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 685 records · Page 38Linked to original sources
PubMed1991

Logistic multiple regression analysis of factors predicting recurrence of hyperthyroidism after medical treatment of toxic diffuse goitre.

In order to determine which factors predict the outcome of short term antithyroid drug treatment we studied 42 patients with diffuse goitre in whom 43 instances of thyrotoxicosis were treated. Treatment duration ranged from 24 to 61 wk (median 30 wk). All patients received high-dose carbimazole and thyroid hormone substitution. Patients in remission were followed for 39 to 134 wk (median 73 wk). The relapse rate at 1 yr and at 2 yr after cessation of antithyroid drug treatment was 51%. Of the parameters studied presence or absence of eye signs and initial serum levels of thyroxine, triiodothyronine and immunoglobulin-G in the relapse group and in the remission group showed significant differences in univariate analysis. No significant differences were found for age, sex, family history of thyroid disease, thyroid gland volume or TSH-receptor stimulating autoantibodies. Linear discriminant analysis shows that of the four remaining factors thyroxine is not important in separating both groups. Cox analysis yields only initial serum triiodothyronine and eye signs as significant prognostic factors. With these two factors 18 out of 23 predictions of remission and 16 out of 18 predictions of relapse in the 41 patients with known initial serum triiodothyronine concentrations are correct. Such predictions can be used in the choice of therapy, short-term medical treatment for patients with a low risk and long-term medical treatment or, at the appropriate time, a destructive form of therapy for patients with a high risk of relapse.

Analysis of Variance↗
PubMed1982

[Statistical regression analysis for four new cephalosporins (author's transl)].

The MIC (agar dilution method) and the zone diameter (disk diffusion) were determined for two hundred and sixteen of the most frequently bacterial strains found in hospitals in order to establish the regression curves for four new cephalosporins : cefotaxime, cefoperazone, cefotiam, moxalactam. The dta (MIC, zone size) were represented by a series of points (Mi (Xi Yi) in semilogarithmic rectangular co-ordinates expressed as : Xi = log2 MIC + 7 ; Yi = zone diameter. Firstly, we tried to fit a regression polynomial to this series of points progressing from degree 1, simple equation (standard regression line) 2, 3, etc. in order to reduce if need be, the deviation between the model (polynomial) and the empirical data. For these four cephalosporins a regression polynomial was obtained. Secondly, the global series was divided into three series of points Mi corresponding to three ranges of MIC : MIC less than or equal to 4 microgram/ml ; 4 less than or equal to 16 micrograms/ml; MIC greater than or equal to 16 microgram/ml. A regression line was obtained for each series and compared with each other and each one with the global regression line (they do not coincide statistically). Concerning these antibiotics, it is better to use a regression polynomial rather than apply a standard regression line.

Bacteria↗
PubMedDate not supplied

A linear regression analysis of the gamma dose in fast neutron beams.

The dual dosimeter technique for determining both the absorbed dose of neutrons and photons in a mixed field has been applied to multiple dosimeter use. The data were analyzed by a linear regression method which yields the neutron dose from the slope and the photon dose from the intercept and an estimation of the uncertainty of the photon dose can also be obtained. Measurements were made on a high energy neutron beam and the photon dose obtained both as a function of field size and depth in a tissue equivalent phantom.

Fast Neutrons↗
PubMedDate not supplied

Prognostic modeling with logistic regression analysis: in search of a sensible strategy in small data sets.

Clinical decision making often requires estimates of the likelihood of a dichotomous outcome in individual patients. When empirical data are available, these estimates may well be obtained from a logistic regression model. Several strategies may be followed in the development of such a model. In this study, the authors compare alternative strategies in 23 small subsamples from a large data set of patients with an acute myocardial infarction, where they developed predictive models for 30-day mortality. Evaluations were performed in an independent part of the data set. Specifically, the authors studied the effect of coding of covariables and stepwise selection on discriminative ability of the resulting model, and the effect of statistical "shrinkage" techniques on calibration. As expected, dichotomization of continuous covariables implied a loss of information. Remarkably, stepwise selection resulted in less discriminating models compared to full models including all available covariables, even when more than half of these were randomly associated with the outcome. Using qualitative information on the sign of the effect of predictors slightly improved the predictive ability. Calibration improved when shrinkage was applied on the standard maximum likelihood estimates of the regression coefficients. In conclusion, a sensible strategy in small data sets is to apply shrinkage methods in full models that include well-coded predictors that are selected based on external information.

Aged↗
PubMed1997

[A forecasting trial of the dento-skeletal and cutaneous facial profile, as a function of sex, in Class II, division 1 malocclusion, using regression analysis of the coordinates].

For a pattern of 36 leucoderm subjects, aged 12 years, with Class II, division 1, prediction of results are been calculated, in each sex, proceeding bony and cutaneous points coordinates, in axis defined by S-N, with "regression" method. Bony, dental and cutaneous profiles expected 3 years afterwards can be established in a personalized way, as well as correspondent security limits. A logical make easy utilisation of this process.

Adolescent↗