SEARCH · PubMed Health

Results for “Logistic Models”

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 631 records · Page 35Linked to original sources

PubMed1997

Regression analysis of interval-censored failure time data.

Interval-censored failure time data often occur, for example, in clinical trials or longitudinal studies. For the regression analysis of such data, there have been a number of methods proposed based on continuous regression models such as Cox's proportional hazards model. In practice, however, observed interval-censored data that arise from clinical trials are often given in a discrete scale due to the nature of clinical trials although the underlying variable may be continuous. It is apparent that in this case, one can better handle analysis of the data with the methods based on discrete models. In this paper, I propose a method based on a discrete logistic model for the regression analysis of interval-censored failure time data with focus on the comparison of failure time distributions among different treatments. I discuss the relationship between the proposed method and existing methods.

Clinical Trials as Topic↗

PubMed1995

Kinetic models of batch and fed-batch culture of single-cell protein with double carbon sources.

Unstructured logistic models were applied to simulate the kinetic process of the SCP batch culture with two species of yeast, in which glucose, xylose, or a mixture of the two were used as carbon sources. A modified logistic growth model was used to describe the fed-batch process of using mixed carbon sources. The model parameters were determined by the simulated model with the experimental data. The simulated curves of the models were in agreement with the experimental data.

Candida↗

PubMed2005

Role of the serotonin transporter gene and family function in adolescent alcohol consumption.

BACKGROUND: That the extent to which a particular individual will engage in problematic behaviors such as delinquency, violence, or drug abuse is determined by the way psychosocial, situational, and hereditary factors interact is widely accepted. However, only recently have researchers begun to investigate the interactions between specific genotypes and psychosocial factors in relation to behavior. The purpose of the present study was to investigate possible interactions between a polymorphism in the promoter region of the serotonin transporter (5-HTT) gene and family relations on adolescent alcohol consumption. METHODS: A cross-sectional study with a randomized sample from a total population of 16- and 19-year-old adolescents from a Swedish county was conducted. Eighty-one male and 119 female adolescents, who volunteered to participate after having answered a questionnaire, were randomly selected from quartiles of volunteers representing various degrees of psychosocial risk behavior. RESULTS: 5-HTT genotype (p=0.029) and family relations (p=0.022) predicted alcohol consumption independently as well as through an interaction with one another (p=0.05). The model explained 11% of the variance in alcohol consumption. In a binary logistic model, we found that adolescents with the LS variant of the 5-HTT gene and with family relations being "neutral" or "bad" had a 12- to 14-fold increased risk for high intoxication frequency. CONCLUSIONS: In sum, our results show that a functional polymorphism of the 5-HTT genotype, family relations, and interactions between these variables predict adolescent alcohol consumption in a randomized sample of adolescents.

Adolescent↗

PubMed1990

Maximum likelihood regression methods for paired binary data.

We discuss maximum likelihood methods for analysing binary responses measured at two times, such as in a cross-over design. We construct a 2 x 2 table for each individual with cell probabilities corresponding to the cross-classification of the responses at the two times; the underlying likelihood for each individual is multinomial with four cells. The three dimensional parameter space of the multinomial distribution is completely specified by the two marginal probabilities of success of the 2 x 2 table and an association parameter between the binary responses at the two times. We examine a logistic model for the marginal probabilities of the 2 x 2 table for individual i; the association parameters we consider are either the correlation coefficient, the odds ratio or the relative risk. Simulations show that the parameter estimates for the logistic regression model for the marginal probabilities are not very sensitive to the parameters used to describe the association between the binary responses at the two times. Thus, we suggest choosing the measure of association for ease of interpretation.

Humans↗

PubMedDate not supplied

[Migration from a rural zone to an urban one is associated with android distribution of body fat in obese women].

INTRODUCTION: Studies about migration to industrialized countries have shown an increased prevalence of diabetes, obesity and dyslipidaemias, all of them related to android body fat distribution. Migration status might be influence body fat distribution but it has not been sufficiently investigated. The aim of this study is to determine the relationship between body fat distribution and migration from rural to urban areas in Mexico. MATERIAL AND METHODS: This sequential sample of 433 women were seen in the outpatient obesity clinic of four federal states: Tabasco (n = 81), Mexico City (n = 166), Coahuila (n = 80), and Yucatan (n = 106). Migration history from rural to urban area, familial history of diabetes, ages of onset of obesity, height and weight circumferences were obtained. A regression logistic model was used and maintained as dependent variable body fat distribution. Age and federal state were considered as confounders and they adjusted the model. RESULTS: Migrating women from rural to urban area were 121 (27.9%). The waist circumference was higher in Tabasco (102.2 +/- 12 cm), and lesser in Yucatan (93.6 +/- 15 cm, p < 0.001); no differences were found for hip circumference. The logistic regression model showed that body fat distribution is associated to migration from rural to urban area, and also to diabetes of mother and age of onset of obesity. CONCLUSIONS: Migrating from rural to urban area is a risk factor for android body fat distribution and this risk increases with age, history of diabetes in mother and adulthood onset o obesity.

Abdomen↗

PubMed1994

Kinetics of fiber digestion from in vitro gas production.

In vitro gas production, measured by computer-interfaced pressure sensors, was used to follow the digestion of a crystalline processed cellulose, a bacterial cellulose, and mixtures of these substrates by mixed ruminal bacteria. A first-order, substrate limited model (simple exponential with lag) and two bacterial growth models (logistic, Gompertz) were tested to fit these data. No single pool model gave an optimal fit to all substrates, but dual pool versions of both the logistic and Gompertz models fitted the data extremely well. Derivations of these models in the context of gas production are presented. The dual pool version of the exponential model commonly used to analyze fiber digestion was not able to reproduce the slope variations seen with mixed substrates. A modified dual pool logistic equation, with a single lag value, was selected to model the in vitro digestion of these substrates. The model was able to predict adequately both the input composition and the kinetic parameters for a defined mixture and gave a good fit (r2 > .995) to data from all the single and mixed substrates tested. This model may be useful for interpreting gas accumulation from natural feedstuffs.

Animal Feed↗

PubMedDate not supplied

Modeling pore size distribution in cellulose rolled stationary phases.

Rolled stationary phases are fabrics (i.e., nonparticulate phases) that rapidly separate proteins from salts on the basis of size exclusion. Pore size and pore size distributions in the stationary phase determine how different size molecules distribute between the stationary and mobile phases in liquid chromatography columns. The potential for size exclusion chromatography by fabrics is not initially obvious because their interlaced structures are atypical for size exclusion supports. A simple logistic model fits the pore size distribution of a rolled stationary phase when pore sizes were measured using PEG, Dextran, D2O, glucose, and NaCl probes. When the fabric is treated with cellulase enzymes, the water-accessible pores uniformly decrease and peak retention is lower. The logistic function model captures this result and enables comparison of pore size distribution curves between enzyme-treated and untreated fabrics in rolled stationary phase columns.

Animals↗

PubMed2004

Binary classification of dyslipidemia from the waist-to-hip ratio and body mass index: a comparison of linear, logistic, and CART models.

BACKGROUND: We sought to improve upon previously published statistical modeling strategies for binary classification of dyslipidemia for general population screening purposes based on the waist-to-hip circumference ratio and body mass index anthropometric measurements. METHODS: Study subjects were participants in WHO-MONICA population-based surveys conducted in two Swiss regions. Outcome variables were based on the total serum cholesterol to high density lipoprotein cholesterol ratio. The other potential predictor variables were gender, age, current cigarette smoking, and hypertension. The models investigated were: (i) linear regression; (ii) logistic classification; (iii) regression trees; (iv) classification trees (iii and iv are collectively known as "CART"). Binary classification performance of the region-specific models was externally validated by classifying the subjects from the other region. RESULTS: Waist-to-hip circumference ratio and body mass index remained modest predictors of dyslipidemia. Correct classification rates for all models were 60-80%, with marked gender differences. Gender-specific models provided only small gains in classification. The external validations provided assurance about the stability of the models. CONCLUSIONS: There were no striking differences between either the algebraic (i, ii) vs. non-algebraic (iii, iv), or the regression (i, iii) vs. classification (ii, iv) modeling approaches. Anticipated advantages of the CART vs. simple additive linear and logistic models were less than expected in this particular application with a relatively small set of predictor variables. CART models may be more useful when considering main effects and interactions between larger sets of predictor variables.

Adult↗

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↗

PubMed1998

Does percent reduction in seizure frequency correlate with plasma concentration of anticonvulsant drugs? Experience with four anticonvulsant drugs.

OBJECTIVE: To investigate the relationship between the percentage reduction in seizure frequency in patients with epilepsy and plasma concentrations after oral administration of 4 anticonvulsant drugs. METHODS: Patients with a minimum of 25% reduction in their seizure frequency from their baseline value were declared responders. The percentage reduction in seizure frequency was plotted against plasma concentrations with use of pharmacodynamic models (linear, log-linear, Emax, and sigmoidal Emax models). In addition to pharmacodynamic models, a logistic regression model was also fitted to the concentration-response data, with a value of 1 for responders and 0 for nonresponders. RESULTS: The concentration-effect relationship could not be adequately described either by the pharmacodynamic models or by the logistic regression analysis. CONCLUSIONS: Based on the results obtained from both pharmacodynamic models and logistic regression analysis the percentage reduction in seizure frequency may not be a true surrogate marker for anticonvulsant drugs to establish a pharmacodynamic relationship with plasma concentrations.

Anticonvulsants↗

PubMed1997

Genetic analysis of families with nonsmoking lung cancer probands.

As part of a genetic epidemiologic study of lung cancer among nonsmokers, we investigated the role of genetic predisposition in familial aggregation. Cases were identified from the Metropolitan Detroit Cancer Surveillance System. Information on lung cancer occurrence, smoking habits (active or passive), and chronic respiratory diseases in first-degree relatives was obtained for 257 nonsmoking lung cancer probands (71 males, 186 females) diagnosed at ages 40-84 years. Among the 2,021 first-degree relatives, 24 (2.6%) males and 10 (1.1%) females were reported as having lung cancer. The occurrence of lung cancer among smoking and nonsmoking relatives was 4.5% and 1.1% in males and 2.8% and 0.4% in females, respectively. To evaluate the role of a putative Mendelian gene (one locus, two alleles) in the presence of other risk factors, we performed complex segregation analyses on the data using two different regressive model approaches [Segregation Analysis of a Discrete Trait Under a Class A Regressive Logistic Model, V4.0 (REGD) and Segregation Analysis of a Truncated Trait, V2.0, Model 1(REGTL)] as implemented in the Statistical Analysis for Genetic Epidemiology (SAGE) program. Using either approach, an environmental model best explained the observed lung cancer aggregation in families ascertained through nonsmoking probands. Based on our final model, only 0.04% of this population had a very high risk and 4.2% had a moderate risk of lung cancer. The rest of the population had virtually no risk of lung cancer during their lifetime unless they have multiple risk factors. Among the high-risk individuals without any risk factor under study, the estimated risks at ages 40, 60, and 80 years in males were 16.7%, 83.6%, and 95.4%, and in females were 14.0%, 72.2%, and 88.0%, respectively. Among at-moderate-risk smokers the estimated risks at the same age and gender groups were essentially the same as in the high-risk nonsmokers. Our results suggest that the pattern of lung cancer occurrence in families of nonsmoking lung cancer patients differs from that in families of smoking lung cancer patients. Despite the profound effect of smoking on the risk of lung cancer, other environmental and/or genetic risk factors need to be identified.

Adult↗

PubMed1998

The incidence of sheep strike by Lucilia sericata on sheep farms in Britain: a simulation model.

Using a combination of a temperature-dependent simulation of the population dynamics of Lucilia sericata and a logistic model of sheep susceptibility, based on patterns of faecal soiling, fleece length and fleece humidity, a deterministic simulation model has been constructed to predict the incidence of sheep strike by L. sericata on lambs and ewes in Britain. Strikes are assigned to individual sheep in the flock according to a negative binomial distribution. The model shows that the majority of the first strikes of the season occur equally on ewes and lambs, when ewe fleeces are long and before lamb susceptibility increases though faecal soiling. With each subsequent generation of gravid blowflies, however, a greater proportion of strikes occur on lambs, while the number of strikes observed on ewes remains relatively consistent throughout the season. The increase in strike of lambs is due to the seasonal increase in lamb faecal soiling, associated with rising helminth infections and lamb fleece growth. The model shows that at the beginning of the season the incidence of strike is limited by the low number of flies present: towards the end of seasons, however, the blowfly population had grown large enough for the number of strikes to be limited by the number of susceptible ewes and lambs. The model shows that the seasonal incidence of strike is highly sensitive to the interactions between temperature, rainfall and would length. Strike was most prevalent under warm, wet conditions. However, there is a critical period during spring, where the short fleeces of lambs and ewes, in the latter case due to shearing, results in the desiccation of egg batches. If this coincides with dry conditions, the high levels of mortality result in suppression of the fly population and a reduction in the subsequent incidence of strike.

Animals↗

PubMed1997

[Clinical assessment systems in the diagnosis of pulmonary thromboembolism].

We proposed developing two symptom-based systems for assessing the presence of pulmonary thromboembolism (TEP) in our practice, using a standardized questionnaire and multivariate models. Data were collected from September 1993 through November 1994 (case reports, physical examination findings and complementary test results) of patients admitted to our ward with a suspicion of TEP. The calculated odds ratio for each of the variables recorded were used as weights to determine their relevance or not for the group at risk for TEP. The yield of the two systems developed (a weights system and a logistical model) were studied by plotting ROC curves. Eighty-two patients (40 women and 42 men, mean age 60.94 +/- 14.39 years) were admitted. The questionnaire had a sensitivity of 88% and a specificity of 75%, a positive predictive value of 94% and a negative predictive value of 60%. The logistical regression model had a sensitivity of 96.3% for a diagnosis of TEP with inclusion of the following variables: female sex, disease-related immobility, presence of deep venous thrombosis (DVT) in the lower extremities and the appearance of unexplained dyspnea. Neither system was clearly superior to the other for arriving at a clinical diagnosis of TEP.

Aged↗

PubMed2006

Revascularization and cardioprotective drug treatment in myocardial infarction patients: how do they impact on patients' survival when delivered as usual care.

BACKGROUND: Randomized clinical trials showed the benefit of pharmacological and revascularization treatments in secondary prevention of myocardial infarction (MI), in selected population with highly controlled interventions. The objective of this study is to measure these treatments' impact on the cardiovascular (CV) mortality rate among patients receiving usual care in the province of Quebec. METHODS: The study population consisted of a "naturalistic" cohort of all patients > or = 65 years old living in the Quebec province, who survived a MI (ICD-9: 410) in 1998. The studied dependant variable was time to death from a CV disease. Independent variables were revascularization procedure and cardioprotective drugs. Death from a non CV disease was also studied for comparison. Revascularization procedure was defined as percutaneous transluminal coronary angioplasty (PTCA) or coronary artery bypass graft (CABG). The exposure to cardioprotective drugs was defined as the number of cardioprotective drug classes (Acetylsalicylic Acid (ASA), Beta-Blockers, Angiotensin-Converting Enzyme (ACE) Inhibitors, Statins) claimed within the index period (first 30 days after the index hospitalization). Age, gender and a comorbidity index were used as covariates. Kaplan-Meier survival curves, Cox proportional hazard models, logistic regressions and regression trees were used. RESULTS: The study population totaled 5596 patients (3206 men; 2390 women). We observed 1128 deaths (20%) within two years following index hospitalization, of them 603 from CV disease. The CV survival rate at two years is much greater for patients with revascularization, regardless of pharmacological treatments. For patients without revascularization, the CV survival rate increases with the number of cardioprotective drug classes claimed. Finally, Cox proportional hazard models, regression tree and logistic regression analyses all revealed that the absence of revascularization and, to a lower extent, absence of cardioprotective drugs were major predictors for CV death, even after adjusting for age, gender and comorbidity. CONCLUSION: Considering usual care management of MI in the province of Quebec in 1998, CV survival is positively correlated to the presence of a revascularization procedure and to the intensity of cardioprotective pharmacological treatment. These results are coherent with data from randomized control trials.

Adrenergic beta-Antagonists↗

PubMed1988

Carrier detection in hemophilia A: ABO blood group, multiple measurements, and application of logistic discrimination.

In healthy 20- to 50-year-old women, the ABO blood group has a significant effect on levels of von Willebrand factor (VWF:Ag, formerly VIIIR:Ag) and on factor VIII activity (F.VIII:C). However, there is no significant effect of ABO group or subject age on the ratio log e(F.VIII:C/VWF:Ag). Multiple measurements of the "ratio" on possible carriers of hemophilia A may be combined with pedigree information using logistic discrimination to yield final risk assessment. To reduce misclassification of carriers as normal women, a lower limit, specified by the logistic model, is set on the logistic carrier probabilities. In this study, the proportion of blood group A for a population of obligate carriers was significantly higher than that expected for the general population (60% vs. 42%); for a population of control women it was lower than expected (22.5 vs. 42%). The effect for the carriers came primarily from daughters of affected fathers, as 81.3% were of blood group A. These observations indicate that a "universal" discriminant should be applied with caution.

ABO Blood-Group System↗

PubMed2005

Bayesian analysis of misclassified binary data from a matched case-control study with a validation sub-study.

Bayesian methods are proposed for analysing matched case-control studies in which a binary exposure variable is sometimes measured with error, but whose correct values have been validated for a random sample of the matched case-control sets. Three models are considered. Model 1 makes few assumptions other than randomness and independence between matched sets, while Models 2 and 3 are logistic models, with Model 3 making additional distributional assumptions about the variation between matched sets. With Models 1 and 2 the data are examined in two stages. The first stage analyses data from the validation sample and is easy to perform; the second stage analyses the main body of data and requires MCMC methods. All relevant information is transferred between the stages by using the posterior distributions from the first stage as the prior distributions for the second stage. With Model 3, a hierarchical structure is used to model the relationship between the exposure probabilities of the matched sets, which gives the potential to extract more information from the data. All the methods that are proposed are generalized to studies in which there is more than one control for each case. The Bayesian methods and a maximum likelihood method are applied to a data set for which the exposure of every patient was measured using both an imperfect measure that is subject to misclassification, and a much better measure whose classifications may be treated as correct. To test methods, the latter information was suppressed for all but a random sample of matched sets.

Bayes Theorem↗

PubMed1995

Standardized estimates from categorical regression models.

We consider the problem of interpreting categorical regression models, such as the polytomous logistic model, the continuation-ratio model, the stereotype model, and the cumulative-odds model. We present a method to convert categorical regression coefficients into estimates of standardized fitted probabilities, probability differences and probability ratios. We use a delta-method approach to estimate standard errors. We then present a small simulation study to compare different transforms for setting confidence limits, and provide an illustration of our approach in an observational study of drug therapy of polymyositis.

Algorithms↗

PubMedDate not supplied

Mathematical modelling of bacterial growth at subinhibitory levels of aminoglycosides.

The subinhibitory effect of antibiotics has often been studied without any clear theoretical framework; we have chosen to use mathematical bacterial growth modelling as a useful tool to analyse these biological states in a more rigorous manner. Since their mode of action and molecular target are relatively well known, aminoglycosides were well suited for this more sophisticated study of subinhibitory action. We have shown that two models (the Monod and the logistic models) regularly used in bacteriology, were adequate to describe these effects in a glucose-limited medium. A change of model, according to antibiotic concentration, revealed the existence of two separate actions. At lower concentrations, inhibition affected mainly glucose use, the substrate remained limiting and growth mode did not change. As soon as the concentration exceeded a threshold, growth was totally disturbed, probably through a physiological "catastrophe". This threshold can be used to estimate bacterial susceptibility to these antibiotics.

Aminoglycosides↗