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 685 records · Page 38Linked to original sources

PubMed2002

A pattern-mixture model for longitudinal binary responses with nonignorable nonresponse.

Longitudinal studies frequently incur outcome-related nonresponse. In this article, we discuss a likelihood-based method for analyzing repeated binary responses when the mechanism leading to missing response data depends on unobserved responses. We describe a pattern-mixture model for the joint distribution of the vector of binary responses and the indicators of nonresponse patterns. Specifically, we propose an extension of the multivariate logistic model to handle nonignorable nonresponse. This method yields estimates of the mean parameters under a variety of assumptions regarding the distribution of the unobserved responses. Because these models make unverifiable identifying assumptions, we recommended conducting sensitivity analyses that provide a range of inferences, each of which is valid under different assumptions for nonresponse. The methodology is illustrated using data from a longitudinal study of obesity in children.

Adolescent↗

PubMed1998

Improved predictability of extracapsular extension and seminal vesicle involvement based on clinical and biopsy findings in prostate cancer in Japanese men.

OBJECTIVES: The accurate preoperative prediction of the extent of cancer by pathologic examination is essential for choosing the optimal treatment for patients with prostate cancer. Currently available clinical staging methods are not adequate and more precise staging is required. METHODS: Using the log likelihood ratio test and receiver operating characteristic (ROC) curve analysis, preoperative variables, including biopsy pathologic findings, were assessed for predicting final pathologic stage in prostate cancer. A multivariate model for predicting disease organ confinement status was established for easy clinical use. RESULTS: The use of the number of cores with cancer and maximum cancer length in conjunction with the three variables (prostate-specific antigen, clinical stage, and biopsy Gleason score) was found to significantly improve predictability of extracapsular extension and seminal vesicle involvement in clinically resectable (n = 96) and localized prostate cancers (n = 81) (P < 0.05). Areas under ROC curves for the above two parameter sets (five- versus three-variable model) were 0.8395 and 0.7109, respectively, for capacity for extracapsular extension prediction in clinically localized cancer. These values for seminal vesicle involvement were 0.7861 and 0.6927, respectively. The logistic model gave positive and negative predictive values of 73% and 78%, and 64% and 83%, respectively, for extracapsular extension and seminal vesicle involvement in clinically localized cancer at a predicted probability of 0.5 or greater. CONCLUSIONS: The present method may be used to predict non-organ-confined prostate cancer with greater accuracy than the previously reported model using three variables.

Aged↗

PubMed2005

Testing threshold and hormesis in a random effects dose-response model applied to developmental toxicity data.

Here we describe a random effects threshold dose-response model for clustered binary-response data from developmental toxicity studies. For our model we assume that a hormetic effect occurs in addition to a threshold effect. Therefore, the dose-response curve is based on two components: relationships below the threshold (hormetic u-shaped model) and those above the threshold (logistic model). In the absence of hormesis and threshold effects, the estimation procedure is straightforward. We introduce score tests that are derived from a random effects hormetic-threshold dose-response model. The model and tests are applied to clustered binary data from developmental toxicity studies of animals to test for hormesis and threshold effects. We also compare the score test and likelihood ratio test to test for hormesis and threshold effects in a simulated study.

Animals↗

PubMed2005

Determining the error of dose estimates and minimum and maximum acceptable concentrations from assays with nonlinear dose-response curves.

A method is described here that uses a modified Monte-Carlo method to provide an improved estimate of the confidence bounds of concentration estimates. This method accommodates even strongly nonlinear curve models, such as the five parameter logistic model, in contrast to the common but often poor approach of linearizing the regression problem and using linear theory to obtain the confidence bounds. The method uses an interpolation technique to reduce artifacts in the precision profile due to small simulation sample sizes and proximity to horizontal asymptotes in the curve model. The paper also describes how to define and calculate the minimum and maximum acceptable concentrations of dose-response curves by locating the concentrations where the size of the error, defined in terms of the size of the concentration confidence interval, exceeds the threshold of acceptability determined for the application.

Algorithms↗

PubMed2000

Incidence and epidemiology of Citrus tristeza virus in the Valencian community of Spain.

The first outbreak of citrus tristeza disease in Spain caused by Citrus tristeza virus (CTV) was recorded in 1957 in the Valencian Community (VC). In total c. 40 million trees, mainly of sweet orange and mandarin grafted on sour orange rootstocks, declined due to CTV. Large-scale surveys in different municipalities of the VC indicated that the disease spread very fast. Incidence increased from 11% in 1989 to 53% in 1998. Toxoptera aurantii and Aphis spiraecola (inefficient aphid vectors of CTV) predominated before 1985-87. Since then the relatively efficient vector Aphis gossypii has become dominant and induced an epidemic that has been modelled. The large number of A.gossypii that visited each clementine tree (estimated to exceed 97000 per year) explained the difference between the temporal pattern of spread of CTV in clementine which followed the Gompertz model and that in sweet orange (logistic model). The susceptibility of the different citrus species to CTV infection by aphids seems to depend on the number of young, succulent shoots produced. The epidemiological data allowed specific recommendations to be made to growers in order to facilitate a change to a modern citrus industry based on the use of selected varieties grafted on tristeza-tolerant rootstocks produced within a certification scheme. This has been done already in almost 90% of the VC citrus-growing area. The tristeza problem has been solved unless more aggressive isolates are introduced and become prevalent.

Animals↗

PubMed2003

Frailty modeling for spatially correlated survival data, with application to infant mortality in Minnesota.

The use of survival models involving a random effect or 'frailty' term is becoming more common. Usually the random effects are assumed to represent different clusters, and clusters are assumed to be independent. In this paper, we consider random effects corresponding to clusters that are spatially arranged, such as clinical sites or geographical regions. That is, we might suspect that random effects corresponding to strata in closer proximity to each other might also be similar in magnitude. Such spatial arrangement of the strata can be modeled in several ways, but we group these ways into two general settings: geostatistical approaches, where we use the exact geographic locations (e.g. latitude and longitude) of the strata, and lattice approaches, where we use only the positions of the strata relative to each other (e.g. which counties neighbor which others). We compare our approaches in the context of a dataset on infant mortality in Minnesota counties between 1992 and 1996. Our main substantive goal here is to explain the pattern of infant mortality using important covariates (sex, race, birth weight, age of mother, etc.) while accounting for possible (spatially correlated) differences in hazard among the counties. We use the GIS ArcView to map resulting fitted hazard rates, to help search for possible lingering spatial correlation. The DIC criterion (Spiegelhalter et al., Journal of the Royal Statistical Society, Series B 2002, to appear) is used to choose among various competing models. We investigate the quality of fit of our chosen model, and compare its results when used to investigate neonatal versus post-neonatal mortality. We also compare use of our time-to-event outcome survival model with the simpler dichotomous outcome logistic model. Finally, we summarize our findings and suggest directions for future research.

Adult↗

PubMed2001

Sequential ordinal modeling with applications to survival data.

This paper considers the class of sequential ordinal models in relation to other models for ordinal response data. Markov chain Monte Carlo (MCMC) algorithms, based on the approach of Albert and Chib (1993, Journal of the American Statistical Association 88, 669-679), are developed for the fitting of these models. The ideas and methods are illustrated in detail with a real data example on the length of hospital stay for patients undergoing heart surgery. A notable aspect of this analysis is the comparison, based on marginal likelihoods and training sample priors, of several nonnested models, such as the sequential model, the cumulative ordinal model, and Weibull and log-logistic models.

Algorithms↗

PubMed1987

Determinants of all causes of death in samples of Italian middle-aged men followed up for 25 years.

A total of 1712 men aged 40 to 59 years in two rural cohorts of northern and central Italy have been followed up for 25 years after an entry examination in 1960. Forty one individual characteristics have been considered as possible predictors of death in the next 25 years. After exclusion of 55 men with life threatening diseases (cardiovascular and cancer) and of 161 men because of missing measurements, 1495 men have been analysed for relation between entry factors and subsequent death (n = 670). Twelve factors eventually emerged as powerful predictors of future death: in hierarchical order, age, blood pressure, forced expiratory volume, cigarette smoking, xanthelasma, mother life-status, arm circumference, father life-status, shoulder-pelvis ratio, vital capacity, arcus senilis, and serum cholesterol. Discrimination as provided by logistic modelling placed 19.6% of all cases in the upper decile of the estimated risk, 36.8% in the upper quintile, 2.5% in the lowest decile, and 7.1% in the lowest quintile. Out of those located in the lowest decile of risk, 11.4% died within 25 years while the corresponding percentage in the upper decile was 87.3%. Use of the Cox model yielded slightly better coefficients than logistic function.

Adult↗

PubMed1991

Structural equation modeling in environmental risk assessment.

Environmental epidemiology requires effective models that take individual observations of environmental factors and connect them into meaningful patterns. Single-factor relationships have given way to multivariable analyses; simple additive models have been augmented by multiplicative (logistic) models. Each of these steps has produced greater enlightenment and understanding. Models that allow for factors causing outputs that can affect later outputs with putative causation working at several different time points (e.g., linkage) are not commonly used in the environmental literature. Structural equation models are a class of covariance structure models that have been used extensively in economics/business and social science but are still little used in the realm of biostatistics. Path analysis in genetic studies is one simplified form of this class of models. We have been using these models in a study of the health and development of infants who have been exposed to lead in utero and in the postnatal home environment. These models require as input the directionality of the relationship and then produce fitted models for multiple inputs causing each factor and the opportunity to have outputs serve as input variables into the next phase of the simultaneously fitted model. Some examples of these models from our research are presented to increase familiarity with this class of models. Use of these models can provide insight into the effect of changing an environmental factor when assessing risk. The usual cautions concerning believing a model, believing causation has been proven, and the assumptions that are required for each model are operative.

Environmental Exposure↗

PubMed1997

Generalized covariance-adjusted discriminants: perspective and application.

When discriminant analysis is used in practice for assessing the usefulness of diagnostic markers, the lack of control over covariates motivates the need for their adjustment in the analysis. This necessity for adjustment arises especially when the researcher's aim is classification based on a set of diagnostic markers and is not based on a set of covariates for which there exists known heterogeneity among the subjects with respect to the groups under consideration. The traditional covariance-adjusted approach is restrictive for such applications in that they assume linear covariates and a normal distribution for the the feature vector. Further, there is no available method for variable selection in using such covariance-adjusted models. In this paper, we generalize the traditional covariance-adjusted model to a general normal and logistic model, where these generalized models not only relax the distributional assumptions on the feature vector but also allow for nonlinear covariates. Exact and asymptotic tests are also derived for the problem of variable selection for these new models. The methodology is illustrated with both simulated data and an actual data set from a psychiatric study on using the Social Rhythm Metric for patients with anxiety disorders.

Activities of Daily Living↗

PubMed2002

The differential diagnostic potential of a panel of tumor markers (CA 125, CA 15-3, and CA 72-4 antigens) in patients with a pelvic mass.

OBJECTIVE: The purpose of this study was to assess the differential diagnostic potential of a combination of CA 125, CA 15-3, and CA 72-4 antigens in the definition of malignant disease, especially ovarian carcinoma in patients with a pelvic mass. STUDY DESIGN: A total of 412 patients were evaluated in a multicenter, retrospective study. RESULTS: Two hundred twenty-six malignant, 171 benign pelvic tumors (of which 129 were benign ovarian tumors), and 15 borderline tumors were evaluated. One hundred thirty-three patients had ovarian carcinoma. In 76 cases (55%), the International Federation of Gynecology and Obstetrics stage was III or IV. Borderline tumors (n = 15) were excluded from the statistical calculations. CA 125 antigen was the most sensitive marker for ovarian carcinoma (81%). The highest specificity and positive predictive value was obtained with CA 15-3 antigen (95% and 92%, respectively). Considering a concomitant elevation of all 3 markers as positive, a positive predictive value of 97% was found. However, only 28% of the patients in the total group and 41% of the patients with ovarian carcinoma had a concomitant elevation of all 3 markers. The combination of all 3 markers with levels below the cut-off resulted in a (false-positive) positive predictive value for malignancy between 12% and 36%. With the use of logistic regression analysis, we found a correct prediction in 73% of the cases. CA 15-3 antigen makes the most significant (P <.0001) contribution to the logistic model in the prediction of malignancy in the total group, with all pelvic masses with an odds ratio of 3.86. CONCLUSION: The combination of a simultaneous elevated level of CA 125, CA 15-3, and CA 72-4 antigens was predictive for malignant disease in almost all cases. However, such concomitant elevation was found in few of the malignant masses. Logistic regression analysis revealed that CA 15-3 antigen makes the most significant contribution to a model for the prediction of malignancy in the total group. The logistic model gave a correct prediction in 73% to 83%. The present tumor marker panel seems inferior to combinations with other test modalities, which include ultrasonography and/or physical examination and/or menopausal status or age.

Adult↗

PubMed2004

Combined logistic and Bayesian modeling of cesarean section risk.

OBJECTIVE: This study was undertaken to develop a simple and robust method for predicting risk of cesarean section. STUDY DESIGN: Retrospective cohort study of singleton births at term between 1992 and1999 in 22 Scottish maternity hospitals among primigravid women induced with prostaglandin. The risk of emergency cesarean section was modeled by using multivariate logistic regression in a development sample (n = 14,968). The output of this model was converted into adjusted likelihood ratios by using a novel method and tested in a validation sample (n = 12,638). RESULTS: Maternal age, height, gestational age, and fetal sex were all predictive of the risk of emergency cesarean section after prostaglandin induction of labor (all P < .001). The area under the receiver operator characteristic (ROC) curve in the development group was 0.677. The derived Bayesian model was comparably predictive of cesarean section risk in the validation sample: ROC of 0.673 (95% CI 0.662-0.684). Among the 994 women (8%) with a predicted cesarean section risk of more than 40%, the expected proportion was 48.2% and the observed proportion was 47.2%. Among the 1439 (11.4%) with a predicted cesarean risk of less than 10%, the expected proportion was 7.9% and the actual proportion 8.6%. CONCLUSION: Women at low or high risk of cesarean section after prostaglandin induction of labor can be identified with the use a novel combination of logistic regression and Bayesian modeling. The method is simple, robust, and may be generally applicable for clinical estimation of risk.

Adult↗

PubMed1998

Doses and models in risk assessment analysis for bronchial hyperresponsiveness.

The aims of this study are: (1) to evaluate whether the estimates of the association of risk factors with bronchial hyperresponsiveness (BHR) depends on the accumulated dose administered in challenge tests; and (2) to verify whether a model developed for survival studies (Weibull regression) is suited to analyze methacholine dose-response curves. For these purposes, 863 challenge tests, from EC Respiratory Health Survey in Italy, up to a cumulative dose of 6 mg methacholine, were analyzed by Weibull regression and by traditional methods (logistic model and linear model), both before and after truncation of the curves at 2 mg. With all methods the main risk factors for BHR were respiratory symptoms and atopy while age and airway caliber exerted a protective action. Our results confirmed that in epidemiological surveys 2 mg methacholine is enough to fully appreciate the effect of risk factors on BHR and showed that the Weibull model explains the observed variability better than linear and logistic regressions.

Adult↗

PubMed2003

Development of a model for case-mix adjustment of pressure ulcer prevalence rates.

BACKGROUND: Acute care hospitals participating in the Dutch national pressure ulcer prevalence survey use the results of this survey to compare their outcomes and assess their quality of care regarding pressure ulcer prevention. The development of a model for case-mix adjustment is essential for the use of these prevalence rates as an outcome measure. OBJECTIVE: The development of a valid model for case-mix adjustment to compare the prevalence rates in the acute care hospitals that participated in the 1998 Dutch pressure ulcer prevalence survey, for the purpose of performance comparisons among the hospitals. DESIGN: Cross-sectional design. SUBJECTS: Subjects were patients residing in the 43 acute care hospitals that participated in the national pressure ulcer prevalence survey on May 26, 1998. MEASURES: The study examined the validity of a model for case-mix adjustment of pressure ulcer prevalence rates and compared hospitals to evaluate the impact of adjusted prevalence rates on their performance. RESULTS: A logistic model was developed for case-mix adjustment, using age, malnutrition, incontinence, activity, mobility, sensory perception, friction and shear, and ward specialty. This model was found to have content, construct, and internal validity. Case-mix adjustment influenced the hospitals' performance. CONCLUSION: The data of the national pressure ulcer prevalence survey can be used to develop a valid model for case-mix adjustment. Conclusions about the quality of care were influenced by the use of case-mix adjusted outcomes as a measure of this quality.

Adolescent↗

PubMed1996

[Prediction of ijime (bullying)].

Questionnaires about Ijime (bullying) and Ijimerare (being bullied) were given to 261 junior high school students and their mothers from July to September in 1995. There were two types of questionnaires, one for the child, which inquired about his experiences of bullying other children and being bullied by other children during the last year, and the other for the mother which assessed the child's emotional and behavioral problems in the last year, was the Rutter Parental Questionnaire (R.P.Q.). These two questionnaires were filled in separately in order to avoid consultation between mother and child about their contents. Analysis of level of recognition of Ijime and Ijimerare by the mother showed a sensitivity level of Ijime and Ijimerare of 11.7% and 33.3%, respectively. The specificity on the part of the mother was 99.0% and 97.6% for Ijime and Ijimerare, respectively. While the specificity was sufficiently high for both phenomena, the sensitivity was low particularly for bullying other children. This underestimation may suggest the mother's psychological tendency to overlook the child's bullying of other children. A cross tabulation of the children's responses to "I bullied other children" and "I was bullied by other children" was made. A total of 254 cases were divided into the ijime group (n = 80) and the non-ijime group (n = 174), and the former was further divided into the aggressor (n = 37), the victim (n = 20) and the mixed group (n = 23). The average R.P.Q. scores of the victim and the mixed group were very similar at 7.4 and 7.1 points, respectively. Those of the aggressor and the non-ijime group were 4.6 and 3.8 points, respectively. A logistic regression analysis was performed and a predicted equation for probability of Ijime and Ijimerare was deduced. Using this logistic model, the specificity and the sensitivity were compared to those estimated by the mother. From this, it could be seen that, although the specificity was high in both estimations, the sensitivity estimated by this model was considerably better than that by the mother, especially in the prediction of Ijimerare which amounted to almost fifty percent accuracy. Because no information other than sex, school year, and the R.P.Q. scores (excluding enuresis and encopresis) are necessary, this model seems to possess convenience and a broad range of adaptability.

Adolescent↗

PubMed2006

Diagnostic value of vestibulo-ocular reflex parameters in the detection and characterization of labyrinthine lesions.

OBJECTIVE: To evaluate the power of various parameters of the vestibulo-ocular reflex (VOR) in detecting unilateral peripheral vestibular dysfunction and in characterizing certain inner ear pathologies. STUDY DESIGN: Prospective study of consecutive ambulatory patients presenting with acute onset of peripheral vertigo and spontaneous nystagmus. SETTING: Tertiary referral center. PATIENTS: Seventy-four patients (40 females, 34 males) and 22 normal subjects (11 females, 11 males) were included in the study. Patients were classified in three main diagnoses: vestibular neuritis: 40; viral labyrinthitis: 22; Meniere's disease: 12. METHODS: The VOR function was evaluated by standard caloric and impulse rotary tests (velocity step). A mathematical model of vestibular function was used to characterize the VOR response to rotational stimulation. The diagnostic value of the different VOR parameters was assessed by uni- and multivariable logistic regression. RESULTS: In univariable analysis, caloric asymmetry emerged as the most powerful VOR parameter in identifying unilateral vestibular deficit, with a boundary limit set at 20%. In multivariable analysis, the combination of caloric asymmetry and rotational time constant asymmetry significantly improved the discriminatory power over caloric alone (p<0.0001) and produced a detection score with a correct classification of 92.4%. In discriminating labyrinthine diseases, different combinations of the VOR parameters were obtained for each diagnosis (p<0.003) supporting that the VOR characteristics differ between the three inner ear disorders. However, the clinical usefulness of these characteristics in separating the pathologies was limited. CONCLUSION: We propose a powerful logistic model combining the indices of caloric and time constant asymmetries to detect a peripheral vestibular loss, with an accuracy of 92.4%. Based on vestibular data only, the discrimination between the different inner ear diseases is statistically possible, which supports different pathophysiologic changes in labyrinthine pathologies.

Adolescent↗

PubMed2004

Building an outcome predictor model for diffuse large B-cell lymphoma.

Diffuse large B-cell lymphoma (DLBCL) patients are treated using relatively homogeneous protocols, irrespective of their biological and clinical variability. Here we have developed a protein-expression-based outcome predictor for DLBCL. Using tissue microarrays (TMAs), we have analyzed the expression of 52 selected molecules in a series of 152 DLBCLs. The study yielded relevant information concerning key biological aspects of this tumor, such as cell-cycle control and apoptosis. A biological predictor was built with a training group of 103 patients, and was validated with a blind set of 49 patients. The predictive model with 8 markers can identify the probability of failure for a given patient with 78% accuracy. After stratifying patients according to the predicted response under the logistic model, 92.3% patients below the 25 percentile were accurately predicted by this biological score as "failure-free" while 96.2% of those above the 75 percentile were correctly predicted as belonging to the "fatal or refractory disease" group. Combining this biological score and the International Prognostic Index (IPI) improves the capacity for predicting failure and survival. This predictor was then validated in the independent group. The protein-expression-based score complements the information obtained from the use of the IPI, allowing patients to be assigned to different risk categories.

Adolescent↗

PubMed2000

Prediction rules for complications in coronary bypass surgery: a comparison and methodological critique.

BACKGROUND: Clinical prediction rules have been developed that use preoperative information to stratify patients according to risk of complications after cardiac surgery. OBJECTIVES: To assess the methodological standards and performance of 7 models. PARTICIPANTS: The validation portion of the Quality Measurement and Management Initiative (QMMI) cohort included a random sample of all adult patients (n = 3,261) who underwent coronary artery bypass grafting (CABG) surgery not involving valvular or other concomitant procedures at 12 medical centers from August 1993 to October 1995. OUTCOME MEASURES: Methodological standards used for model comparison were adapted from published criteria. Model performance was assessed by receiver-operating characteristic (ROC) analysis, and calibration was evaluated with the Hosmer-Lemeshow (HL) statistic and observed-expected plots. METHODS: We performed cross-validation by applying the published criteria for the development of each model to the validation subset of the QMMI cohort and by assessing the performance of each model in discriminating outcomes. RESULTS: Wide variations existed in the methodologies used to develop and validate the 5 additive scores evaluated. Cross-validation of all 5 additive scores revealed degradation in their abilities to discriminate outcomes. The 2 logistic models examined performed similarly to the additive scores examined in predicting mortality. CONCLUSIONS: Substantial variation existed both in the methodologies used to develop models and in the ability of the models to predict outcomes. Models developed at single institutions or using fewer patients may be less generalizable when applied to diverse clinical settings. Additive and logistic regression models performed similarly, as assessed by ROC and HL analyses.

Adult↗