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 397 records · Page 22Linked to original sources
PubMed1986

Logistic regression methods for retrospective case-control studies using complex sampling procedures.

There are a number of possible designs for case-control studies. The simplest uses two separate simple random samples, but an actual study may use more complex sampling procedures. Typically, stratification is used to control for the effects of one or more risk factors in which we are interested. It has been shown (Anderson, 1972, Biometrika 59, 19-35; Prentice and Pyke, 1979, Biometrika 66, 403-411) that the unconditional logistic regression estimators apply under stratified sampling, so long as the logistic model includes a term for each stratum. We consider the case-control problem with stratified samples and assume a logistic model that does not include terms for strata, i.e., for fixed covariates the (prospective) probability of disease does not depend on stratum. We assume knowledge of the proportion sampled in each stratum as well as the total number in the stratum. We use this knowledge to obtain the maximum likelihood estimators for all parameters in the logistic model including those for variables completely associated with strata. The approach may also be applied to obtain estimators under probability sampling.

Clinical Trials as Topic↗
PubMed1992

Toward improved empiric management of moderate to severe urinary tract infections.

BACKGROUND: Guidelines to show whether a patient hospitalized because of a urinary tract infection (UTI) has a severe infection, and whether he or she is at high risk for harboring a multiresistant pathogen, are scant. The aims of the present study were to find (1) clinical and laboratory variables known within 24 hours of admission that, combined in a logistic model, will point to a high or low probability of bacteremia and (2) variables that can be used to define patients at high risk for the subsequent isolation of a multiresistant uropathogen. METHODS: In a set of patients consecutively admitted to a department of medicine because of UTI, we compared bacteremic vs nonbacteremic patients, and patients with a multiresistant uropathogen vs others, on logistic regression analysis. The logistic models derived were validated in a second set of patients with UTI. RESULTS: Among 247 patients with UTI (median age, 75 years), 80 of them with bacteremia, five factors were significantly and independently associated with bacteremia: serum creatinine level, leukocyte count, temperature, diabetes mellitus, and low serum albumin level. A logistic model incorporating those factors was used to divide the patients into three groups with increasing prevalence of bacteremia (6%, 39%, and 69%) and of death (3%, 6%, and 20%). Three factors were predictive of the subsequent isolation of a resistant uropathogen: use of antibiotics before admission, advanced age, and male gender. The combination of those factors was used to divide patients into two groups, with resistance to cefuroxime of 9% vs 28%, to gentamicin of 7% vs 20%, and to sulfamethoxazole-trimethoprim of 30% vs 50%. In a second set of 144 patients with UTI, the percentages of bacteremia in the three groups were 5%, 16%, and 55%, and those of death, 2%, 6%, and 17%. When divided by the second model, the resistance to cefuroxime in the two groups was 16% vs 30%; to gentamicin, 16% vs 28%; and to sulfamethoxazole-trimethoprim, 28% vs 59%. CONCLUSIONS: If prospectively validated in other settings, the models can be used to define groups of patients with UTI at low and high risk for bacteremia, and to help in the choice of empiric antibiotic treatment.

Adolescent↗
PubMed2005

Single and repeated GnRH agonist stimulation tests compared with basal markers of ovarian reserve in the prediction of outcome in IVF.

PURPOSE: To study the value of a single or repeated GnRH agonist stimulation test (GAST) in predicting outcome in IVF compared to basal ovarian reserve tests. METHODS: A total of 57 women was included. In a cycle prior to the IVF treatment, on day 3, an antral follicle count (AFC) was performed and blood taken for basal FSH, inhibin B and E2 measurements, followed by a subcutaneous injection of 100 microg triptorelin for the purpose of the GAST. Twenty-four hours later blood sampling was repeated. All the tests were repeated in a subsequent cycle. From the GAST E2 and inhibin B response were used as test parameters. The outcome measures were poor ovarian response and ongoing pregnancy. Group comparisons were done using the Mann-Whitney or chi-square test. Univariate and multivariate logistic regression was applied to assess which test revealed the highest predictive accuracy as expressed in the area under receiver-operating characteristic curve (ROC(AUC)). Clinical value was compared by calculating classical test characteristics for the best logistic models. RESULTS: All the basal and GAST variables were significantly different in the poor responders (n = 19) compared to normal responders (n = 38). In the univariate analysis on cycle 1 tests the AFC was the best predictor for poor ovarian response, while in cycle 2 the E2 response in the GAST performed best (ROC(AUC) of 0.91 for both). Multivariate analysis of the basal variables led to the selection of AFC and inhibin B in cycle 1, yielding a ROC(AUC) of 0.96. Mean E2 response was selected in a multivariate analysis of the repeated GAST variables (ROC(AUC) 0.91). At a specificity level of -0.90, several logistic models including GAST variables appeared to have a sensitivity (-0.80), positive predictive value (-0.82) and false positive rate (-0.18), comparable to a logistic model containing AFC and inhibin B. None of the test variables showed a significant relation with ongoing pregnancy. CONCLUSIONS: The GAST has a rather good ability to predict poor response in IVF. However, comparing the predictive accuracy and clinical value of the GAST with a day 3 AFC and inhibin B, it appeared that neither a single nor a repeated GAST performed better. In addition, the predictive ability towards ongoing pregnancy is poor. Therefore, the use of the GAST as a predictor of outcome in IVF should not be advocated.

Adult↗
PubMed1997

Random effects probit and logistic regression models for three-level data.

In analysis of binary data from clustered and longitudinal studies, random effect models have been recently developed to accommodate two-level problems such as subjects nested within clusters or repeated classifications within subjects. Unfortunately, these models cannot be applied to three-level problems that occur frequently in practice. For example, multicenter longitudinal clinical trials involve repeated assessments within individuals and individuals are nested within study centers. This combination of clustered and longitudinal data represents the classic three-level problem in biometry. Similarly, in prevention studies, various educational programs designed to minimize risk taking behavior (e.g., smoking prevention and cessation) may be compared where randomization to various design conditions is at the level of the school and the intervention is performed at the level of the classroom. Previous statistical approaches to the three-level problem for binary response data have either ignored one level of nesting, treated it as a fixed effect, or used first- and second-order Taylor series expansions of the logarithm of the conditional likelihood to linearize these models and estimate model parameters using more conventional procedures for measurement data. Recent studies indicate that these approximate solutions exhibit considerable bias and provide little advantage over use of traditional logistic regression analysis ignoring the hierarchical structure. In this paper, we generalize earlier results for two-level random effects probit and logistic regression models to the three-level case. Parameter estimation is based on full-information maximum marginal likelihood estimation (MMLE) using numerical quadrature to approximate the multiple random effects. The model is illustrated using data from 135 classrooms from 28 schools on the effects of two smoking cessation interventions.

Clinical Trials as Topic↗
PubMed1996

Previous ectopic pregnancy should be considered a contraindication for microsurgery.

BACKGROUND: To estimate the risk of subsequent ectopic pregnancy (EP) after tubal surgery, given that the woman becomes pregnant, by means of a logistic model, a retrospective study was initiated. METHODS: During the period 1986-1990, 221 women with tubal infertility underwent microsurgery. Subsequent fertility was evaluated in 1991. Ninety women conceived, of whom 84 were included in the study (30 with EP and 54 with intra-uterine pregnancy as the only outcome). Clinical background factors of importance, surgical procedures used, scoring systems for tubal lesions, adnexal adhesions and risk of EP were analysed for possible correlation to subsequent EP. These factors were further used in a logistic model to estimate the risk of subsequent EP as only outcome. RESULTS: The risk of EP after microsurgery is minimum 15% without any risk factors. Previous EP and endometriosis could be identified as factors with prognostic power in the logistic model. One previous EP implies a 60% risk, whereas two previous EPs and endometriosis increase the risk to 95%. CONCLUSION: Patients with previous EP should generally not be considered for microsurgery owing to the high risk of recurrence and to the reduced chance of intra-uterine pregnancy.

Adult↗
PubMed1990

Use of a multiple logistic regression model to determine prognosis of dairy cows with right displacement of the abomasum or abomasal volvulus.

Data at admission and at surgery were collected on 458 cows with right displacement of the abomasum or abomasal volvulus, to derive multiple logistic regression models for predicting postsurgical outcome (productive, salvaged, or terminal). The derived models contained few and easily obtained variables. The weight associated with each variable was determined objectively. Three admission variables (heart rate, base excess, and plasma chloride concentration), and 5 surgical variables (heart rate, base excess, diagnosis, method of decompression used, and appearance of abomasal serosa) were used in the final models. Predicted outcomes that used the admission and surgical models were closely related with actual outcomes. Total correct classification for satisfactory (productive) versus unsatisfactory outcome (salvaged and terminal) was 78.2% for the admission model and 82.7% for the surgical model. Combining data on cows with productive and salvaged outcomes as satisfactory outcome, and terminal as unsatisfactory outcome, total correct classification was 90.7% for the admission model and 93.2% for the surgical model. Using predicted probabilities, the market value of productive and salvaged cows, and the medical and surgical costs, one can calculate the expected economic value of each outcome. Treatment can be justified if the sum of the expected value of productive and salvaged outcome exceeds the sum of the medical and surgical costs and the expected salvaged value of the cow that was not treated surgically.

Abomasum↗
PubMed2003

[Usefulness of logistic regression model to predict the endometrial carcinoma based on blood flow indices measured wit the of three-dimensional Doppler sonography].

AIM: Construction and prospective verification of predictive model permitting to rate individual probability of existence of endometrial carcinoma with use of three-dimensional Doppler Doppler sonography. MATERIAL AND METHODS: We analyzed the results of 3D sonography of 123 women (mean age 53.8 +/- 10.6; mean BMI--28.2 +/- 5.4). We estimated: endometrial thickness and volume, blood flow indices. All ultrasound measurements were verified by histology. In aim of finding the best combination of features essentially affect on endometrial cancer's risk and for estimate individual probability of endometrial cancer we use a logistic regression analysis. The obtained model was verified on 20 new cases. RESULTS: There were 24 women with endometrial cancer, 59 women with endometrial hyperplasia and 40 women without endometrial changes. We affirmed that only three variables had statistical significant influence on the constructed predictive model. Probability of endometrial carcinoma was: P(X) = 1/(1 + e-z), where "e" is mathematical constant and z = 0.12 x age + 0.16 x endometrial thickness + 0.47 x VI -11.3. The best sensitivity and specificity were 70.8% and 98.9%. The sensitivity and specificity in 20 new cases was adequately 75% and 91.6%. CONCLUSION: The usefulness of a predictive model built with the help of logistic regression analysis increase sonographic diagnostic precision. The usefulness of endometrial blood flow indices permit to estimate the individual risk of endometrial cancer in women examined with three-dimensional sonography.

Adult↗
PubMed1994

Predictors of employment status of treated patients with DSM-III-R diagnosis. Can logistic regression model find a solution?

To investigate the predictors of employment status of patients with DSM-III-R diagnosis, 55 patients were selected by a simple random technique from the main psychiatric clinic in Al Ain, United Arab Emirates. Structured and formal assessments were carried out to extract the potential predictors of outcome of schizophrenia. Logistic regression model revealed that being married, absence of schizoid personality, free or with minimum symptoms of the illness, later age of onset, and higher educational attainment were the most significant predictors of employment outcome. The implications of the results of this study are discussed in the text.

Adolescent↗
PubMed2005

The severity of Minamata disease declined in 25 years: temporal profile of the neurological findings analyzed by multiple logistic regression model.

Minamata disease (MD) was caused by ingestion of seafood from the methylmercury-contaminated areas. Although 50 years have passed since the discovery of MD, there have been only a few studies on the temporal profile of neurological findings in certified MD patients. Thus, we evaluated changes in neurological symptoms and signs of MD using discriminants by multiple logistic regression analysis. The severity of predictive index declined in 25 years in most of the patients. Only a few patients showed aggravation of neurological findings, which was due to complications such as spino-cerebellar degeneration. Patients with chronic MD aged over 45 years had several concomitant diseases so that their clinical pictures were complicated. It was difficult to differentiate chronic MD using statistically established discriminants based on sensory disturbance alone. In conclusion, the severity of MD declined in 25 years along with the modification by age-related concomitant disorders.

Adolescent↗
PubMed1987

Use of the logistic regression model for the analysis of proportionate mortality data.

A new statistical analysis strategy for proportionate mortality data is proposed. It is assumed that the occupational exposure, if it has an effect on mortality, increases the rate of death for some subset of causes by a multiplicative factor while not affecting the rates for the remaining causes of death. The unconditional logistic regression model is shown to provide a structure for the data analysis, with one of the predictors being the logit of the probability in the reference population that death was due to the affected causes. Using this model, one can estimate the effect of exposure while simultaneously controlling for a number of potential confounding and selection variables. Also, this model avoids the problems of comparing standardized proportionate mortality ratios, which are indirectly standardized measures. The model is demonstrated on a set of proportionate mortality data for factory workers from the northeastern United States.

Epidemiologic Methods↗
PubMed1997

[A logistic regression model applied to Chagas' disease]

The focus of this paper is the application of statistical models to the study of socioeconomic conditioning factors in perinatal Chagas' disease conducted in Rosario, Argentina. A case (154) and control (158) design was applied to investigate socioeconomic and cultural differences in pregnant women in Hospital Roque Sáenz Peña as to their infection status. Logistic regression models were used to evaluate the importance of antecedents linked to the infection and socioeconomic and cultural factors for infection status. For pregnant women, the importance of antecedents linked to the infection was confirmed and the women's level of schooling stood out as the predominant socioeconomic condition associated with infection. Log-linear models were used to explore the associations between certain explanatory variables. This approach pointed up the most relevant associations between such factors and Chagas' disease and provided a better understanding of the framework of relationships among them.

Journal Article↗
PubMed1996

[Dichotomization of continuous variables in logistic regression models].

It is highly to observe in the biomedical literature, that the continuous variables such as systolic blood pressure and cholesterol, are dichotomized and used in such manner in a given statistical analysis. The consequences of such transformation can be very varied. In this article, how the dichotomization of a continuous exposure variable affects the quality of the prediction of a response using the logistic regression is examined. One can conclude that, in almost all the studied stimulation, the percentages of misclassification can increase drastically, reaching over there times the probability of misclassification of the ones using the original variable. Therefore, we recommend to avoid dichotomization.

Logistic Models↗
PubMed1987

Analysis of proportionate mortality data using logistic regression models.

When only proportionate mortality data are available to an investigator studying the effect of an exposure on a particular cause of death, controls must be selected from among persons dying of other causes believed to be uninfluenced by the exposure under study. When qualitative or quantitative estimates of exposure history can be obtained for the deceased individuals, it is shown that one can use logistic regression models for the mortality odds to efficiently estimate the effect of exposure while controlling for relevant confounding factors by incorporating a priori information on baseline mortality rates available from US life tables. The proposed method is used to reanalyze data from a cohort of arsenic-exposed workers in a Montana copper smelter.

Adult↗
PubMed1985

Application of the log-linear and logistic regression models in the prediction of systemic lupus erythematosus in the dog.

This study sought to mathematically define canine systemic lupus erythematosus (SLE) by unifying diagnostic criteria proposed by others. Thirty-one cases of canine SLE were selected for modeling when 4 different published schemes agreed on the diagnosis, and 122 controls were selected when a patient's status met no scheme's criteria. The log-linear method showed an association between SLE and polyarthritis, hematologic abnormalities, renal damage, dermatologic disorders, and antinuclear antibody test response (positive). Logistic regression was then used to derive a predictive algorithm that could identify cases and controls with which all published criteria would be in accordance. The final equation correctly classified 93.5% of the affected dogs and 98.4% of the controls. It was concluded that the log-linear and logistic regression models are useful for the diagnosis of clinically similar, but distinguishable, disease states.

Animals↗
PubMedDate not supplied

An empirical method to refine personality disorder classification using stepwise logistic regression modeling to develop diagnostic criteria and thresholds.

This study of DSM-III-R personality disorder (PD) classification provides an empirical approach to determine (1) the discriminative power of each criterion and (2) the optimal number of criteria needed to diagnose the presence of each PD. A semistructured assessment of 110 outpatients was performed for the 11 PDs and their 104 diagnostic criteria. Sensitivity, specificity, and predictive powers were calculated for each criterion item. Logistic regression was performed to determine (1) univariate weightings of the individual criteria as applied to a given diagnosis, and (2) multivariate measures of the criteria that significantly improved the chi-square value in a stepwise fashion. The significant items were then equally weighted to calculate the optimal number needed to diagnose category membership. Of 104 PD criteria, 41 discriminated at a significance level of .05 or less, and each PD could be optimally diagnosed with fewer criteria than currently required. We can empirically reduce the number of criteria combinations comprising individual categories, decrease heterogeneity, and narrow diagnostic boundaries. This increases the likelihood of identifying etiological factors, predictors of clinical course, specific treatments, familial aggregation, and neurobiological correlates for the PD taxa.

Adolescent↗
PubMed2001

[A study on affecting factors on dental care demands by logistic regression model].

OBJECTIVE: To develop probability model for dental visits based on analysis of the factors affecting people's dental services utilization in urban area of Beijing, thus providing some evidence for forecasting further demanded dental care and building the more efficient oral health care delivery system. METHODS: A cross-sectional survey was conducted amnog 1,517 subjects of all age groups in Beijing selected by stratified, clustering, random sampling. The first model of dental care demand--probability model for dental visits was established with logistic regression. RESULTS: Awareness of oral health showed the most important relationship. Different social background and economic factors also had great effects on demand. Higher demand for care were existed among those who suffered from a disease that was painful or resulted in poor oral function. CONCLUSIONS: It is suggested that dental care has different characters from general health service. Utilization of dental service is largely determined by people's awareness, income and insurance system. So oral health education should be extended in order to promote the effective demands and reduce the potential demands. Rational oral health insurance system should be set up to provide some priority to certain people, thus stimulating the supply of dental care and improving the utilization of dental service. Since raising price for care will restrain the low-incomer's basic demand, and it will neither add the profit nor improve people's oral health, the price decision should be considered carefully. At last, oral health care should be emphasized on children and adolescent as priority age groups based on the conclusion from demand model.

Dental Care↗
PubMed2005

Characteristics of rear-end accidents at signalized intersections using multiple logistic regression model.

Multi-vehicle rear-end accidents constitute a substantial portion of the accidents occurring at signalized intersections. To examine the accident characteristics, this study utilized the 2001 Florida traffic accident data to investigate the accident propensity for different vehicle roles (striking or struck) that are involved in the accidents and identify the significant risk factors related to the traffic environment, the driver characteristics, and the vehicle types. The Quasi-induced exposure concept and the multiple logistic regression technique are used to perform this analysis. The results showed that seven road environment factors (number of lanes, divided/undivided highway, accident time, road surface condition, highway character, urban/rural, and speed limit), five factors related to striking role (vehicle type, driver age, alcohol/drug use, driver residence, and gender), and four factors related to struck role (vehicle type, driver age, driver residence, and gender) are significantly associated with the risk of rear-end accidents. Furthermore, the logistic regression technique confirmed several significant interaction effects between those risk factors.

Accidents, Traffic↗