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 739 records · Page 41Linked to original sources

PubMed1999

Logistic regression analysis for more than one characteristic of exposure.

When more than one characteristic of an exposure is under study, it is easy to misinterpret the results of a logistic regression analysis that incorporates terms for each characteristic. For example, in a study of the risk of endometrial cancer in relation to the duration and the recency of use of estrogen replacement therapy (ERT), simultaneously including terms for duration and recency of exposure to ERT in a logistic model may leave the mistaken impression that it is possible to adjust for recency when a given duration of ERT use is compared with no use. In this article, the authors show why such an adjusted comparison is impossible, and they discuss several pitfalls in the interpretation of logistic regression coefficients when two or more characteristics of exposure are under study. They also suggest a method for avoiding these pitfalls.

Confounding Factors, Epidemiologic↗

PubMed1995

A multivariate analysis of factors related to the mortality of blunt trauma admissions to the North Staffordshire Hospital Centre.

OBJECTIVE: To identify characteristics of blunt trauma admissions associated with mortality and to derive a linear logistic model for predicting the probability of mortality. DESIGN: A prospective observational study. MATERIALS AND METHODS: Data were collected on blunt trauma patients admitted via the Accident and Emergency Department, North Staffordshire Hospital Centre. Univariate and multivariate analyses were performed of potential risk factors associated with mortality. MEASUREMENTS AND MAIN RESULTS: The main outcome variable was survival or death occurring in hospital within 30 days of admission. Injury Severity Score, age, Revised Trauma Score, and place of injury (road traffic, home, or elsewhere) were independently related to mortality. The effect of age on mortality was best described using three categories: 0-64 excluding 15-24, 15-24, 65+. The survival probability model suggests that a person whose injury occurs in the home has a lower probability of survival than a patient with similar age and trauma scores who is involved in a road traffic accident. A model validation check indicated good agreement between model predictions and patient outcomes. CONCLUSIONS: The authors propose that use of this model may provide a more accurate evaluation of the mortality of British trauma admissions than would be obtained using models based on data from American trauma cases.

Adolescent↗

PubMed1993

Predicted probabilities of hospital death as a measure of admission severity of illness.

This paper evaluates a new method for assessing hospital admission severity of illness based on disease-specific models (logistic regression) of the probability of in-hospital mortality. Results for the 26 disease groups in MDC 4--Diseases of the Respiratory System, MDC 5--Diseases of the Circulatory System, and MDC 6--Diseases of the Digestive System are presented using data on all 1991 admissions from 111 hospitals throughout the United States. These disease models are empirically derived using clinical findings from laboratory, radiology, pathology, diagnostic procedures, patient history and physical exam, as well as patient age and sex. Each predictive algorithm is presented, and the strong predictive performance of these models is indicated by the average C statistic of .870. A predicted probability of death is calculated for each hospital patient in the study sample, and these probabilities comprise a continuous variable that indicates admission severity of illness.

Algorithms↗

PubMed2006

Crowding effect on adult growth, pre-patent period and egg shedding of Fasciola hepatica.

Fascioliasis pathogenesis depends on fluke burden. In human hyperendemic zones, individual infection intensities reach very high levels and the majority of infected subjects should be in the advanced chronic phase. The rat model offers a useful approach for pathological research in the advanced chronic period. The influence of infection intensity per rat on fluke development, pre-patent period and egg shedding (eggs/g faeces/worm) was analysed in 3 groups (I: 1-3 worms/rat; II: 4-6; III: 7-9). Ontogenetic trajectories of fluke body measures followed a logistic model. Results showed that when the burden increases, the maximum values of fluke measures decrease. The crowding effect is manifest when fluke measures approximate their maximums in the advanced chronic stage. The pre-patent period and egg production decrease when the burden increases. This means that measurements of eggs per gramme of faeces tend to underestimate the fluke burden. The present study demonstrates how to quantify the fascioliasis experimental rat model crowding effect on adult growth, pre-patent period and egg production. This quantification may be of great interest in epidemiological studies and in experimental research on the in vivo actions of different anthelminthic drugs and vaccines, pathology, immunology and resistance studies.

Animals↗

PubMed2006

Hepatitis A prevalence among injection drug users.

OBJECTIVE: The purpose of this study was to develop a descriptive model of the association between injection drug use and hepatitis A (HAV) in a sample of injection drug users (IDUs). DESIGN: From May 1997 to July 1999, 493 subjects were administered the NIDA Risk Behavior Assessment (RBA). Participants had blood drawn; sera were tested for antibodies to HAV, hepatitis B core (HBcAB), and hepatitis C. The principal method of analysis was logistic regression. SETTING: The study took place in a community-based field station in Anchorage, Alaska. PARTICIPANTS: Eligibility was determined using the following criteria: a) age greater than 17 years, b) possession of picture identification, c) positive urinalysis for cocaine metabolites, morphine, and/or amphetamines using the ONTRAK system (Roche Diagnostics), and d) injection drug use in the last six months as confirmed by presentation of track marks. MAIN OUTCOME MEASURE: Presence of antibodies to HAV infection. RESULTS: The prevalence of total HAV antibody in our sample was 33% (161/493). The final multivariate logistic model, using positive HAV serostatus as the outcome, included positive HBcAB serostatus (OR = 3.43; 95% CI, 2.22-5.30), less than high school education (vs. high school or greater education) (OR = 2.05; 95% CI, 1.33-3.17), age (OR = 1.06 (each year); 95% CI, 1.03-1.09), number of days injected heroin in the last 30 days (OR = 1.05 (each day), 95% CI, 1.01-1.08), and race (White vs. all other race/ethnicities) (OR = 0.49; 95% CI, 0.32-0.75). CONCLUSIONS: A model including both demographic and drug use variables best describes HAV prevalence in this sample. Findings suggest that IDUs are targets for interventions focusing on hepatitis vaccinations and hygiene practices. Further research is needed to understand the association of HAV with hepatitis B infection.

Adult↗

PubMed2000

Demographic models of birth outcomes and infant mortality: an alternative measurement approach.

Most demographic studies use 2,500 grams of birth weight and 37 weeks of gestation as cutpoints for evaluating the effects of adverse birth outcomes on infant mortality. We propose an alternative strategy, which relies on continuous measures of birth outcomes, identifies an optimal combination of birth weight and gestational age for infant survival, and estimates the effects of adverse birth outcomes in terms of their departure from this "optimal point." We illustrate the advantages of this approach by estimating a logistic model using data from the 1989-1991 NCHS linked birth/infant death files. Finally, we discuss future applications and methodological issues to be resolved in subsequent research.

Birth Weight↗

PubMed1999

Study on the relationship between milk immune factors and Staphylococcus aureus intramammary infections in dairy cows.

The distribution of Staphylococcus aureus within herds seems to be related to interactions among the shedding characteristics of the bacteria, their pathogenicity and mammary gland immune status. The aim of the present study was to investigate the relationships between selected mammary gland immune factors and intramammary infections associated with Staph. aureus. Overall, 70 cows from five commercial dairy herds were included in the study and quarter milk samples were assessed using bacteriological and cytological tests. We evaluated differential cell count, lysozyme concentration, N-acetyl-beta-glucosaminidase (NAGase) activity, cell viability and respiratory burst activity in randomly chosen quarter milk samples from each cow. Staph. aureus intramammary infection elicited different responses in the mammary gland immune defences investigated. Polymorphonuclear leucocytes (PMN) as a proportion of total somatic cells in milk, cell viability and NAGase activity were higher in infected quarters, while the proportions of macrophages and lymphocytes, respiratory burst activity and lysozyme levels were lower. Mean values differed among herds, but the differences were not significant. These changes were associated with Staph. aureus infection. The reduced respiratory burst activity together with the increase in the proportion of PMN suggests that both the number and activity of PMN could influence the susceptibility of the mammary gland to pathogens. Indeed, the logistic model adopted suggests that impairment of milk immune factors could be concurrent with the development of an infection.

Acetylglucosaminidase↗

PubMed1991

Collagenase activity in recurrent periodontitis: relationship to disease progression and doxycycline therapy.

Previous reports have suggested that active progression of periodontitis may be correlated with increased collagenolytic activity, and that improved clinical conditions after tetracycline treatment may be explained by inhibition of host collagenase. Eighty-two patients with a recent history of periodontal abscesses and/or loss of gingival attachment level (GAL) despite active periodontal therapy were enrolled in a double-blind, randomized, placebo-controlled trial. Clinical measurements, sampling of gingival crevicular fluid (GCF) and subgingival scaling were performed every 2 months. If any site exhibited greater than 2 mm loss of GAL or a periodontal abscess, patients were administered either 100 mg doxycycline per day for 3 weeks or placebo. During 12 months of monitoring, 55 patients exhibited recurrent active disease and were then randomly assigned to either the doxycycline (n = 30) or placebo (n = 25) groups. Analysis of active collagenase and latent collagenase in GCF samples were determined by functional assays and quantitated after SDS-PAGE and fluorography. Collagenase activities were assayed at sites exhibiting active destruction (study site), at sites with pocket depth comparable to the study site but without active destruction, and at healthy sites. Clinical measurements of GAL and collagenase activity were made at intervals between 1 wk and 7 months after completion of the drug regime. Within 7 months, 15 out of 19 patients on placebo exhibited recurrent disease compared to 13 out of 29 patients on doxycycline. Collagenase activity exhibited large variations among patients and was analyzed as presence or absence of active collagenase with a logistic model.(ABSTRACT TRUNCATED AT 250 WORDS)

Analysis of Variance↗

PubMed1985

Long-term results and prognostic factors in patients with differentiated thyroid carcinoma.

A multivariate analysis of the prognostic factors was carried out on a series of 546 differentiated thyroid cancers followed for 8 to 40 years. For survival, the highest risk factor was associated with age; tumors diagnosed in patients younger than 45 years had higher relapse-free survival (RFS) and total survival (TS) rates and a slower growth rate. In children, although the RFS and TS at 15 years were high, they decreased later. The second independent prognostic factor was histology. There was no difference between papillary and follicular well-differentiated (FWD) tumors, but follicular moderately differentiated (FMD) had lower TS and RFS. Among FMD cancers, relapses occurred earlier and the interval between relapse and death was shorter. The third factor was sex. Tumors tended to disseminate more in male than in female patients. The survival rate after relapse was the same, however, suggesting that the growth rates are not different. The presence of palpable lymph nodes also had a significant independent impact on both TS and RFS. Patients treated after 1960 have a better outcome than patients treated earlier, although they did not differ in age distribution, histologic characteristics, sex ratio, or incidence of palpable lymph nodes. The distribution of time intervals between treatment and relapse was not compatible with an exponential failure time model but fit with a log-logistic model. Relapses can occur as late as 30 years or more after initial treatment. Elevated levels of circulating thyroglobulin have been observed in about 12% of the patients who had been in complete remission for longer than 20 years.

Adolescent↗

PubMed2005

Modeling Staphylococcus aureus growth and enterotoxin production in milk.

Staphylococcus aureus growth and its enterotoxin production in sterilized milk were modeled with a modification of a new logistic model recently developed by us. The modified model and the Baranyi model described the early exponential phase of a growth curve more accurately than the previous model, at constant temperatures from 14 to 36.5 degrees C. The amount of toxin in milk increased linearly with time from the time the cell population reached about 10(6.5) cfu/ml. The rate of toxin production linearly increased at temperatures between 14 and 32 degrees C. From parameter values obtained at the constant temperatures, the model successfully predicted bacterial growth in the milk at a varying temperature. For toxin level estimation, we postulated that the rate of toxin production might be regulated with the temperature after the cell concentration reached 10(6.5) cfu/ml; the time point when the cell concentration reached that value was predicted with the modified growth model. Introduction of a correction factor in the toxin estimation successfully predicted the toxin level in milk at a varying temperature. These results showed that this prediction system consisting of the modified model and the toxin production algorithm might be a useful tool for modeling bacterial growth and its metabolite production in liquid foods.

Animals↗

PubMed1997

Prediction of early death in end-stage renal disease patients starting dialysis.

Demand for dialysis for patients with end-stage renal disease is growing, as is the comorbidity of dialysis patients. Accurate prediction of those destined to die quickly despite dialysis could be useful to patients, providers, and society in making decisions about starting dialysis. To determine whether age and comorbidity accurately predict death within 6 months of first dialysis for end-stage renal disease, a prospective cohort study of 822 patients starting dialysis at one of 11 Canadian centers was performed. Patient characteristics were recorded at first dialysis. Follow-up continued until death or study end (at least 6 months after enrollment). One hundred thirteen of 822 (13.7%) patients died within 6 months. Although an existing scoring system predicted prognosis, adverse scores greater than 9 were found in only 9.7% of those who died; only 52% of those who scored higher than 9 died within 6 months. No score cutoff point combined high true-positive and low false-positive rates for predicting early death. Age, severity of heart failure or peripheral vascular disease, arrhythmias, malnutrition, malignancy, or myeloma were independent prognostic factors identified in multivariate models. However, the best fit discriminant and logistic models were also unable to accurately predict death within 6 months. Clinicians were very accurate in assigning patients to prognostic groups up to a 50% risk of death by 6 months, above which they tended to overestimate risk. However, clinicians were only marginally better than the predictive models in determining whether a given high-risk patient would die. The inability of a scoring system or clinical intuition to accurately predict death soon after starting dialysis for end-stage renal disease suggests that limiting access to dialysis on the basis of likely short survival may be inappropriate in Canada.

Aged↗

PubMed2005

Analysis of alcohol dependence phenotype in the COGA families using covariates to detect linkage.

Linkage analysis methods that incorporate etiological heterogeneity of complex diseases are likely to demonstrate greater power than traditional linkage analysis methods. Several such methods use covariates to discriminate between linked and unlinked pedigrees with respect to a certain disease locus. Here we apply several such methods including two mixture models, ordered subset analysis, and a conditional logistic model to genome scan data on the DSM-IV alcohol dependence phenotype on the Collaborative Studies on Genetics of Alcoholism families, and compare the results to traditional nonparametric linkage analysis. In general, there was little agreement among the various covariate-based linkage statistics. Linkage signals with empirical p-values less than 0.001 were detected on chromosomes 3, 4, 7, 10, and 12, with the highest peak occurring at the GABRB1 gene using the ecb21 covariate.

Alcoholism↗

PubMed2004

Women with coronary artery disease report worse health-related quality of life outcomes compared to men.

BACKGROUND: Although there have been substantial medical advances that improve the outcomes following cardiac ischemic events, gender differences in the treatment and course of recovery for patients with coronary artery disease (CAD) continue to exist. There is a general paucity of data comparing the health related quality of life (HRQOL) in men and women undergoing treatment for CAD. The purpose of this study was to compare HRQOL outcomes of men and women in Alberta, at one-year following initial catheterization, after adjustment for known demographic, co-morbid, and disease severity predictors of outcome. METHOD: The HRQOL outcome data were collected by means of a self-reported questionnaire mailed to patients on or near the one-year anniversary of their initial cardiac catheterization. Using the Seattle Angina Questionnaire (SAQ), 5 dimensions of HRQOL were measured: exertional capacity, anginal stability, anginal frequency, quality of life and treatment satisfaction. Data from the APPROACH registry were used to risk-adjust the SAQ scale scores. Two analytical strategies were used including general least squares linear modeling, and proportional odds modeling sometimes referred to as the "ordinal logistic modeling". RESULTS: 3392 (78.1%) patients responded to the follow-up survey. The adjusted proportional odds ratios for men relative to women (PORs > 1 = better) indicated that men reported significantly better HRQOL on all 5 SAQ dimensions as compared to women. (PORs: Exertional Capacity 3.38 (2.75-4.15), Anginal Stability 1.23 (1.03-1.47), Anginal Frequency 1.70 (1.43-2.01), Treatment Satisfaction 1.27 (1.07-1.50), and QOL 1.74 (1.48-2.04). CONCLUSIONS: Women with CAD consistently reported worse HRQOL at one year follow-up compared to men. These findings underline the fact that conclusions based on research performed on men with CAD may not be valid for women and that more gender-related research is needed. Future studies are needed to further examine gender differences in psychosocial adjustment following treatment for CAD, as adjustment for traditional clinical variables fails to explain sex differences in health related quality of life outcomes.

Aged↗

PubMed2005

Improved empirical models describing hormesis.

During the past two decades, the phenomenon of hormesis has gained increased recognition. To promote research in hormesis, a sound statistical quantification of important parameters, such as the level and significance of the increase in response and the range of concentration where it occurs, is strongly needed. Here, we present an improved statistical model to describe hormetic dose-response curves and test for the presence of hormesis. Using the delta method and freely available software, any percentage effect dose or concentration can be derived with its associated standard errors. Likewise, the maximal response can be extracted and the growth stimulation calculated. The new model was tested on macrophyte data from multiple-species experiments and on laboratory data of Lemna minor. For the 51 curves tested, significant hormesis was detected in 18 curves, and for another 17 curves, the hormesis model described that data better than the logistic model did. The increase in response ranged from 5 to 109%. The growth stimulation occurred at an average dose somewhere between zero and concentrations corresponding to approximately 20 to 25% of the median effective concentration (EC50). Testing the same data with the hormesis model proposed by Brain and Cousens in 1989, we found no significant hormesis. Consequently, the new model is shown to be far more robust than previous models, both in terms of variation in data and in terms of describing hormetic effects ranging from small effects of a 10% increase in response up to effects of an almost 100% increase in response.

Animals↗

PubMed1999

[Analysis of longitudinal Gaussian data with missing data on the response variable].

BACKGROUND: Using an application and a simulation study we show the bias induced by missing data in the outcome in longitudinal studies and discuss suitable statistical methods according to the type of missing responses when the variable under study is gaussian. METHOD: The model used for the analysis of gaussian longitudinal data is the mixed effects linear model. When the probability of response does not depend on the missing values of the outcome and on the parameters of the linear model, missing data are ignorable, and parameters of the mixed effects linear model may be estimated by the maximum likelihood method with classical softwares. When the missing data are non ignorable, several methods have been proposed. We describe the method proposed by Diggle and Kenward (1994) (DK method) for which a software is available. This model consists in the combination of a linear mixed effects model for the outcome variable and a logistic model for the probability of response which depends on the outcome variable. RESULTS: A simulation study shows the efficacy of this method and its limits when the data are not normal. In this case, estimators obtained by the DK approach may be more biased than estimators obtained under the hypothesis of ignorable missing data even if the data are non ignorable. Data of the Paquid cohort about the evolution of the scores to a neuropsychological test among elderly subjects show the bias of a naive analysis using all available data. Although missing responses are not ignorable in this study, estimates of the linear mixed effects model are not very different using the DK approach and the hypothesis of ignorable missing data. CONCLUSION: Statistical methods for longitudinal data including non ignorable missing responses are sensitive to hypotheses difficult to verify. Thus, it will be better in practical applications to perform an analysis under the hypothesis of ignorable missing responses and compare the results obtained with several approaches for non ignorable missing data. However, such a strategy requires development of new softwares.

Data Interpretation, Statistical↗

PubMed1978

The analysis of pair-matched case-control studies, a multivariate approach.

In matched case-control studies one frequently must consider more than one variable in the analysis and in this paper a log-linear model is presented to meet this objective. A conditional argument yields a method for making inferences on the parameters measuring the association between the variables and disease. The result is similar to the problem of fitting a Bradley-Terry model when one has paired comparisons. Two methods of obtaining maximum conditional likelihood estimates of the parameters are possible: (i) fitting a quasi-independence model in the usual log-linear models context and (ii) fitting a linear logistic model. The results are illustrated with examples.

Adult↗

PubMedDate not supplied

Effect of gender on suicide attempters versus nonattempters in an adolescent inpatient unit.

BACKGROUND: Although gender differences have been noted in the risk factors for suicide and attempted suicide, comparative studies to date have used only 2 groups and a limited number of measures. The present study compared the effect of gender on suicide among 4 groups of psychiatrically hospitalized adolescents using a cross-sectional design. METHODS: The study sample consisted of 404 patients, aged between 12 and 21, who were divided into 4 groups: 76 male suicide attempters, 103 male nonattempters, 143 female suicide attempters, and 82 female nonattempters. Patients were tested for life events, affective disorders, aggression, impulsivity, ego defense mechanisms, and death perception with the Child Suicide Potential Scale, Beck Depression Inventory, State-Trait Anxiety Inventory, Overt Aggression Scale, Multidimensional Anger Inventory, Impulsivity Control Scale, and Life Style Index. Findings were analyzed by multivariate regression with stepwise logistic models. RESULTS: Depression and anxiety were more prevalent in female nonattempters than in male nonattempters; there were no such gender differences among the attempters. Antisocial behavior was more prevalent in male attempters than in female attempters; there were no gender differences on this aspect among the nonattempters. There were gender differences for defense mechanisms in the attempters. Logistic regression models for men and women separately revealed that antisocial behavior and anxiety were common predictors of suicide attempt, that destructiveness was a predictor in women only, and that depression was associated with suicide attempt in men only. CONCLUSIONS: Suicide-prone female and male adolescent inpatients show distinct differences in psychopathology, ego defense mechanisms, and life events compared to psychiatrically hospitalized adolescents without any history of suicide attempt. Any deviation from a gender-specific behavior must raise suspicion of a risk of attempted suicide.

Adolescent↗

PubMed1998

Smoking and suicidality among psychiatric patients.

OBJECTIVE: This study investigated the relationship between smoking and suicidality among psychiatric patients. METHOD: All psychiatric patients (N = 1,217) from Kuopio University Hospital in eastern Finland were examined for current smoking, suicidal ideation, previous suicide attempts, and other risk factors by using a cross-sectional, multivariate design. RESULTS: The probability of at least one previous suicide attempt was 100% higher in current smokers than in nonsmokers. Smokers also had a 43% higher risk of experiencing mild to severe suicidal ideation than nonsmokers. CONCLUSIONS: Current smoking was significantly related to suicidality in multiple logistic models, even after several confounding factors were controlled. The independence of this association seems possible, but the subject requires further research.

Adult↗