cepstral analysis as a component of that software package. The cepstral peak, in this instance, is calculated by the voice clinician analyzing the phonatory sample by subtracting the value of the peak from the apparent baseline signal. This study compares the ability of cepstral peak values calculated from the CSL software to predict dysphonia reliably with that of the values produced by the smoothing algorithm and linear regression analysis of Hillenbrand. The results of this study show that linear regression analysis is an important step in calculating the cepstral peak prominence, thus limiting the usefulness of software programs that do not employ this step.

Algorithms↗

PubMed1998

Performance of Deming regression analysis in case of misspecified analytical error ratio in method comparison studies.

Application of Deming regression analysis to interpret method comparison data presupposes specification of the squared analytical error ratio (lambda), but in cases involving only single measurements by each method, this ratio may be unknown and is often assigned a default value of one. On the basis of simulations, this practice was evaluated in situations with real error ratios deviating from one. Comparisons of two electrolyte methods and two glucose methods were simulated. In the first case, misspecification of lambda produced a bias that amounted to two-thirds of the maximum bias of the ordinary least-squares regression method. Standard errors and the results of hypothesis-testing also became misleading. In the second situation, a misspecified error ratio resulted only in a negligible bias. Thus, given a short range of values in relation to the measurement errors, it is important that lambda is correctly estimated either from duplicate sets of measurements or, in the case of single measurement sets, specified from quality-control data. However, even with a misspecified error ratio, Deming regression analysis is likely to perform better than least-squares regression analysis.

Bias↗

PubMed1980

Multiple regression analysis predictor models in exotropia surgery.

We developed a statistical model based on multiple regression analysis from a sample population of 30 patients with intermittent exotropia who underwent bilateral recession of the lateral recti muscles. We also identified predictor variables, which accounted for most of the change in deviation per millimeter of surgery performed. The application of multiple regression analysis to the problem of exotropia may be limited by variation in predictor variables according to sample population, multicollinearity, measurement unreliability, and the unproven assumption that the preoperative calculation of effect per millimeter correlates well with postoperative clinical success. Controlled clinical trials are needed to determine the value of multiple regression analysis in exotropia surgery. Discriminant analysis may represent an alternative statistical technique with application to this particular problem.

Humans↗

PubMed2002

[Sonoangiography and logistic regression analysis in the preoperative differentiation of ovarian tumors].

OBJECTIVE: To apply logistic regression analysis for several clinical and sonographic data for the construction of a predictive model that could be helpful in the preoperative differentiation of adnexal masses. MATERIALS AND METHODS: Two hundred and eight women with tumors thought to be of adnexal origin were examined preoperatively. Initial analysis included age and menopausal status, ultrasound derived morphological features of adnexal masses (unilateral/bilateral tumors, papillae, septae, tumor size and volume) as well as color Doppler criteria such as PI, RI, Peak Systolic Velocity, PSV assessment. In all examinations we used B&K 2002 ADI (Denmark) and Kretz Voluson V730 (Austria) scanners with transvaginal probes 5-9 MHz. Stepwise logistic regression analysis was used to construct a predictive model that would allow probability of malignancy calculation for individual patient. RESULTS: There were 159 benign and 49 malignant masses. Seven cancers were in FIGO stage one. Statistical analysis revealed that only 5 of initially tested 14 variables had significant influence on the regression equation. These were: age, bilateral mass, presence of septa > 3 mm, papillary projections > 3 mm in the tumor wall and subjective color scale assessment according to Timmerman et al. (1999). Sensitivity and specificity at the 50% probability level of malignancy in the studied tumor were 77.5% and 96.8%, respectively. When 25% cut-off probability level was used, sensitivity increased to 87.7% and specificity dropped to 89.9%. Prospective testing in a new group of 30 patients (5 ovarian cancers) gave sensitivity of 80% and specificity of 100%. CONCLUSIONS: The use of logistic regression analysis can help in modeling clinical and sonographic data. Our model had better predictive value than individual tests and allowed to calculate true probability figure of ovarian malignancy for any given patient with adnexal mass.

Adolescent↗

PubMed1989

A method to cope with the random errors of observed accident rates in regression analysis.

The paper is concerned with linear multiregression analysis on accident rates related to road geometric design elements. Supposing that a data set of accident records and geometric design elements of a certain stretch of a road is given, there are two steps for regression analysis: first, division of the road into a number of segments; and second, application of regression analysis to the set of segments. The main interest of the present paper is the first step. Occurrence of a traffic accident in a road segment is a stochastic event and an observed accident rate in a segment contains a certain magnitude of random error that deteriorates the explanatory power and reliability of the regression analysis. Random errors are required to be appropriately controlled for an effective regression analysis. The first part of the paper discusses how to evaluate a random error contained in an accident rate of a road segment and shows that a random error depends on the number of accidents and vehicle-kilometerage of the segment. It is then shown that random errors of the segment should be as much as possible equal to each other and small enough compared with the accident rate variance based on the discussion of how the random errors affect the efficiency of regression analysis. Several alternative criteria on the random errors for dividing a road into segments are proposed and numerical examples of Tokyo-Kobe Expressway are presented to examine the appropriateness of the alternative criteria. One of them is finally recommended as the most practically useful criterion.

Accidents, Traffic↗

PubMed1998

Three approaches to regression analysis of receiver operating characteristic curves for continuous test results.

The accuracy of a medical diagnostic test is typically summarized by the sensitivity and specificity when the test result is dichotomous. Receiver operating characteristic (ROC) curves are measures of test accuracy that are used when test results are continuous and are considered the analogs of sensitivity and specificity for continuous tests. ROC regression analysis allows one to evaluate effects of factors that may influence test accuracy. Such factors might include characteristics of study subjects or operating conditions for the test. Unfortunately, regression analysis methods for ROC curves are not well developed and methods that do exist have received little use to date. In this paper, we propose and compare three very different regression analysis methods. Two are modifications of methods previously proposed for radiology settings. The third is a special case of a general method recently proposed by us. The three approaches are compared with regard to settings in which they can be applied and distributional assumptions they require. In the setting where test results are normally distributed, we elucidate the correspondence between regression parameters in the different models. The methods are applied to simulated data and to data from a study of a new diagnostic test for hearing impairment. It is hoped that the presentation in this paper will both encourage the use of regression analysis for evaluating diagnostic tests and help guide the choice of the most appropriate regression analysis approach in applications.

Biometry↗

PubMed2004

Regression analysis of restricted mean survival time based on pseudo-observations.

Regression models for survival data are often specified from the hazard function while classical regression analysis of quantitative outcomes focuses on the mean value (possibly after suitable transformations). Methods for regression analysis of mean survival time and the related quantity, the restricted mean survival time, are reviewed and compared to a method based on pseudo-observations. Both Monte Carlo simulations and two real data sets are studied. It is concluded that while existing methods may be superior for analysis of the mean, pseudo-observations seem well suited when the restricted mean is studied.

Biometry↗

PubMed2003

Principal component regression analysis with SPSS.

The paper introduces all indices of multicollinearity diagnoses, the basic principle of principal component regression and determination of 'best' equation method. The paper uses an example to describe how to do principal component regression analysis with SPSS 10.0: including all calculating processes of the principal component regression and all operations of linear regression, factor analysis, descriptives, compute variable and bivariate correlations procedures in SPSS 10.0. The principal component regression analysis can be used to overcome disturbance of the multicollinearity. The simplified, speeded up and accurate statistical effect is reached through the principal component regression analysis with SPSS.

Regression Analysis↗

PubMed2001

[Estimation of low density lipoprotein cholesterol concentration: regression analysis versus Friedewald's formula].

BACKGROUND: The Friedewald formula is used to estimate cholesterol of low density lipoprotein (LDL) from total cholesterol (CT), cholesterol of high density lipoprotein (HDL) and triglycerides (TG), but there are doubts about its precision. AIM: To compare Friedewald formula and regression analysis for the calculation of LDL cholesterol. MATERIAL AND METHODS: One hundred and fifty plasma samples from asymptomatic adults (aged 47.7 +/- 13 years, 50.6% male) were analyzed. CT, HDL, LDL and TG were determined by enzymatic methods. Friedewald formula (LDLc = CT-HDL-(TG/5)) and multiple regression analysis were applied to estimate LDL concentration. RESULTS: Mean total cholesterol was 175.3 +/- 39.7 mg/dl, HDL cholesterol was 35.57 +/- 0.8 mg/dl and TG was 128.4 +/- 65.4 mg/dl. Mean values for LDL cholesterol were significantly higher than those estimated by the Friedewald formula (136.4 +/- 37.9 mg/dl and 114.1 +/- 37.4 mg/dl respectively, p < 0.001) with a mean underestimation of 16.4 +/- 11.7%. LDL cholesterol values were directly proportional to TG concentration. Multiple regression analysis (LDLr = -14.376 + (age x 0.198) + (CT x 0.949) + (HDL x -0.474) + (TG x -0.064) showed no statistical differences with those obtained by the enzymatic method. CONCLUSIONS: These results confirm the underestimation of LDL concentration by the Friedewald formula despite normal range of TG concentration. A multiple regression analysis should be used to estimate LDL concentration with precision.

Adult↗

PubMed2004

Artificial neural network as an alternative to multiple regression analysis in optimizing formulation parameters of cytarabine liposomes.

The objective of the study was to optimize the formulation parameters of cytarabine liposomes by using artificial neural networks (ANN) and multiple regression analysis using 3(3) factorial design (FD). As model formulations, 27 formulations were prepared. The formulation variables, drug (cytarabine)/lipid (phosphatidyl choline [PC] and cholesterol [Chol]) molar ratio (X1), PC/Chol in percentage ratio of total lipids (X2), and the volume of hydration medium (X3) were selected as the independent variables; and the percentage drug entrapment (PDE) was selected as the dependent variable. A set of causal factors was used as tutorial data for ANN and fed into a computer. The optimization was performed by minimizing the generalized distance between the predicted values of each response and the optimized one that was obtained individually. In case of 3(3) factorial design, a second-order full-model polynomial equation and a reduced model were established by subjecting the transformed values of independent variables to multiple regression analysis, and contour plots were drawn using the equation. The optimization methods developed by both ANN and FD were validated by preparing another 5 liposomal formulations. The predetermined PDE and the experimental data were compared with predicted data by paired t test, no statistically significant difference was observed. ANN showed less error compared with multiple regression analysis. These findings demonstrate that ANN provides more accurate prediction and is quite useful in the optimization of pharmaceutical formulations when compared with the multiple regression analysis method.

Artificial Intelligence↗

PubMed1997

Sonographic prediction of malignancy in adnexal masses using multivariate logistic regression analysis.

The aim of the study was to assign a probability of malignancy for any patient with an adnexal tumor by the application of multivariate logistic regression analysis to variables recorded at the time of pelvic sonography. Sixty-seven women with known adnexal masses were examined using transvaginal B-mode and color Doppler imaging. For each patient the variables included: (1) age, (2) maximum tumor diameter, (3) tumor volume, (4) unilocularity (presence (0) or absence(1)), (5) papillary projections (presence (1) or absence (0)), (6) random echogenicity (presence (1) or absence (0)), (7) highest peak systolic velocity (PSV), (8) time-averaged maximum velocity (TAMXV), (9) pulsatility index (PI) and (10) resistance index (RI). The TAMXV, PI and RI were those associated with the highest PSV. These ten independent variables and the final histological diagnosis for each patient (the dependent variable) were used for the regression analysis. Approximately 75% of the entire dataset was randomly selected for generating the regression model. The remaining 25% was used as the testing set for cross-validation of the model. In the entire dataset there were 52 women with benign, three with borderline and 12 with invasive ovarian tumors. Regression analysis on the ten variables resulted in the retention of only 'age', 'papillary projection score' and 'TAMXV' as significantly contributing to predicting the presence or absence of malignancy. The probability of malignancy for any patient was given by solving the equation: Probability = 1/(1 + e-z) where e is the base value for natural logarithms and z = (0.1273 x Age) + (0.2794 x TAMXV) + (4.4136 x Papillary projections score) - 14.2046. Cross-validation of the model on the test set of data gave a 100% sensitivity and specificity. However, for the entire dataset the best sensitivity and specificity were 93.3 and 90.4%, respectively, at a cut-off value of 25% probability of malignancy. In conclusion, multivariate logistic regression analysis enables the calculation of probability of malignancy for any patient with a known adnexal mass. The accuracy of this prediction appears to be better than that of morphological or Doppler criteria when the latter are used independently. The value of this model needs to be tested prospectively.

Adult↗

PubMed2007

Methods for improving regression analysis for skewed continuous or counted responses.

Standard inference procedures for regression analysis make assumptions that are rarely satisfied in practice. Adjustments must be made to insure the validity of statistical inference. These adjustments, known for many years, are used routinely by some health researchers but not by others. We review some of these methods and give an example of their use in a health services study for a continuous and a count outcome. For the continuous outcome, we describe re-transformation using the smear factor, accounting for missing cases via multiple imputation and attrition weights and improving results with bootstrap methods. For the count outcome, we describe zero inflated Poisson and negative binomial models and the two-part model to account for overabundance of zero values. Recent advances in computing and software development have produced user-friendly computer programs that enable the data analyst to improve prediction and inference based on regression analysis.

Data Interpretation, Statistical↗

PubMed1999

Omeprazole plus antibiotics in the eradication of Helicobacter pylori infection: a meta-regression analysis of randomized, controlled trials.

This article presents a meta-regression analysis of published studies of omeprazole plus antibiotics (amoxicillin, clarithromycin, or an imidazole derivative) in the treatment of Helicobacter pylori. Eligible studies were all randomized, controlled trials published through April 1996 with 10 or more patients receiving omeprazole plus antibiotics for 5 or more days and testing for H. pylori eradication 4 weeks or more after treatment. Probability of eradication was calculated for each treatment arm, and logistic regression was performed using study characteristics as covariates. Seventy-four studies involving 117 treatment arms with 4,769 patients were identified. The eradication rate was 76% for omeprazole plus clarithromycin and 65% for omeprazole plus amoxicillin dual regimens (P <.0001). Eradication rates for triple regimens were 82%, omeprazole plus amoxicillin plus clarithromycin; 83%, omeprazole plus amoxicillin plus imidazole; and 89%, omeprazole plus clarithromycin plus imidazole. In a multiple logistic regression analysis, significant factors were antibiotic, disease, omeprazole dose, and whether treatment was followed by maintenance omeprazole. A systematic overview of the best available evidence suggests that dual therapy with omeprazole plus clarithromycin is superior to omeprazole plus amoxicillin. Triple therapy is better than dual therapy. Treatment works better on ulcers than on nonulcer dyspepsia. Higher doses of omeprazole give better results. Additional trials exploring higher omeprazole doses for varying durations as well as cost, side effects, and compliance trade-offs with efficacy are recommended.

Anti-Bacterial Agents↗