Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Statistical Regression”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 865 records · Page 48Linked to original sources
PubMed2002

Dental age estimation and third molars: a preliminary study.

The aim of the present investigation was to reconstruct the chronological age based on the dental developmental stages of third molars evaluated on orthopantomograms. A total of 1175 orthopantomograms were assembled from patients of Caucasian origin between 16 and 22 years of age. Each third molar present was scored by two observers according to a 10-stage developmental scale. The kappa statistics measured the intra- and inter-observer reliability. The general statistical analysis was based on multiple regression analysis in order to obtain multiple regression formulas for dental age estimation based on the number of third molars present on the evaluated orthopantomogram. No regression models were statistically significant when there was only one-third molar present because of the relatively small number of orthopantomograms that fitted this criterion. A strong agreement was found between the intra- and inter-observer measurements. The statistical analysis revealed both for males and females high Pearson correlation coefficients between contralateral third molars and smaller coefficients between antimeres. The obtained multiple regression formulas are only applicable in certain specified conditions, for instance when four third molars are present the following formulas should be used in a Caucasian male "age=10.2000+0.5122UL+0.5273LL" (developmental stage of upper and lower left third molars) or in case of a female "age=13.6206+0.1933UR+0.5080LR" (developmental stage of upper and lower right third molars). This investigation revealed that the chronological age of a Caucasian individual may be estimated based on regression formulas with a S.D. of 1.52 or 1.56 years for males and females, respectively, when all four third molars are present.

Adolescent↗
PubMed2004

Bacterial cupredoxin azurin as an inducer of apoptosis and regression in human breast cancer.

Azurin, a copper-containing redox protein released by the pathogenic bacterium Pseudomonas aeruginosa, is highly cytotoxic to the human breast cancer cell line MCF-7, but is less cytotoxic toward p53-negative (MDA-MB-157) or nonfunctional p53 cell lines like MDD2 and MDA-MB-231. The purpose of this study was to investigate the underlying mechanism of the action of bacterial cupredoxin azurin in the regression of breast cancer and its potential chemotherapeutic efficacy. Azurin enters into the cytosol of MCF-7 cells and travels to the nucleus, enhancing the intracellular levels of p53 and Bax, thereby triggering the release of mitochondrial cytochrome c into the cytosol. This process activates the caspase cascade (including caspase-9 and caspase-7), thereby initiating the apoptotic process. Our results indicate that azurin-induced cell death stimuli are amplified in the presence of p53. In vivo injection of azurin in immunodeficient mice harboring xenografted human breast cancer cells in the mammary fat pad leads to statistically significant regression (85%, P = 0.0179, Kruskal-Wallis Test) of the tumor. In conclusion, azurin blocks breast cancer cell proliferation and induces apoptosis through the mitochondrial pathway both in vitro and in vivo, thereby suggesting a potential chemotherapeutic application of this bacterial cupredoxin for the treatment of breast cancer.

Antineoplastic Agents↗
PubMed1998

Using logistic regression to describe the length of breastfeeding: a study in Guadalajara, Mexico.

This study seeks, through a logistic regression model, to describe the pattern of breastfeeding duration in Guadalajara, Mexico, during 1993. A multistage random sample of children under 1 year of age (n = 1036) was studied; observational data regarding breastfeeding duration, obtained through a "status quo" procedure, were compared with prevalence rates obtained from the logistic regression model. Modeling the duration of breastfeeding during the first year of life rather than only analyzing observational data helps researchers to understand this process in a dynamic and quantitative way. For example, uncommon indicators of breastfeeding were derived from the model. These indicators are impossible to obtain from observational data. The prevalence curve estimated through the logistic model was adequately fitted to observed data: there were no significant differences between the number or distribution of breastfed infants observed and those predicted by the model. Moreover, the model revealed that less than 40% of the children were breastfed in the fourth month of life; the median age for weaning was 39.3 days; 55% of the potential breastfeeding in the first 4 months did not occur; and the greatest abandonment of breastfeeding in the first 4 months was observed in the first 60 days. Thus, logistic regression seems a suitable option to construct a population-based model that describes breastfeeding duration during the first year of life. The indicators derived from the model offer health care providers valuable information for developing programs that promote breastfeeding.

Adult↗
PubMed2001

Unstimulated salivary flow rates of young children.

OBJECTIVE: Studies on salivary flow rates in human beings have mainly been carried out with adults. The purpose of this study was to determine the unstimulated salivary flow rates of children 4 to 7 years old. In addition, the relative contributions of the variables age, gender, race, height, body weight, dentition status, use of prescription medication, and health status (information obtained from parents) to the unstimulated salivary flow rates of children were also studied. STUDY DESIGN: Data were obtained from children (n = 447) at 2 sites in the United States (site 1, southeast Michigan; site 2, northern Michigan) and at 5 sites in Brazil (site 3, Porto Alegre; site 4, São Paulo; site 5, Belém; and sites 6 and 7, sites in Rio de Janeiro). In northern Michigan (site 2) the participants were cognitively or developmentally disabled, or both. In Rio de Janeiro (site 7), a group of 8- to 12-year-olds served as a control group. Saliva samples were collected for 3 minutes between 9 AM and noon in the spring or summer, and the saliva rate was determined gravimetrically. Data were analyzed by analysis of variance, bivariate analysis, and regression analysis. RESULTS: The secretion rates at the 7 sites were (in milliliters per minute) 0.19 +/- 0.15, 0.23 +/- 0.28, 0.34 +/- 0.23, 0.48 +/- 0.37, 0.25 +/- 0.27, 0.37 +/- 0.28, and 0.61 +/- 0.34, respectively. There were significant differences among sites (P <.0001). The older group (site 7) had flow rates that were significantly higher than the flow rates of any other group. In addition, children from Michigan (sites 1 and 2) had significantly lower rates than most groups of children in Brazil. Girls had lower unstimulated salivary flow rates than boys did at all the sites, but the differences were not statistically significant. Race was shown not to affect the flow rates. The use of any prescription medication by children in the previous 3 months was associated with lower salivary flow rates than were found in children not using prescription medication. Children who were in good health and who had no previous medical conditions had higher flow rates--but not significantly so. Higher flow rates occurred in children with mixed dentition than in children with primary dentition, although again the differences were not statistically significant. Regression analysis revealed weight to be of significance in explaining the variability of the unstimulated salivary flow rates at 2 sites, height at 1 site, the use of prescription medication at 2 sites, and age at 1 site. CONCLUSIONS: The unstimulated salivary flow rates in children in the northern United States are comparable with those reported for Japanese children, whereas the flow rates of children in Brazil are comparable with those reported for North American and European adults. In addition, none of the demographic variables/parameters tested contributed consistently to the variability of the unstimulated salivary flow rates in children at the 7 sites assessed in this study.

Age Factors↗
PubMed1991

The pre- and poststerilization predictors of poststerilization regret in husbands and wives.

Husbands and wives from 141 tubal sterilization couples and 162 vasectomy couples were interviewed just prior to sterilization and then again 1 and 2 years later. We conducted linear regression analyses to determine the pre- and poststerilization predictors of poststerilization regret in each of the four gender x method groups (tubal husbands, tubal wives, vasectomy husbands, vasectomy wives). We confirmed a number of hypotheses based on the research literature and our own earlier work. Both individual and couple factors contributed to the development of regret, as did both pre- and poststerilization factors. An important finding was the degree to which regret among the nonsterilized respondents (tubal husbands, vasectomy wives) was affected by pre- and poststerilization interaction with their spouses.

Adult↗
PubMed1995

Predictors of success of reversal of sterilization.

OBJECTIVE: To determine the prognostic variables effecting the successful pregnancy outcome of reversal of sterilization. DESIGN: Demographic and clinical history data were collected prospectively. SETTING: Division of Infertility and Reproductive Endocrinology, Department of Obstetrics and Gynecology, University of British Columbia, Vancouver, Canada. PATIENTS AND INTERVENTION: Between 1981 and 1992, 217 consecutively referred patients underwent reversal of sterilization by a single surgeon using microsurgical techniques. MAIN OUTCOME MEASURES: Prognostic variables associated with success were examined using logistic regression and expressed as odds ratios with corresponding 95% confidence intervals. RESULTS: Age at reversal was a significant factor with the odds of a successful outcome for those < or = 35 years compared with those > 35 years being 2.3 with a 95% confidence interval of 1.3 to 4.1. There was some evidence that average tubal length as categorized in 2-cm intervals was a significant prognostic factor with the odds of a successful outcome for those with an average length of > 4 cm to those < or = 4 cm being 5.3 with a 95% confidence interval of 1.4 to 20.0. CONCLUSIONS: Nonsubjective analysis of the prognostic variables of reversal of sterilization associates only age and tubal length of > 4 cm with intrauterine pregnancy.

Adult↗
PubMed1994

Correlates of reactivity: problems with regression based methods.

Computer simulations were used to compare change score and regression methods for identifying relationships between a third variable and reactivity. When the third variable was not correlated with baseline, the two methods were equally effective for detecting relationships between the third variable and change. However when there was a correlation between the third variable and baseline, regression measures of change yielded a high rate of Type one errors. Regression methods also had less power for detecting a relationship between the third variable and change when the relationship was in the direction opposite to the sign of the correlation with baseline. The implication of these findings for two areas of research are discussed: the relationship of fitness to heart rate reactivity and the relationship of hypertension to blood pressure reactivity.

Computer Simulation↗
PubMed2003

Environmental, dietary, demographic, and activity variables associated with biomarkers of exposure for benzene and lead.

Classification and regression tree methods represent a potentially powerful means of identifying patterns in exposure data that may otherwise be overlooked. Here, regression tree models are developed to identify associations between blood concentrations of benzene and lead and over 300 variables of disparate type (numerical and categorical), often with observations that are missing or below the quantitation limit. Benzene and lead are selected from among all the environmental agents measured in the NHEXAS Region V study because they are ubiquitous, and they serve as paradigms for volatile organic compounds (VOCs) and heavy metals, two classes of environmental agents that have very different properties. Two sets of regression models were developed. In the first set, only environmental and dietary measurements were employed as predictor variables, while in the second set these were supplemented with demographic and time-activity data. In both sets of regression models, the predictor variables were regressed on the blood concentrations of the environmental agents. Jack-knife cross-validation was employed to detect overfitting of the models to the data. Blood concentrations of benzene were found to be associated with: (a) indoor air concentrations of benzene; (b) the duration of time spent indoors with someone who was smoking; and (c) the number of cigarettes smoked by the subject. All these associations suggest that tobacco smoke is a major source of exposure to benzene. Blood concentrations of lead were found to be associated with: (a) house dust concentrations of lead; (b) the duration of time spent working in a closed workshop; and (c) the year in which the subject moved into the residence. An unexpected finding was that the regression trees identified time-activity data as better predictors of the blood concentrations than the measurements in environmental and dietary media.

Activities of Daily Living↗
PubMed1990

Appropriateness of composites in structural equation models.

The appropriateness of using composites instead of multiple indicators in a structural model of physical health was evaluated. Liang's (1986) specification of self-reported physical health is a relatively complex multiple indicators model which may not be practical in actual application. To simplify this formulation, a two-stage strategy was used. First, reliability was estimated for each composite to fix the measurement error variance and the regression of the composite on the latent variable. Second, the model was reestimated by constraining these parameters. Regression analyses were undertaken to assess the impact of using composites instead of multiple indicators. Parameter estimates for causal linkages and residual error variances based on multiple indicators approach were closely reproduced by using composites, thus providing justification for the proposed strategy.

Activities of Daily Living↗
PubMed2005

Acute stress disorder and posttraumatic stress disorder in children and adolescents involved in assaults or motor vehicle accidents.

OBJECTIVE: The authors investigated acute stress disorder and later posttraumatic stress disorder (PTSD) in children and adolescents who had been involved in assaults or motor vehicle accidents. METHOD: They interviewed 93 patients 10-16 years old who were seen in an emergency department for having been assaulted or involved in a motor vehicle accident within 4 weeks after the assault or accident to assess acute stress disorder. At 6 months, they reinterviewed 64 (68.8%) of the patients to assess PTSD. RESULTS: At initial interview, 18 (19.4%) of the 93 patients had acute stress disorder and 23 (24.7%) met all acute stress disorder criteria except dissociation. At 6 months, eight of the 64 patients (12.5%) had PTSD. Acute stress disorder and PTSD did not differ in prevalence between patients who had been assaulted and those who had been in accidents. Sensitivity and specificity statistics and regression modeling revealed that the diagnosis of acute stress disorder was a good predictor of later PTSD but that dissociation did not play a significant role. CONCLUSIONS: Acute stress disorder has merit as a predictor of later PTSD in children and adolescents, but dissociation has questionable utility.

Accidents, Traffic↗
PubMed2000

[Cholera and living conditions, Brazil].

INTRODUCTION: Factors associated with precarious living and environmental conditions are frequently cited as major obstacles for the control of cholera outbreaks and epidemics. The purposes of the study are to evaluate the contribution of factors associated with the population living conditions and correlate the environmental problems with the onset of cholera and its subsequent impact. METHODS: Using a multiple linear regression by the backward stepwise method, and with the researcher's interaction, the study correlated socioeconomic indicators with cholera incidence rates in some counties of Pernambuco State, Brazil, during the year of 1992. RESULTS/CONCLUSIONS: The results of the adjusted model showed that the proportion of households without tap water was the variable that contributed the most to the increasing fluctuation of cholera incidence rates. Two other factors, the proportion of households without sewage and the proportion of householders with an income less than or equal to the minimum wage, also revealed a positive association with cholera incidence rates with statistically significant regression coefficients. The proportion of households with no sanitary installations whatsoever showed a negative association with cholera incidence rates, suggesting that sewage disposal, such as open-air sewage ditches, that is not part of the public sewage disposal system, increases the risk of environmental contamination The results indicate that having an adequate tap water supply is of maximum priority for cholera prevention.

Brazil↗
PubMed2001

Drug therapy and prevalence of erectile dysfunction in the Massachusetts Male Aging Study cohort.

STUDY OBJECTIVE: To examine the association of commonly used drugs with erectile dysfunction (ED) at two time points. DESIGN: Population-based, cross-sectional, survey analysis. PARTICIPANTS: Randomly selected cohort of men in the Massachusetts Male Aging Study (MMAS) that included 1476 men for the baseline (1987-1989) and 922 for the follow-up (1995-1997) analyses. INTERVENTION: Crude associations between specific drug categories were examined with chi2 statistics. Logistic regression analysis was used to separate the effect of drugs from the influence of heart disease, hypertension, untreated diabetes, or depressive symptoms. MEASUREMENTS AND MAIN RESULTS: In the MMAS, medical history, current drug use, and erectile function status were ascertained with in-home interviews. In unadjusted analyses, thiazide and nonthiazide diuretics, beta-blockers, calcium channel blockers, angiotensin-converting enzyme inhibitors, benzodiazepines, digitalis, nitrates, 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors, and histamine2 receptor antagonists were associated with prevalent ED. Adjustment for comorbidities and health behaviors attenuated these associations, with only nonthiazide diuretics and benzodiazepines remaining statistically significant. CONCLUSION: Several common drugs may increase prevalence of ED; however, additional data from larger populations are needed to determine whether these associations are independent of underlying health conditions and to explore the effects of dosage and duration of use.

Adult↗
PubMed2002

Median regression with censored cost data.

Because of the skewness of the distribution of medical costs, we consider modeling the median as well as other quantiles when establishing regression relationships to covariates. In many applications, the medical cost data are also right censored. In this article, we propose semiparametric procedures for estimating the parameters in median regression models based on weighted estimating equations when censoring is present. Numerical studies are conducted to show that our estimators perform well with small samples and the resulting inference is reliable in circumstances of practical importance. The methods are applied to a dataset for medical costs of patients with colorectal cancer.

Biometry↗
PubMed1997

Hierarchical modeling of gene-environment interactions: estimating NAT2 genotype-specific dietary effects on adenomatous polyps.

Data sparseness currently limits gene-environment interaction estimation. To improve effect estimates of gene-environment interactions, we give an overview of one approach, hierarchical modeling, and propose a two-stage hierarchical model. The first stage is a logistic model for the joint effects of the genetic and environmental factors. The second stage regresses the joint effects on genotype-specific enzymatic activity of the environmentally derived substrate. The model is illustrated using a case-control study of adenomas of the large bowel, for which NAT2 genotype and dietary data were collected. The first-stage interactions of dietary components and genotype were regressed on initial conversion rates of dietary heterocyclic amines to aryl nitrenium ions. We fit the hierarchical model by penalized likelihood. Compared to effect estimates from maximum-likelihood logistic regression, hierarchical results are more reasonable and precise. These results lend further support to previous observations that hierarchical regression is preferable to ordinary logistic regression when multiple factors and their interactions are being studied. We propose that hierarchical modeling can act as a bridge between molecular epidemiology studies and laboratory data, combining both efficiently.

Adenomatous Polyps↗
PubMed1990

Factors influencing success or failure after reversal of sterilization: a multivariate approach.

Reversal of sterilization was performed in 215 patients after unipolar coagulation, Fallope ring or clip sterilization, or bipolar coagulation or a Pomeroy sterilization. All patients had a follow-up of at least 1 year. Prognostic variables were derived from the previous history, the preoperative fertility assessment, and the actual operation. Some variables were combined into a fertility reducing factor. A first selection was performed by comparing the preoperative and perioperative variable scores of the successes (defined as all intrauterine pregnancies within 1 year after surgery) with those of the failures (no intrauterine pregnancy within 1 year). Eight variables appeared to have prognostic potential. Taking account of the inter-relations of those variables by using logistic regression analysis, the localization of the anastomosis, the number of anastomosed tubes, and the presence of a fertility reducing factor were selected as the most predictive ones. Using this logistic model, an estimate of the pregnancy outcome could be calculated for each individual patient.

Adult↗
PubMedDate not supplied

A regression approach to the projection of U.S. fertility based on past fertility data.

The method of fertility projection used by the U.S. Bureau of the Census involves assumptions about the ultimate cohort total fertility rate and the ultimate cohort mean age at childbearing based on recent levels of fertility and women's birth expectations. This paper provides an outline of a general regression approach to fertility projection based on past data which would generate these two ultimate cohort characteristics. The technique is illustrated by using the U.S. single-year age-specific fertility rates up to 1986 for total women and projecting them indefinitely into the future until they become stable for both calendar years and cohorts.

Adolescent↗
PubMed1992

Predictors of repeat pregnancy outcome among black and Puerto Rican adolescent mothers.

This prospective study investigated predictors of repeat pregnancies by 12 months after the delivery of a first child and their outcomes in inner-city adolescent mothers. The sample included four groups: those who had therapeutic abortions, miscarriages, full-term deliveries, and no repeat pregnancy. The therapeutic abortion group had more pregnancies before their first delivery (41%) than did full term (20%) and no repeat (15%), p less than .01. More delayed grade placement was found in therapeutic abortion -1.6 years (1.3) and full term -1.8 years (.9) than in no repeat pregnancy -.6 years (1.1), p less .001. Reading achievement scores were higher in no repeat 86.3 (17.1) than in full term 75.0 (16.5), p less than .05. School attendance was higher in no repeat (65%) than in therapeutic abortion (35%) and full-term (24%) p less than .01 groups. Depressive symptoms at baseline were higher among therapeutic aborters 18.9 (9.9) than among full term 10.2 (8.2) and no repeat pregnancy groups 12.2 (6.2). Logistic regression analyses identified delayed grade placement as the most important predictor of pregnancy outcome.

Achievement↗
PubMed2005

Three-dimensional texture analysis of cancellous bone cores evaluated at clinical CT resolutions.

The objective of this study was to determine if three-dimensional (3D) Haralick co-occurrence texture measures calculated from low-resolution CT images of trabecular bone correlate with 3D structural indices measured from high-resolution CT images. Thirty-three cubical regions of trabecular bone from human calcanei were analyzed using images obtained from a micro-computed tomography (micro-CT) scanner. 3D measures of bone architecture were calculated. The original images were then subsampled by factors of 5, 10, 15, and 20, and 3D texture features were calculated for each set of subsampled images. Linear regression models showed that co-occurrence texture features were significantly correlated with structural indices. Over 90% of the variation in three different structural indices was explained in two-variable regression models using texture features as predictors when the voxel side length was reduced by a factor of 10. Texture features calculated from clinical images may increase our ability to obtain trabecular bone architectural information when high-resolution images are unobtainable.

Aged↗