PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “Conditional logit”

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 55 records · Page 3Linked to original sources

A transitional model of barrier methods compliance with unbalanced loss to follow-up.

Differential loss to follow-up can substantially influence the evaluation of treatment effects on a dichotomous outcome of interest in longitudinal trials. The use of transitional models incorporating loss to follow-up as an additional category of response and the nature of the correlated responses can provide a comprehensive view of a trial with unbalanced loss to follow-up. Under the Markov assumption, transitional models estimate the probability of changing from one outcome to another outcome between follow-up visits. Patterns of the response variable can be described by the estimated transition probabilities. The effects of intervention and covariates on the outcome of interest can also be estimated using a conditional likelihood function or a multinomial logit regression. Data from a randomized barrier method study designed to compare the proportion of participants using barrier methods consistently in two counselling groups are used to illustrate the proposed model.

Adolescent↗

Size at birth and neonatal and postneonatal mortality.

UNLABELLED: Birthweight is a crude indicator of size at birth. Some neonatologists and obstetricians have advocated the use of ponderal index and birthlength to characterize size at birth. This paper examines the associations between various size-at-birth indicators and neonatal and postneonatal mortality, with an emphasis on ponderal index and birthlength. Size at birth, gestational age and mortality data for about one million babies born alive in Sweden between 1987 and 1995 were collected from the Swedish Medical Birth Registry. A multinomial logit regression was used to estimate conditional odds ratios. Birthlength and ponderal index were independently associated with neonatal and postneonatal mortality. In the latter period, ponderal index was only weakly associated with mortality. The associations were not sensitive to exclusion of cases of congenital anomalies and adjustment for gestational age. CONCLUSION: Birthlength is strongly associated with both neonatal and postneonatal mortality; ponderal index is strongly associated with neonatal, but weakly with postneonatal mortality. The findings are consistent with previous hypotheses about a transient effect of ponderal index and a persistent effect of birthlength.

Birth Weight↗

Behavioral responses of bison and elk in Yellowstone to snowmobiles and snow coaches.

Managers of public lands are charged with protecting some of our most important natural resources and ecosystems, while providing for their use and enjoyment by visitors. Almost one million visitors entered Yellowstone National Park by motorized means on snowmobiles (87%) or snow coaches (13%) during 1992-2003. Most vehicles toured the central portion of the park where bison (Bison bison) and elk (Cervus elaphus) concentrate in geothermal areas. We sampled >6500 interactions between groups of these species and groups of snowmobiles and snow coaches (collectively, OSV, over-snow vehicles) during five winters (1999-2000, 2002-2004). Multinomial logits models were used to identify conditions leading to behavioral responses. Elk responded three times as often (52%) as bison (19%) during interactions with groups of snowmobiles and snow coaches due to increased vigilance responses (elk, 44%; bison, 10%). However, the frequency of higher-intensity movement responses by bison and elk were similar (6-7% travel, 1-2% flight, <1% defense) and relatively low compared to other studies of ungulates and snowmobile disturbance. The likelihood of active responses by bison and elk increased significantly if animals were on or near roads, groups were smaller, or humans approached. The likelihood of an active response by bison decreased within winters having the largest visitation, suggesting some habituation to snowmobiles and snow coaches. There was no evidence that snowmobile use during the past 35 years affected the population dynamics or demography of bison or elk. Thus, we suggest that regulations restricting levels and travel routes of over-snow vehicles (OSVs) were effective at reducing disturbances to bison and elk below a level that would cause measurable fitness effects. We recommend park managers consider maintaining OSV traffic levels at or below those observed during our study. Regardless, differing interpretations of the behavioral and physiological response data will continue to exist because of the diverse values and beliefs of the many constituencies of Yellowstone.

Animals↗

[The discrimination of mono-syllable words in noise in listeners with normal hearing].

The discrimination of mono-syllable words (67S word-list) pronounced by a male and a female speaker was investigated in noise in 39 normal hearing subjects. The subjects listened to the test words at a constant level of 62 dB together with white or weighted noise in four S/N conditions. By processing the data with logit transformation, S/N-discrimination curves were presumed for each combination of a speech material and a noise. Regardless of the type of noise, the discrimination scores for the female voice started to decrease gradually at a S/N ratio of +10 dB, and reached 10 to 20% at-10 dB. For the male voice in white noise, the discrimination curve was similar to those for the female voice. On the contrary, the discrimination score for the male voice in weighted noise declined rapidly from a S/N ratio of +5 dB, and went below 10% at -5 dB. The discrimination curves seem to be shaped by the interrelations between the spectrum of the speech material and that of the noise.

Adult↗

Penalized binary regression for gene expression profiling.

OBJECTIVES: A typical bioinformatics task in microarray analysis is the classification of biological samples into two alternative categories. A procedure is needed which, based on the expression levels measured, allows us to compute the probability that a new sample belongs to a certain class. METHODS: For the purpose of classification the statistical approach of binary regression is considered. Highdimensionality and at the same time small sample sizes make it a challenging task. Standard logit or probit regression fails because of condition problems and poor predictive performance. The concepts of frequentist and of Bayesian penalization for binary regression are introduced. A Bayesian interpretation of the penalized log-likelihood is given. Finally the role of cross-validation for regularization and feature selection is discussed. RESULTS: Penalization makes classical binary regression a suitable tool for microarray analysis. We illustrate penalized logit and Bayesian probit regression on a well-known data set and compare the obtained results, also with respect to published results from decision trees. CONCLUSIONS: The frequentist and the Bayesian penalization concept work equally well on the example data, however some method-specific differences can be made out. Moreover the Bayesian approach yields a quantification (posterior probabilities) of the bias due to the constraining assumptions.

Bayes Theorem↗

Modelling ordinal responses from co-twin control studies.

The co-twin control design has been widely used in studying the effects of environmental factors on the development of diseases. For binary outcomes that arise from co-twin control studies, the conditional likelihood method is commonly used. This approach, however, does not readily extend to ordinal response data because the standard conditional likelihood does not exist for cumulative logit or proportional odds models. In this paper, we investigate the applicability of the random-effects and GEE approaches in analysing ordinal response data from co-twin control studies. Using both approaches, we re-analyse data from a co-twin control study of the impact of military services during the Vietnam era on post-traumatic stress disorders (PTSD). The ordinal models have considerably increased power in detecting the effects of exposure when compared to the analyses using a dichotomized response. We discuss the interpretation of the estimates from GEE and random-effect models in the context of the twin data.

Cohort Studies↗

Poverty, affluence, and income inequality: neighborhood economic structure and its implications for health.

In this paper, we attempt to verify that neighborhood economic structure influences individual health over and above other individual characteristics, and to comparatively evaluate the effects of neighborhood concentrated affluence, concentrated poverty and income inequality in relation to individual health in the USA. We also explore physical environment, health-enhancing services, social hazards (crime) and social resources as mechanisms operating at the neighborhood level that may help to explain the influence of structural economic conditions on health. We use Hierarchical Ordinal Logit Models to examine a rich multi-level data set. Results indicate that affluence exerts significant contextual effects on self-rated health while poverty and income inequality at the neighborhood level are not important structural factors. Moreover, we find that a composite measure of social resources distinguishes itself in both explaining the impact of concentrated affluence and exerting an independent contextual effect on individual health. Physical environment, or the level of physical disorder in the neighborhood, also mediates the effect of affluence on self-rated health, although to a lesser degree than social resources. Our empirical findings suggest that different dimensions of economic structure do not in fact have unique and additive contributions to individual health; the presence of affluent residents is essential to sustain neighborhood social organization which in turn positively affect health.

Adult↗

Random effects modeling of multiple binomial responses using the multivariate binomial logit-normal distribution.

The multivariate binomial logit-normal distribution is a mixture distribution for which, (i) conditional on a set of success probabilities and sample size indices, a vector of counts is independent binomial variates, and (ii) the vector of logits of the parameters has a multivariate normal distribution. We use this distribution to model multivariate binomial-type responses using a vector of random effects. The vector of logits of parameters has a mean that is a linear function of explanatory variables and has an unspecified or partly specified covariance matrix. The model generalizes and provides greater flexibility than the univariate model that uses a normal random effect to account for positive correlations in clustered data. The multivariate model is useful when different elements of the response vector refer to different characteristics, each of which may naturally have its own random effect. It is also useful for repeated binary measurement of a single response when there is a nonexchangeable association structure, such as one often expects with longitudinal data or when negative association exists for at least one pair of responses. We apply the model to an influenza study with repeated responses in which some pairs are negatively associated and to a developmental toxicity study with continuation-ratio logits applied to an ordinal response with clustered observations.

Abnormalities, Drug-Induced↗

Clinical chemical diagnosis of diseases assisted by logistic regression illustrated by diagnosis of canine primary and secondary hepatobiliary diseases.

OBJECTIVE: The purpose of the present study was to demonstrate the use of logistic regression models in the prediction of diseases using the prediction of canine primary and secondary hepatobiliary diseases as an example. Briefly, in a logistic regression model independent variables (i.e. the analytical results) are combined in a linear equation that is used to estimate the logarithm of the odds (logit) of an event (i.e. having primary or secondary hepatobiliary disease). From the estimated logit given by the logistic regression model, a conditional probability of the event (i.e. having primary or secondary hepatobiliary disease) can be calculated. STUDY DESIGN: Twenty-six dogs with verified primary and secondary hepatobiliary diseases and 19 dogs, initially suspected to have hepatobiliary diseases, but with apparently other diseases, were included in the study. The following clinical chemical parameters were measured: alanine aminotransferase (ALAT), aspartate aminotransferase (ASAT), alkaline phosphatase (AP), bilirubin(Total) (TB), urea, glucose, retention of bromosulphthalein (BSP), fasting and postprandial total serum bile acid concentration (FSBA and PSBA). Logistic regression analysis, using the CATMOD procedure in SAS, was used to select which of the measured parameters should be included in the model, and to derive a logistic regression model using the selected parameters. To observe more closely the potential of the logistic regression model, the model was also used to classify a test group consisting of 13 dogs (6 dogs with hepatobiliary diseases and 7 dogs with other diseases). RESULTS: By logistic regression analysis, ASAT and PSBA were selected to be included in the final model, and the final logistic regression model was Y = -3.194 + 0.044.PSBA + 3.251.ASAT. The logistic regression model classified correctly 38 (84%) of 45 dogs in the present study. Specifically, 21 (81%) of 26 dogs with verified primary or secondary hepatobiliary diseases and 17 (90%) of 19 dogs with various other diseases were correctly classified by the logistic regression model. When the model was used on the test group, 5 (83%) of 6 dogs with hepatobiliary diseases and 7 dogs (100%) of 7 dogs with other diseases were correctly classified. CONCLUSIONS: Even though the logistic regression model derived in the present study only serves as an example, thus reducing the practical usefulness of the derived logistic regression model, the present study indicates a great potential of logistic models for the diagnosis of diseases.

Animals↗

[Heritability of strabismus convergens with exophthalmos in cattle].

Six pedigrees including 107 animals of the breed German Brown Swiss were available to test for a single gene conditioning for convergent strabismus with exophthalmus. The regressive logit models of the segregation analysis showed that a major gene model with additively acting genes explained the segregation of affected animals in the pedigrees in the best way. Additionally, polygenic and environmental effects might be of importance in the occurrence of convergent strabismus with exophthalmus.

Animals↗

Statistical analysis of accident severity on rural freeways.

The growing concern about the possible safety-related impacts of Intelligent Transportation Systems (ITS) has focused attention on the need to develop new statistical approaches to predict accident severity. This paper presents a nested logit formulation as a means for determining accident severity given that an accident has occurred. Four levels of severity are considered: (1) property damage only, (2) possible injury, (3) evident injury, and (4) disabling injury or fatality. Using 5-year accident data from a 61 km section of rural interstate in Washington State (which has been selected as an ITS demonstration site), we estimate a nested logit model of accident severity. The estimation results provide valuable evidence on the effect that environmental conditions, highway design, accident type, driver characteristics and vehicle attributes have on accident severity. Our findings show that the nested logit formulation is a promising approach to evaluate the impact that ITS or other safety-related countermeasures may have on accident severities.

Accidents, Traffic↗

The impact of children on the labour supply off married women: comparative estimates from European and U.S. data.

This paper presents estimates of the impact of the age pattern and level of fertility on the probability of labor force participation by married and cohabiting women in 12 eastern and western European countries and the US. Logit models for labor force participation probabilities are estimated conditional on age, age at marriage or union, educational attainment, current parity, and number of years in parity, using data on married and cohabiting women from the UN Economic Commission for Europe (UNECE) Comparative Study of World Fertility Surveys. The estimated models are used to simulate the age profiles of labor force participation conditional on the level and timing of fertility. The simulation results are presented in a series of charts grouped according to similarities in the effects of fertility on the age profiles of labor force participation. Four distinct patterns are identified that depend on the empirical significance of distinct number-of-children and age-of-youngest-child effects. The role of family policies and the extent to which the labor supply reductions associated with childbearing can be interpreted as opportunity costs are considered.

Age Factors↗

The effectiveness of helmets in reducing all-terrain vehicle injuries and deaths.

This article examines the effectiveness of helmets in reducing all-terrain vehicle (ATV) related deaths and head injuries, conditional on the occurrence of injury producing accidents. A logit regression model is used to analyze cross-section data on ATV-related fatal and nonfatal injuries, and to determine the factors that are associated with deaths and head injuries. The results suggest that, given an accident resulting in injury or death, helmet use reduces the risk of death by about 42%, and could reduce the likelihood that a given nonfatal injury involves the head by about 64%. Other factors that are associated with the risk of fatality for injury accidents include the use of alcohol or drugs, driving on paved roads, the driver's age and sex, and the vehicle's engine size. A benefit-cost analysis of helmet use is conducted and policy implications are discussed.

Adolescent↗

Risk factors for heart disease and stroke among diabetic persons, by disability status.

OBJECTIVE: To determine whether disabled diabetic persons have a higher prevalence of risk factors for heart disease and stroke than do diabetic persons without disability. RESEARCH, DESIGN, AND METHODS: Data were analyzed for noninstitutionalized adults in 27 states and the District of Columbia that participated in the Behavioral Risk Factor Surveillance System (BRFSS) in 2001 and/or 2003. Logistic regression analysis was used to estimate the adjusted prevalence and odds ratios of disabled diabetic persons, by sociodemographic characteristics. The logit form of each model was used to estimate conditional marginal probabilities of risk factors for heart disease and stroke among diabetic persons, by disability status. RESULTS: Diabetic persons with disability were more likely than those without disability to have more risk factors for heart disease and stroke, including insufficient leisure-time physical activity or inactivity (adjusted prevalence: 75.2% vs. 63.3%; P<.001), obesity (58.9% vs. 43.3%; P<.001), hypercholesterolemia (52.6% vs. 48.4%; P=.038), and hypertension (63.9% vs. 56.6%; P<.001). They were also more likely to have one or more, two or more, three or more, and four or more risk factors (97.2% vs. 95.6%, 83.5% vs. 74.0%, 56.5% vs. 41.1%, and 22.2% vs. 13.6%, respectively; P< or =.005). CONCLUSIONS: Diabetic persons with disability are more likely than those without disability to have clusters of risk factors for heart disease and stroke. Health care guidelines specifically targeting diabetic patients with disability may be needed to aid health care providers in addressing these risk factors.

Adult↗

Logistic representation of the sensitometric response of screen-film systems: empirical validation.

An empirical logistic model that linearizes the sensitometric response data of screen-film systems over the entire dynamic range of exposures is presented. This linearization is evident when the net optical densities are scaled as fractions of the net saturation density and plotted on commercial logit, probit, or double-log paper, except under conditions of reciprocity law failure. Weighted linear regression analysis shows that the intercept, but not the slope, depends on the screen-film system used. Previous work indicates that the slope is also independent of development time and photon energy. The model is verified through an analysis of tabulated sensitometric data published by the Center for Devices and Radiological Health.

Logistic Models↗

Why do the sick not utilise health care? The case of Zambia.

When ill the individual faces the options of seeking health care, using self-medication or doing nothing. In an economic perspective, an individual's propensity to utilise health care is determined by the costs of utilisation and the perceived benefits of health care. The propensity to utilise health care may hence be expected to vary between individuals. In this paper we attempt on the one hand to determine what factors influence sick individuals' propensity to seek health care at a health facility or use self-medication (or do nothing), and on the other hand attempt to determine the factors that influence the magnitude of their expenditures for health care, in particular what other factors than just health status influence utilisation. For the empirical analysis, data, covering 9700 individuals, from the 1998 Living Conditions Monitoring Survey (LCMS) is used. We use a Multinomial Logit selection model to estimate the equation, which allows us to analyse health-care utilisation through two separate processes, the decision to seek care and the magnitude of expenditures incurred. In general, we find that the individuals are influenced by income, insurance, type of illness and access variables such as distance and owning a vehicle.

Cost of Illness↗

Cholinergic sensitivity predicts severity of mania.

Our laboratory and others have reported that pupillary constrictions following application of the cholinergic agonist pilocarpine are increased in depressed patients. Moreover, mood improvements in manic patients, given lithium or Depakote, are also correlated with increases in pupil sensitivity. The present report describes the relationship between symptom severity and cholinergic sensitivity in a larger group (N=20) of manic patients (bipolar I; 296.4x). Pupil responses to pilocarpine eye drops (0-2%) were recorded using infrared pupillometry. The results were compared with pupil sizes measured under conditions of cholinergic blockade (0.5% tropicamide). Pupil responses were computed as percentages of the maximal range of areas measured under saturating agonist and antagonist conditions. Dose response curves were subjected to a log-logit transformation and ED(50) values were determined by weighted least squares regression. Bech-Rafaelsen mania ratings were found to be linearly related to ED(50) values (r=0.48). Patients with more severe mania required higher concentrations of pilocarpine in order to elicit a 50% reduction in pupil size. The present findings support a putative cholinergic role in the regulation of mood state. Moreover, the results suggest that pupillary responses may provide a simple and non-invasive means to evaluate cholinergic sensitivity in patients with affective disorders.

Adult↗

New directions in the modelling of interregional migration.

"This paper suggests two new approaches to modeling job-related migration. Both approaches are designed to simultaneously capture the inertia of past flows and information pertaining to economic conditions. One approach makes use of the incremental form of the logit model, and the other models the job application and competition process. Both techniques are illustrated using interregional migration flow data for the United States, and the results are compared with other migration models."

Americas↗