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 793 records · Page 44Linked to original sources
PubMed1999

Studying seasonality by using sine and cosine functions in regression analysis.

STUDY OBJECTIVE: A statistical test that allows for adjustment of confounding can be helpful for the study of seasonal patterns. The aim of this article is to supply a detailed description of such a method. An example of its application is given. DESIGN: A statistical test is presented that retains the information on the connection of time periods by describing the seasonal pattern as one sine and one cosine function. Such functions can be included into a regression model. The resulting form of the seasonal pattern follows a cosine function with variable amplitude and shift. MAIN RESULTS: The test is shown to be applicable to test for seasonality. Not only one cosine function per time period, but also a mixture of cosine functions can be used to describe the seasonal pattern. Adjustment for confounding effects is possible. CONCLUSIONS: This method for studying seasonal patterns can be applied easily in a regression model. Adjusted prevalences and odds ratios can be calculated.

Confounding Factors, Epidemiologic↗
PubMed2005

Prediction methods for babies' birth weight using linear and nonlinear regression analysis.

The aim of this study is to determine more accurate prediction methods between linear and non-linear methods for prediction of babies' birth weight among maternal demographic characteristics. Three hundred pregnant women were included in the study. Blood glucose level before and after ingestion of glucose load, age, body mass index, % of change in weight during pregnancy, height, gestational age, parity, and fetal sex were collected as independent variables and baby birth weight as dependent variable. In linear regression, least squares estimation method was used to estimate parameters. Non-linear regression method was performed using neural network model with multilayer perceptrons, back propagation method was preferred as learning algorithm. Coefficient of determination, R2, of the linear regression equation was found 59.8% and the standard error of the estimate was calculated as 325.69 gr. In non-linear regression method R2 value was also found 59.8% and standard error of estimate was calculated as 320.30 gr. According to the results of the present study, one method is not significantly better than the other. When "accuracy in prediction" is aimed, it is better to use the two methods and compare the results, and then decide on the selection of the favourable method.

Algorithms↗
PubMed2002

Regression analysis based on pairwise ordering of patients' clinical histories.

When a medical treatment influences a variety of outcomes, describing the global effect of treatment can be difficult. Traditional approaches specify how treatment affects each separate outcome. This can be done with separate models for each outcome, or by using a combined multivariate model. Describing the overall effect of a treatment thus requires combining these separate effects in some fashion and can be difficult to explain. In this paper, I specify a regression model for use with multiple outcomes where the outcome histories for each pair of patients are ranked. Pairs of patients with different lengths of follow-up are evaluated solely over the common follow-up interval. The logit of the probability that the outcome for patient i is better than that of patient j is assumed to depend on a linear function of the difference of the covariate vectors (for example, treatment indicators) for persons i and j. Thus covariates directly affect the entire clinical history, rather than directly affecting specific outcomes that comprise the history. The idea is that ranking outcomes is more relevant and interpretable than statistically combining separate effects. An estimating equations approach for estimation is described and an example of a clinical trial involving patients with heart failure is provided.

Angiotensin-Converting Enzyme Inhibitors↗
PubMed2002

Real-time water quality monitoring and regression analysis to estimate nutrient and bacteria concentrations in Kansas streams.

An innovative approach currently is underway in Kansas to estimate and monitor constituent concentrations in streams. Continuous in-stream water-quality monitors are installed at selected U.S. Geological Survey stream-gaging stations to provide real-time measurement of specific conductance, pH, water temperature, dissolved oxygen, turbidity, and total chlorophyll. In addition, periodic water samples are collected manually and analyzed for nutrients, bacteria, and other constituents of concern. Regression equations then are developed from measurements made by the water-quality monitors and analytical results of manually collected samples. These regression equations are used to estimate nutrient, bacteria, and other constituent concentrations. Concentrations then are available to calculate loads and yields to further assess water quality in watersheds. The continuous and real-time nature of the data may be important when considering recreational use of a water body; developing and monitoring total maximum daily loads; adjusting water-treatment strategies; and determining high constituent concentrations in time to prevent adverse effects on fish or other aquatic life.

Enterobacteriaceae↗
PubMed1993

Nonlinear regression analysis of the time course of ligand binding experiments.

The binding and release of hormones and growth factors are often relatively slow processes under biological conditions. Consequently, a knowledge of the underlying rate constants may be of greater physiological relevance than the equilibrium constant. Here we show how, by following a single time course of binding, the rate constants for both binding and release can be determined. The ratio of these rate constants allows the binding constant to be calculated. A nonlinear regression computer program is described which facilitates these calculations and which provides estimates and standard errors of the constants determined. The method is illustrated by the binding of human growth hormone to the human growth hormone binding protein, and the binding of ovine prolactin to the rabbit prolactin receptor.

Animals↗
PubMed1992

Auditing quality control procedures in a chemical pathology laboratory--a multiple regression analysis.

We undertook a retrospective analysis of the monthly test rejection rates and the monthly external quality assessment scheme performance indices for our laboratory's two automated analysers, and examined the association of these variables with measures of laboratory workload, manpower, staff training, instrument servicing, seasonal and temporal factors and changes of calibration, method and assigned internal quality control values. Using multiple linear regression and stepwise multiple linear regression, we found that test rejection rates differed significantly between instruments, and were highest on the instrument performing the widest variety and lowest volume of tests. On that instrument, rejection rates were significantly associated with the introduction of new staff and laboratory manpower levels, and also showed a highly significant trend upwards over the study period, independent of the effects of the other variables examined. External quality assessment scheme performance indices showed small trends over the study period. They were not related to the test rejection rates on either analyser but also showed a significant association with the introduction of new staff and a small but significant association with laboratory workload. We conclude that the training and introduction of new staff and decreased laboratory manpower levels may significantly increase the level of test rejection, and adherence to appropriate quality control protocols effectively maintains the quality of the laboratory's results, but may not be completely successful in filtering out the effects of some assignable causes of variation in test results. It is suggested that clinical laboratories use the statistical approach adopted here to identify factors which may be adversely affecting quality performance and running costs and to provide evidence that quality control procedures are both cost- and quality-effective.

Clinical Laboratory Techniques↗
PubMed2007

Cervical spine curvature and craniofacial morphology in an adult Caucasian group: a multiple regression analysis.

The aim of this study was to investigate the relationship between cervical curvature and skeletal facial morphology. Pre-treatment lateral cephalograms in natural head position were obtained from 98 orthodontic adult subjects (56 males and 42 females; mean age = 31.5 years, standard deviation +/-5.8 years). The amount of concavity of the cervical curvature was calculated by a second-order quadratic interpolation of the most infero-posterior points of the bodies of the seven cervical vertebrae on the radiographs. Linear regression with stepwise elimination was carried out to evaluate the sagittal and vertical craniofacial dimensions. The results revealed that the amount of curvature was related to (1) the horizontal position of the upper incisors with respect to the maxillary and mandibular skeletal bases and (2) the sagittal millimetric dimension of the maxillary base. These cephalometric variables explained 51 per cent of the total variance of cervical curvature. There was no significant correlation between cervical curvature and any other cephalometric variable. Cervical curvature was not influenced by age or gender.

Adolescent↗
PubMedDate not supplied

[Correlation-regression analysis of infant mortality].

Correlations between neonatal mortality coefficient and a group of socioeconomic and ecological parameters are analyzed for some economic regions and districts of Russia. At the level of regions, the strongest correlations were observed between birth rates, maternal mortality, subsistence minimum, level of education, incidence of narcomania and toxicomania (including that among adolescents), release of hard harmful substances in the atmosphere and of chlorine and sulfates in surface water bodies (r = 0.4-0.8). Multifactorial model was derived by the step regression method. This model can be used for predicting the level of neonatal mortality in the nearest 5.6 years on the basis of the negative or positive shifts in the socioeconomic status of Russia. In addition, several unifactorial exponential models at a district level were obtained, reflecting the relationship between neonatal mortality and birth rates, unemployment and incidence of narcomania and toxicomania.

Humans↗
PubMed1998

[The correlation-regression analysis of the morphometric indices of tibial status after experimental osteotomy].

Dynamics of morphometric parameters which characterize the processes and extent of ischemic lesion and bone posttraumatic reparative regeneration was defined in rats following experimental tibial osteotomy within time intervals from 3 to 60 days. Signs of microcirculatory bed function exclusion and of the formation of foci of bone and bone marrow infarction were found to increase in the terminal parts of the fragments by the end the 1st posttraumatic day and then regress with the development of the process of substitution and repair. Pairs of most closely correlating morphometric characteristics of lesion at different time intervals of posttraumatic period were determined separately for proximal and distal bone fragments, regression type was determined and regression equations were deduced.

Animals↗
PubMed1988

Cytologic criteria for fibroadenoma. A step-wise logistic regression analysis.

The authors reviewed the fine-needle aspiration biopsy smears from 62 subjects with proven fibroadenoma, 60 subjects with proven ductal carcinoma, and 42 subjects with proven fibrocystic disease. All smears were coded as to the presence or absence of the following variables: epithelial cells, stroma, honeycomb sheets, antler horn clusters, naked nuclei, nucleoli, marked cellularity, foam cells, apocrine cells, anisokaryosis, atypical nuclear hyperchromasia, single cells with cytoplasm, mitotic figures, and nuclei greater than two red blood cell diameters. Step-wise logistic regression analyses were performed to determine the variables predictive of fibroadenoma. The statistical analyses selected stroma, antler horn clusters, and marked cellularity as the key cytologic criteria to differentiate fibroadenomas from fibrocystic disease. The statistical analyses selected stroma, antler horn clusters, and honeycomb sheets as the key cytologic criteria to differentiate fibroadenoma from ductal carcinoma.

Adenofibroma↗
PubMed1994

[A community cohort study on risk factors of strokes in Shanghai: a Cox regression analysis on 15,885 subjects].

Baseline investigation on 15,885 subjects who were over thirty-five years old had been launched since July 1987. All of them were selected from eighteen neighbourhoods of three districts in Shanghai by using multiple stage cluster sampling procedure. All of them were followed up till the end of 1989. Each new case of stroke was randomly matched with about twenty controls by age and sex. The total 2,682 controls together with 127 new cases of strokes were analyzed by using Cox regression model for risk factors. The population attributable risk proportion (PARP) of factors was also estimated.

Acute Disease↗
PubMed1984

Multiple regression analysis of the cholinesterase activity with certain physiochemical factors.

Nineteen hundred sixteen agricultural workers who have been living in a rural area in central Japan were studied. Their age, blood pressure, Broca index, hemoglobin content, serum total cholesterol, and activity of serum cholinesterase and transaminase (GOT and GPT) were determined. The relation between the cholinesterase activity and certain physiochemical factors of the subjects was studied. Cholinesterase activity (ChE) is related to certain factors such as age, hemoglobin content (Hb), serum total cholesterol (TCh), transaminase activity (GPT), and Broca index (BI). A multiple regression equation exists between the cholinesterase activity and these factors: ln ChE = a (age) + b (Hb2) + c X ln GPT + d X ln TCh + e BI + f, where a, b, c, d, e, and f are constants. The estimated value of cholinesterase activity agrees with its measured activity. The partial correlation coefficients of the equation can be classified into the following three classes: (1) The partial correlation coefficients of total cholesterol and Broca index are nearly constant without distinction of sex and season. (2) The coefficient of hemoglobin content has a small seasonal and sexual difference. (3) The coefficients of age and GPT have a great seasonal and sexual difference. Using the equation, the most probable value of cholinesterase activity can be estimated. Therefore, the significant changes of its activity may be attributed to the toxic effects of insecticides or the abnormality of liver function.

Adult↗
PubMed1980

The Tromsø heart study. A multiple regression analysis of the relationship between coronary risk factors and some physical and social variables.

Univariate analysis of the cross-sectional data of the Tromsø Heart Study revealed group differences in the levels of blood pressure and serum lipids. In order to isolate the contribution from each of the physical and social variables which characterized these groups, a multiple variate analysis was carried out with serum triglycerides, serum cholesterol, systolic and diastolic blood pressure as dependent variables. Besides age, height and weight, seasonal variation contributed substantially to the explanation of the variance of the dependent variables. The influence of the social variables was somewhat reduced compared to the univariate analysis, though most of them did remain significant also in the multivariate analysis. The explained percentage of the total variance ranged from 13% for systolic blood pressure to 26% for serum cholesterol.

Adult↗
PubMed2003

[Cox regression analysis of the prognostic factors of unresectable esophageal carcinoma after stenting].

BACKGROUND & OBJECTIVE: There are many factors affect the prognosis of the patients with unresectable esophageal carcinoma who underwent intubation, however, it is unclear which ones are main causes so far. This study was designed to analyze the prognostic factors of unresectable esophageal carcinoma after stenting in order to find the reasonable modalities of palliative therapy. METHODS: Consecutive 102 patients with unresectable esophageal carcinoma who were eligible for inclusion criteria were analyzed after stenting. Twelve factors including gender, age tumor site, tumor length, stricture degree in diameter, pathologic type, grade of cell differentiation, clinical tumor stage (T, N, M), pre-stenting therapy and post-stenting therapy (radiotherapy and/or chemotherapy) were used for Cox regression model analysis. The survival rate was calculated by life table. RESULTS: The technical sucess rate of stenting was 98.3%. There were significant improvement on the grade of swallowing function (P = 0.000) and quality of life (KPS scores, P = 0.000). The incidence rate of complications was 43.1% (44/102). The survival rates of 3, 6, 9, 12 months were 67.53%, 40.59%, 27.43%, and 18.65%, respectively. The results of Cox regression showed that invasion degree of primary tumor (T, P = 0.0410) and distal metastasis (M, P = 0.006) were the statistically significant prognostic factors. The odds ratio(ORT) was 1.750 (95% CI, 0.996-3.074) on T stage of primary carcinoma, ORM was 1.527 (95% CI, 1.126-2.069) on M stage. There was no statistical significance in the survival of the patients affected by radiotherapy and/or chemotherapy after intubation. CONCLUSION: T stage and M stage are the major prognostic factors affecting the survival of patients with unresectable esophageal carcinoma after stenting. There is no benefit for survival of patients treated with radiotherapy and/or chemotherapy after intubation.

Adult↗
PubMed2002

Prediction of human clearance from animal data and molecular structural parameters using multivariate regression analysis.

The aim of the study reported here was to develop a method for predicting human clearance that can be applied to various kinds of drugs using clearance values for rats and dogs and some molecular structural parameters. The clearance data for rats, dogs, and humans of 68 drugs were obtained from literature. The compounds have various structures, pharmacological activities, and pharmacokinetic characteristics. In addition, molecular weight, c log P, and the number of hydrogen bond acceptors were used as possible descriptors related to the human clearance value for each drug. Three types of regression methods, multiple linear regression (MLR) analysis, partial least squares (PLS) method, and artificial neural network (ANN), were used to predict human clearance, and their predictive performances were compared with allometric approaches, which have been widely used in interspecies scaling. In MLR and PLS analyses, interaction terms were introduced to evaluate the nonlinear relationships. For the data sets used in the present study, MLR and PLS with quadratic terms gave the same equation and the best predictive performance. The value of the squared cross-validated correlation coefficient (q(2)) was 0.682. In conclusion, the MLR method using animal clearance data from only two species and using easily calculated structural parameters can generally predict human clearance better than allometric methods. This approach can be applied to drugs with various characteristics.

Animals↗