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 343 records · Page 19Linked to original sources

PubMed1988

Multiple regression analysis, using body size and cardiac cycle length, in predicting echocardiographic variables in dogs.

A significant (P less than 0.0001) positive correlation was demonstrated between left ventricular internal chamber dimension in diastole or systole and body weight, body surface area, cycle length, and the square root of cycle length. On the basis of adjusted coefficients of determination, multiple regression analysis, using body weight or body surface area and cycle length or the square root of cycle length, was superior to separate simple regression with these variables in accounting for variations in left ventricular internal chamber dimensions. Shortening fraction had a significant (P less than 0.0001) negative correlation and left ventricular free wall measurements had a significant (P less than 0.0001) positive correlation to body weight and body surface area. For these echocardiographic variables, correlation to the square root of cycle length was insignificant (P greater than 0.05), and a multiple regression model was not helpful in developing confidence intervals. Septal wall measurements were not correlated with body weight, body surface area, cycle length, or the square root of cycle length.

Animals↗

PubMedDate not supplied

The use of multivariate regression analysis in contrast-detail studies of CT scanners.

Previous studies of the imaging performance of computed tomography (CT) scanners, and other imaging modalities, have failed to apply appropriate statistical methods to data analysis, thus impairing the accuracy and significance of results. Given that imaging performance involves a number of interrelated variables and an element of randomness, its empirical assessment requires multivariate regression analysis. This method is used here to analyze anew a set of contrast-detail data from a previous study on CT scanners. The main issues considered are the specification of the proper functional form linking perceptibility, dose and contrast, the estimation of the contrast and dose coefficients, and of scanner-specific constants to be used in computing indices of imaging quality. One of the main empirical findings is that the dose coefficient of the CT scanners studied is significantly less than that predicted by the theoretical model: 1/5 instead of 1/3. This result suggests that actual dose used in routine clinical studies could be reduced substantially without impairing much the quality of the images. On the other hand, the coefficient of contrast does correspond to its predicted value, i.e., 2/3. The methodology used here is not limited to the contrast-detail framework, but is applicable to, and indeed essential in, empirical studies of the performance of any imaging modality.

Mathematics↗

PubMed1985

A semiparametric model for regression analysis of interval-censored failure time data.

Left-, right-, and interval-censored response time data arise in a variety of settings, including the analyses of data from laboratory animal carcinogenicity experiments, clinical trials, and longitudinal studies. For such incomplete data, the usual regression techniques such as the Cox (1972, Journal of the Royal Statistical Society, Series B 34, 187-220) proportional hazards model are inapplicable. In this paper, we present a method for regression analysis which accommodates interval-censored data. We present applications of this methodology to data sets from a study of breast cancer patients who were followed for cosmetic response to therapy, a small animal tumorigenicity study, and a clinical trial.

Animals↗

PubMed1987

Risk factors for mortality after bereavement: a logistic regression analysis.

A national sample of elderly widowed people was followed up for six years. Excess mortality was found for men aged 75 years and over in the first six months of bereavement compared with men of the same age in the general population. Logistic regression analysis, controlling for age and sex together, demonstrated that the best independent predictors of mortality among the elderly widowed were: interviewer assessment of low happiness level; interviewer assessed and self-reported problems with nerves and depression; and lack of telephone contacts. The general practitioner is well placed to assess levels of depression and unhappiness among the widowed and to check that they have adequate social support.

Aged↗

PubMed1980

An electrocardiographic estimation of the left and right ventricular weight of the heart. Regression analysis based on the deviation from normal standards.

The deviation of the electrocardiographic deflections from normal standards and a grading of the ST segments were obtained from the ECGs of subjects who were subsequently examined by cardiac partitioning at autopsy. A stepwise, multiple, linear regression analysis was calculated, where the ventricular weights were the determinate variables, and the electrocardiographic deviation from normal standards, the grading of the ST segments, age and weight were the indeterminate variables. The results for the entire series of cases and subgroups were statistically significant at P less than 0,001. The wide spectrum of left and right ventricular weights required the establishment os subgroups, in which the variation in ventricular weight explained by the indeterminate variables was of a high order. The results were similar to estimations based on the actual electrocardiographic measurements rather than on their deviation from normal standards.

Cardiomegaly↗

PubMed2004

Asbestos exposure and cancer mortality among petroleum refinery workers: a Poisson regression analysis of updated data.

The authors investigated the relationship between asbestos exposure and respiratory cancer mortality among maintenance workers and other blue-collar workers at an Italian oil refinery. The cohort contained 931 men, 29,511 person-years, and 489 deaths. Poisson regression analysis using white-collar workers as an internal referent group provided relative risk estimates (RRs) for main causes of death, adjusted for age, age at hiring, calendar period, length of exposure, and latency. Among maintenance workers, RRs for all tumors (RR = 1.50), digestive system cancers (RR = 1.41), lung cancers (RR = 1.53), and nonmalignant respiratory diseases (RR = 1.71) were significantly increased (p < 0.05); no significant excess was found for all causes and among maintenance (RR = 1.12) and other blue-collar workers (RR = 1.01). Results confirm the increased risk of death from respiratory diseases and cancer among maintenance workers exposed to asbestos, whereas other smoking-related diseases (circulatory system) were not statistically different among groups.

Adult↗

PubMed2000

Predictive value of pharmacological activity for the relative efficacy of antidepressant drugs. Meta-regression analysis.

BACKGROUND: There is uncertainty about the contribution of specific pharmacological properties to the efficacy of antidepressants. AIMS: To assess whether specific pharmacological characteristics of alternative antidepressants resulted in altered efficacy compared to that of selective serotonin reuptake inhibitors in the treatment of major depression. METHOD: Meta-regression analysis of randomised trials that compare treatment with a selective serotonin reuptake inhibitor and an alternative antidepressant. RESULTS: One-hundred-and-five randomised trials were included. None of the factors identified a priori predicted a statistically significant improvement in outcome across the trials. CONCLUSIONS: This analysis does not provide evidence that antidepressants acting at more than one pharmacological site differ in efficacy from drugs selective for serotonin reuptake in the treatment of major depression.

Adrenergic Uptake Inhibitors↗

PubMed2001

Stepwise regression analysis to study male and female factors impacting on pregnancy rate in an intrauterine insemination programme.

The aim of this study was to evaluate the impact of male and female factors on the pregnancy rate in an intrauterine insemination (IUI) programme. Data on 522 cycles were retrospectively studied. All patients 39 years or younger were included in the study where data were available on male and female diagnosis, as well as on ovulation induction methodology. Regression analysis was possible on 495 cycles to study different factors affecting the pregnancy rate per treatment cycle. Logistic regression identified variables which were related to outcome and were subsequently incorporated into a statistical model. The number of follicles was found to have a linear association with the risk ratio (chance) of pregnancy. The age of the woman was also found to have a linear (negative) association with pregnancy. The percentage motility and percentage normal morphology (by strict criteria) of spermatozoa in the fresh ejaculate were the male factors that significantly and independently predicted the outcome. Percentage motility > or = 50 was associated with a risk ratio of pregnancy of 2.95 compared to percentage motility < 50. Percentage normal sperm morphology > 14% was associated with a risk ratio of pregnancy of 1.8 compared to percentage normal morphology < or = 14%. Female patients with idiopathic infertility were divided into three groups according to normal sperm morphology. The pregnancy rate per cycle was 2.63% (1/38) for the P (poor) pattern group (0-4% normal forms), 11.4% (17/149) for the G (good) pattern group (5-14%), and 24% (18/75) for the N (normal) pattern group (> 14% normal forms). A female diagnosis of endometriosis or tubal factor impacted negatively on the probability of pregnancy (risk ratio of 0.17), compared with other female diagnoses. Male and female factors contribute to pregnancy outcome, but the clinician can influence prognosis by increasing the number of follicles, especially in severe male factor cases.

Adult↗

PubMed1985

Regression analysis of total fatty acids from human whole brain according to age and sex.

Total fatty acids processed from whole brain samples removed at autopsy from men and women, 26-87 years of age, were esterified, the methyl esters analyzed by gas chromatography, and the respective profiles deduced. In the study of the effect of age and sex in relation to the fatty acid levels, regression analysis was applied to smoothened data to take care of extraneous effects or variables. Toward this end, five age groupings were formed and the regression relationships explored for each group per sex. Generally, the saturated homologs occurred in higher amounts among males of advancing age but decreased with age among females. This effect stemmed from the even-carbon acids. For the unsaturated fatty acids, the reverse behavior appeared evident. For the prominent acids, 16:0 and 18:0, the differences were moderately significant for cases 50 years and older, for 18:2, significance based on either sex was noted with all age groups.

Adult↗

PubMed1999

Regression analysis of doubly censored failure time data with applications to AIDS studies.

In many epidemiological studies, the survival time of interest is the elapsed time between two related events, the originating event and the failure event, and the times of the occurrences of both events are right or interval censored. We discuss the regression analysis of such studies and a simple estimating equation approach is proposed under the proportional hazards model. The method can easily be implemented and does not involve any iteration among unknown parameters, as full likelihood approaches proposed in the literature do. The asymptotic properties of the proposed regression coefficient estimates are derived and an AIDS cohort study is analyzed to illustrate the proposed approach.

Acquired Immunodeficiency Syndrome↗

PubMed1996

Regression analysis of biomedical research data based on a repeated measure or cluster sample.

Research in biology and medicine often entails estimating the effect of an exposure variable X on a response variable Y from a cluster sample, that is, where X and Y may be measured repeatedly from the same subject (cluster), or X and Y may be measured from two or more subjects based on some related grouping such as a litter or household (cluster). Typically, regression analysis of the data is performed ignoring the subject or grouping (cluster) identity. This analytical approach has two drawbacks. First, the statistical inference of the regression coefficient of Y on X (beta) ignored cluster identity will likely be biased. More serious though, is that beta ignored cluster identity will likely be quite discrepant from the average within-cluster beta. It is the latter that is of relevance to the research question. Indeed it is not clear what beta ignored cluster identity really conveys. We describe a multiple regression model for the analysis of data from a cluster sample. The model treats the cluster as a nominal confounding variable to be adjusted. The idea is to represent the cluster by a set of dummy variables to be included as explanatory variables in the regression model. This model gives the average within-cluster beta with valid statistical inference. Numeric examples were used to illustrate the application of the dummy variable multiple regression model. This statistical method can be implemented by any software package that includes multiple regression, and virtually all commercial packages include this procedure.

Cluster Analysis↗

PubMed1990

Use of a Lotus 1-2-3 spreadsheet to plot the confidence and the prediction intervals in linear regression analysis.

Computer-derived spreadsheet programs such as Lotus 1-2-3 are being used by many investigators for individualized statistical calculations. Spreadsheet templates have been described for the performance of simple statistical tests by non-statisticians with relatively limited computer experience. This paper describes a template for use with Lotus 1-2-3 for automatic graphic display of confidence and prediction intervals of simple linear regression analysis. The graph can be printed using the Lotus printgraph program. The program also allows comparison of the slope and the intercept of two regression lines.

Computer Graphics↗

PubMed2005

Increasing prevalence of Japanese cedar pollinosis: a meta-regression analysis.

BACKGROUND: Japanese cedar pollinosis, caused by the pollen of the Japanese cedar tree (Cryptomeria japonica), is the commonest seasonal allergic disease in Japan. A number of epidemiological surveys have been reported on Japanese cedar pollinosis, but it has never been assessed systematically or quantitatively. To confirm the increasing prevalence of Japanese cedar pollinosis and related factors, we conducted a meta-regression analysis on population-based surveys in Japan. METHODS: We searched for data from population-based surveys in which serological methods were used to test all participants. Weighted regression of logit-transformed prevalence and sensitization rates were used to evaluate the effects of the year of survey, age, and degree of urbanization. We also analyzed the relationship between prevalence and sensitization rate. RESULTS: Thirty-eight reports with 27 subgroups for prevalence and 134 subgroups for sensitization rate were selected from the literature published in the years between 1986 and 2000. The Japanese cedar pollen sensitization rate was found to be significantly correlated with the year of survey, age, and degree of urbanization (adjusted R(2) = 0.55). The coefficient for the correlation between the prevalence and the sensitization rate revealed a statistically significant correlation (Pearson's r = 0.70, p < 0.001). CONCLUSIONS: The prevalence of Japanese cedar pollinosis among adolescents was predicted to be 28.7% in metropolitan areas and 24.5% in the general population in urban areas in the year 2004, derived from the estimated sensitization rate and the relationship between sensitization rate and prevalence. The prevalence of Japanese cedar pollinosis increased 2.6-fold between 1980 and 2000, and the prevalence differed considerably according to age and degree of urbanization.

Adolescent↗

PubMed1983

[Evaluation of prognostic factors in hepatocellular carcinoma using multiple regression analysis].

We analysed the relationship between 19 prognostic factors and the life span of patients with hepatocellular carcinoma. Among 52 patients who had been admitted to the Gifu Prefectural Gifu Hospital, in 46 the pathological diagnosis was confirmed. We selected 9 of 19 variables by the stepwise forward regression method. These were Arels Classification, ascites, HBsAg, performance status, method of therapy, sex, GOT, albumin and bilirubin. Results of multiple regression analysis demonstrated that the presence of ascites is the most important factor; other significant factors are Ariel's Classification, performance status, positivity of HBsAg, method of therapy, and GOT. These parameters were paralleled with the survival curves, most of which had been categorized. Although intra-arterial infusion of ADM was considered the most favourable treatment method, the survival curves suggested that other methods were equally effective.

Adult↗

PubMed1987

Approach to evaluation of impairment in rheumatoid arthritis--a study by multiple regression analysis.

We investigated the weight of various factors involved in the activity of rheumatoid arthritis (RA) and joint deformity in an attempt to establish a more objective method of evaluation of the impairment in RA using these factors as parameters. The subjects were 100 patients with RA. Multiple regression analysis was undertaken using Lansbury parameters and Stage as predictor variables and the classification of functional capacity (Class) as a criterion variable. The results of analysis showed that the weight was greatest, 32% for the time of appearance of fatigue, followed by 29% for the Stage, 16% for joint index, 10% for grasping power, 7% for ESR, 5% for morning stiffness and 0% for the dose of aspirin used. On the basis of these results, an original method of evaluation of impairment in RA was proposed. There was a close correlation (R = 0.82) between the Class and the total score obtained by this method.

Activities of Daily Living↗

PubMed1989

Polytomous logistic regression analysis of the General Health Questionnaire and the Present State Examination.

First, two examples of dichotomous logistic regression analysis are presented. The probability of being a psychiatric case according to the Present State Examination is predicted from the total score on the General Health Questionnaire and from the general practitioner's judgement on the presence of a mental health problem. Subjects were 292 primary care attenders. Results are compared with those from prior studies. Next, the extension to the polytomous case is demonstrated. The probability of being at any given level of the Index of Definition (computed from PSE data) is estimated from the General Health Questionnaire total score by an ordered polytomous logistic regression model. Several applications of the polytomous logistic regression model are discussed. These range from estimating the proportion of psychiatric cases among individuals who refuse to be interviewed to the formulation of sampling schemes which can be expected to reduce costs while at the same time yielding optimal information for testing specific hypotheses.

Family Practice↗

PubMed1991

[A logistic regression analysis of relations between juvenile myopia and TV-watching, trace elements, and psychological characteristics].

With a paired study, 23 factors were investigated in 204 school children aged 9-14 in Taiyuan. The results of analysis of logistic regression showed that there was a close relationship between juvenile myopia and TV-watching distance, myopia in parents, Zn: Cu ratio in hair, N score of EPQ, Pscore of EPQ, and verbal IQ. The relations between myopia and some factors were discussed.

Adolescent↗

PubMed1989

Serum alpha-fetoprotein levels in normal infants: a reappraisal of regression analysis and sex difference.

Age-dependent changes of serum alpha-fetoprotein (AFP) levels were analyzed in 284 normal infants, including 113 full-term neonates and 171 infants between 1-12 months old. Serial statistical analyses attempted to clarify some confusing and conflicting observations presently existing in the literature. The results showed the following: (a) variations of AFP values were homogeneous throughout infancy, after the logarithmic transformation that is essential in dealing with AFP data. (b) To obtain an approximately 95% prediction band of normal AFP levels in infancy, the regression analysis should be based on the power regression rather than the exponential progression model: log Y = 7.397 - 2.622.log (X + 10), where X = age in days and Y = AFP level in nanograms per milliliter. (c) A sex-related difference existed with higher mean values for male babies between 2-4 days of age, a fact that may be missed with casual analysis. In addition, the mean AFP level was found to be relatively constant at some time during the 1st week of life, in contrast to its subsequent extremely rapid decline. These characteristic changes deserve further investigation.

Female↗