PubMed Health⌕ Search

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 451 records · Page 25Linked to original sources
PubMed2004

Milk protein synthesis as a function of amino acid supply.

Most prediction schemes of milk protein secretion overestimate milk protein yield from dairy cows at high protein intakes, thereby overestimating milk protein yield response to protein supplementation. This study was conducted to determine factors contributing to such an overestimation. Using published studies, a database was constructed that was limited to amino acid (AA) infusion studies, as then only the digestible amino acid of dietary origin needed to be estimated, whereas the amount infused was known exactly, thereby reducing the dependence on estimated values. Although milk protein yield was positively related with total energy supply, and both digestible duodenal supply and infused AA, in this database there was no relationship between milk protein yield response above control treatments and the nutrient status of the cows (energy or protein). Total milk protein yield was defined as a function of individual AA supply, using a segmented-linear and a logistic model to obtain estimates of the efficiency of conversion of AA into milk protein. Except for Lys and Met supply, the segmented-linear model yielded lower root mean square error and better correlation, but both models were similar in their reliability. For both models, the estimated efficiency of conversion of AA to milk differed among AA. Estimations of the ideal profile of AA for lactating dairy cows were similar between models, with requirements for Lys and Met in line with 2001 National Research Council recommendations. The major difference is that the segmented-linear model yields a constant efficiency of conversion of an AA until requirements are met, with zero efficiency beyond this point. The logistic model allows for an estimation of the decreasing marginal efficiency of conversion of AA as the supply approaches the requirements. The use of variable efficiency factors should improve our ability to predict protein yield in response to supplemental protein.

Amino Acids↗
PubMed1995

Analysis of ordinal data in a study of endometrial cancer under a matched pairs case-control design.

While estimating odds ratios (ORs) in the context of dose levels of conjugated oestrogen exposure and development of endometrial cancer, the categories formed by the levels of the exposure are ordinal in nature. In the literature, the binary logistic model is used for estimating OR for each category relative to the baseline category. We describe the use of two ordinal logistic models, the cumulative logit and continuation-ratio logit models, to estimate the ORs for the matched pairs case-control data set of the Los Angeles endometrial cancer study. A test for equality of the cumulative ORs across the exposure levels is proposed. The test statistic follows asymptotically the chi-square distribution.

Case-Control Studies↗
PubMed2002

Inhibin B on the day of HCG-administration is a predictive factor for failure in assisted reproduction cycles.

MATERIALS AND METHODS: This retrospective study was carried out in order to assess the value of inhibin B as a predictive factor for the assisted reproduction-outcome. Inhibin B on the day of HCG-administration (ovulation induction) was measured in 15 pregnant and 16 non-pregnant patients (defined as positive serum-HCG) and compared by single and multiple logistic regression analysis with classical predictive factors such as estrogen and oocyte number. Both groups were similar with respect to age and cause of infertility. RESULTS: While estrogen, FSH and LH at the HCG-day did not differ, inhibin B, the number of cumulus oocyte complexes and metaphase II -oocytes were statistically significantly different. In a single variable logistic model inhibin B, cumulus oocyte complexes and metaphase II oocytes showed significant correlation with pregnancy. When reassessed in a multiple variable logistic model only inhibin B maintained a considerable influence. Further inspection revealed, that the predictive inhibin B value at HCG-day <1200 pg/ml for failure of ART (assisted reproduction techniques) was 91%. Inhibin >1200 pg/ml showed 70% predictivity for pregnancy. CONCLUSION: In comparison to classical parameters inhibin B at HCG-day seems to be the better prognostic factor for the outcome of ART. Inhibin B <1200 pg/ml seems to be highly predictive for failure of ART.

Chorionic Gonadotropin↗
PubMed1998

Patterns and associates of hyperphagia in patients with dementia.

This study examined patterns and associates of excessive eating (hyperphagia) in a community-based registry of patients with dementia. From patients enrolled in the Mayo Clinic Alzheimer's Disease Patient Registry (n = 439), 39 were identified with excessive eating reported on the Behavior Symptom Checklist at some time during their illness. They were matched for age, gender, duration of disease, and Global Deterioration Scale (GDS) score to "normal eaters." Annualized weight change was determined based on weight from the 6 months before the evaluation to weight 6 months after the evaluation. Annualized weight change scores were not significantly different between excessive eaters and normal eaters nor between wanderers and nonwanderers. In cross-sectional analysis, univariate modeling suggested age at onset, GDS, and Mini-Mental State Examination score to be significant predictors of excessive eating. Using multivariate logistic model with backward elimination, only age of onset and GDS were retained as associates of excess eating. Rater type also emerged as a significant predictor for excessive eating with family raters reporting this behavior in 16% of patients compared to 5% for other raters. In chi-square analyses excessive eating was associated with greater frequency of wandering, unpredictable behavior, inappropriate dressing, inappropriate bodily concerns, and threatening self-harm. Associates of excess eating were subsequently examined separately in wandering and nonwandering excessive eaters. Logistic modeling suggested that among nonwanderers, patients who were younger but more severely demented were likely to have reported excessive eating. These results suggest hyperphagia to be present in approximately 10% of a community-based cohort of patients with dementia and associated with increasing functional decline. Excessive eating does not appear to arise from memory dysfunction, but for wanderers may result from needing increased caloric intake because of increased activity levels. Thus, for wandering excessive eaters, it may be appropriate to endure the eating to ensure appropriate caloric intake. Nonwandering excessive eaters were younger, had greater dementia severity, and had more unpredictable behavior. They may have dementia with prominent frontal lobe involvement and may respond to any food stimulus respective of hunger. Restricting food exposure may be an effective management intervention for them.

Aged↗
PubMed2003

An EM-based semi-parametric mixture model approach to the regression analysis of competing-risks data.

We consider a mixture model approach to the regression analysis of competing-risks data. Attention is focused on inference concerning the effects of factors on both the probability of occurrence and the hazard rate conditional on each of the failure types. These two quantities are specified in the mixture model using the logistic model and the proportional hazards model, respectively. We propose a semi-parametric mixture method to estimate the logistic and regression coefficients jointly, whereby the component-baseline hazard functions are completely unspecified. Estimation is based on maximum likelihood on the basis of the full likelihood, implemented via an expectation-conditional maximization (ECM) algorithm. Simulation studies are performed to compare the performance of the proposed semi-parametric method with a fully parametric mixture approach. The results show that when the component-baseline hazard is monotonic increasing, the semi-parametric and fully parametric mixture approaches are comparable for mildly and moderately censored samples. When the component-baseline hazard is not monotonic increasing, the semi-parametric method consistently provides less biased estimates than a fully parametric approach and is comparable in efficiency in the estimation of the parameters for all levels of censoring. The methods are illustrated using a real data set of prostate cancer patients treated with different dosages of the drug diethylstilbestrol.

Algorithms↗
PubMed1998

Development of a clinical prediction model for an ordinal outcome: the World Health Organization Multicentre Study of Clinical Signs and Etiological agents of Pneumonia, Sepsis and Meningitis in Young Infants. WHO/ARI Young Infant Multicentre Study Group.

This paper describes the methodologies used to develop a prediction model to assist health workers in developing countries in facing one of the most difficult health problems in all parts of the world: the presentation of an acutely ill young infant. Statistical approaches for developing the clinical prediction model faced at least two major difficulties. First, the number of predictor variables, especially clinical signs and symptoms, is very large, necessitating the use of data reduction techniques that are blinded to the outcome. Second, there is no uniquely accepted continuous outcome measure or final binary diagnostic criterion. For example, the diagnosis of neonatal sepsis is ill-defined. Clinical decision makers must identify infants likely to have positive cultures as well as to grade the severity of illness. In the WHO/ARI Young Infant Multicentre Study we have found an ordinal outcome scale made up of a mixture of laboratory and diagnostic markers to have several clinical advantages as well as to increase the power of tests for risk factors. Such a mixed ordinal scale does present statistical challenges because it may violate constant slope assumptions of ordinal regression models. In this paper we develop and validate an ordinal predictive model after choosing a data reduction technique. We show how ordinality of the outcome is checked against each predictor. We describe new but simple techniques for graphically examining residuals from ordinal logistic models to detect problems with variable transformations as well as to detect non-proportional odds and other lack of fit. We examine an alternative type of ordinal logistic model, the continuation ratio model, to determine if it provides a better fit. We find that it does not but that this model is easily modified to allow the regression coefficients to vary with cut-offs of the response variable. Complex terms in this extended model are penalized to allow only as much complexity as the data will support. We approximate the extended continuation ratio model with a model with fewer terms to allow us to draw a nomogram for obtaining various predictions. The model is validated for calibration and discrimination using the bootstrap. We apply much of the modelling strategy described in Harrell, Lee and Mark (Statist. Med. 15, 361-387 (1998)) for survival analysis, adapting it to ordinal logistic regression and further emphasizing penalized maximum likelihood estimation and data reduction.

Chi-Square Distribution↗
PubMed2000

Prediction of vesico-ureteric reflux in childhood urinary tract infection: a multivariate approach.

UNLABELLED: In this study, independent predictors obtained from patient history, physical examination and laboratory results for vesico-ureteric reflux (VUR) in children of 0-5 y with a first urinary tract infection (UTI) were assessed and the added value of renal ultrasound (US) investigated. Information was collected from children visiting the paediatric outpatient department with a first proven UTI, defined as a urine monoculture with > or = 10(5) organism/ml, with clinical symptoms and possible white cell count > or = 20 per high-power field of spun fresh urine. Children with neurologic bladder dysfunction were excluded. VUR was determined by voiding cystourethrography (VCUG) and graded from I to V. The diagnostic value of predictors was judged using multivariate logistic modelling with the area under the receiver operating characteristic (ROC area). A risk score was derived based on the regression coefficients of the independent predictors in the logistic model. In 140 children (51 boys and 89 girls) VUR was diagnosed in 37. Independent predictors for VUR were male gender, age, family history for uropathology, serum C-reactive protein level (CRP) and dilatation of the urinary tract on US. The ROC area of this model was 0.78 (95% CI: 0.69-0.87). This prediction model identified 12% (95% CI: 7-18) of the patients without VUR without missing one case of VUR. If we used VUR > or = grade 3 as a threshold, the model assessed VUR to be absent in 34% (95% CI: 26-42). CONCLUSION: A prediction rule based on age, gender, family history, CRP and US results is useful in assessing the probability of VUR in the individual child with a first UTI and may help the physician to make decisions about performing additional imaging techniques. Prospective validation of the model in future patients, however, will be necessary before applying the rule in practice.

Age Factors↗
PubMed1999

Bivariate logistic regression: modelling the association of small for gestational age births in twin gestations.

Clustered binary responses, such as disease status in twins, frequently arise in perinatal health and other epidemiologic applications. The scientific objective involves modelling both the marginal mean responses, such as the probability of disease, and the within-cluster association of the multivariate responses. In this regard, bivariate logistic regression is a useful procedure with advantages that include (i) a single maximization of the joint probability distribution of the bivariate binary responses, and (ii) modelling the odds ratio describing the pairwise association between the two binary responses in relation to several covariates. In addition, since the form of the joint distribution of the bivariate binary responses is assumed known, parameters for the regression model can be estimated by the method of maximum likelihood. Hence, statistical inferences may be based on likelihood ratio tests and profile likelihood confidence intervals. We apply bivariate logistic regression to a perinatal database comprising 924 twin foetuses resulting from 462 pregnancies to model obstetric and clinical risk factors for the association of small for gestational age births in twin gestations.

Adolescent↗
PubMed2005

Modelling the hierarchical structure of road crash data--application to severity analysis.

Road crashes have an unquestionably hierarchical crash-car-occupant structure. Multilevel models are used with correlated data, but their application to crash data can be difficult. The number of sub-clusters per cluster is small, with less than two cars per crash and less than two occupants per car, whereas the number of clusters can be high, with several hundred/thousand crashes. Application of the Monte-Carlo method on observed and simulated French road crash data between 1996 and 2000 allows comparing estimations produced by multilevel logistic models (MLM), Generalized Estimating Equation models (GEE) and logistic models (LM). On the strength of a bias study, MLM is the most efficient model while both GEE and LM underestimate parameters and confidence intervals. MLM is used as a marginal model and not as a random-effect model, i.e. only fixed effects are taken into account. Random effects allow adjusting risks on the hierarchical structure, conferring an interpretative advantage to MLM over GEE. Nevertheless, great care is needed for data coding and quite a high number of crashes are necessary in order to avoid problems and errors with estimates and estimate processes. On balance, MLM must be used when the number of vehicles per crash or the number of occupants per vehicle is high, when the LM results are questionable because they are not in line with the literature or finally when the p-values associated to risk measures are close to 5%. In other cases, LM remains a practical analytical tool for modelling crash data.

Accidents, Traffic↗
PubMed1999

Nursing home resident use of care directives.

BACKGROUND: The Patient Self-Determination Act of 1991 requires that nursing homes reimbursed by Medicare or Medicaid inform all residents upon admission of their rights to enact care directives in the event of terminal illness. This study investigated the relationship between care directive use and resident functional status. METHODS: We analyzed a version of the Minimum Data Set (MDS+) from a single state. We selected residents who were admitted to a nursing home in the first half of 1993 and followed them in the nursing home through the end of 1994. We created logistic models to examine independent correlates associated with having an advance directive or a do-not-resuscitate (DNR) order on admission. We then created similar logistic models to examine independent correlates associated with writing an advance directive or DNR order subsequent to admission. RESULTS: Of the 2,780 residents, 11% (292) had advance directives and 17% (466) had DNR orders upon admission. Of those without care directives upon admission, 6% (143) subsequently had an advance directive and 15% (339) subsequently had a DNR order. Cross-sectionally, older individuals and whites were more likely to have a care directive. Having poor cognitive and physical function was associated with having a DNR order upon admission. Longitudinally, longer stayers and whites were more likely to have an advance directive. Residents who lost physical function were more likely to have an advance directive and those who lost cognitive function were more likely to have a DNR order. CONCLUSIONS: Care directive use is influenced by a number of sociodemographic and functional characteristics.

Activities of Daily Living↗
PubMed1993

Facts versus values: why legislators vote against injury control laws.

BACKGROUND: Control of motor vehicle-related injuries depends upon passage of mandatory safety belt and other injury control laws. Unfortunately, state legislators often oppose these laws. METHODS: In 1988, a 62-item questionnaire was mailed to the 97 Colorado legislators who voted on a 1987 safety belt law to identify factors (knowledge, experiences, attitudes, and beliefs) associated with "yes" and "no" votes. To test for associations between these attributes and the legislators' recorded votes, odds ratios (OR) and 95% confidence intervals (CIs) were calculated. A stepwise logistic regression identified independent predictors of "vote." RESULTS: Fifty-three (55%) of the legislators responded. Responders and nonresponders were demographically similar. "Vote" was not associated with age; sex; having young children in the family; perceived injury risk; recent traffic tickets; family or personal crash experience; or knowledge of the fatality risk reductions attributable to wearing safety belts. Ninety-six percent of the legislators knew that safety belts reduce the risk of death and 87% believed a safety belt law would save lives. The strongest predictors of a "yes" vote were impression that constituents favored the law (OR = 31, CI 95 = 3.5, 270); belief that a mandatory safety belt law will save lives (OR = 20, CI 95 = 2.1, 203); and "extreme" importance paid in the voting decision to effectiveness of the law in reducing deaths (OR = 19, CI 95 = 3.5, 107). Legislators who considered restrictions on individual freedoms an "extremely" important decision criterion were 43 times (CI 95 = 7, 267) more likely to vote "no." In the logistic model only extreme importance assigned to individual freedoms (beta = 3.7; OR = .025; p = 0.002) and policy effectiveness (beta = +3.1; OR = 22; p = 0.01) predicted "vote." The logistic model correctly predicted 90% of legislators' votes. CONCLUSIONS: In this study the strongest predictors of voting behavior were concern for individual freedoms, perceived constituents' support and attention paid to policy effectiveness. Those seeking to persuade legislators to vote for mandatory safety belt laws must pay attention to attitudes and values in addition to scientific facts.

Accidents, Traffic↗
PubMed2001

[Hemorrhage in arteriovenous malformations: clinical and anatomic data].

BACKGROUND AND PURPOSE: Potential severity of hemorrhage often leads to treat a cerebral arteriovenous malformation. Consequences can be very various and serious. Our first purpose is to define the different types of hemorrhage. Our second purpose is to appreciate more precisely individual hemorragic risk of a cerebral arteriovenous malformation with the study of his angioarchitecture. We performed a prospective study in order to validate a logistic model and a classification previously described. PATIENTS: and method. From the whole series of 705 patients, 57% (n=394) suffered a parenchymal, subarachnoid or ventricular hemorrhage. Logistic model and classification of the hemorrhagic risk were prospectively tested on a consecutive population of 78 patients. Comparisons of theorical (calculated hemorrhagic risk) and real (hemorrhage or not) were performed by non parametric tests. RESULTS: Characteristics and clinical consequences of the hemorrhage were analyzed. Results of the prospective study confirmed data of the classification and showed a hemorrhage risk increasing with the grade: grade Ia 0%, grade Ib 30%, grade II 44%, grade III 57%, grade IV 88%. CONCLUSION: The study of the angioarchitecture of a cerebral arterio-venous malformation allowed to assess with accuracy his individual hemorrhagic risk. However, this precision may be improved by the study of other parameters of intracranial arteriovenous malformation.

Adolescent↗
PubMedDate not supplied

Esophageal invasion by thyroid carcinomas: prediction using magnetic resonance imaging.

PURPOSE: We evaluated the accuracy of magnetic resonance imaging (MRI) in predicting esophageal invasion by thyroid carcinomas and established an optimal criterion for diagnosing esophageal invasion. METHOD: The MRI findings (size and margins of tumor, ratio of tumor contact to the esophagus, shape and displacement of the esophagus, and tumor invasion to the outer and inner layers of the esophagus) in 67 patients with thyroid carcinomas were retrospectively reviewed and correlated with surgical and pathologic findings. Logistic modeling was used to determine the significant factors for predicting esophageal invasion. RESULTS: Seventeen (34%) of the 67 patients had pathologically or surgically verified esophageal invasion. The logistic modeling revealed that outer layer invasion (P < 0.001) and poorly defined margins (P = 0.001) were the significant factors. The outer layer invasion showed the highest accuracy of 91%, with 82% sensitivity and 94% specificity. The addition of poorly defined margins to this criterion did not improve its accuracy. CONCLUSION: Esophageal invasion by thyroid carcinoma was accurately predicted with MRI, and an MRI finding of outer layer invasion was optimal for diagnosing esophageal invasion.

Adult↗
PubMed2000

Analysis of switching insurance plan type. Comparison of two statistical methods.

PURPOSE: To compare results of 2 statistical methods for identifying factors in claims data that are associated with switching insurance plans between managed care (MC) and indemnity (IN).METHODS: Using claims data from 2 insurance providers in a northeastern city, we analyzed patients aged 18+ with diabetes, asthma, or congestive heart failure (CHF) who were covered any time in 1993-1997 (N = 88,917). Stratifying by initial plan type, we examined predictors of switching from the initial plan type using logistic regression and survival analysis. Covariates included age, time in study (for logistic models), gender, diabetes (yes/no), CHF (yes/no), and asthma (yes/no). Survival analysis accounted for time to switch and allowed time-varying covariates.RESULTS: In logistic regression models, older individuals who were in IN were much less likely to switch into MC. Those in MC were more likely to switch to IN, with the greatest likelihood of switching in ages 60-69 (OR = 4.00, 95% CI = 3.32-4.83). Females were less likely to switch from IN to MC (OR = 0.92, 95% CI = 0.87-0.98), CHF patients were less likely to switch from IN to MC (OR = 0.75, 95% CI = 0.68-0.83), and diabetes patients were less likely to switch from MC to IN (OR = 0.77, 95% CI = 0.62-0.96). Hazard ratios calculated using Cox regression were similar to odds ratios for most covariates. However, some coefficients for diseases were significant in Cox models but not in the logistic models. Cox models took 45 times longer in CPU time than logistic regression models.CONCLUSIONS: Logistic regression was a good approximation to Cox regression in identifying many of the factors in switching insurance plan in these data, at a fraction of the computing time. However, Cox models allowed diseases to be time-varying, and so was more sensitive to identifying significant relationships with disease.

Journal Article↗
PubMed2001

Sociodemographic and clinical factors associated with low quality of life one year after coronary bypass operations: the Israeli coronary artery bypass study (ISCAB).

OBJECTIVES: We sought to examine the effect of sociodemographic characteristics and perioperative clinical factors 1 year after coronary bypass operations on low health-related quality of life. We also sought to assess the usefulness of an additional single question on overall health for identifying patients with low health-related quality of life. METHODS: This report is part of the Israeli coronary artery bypass study of 1994, in which every patient undergoing isolated coronary bypass grafting in Israel was included. The target population for this report comprised all survivors beyond 1 year who were 45 to 65 years of age. Patients were interviewed before the operations. Self-administered questionnaires regarding health-related quality of life (SF-36) were sent to 1724 patients who were successfully located 1 year postoperatively, and 1270 questionnaires were completed. Low health-related quality of life was defined as the lowest tertile of the distribution of scores for the 2 summary components of the SF-36 and the single question on overall health. Logistic models were constructed for each of the 3 outcomes. RESULTS: Female sex and low socioeconomic background were associated with low health-related quality of life in the logistic models. Other significant factors were symptoms of angina, sleep disturbances, hypertension, high severity of illness scores, hospital readmission, no rehabilitation, and hospitals with high perioperative mortality. Of the 3 study outcomes, the model for the single question on overall health was the most discriminating (C statistic = 0.76 vs 0.70 and 0.70, respectively). CONCLUSIONS: The study identifies patients who would most benefit from posthospitalization community support after bypass operations. Under circumstances of limited resources, these disadvantaged groups should be targeted as a priority. Encouraging participation in existing rehabilitation programs or introducing telephone hotlines could improve health-related quality of life after coronary bypass grafting without large investments.

Aged↗
PubMed2006

Optimization of elastolysis conditions and elastolytic kinetic analysis with elastase from Bacillus licheniformis ZJUEL31410.

The solubilization of elastin by Bacillus licheniformis elastase cannot be analyzed by conventional kinetic methods because the biologically relevant substrate is insoluble and the concentration of enzyme-substrate complex has no physical meaning. In this paper we report the optimization of elastolysis conditions and analysis of elastolytic kinetics. Our results indicated that the hydrolyzing temperature and time are very important factors affecting elastolysis rate. The optimized conditions using central composite design were as follows: elastolysis temperature 50 degrees C, elastase concentration 1 x 10(4) U/ml, elastin 80 mg, elastolytic time 4 h. Investigation of the effects of substrate content, elastase concentration and pH was also revealed that low or high elastin content inhibits the elastolysis process. Increasing elastase improves elastin degradation, but high elastase may change the kinetics characterization. Alkaline environment can decrease elastin degradation rate and pH may affect elastolysis by changing elastase reaction pH. To further elucidate the elastolysis process, the logistic model was used to elastolysis kinetics study showing clearly that the logistic model can reasonably explain the elastolysis process, especially under lower elastase concentration. However, there is still need for more investigations with the aid of other methods, such as biochemical and molecular methods.

Bacillus↗
PubMedDate not supplied

Predictors of sharing drugs among injection drug users in the South Bronx: implications for HIV transmission.

HIV may be transmitted in the process of sharing injected drugs, even if all participants have their own syringes. In an effort to gain understanding of the extent and predictors of drug sharing, data were obtained via personal interviews with 1,024 injection drug users from four neighborhoods in the South Bronx. The relationship between drug-sharing and demographic, sexual, and drug-related variables was first examined in a bivariate analysis, and then via multiple logistic regression. Individuals who split drugs were more likely to be female, have had sex with a casual partner, exchanged sex for drugs or other needs, recently smoked crack cocaine, and shared needles. They were less likely to live or inject at their own home or have used a new needle the last time they injected. In a final logistic model, correlates of drug sharing included trading sex, injecting outside one's home, and using borrowed, rented or shared needles. Despite the lack of significance for gender in the final logistic model, females were at high risk of drug sharing because they constituted the great majority of those who exchanged sex. Continuing research is needed to understand how drug-sharing contributes to the spread of HIV and other infections, as are studies of approaches to reducing drug sharing. Prevention strategists and outreach organizations should be aware of the HIV risks inherent in the widespread practice of drug sharing.

Adult↗