SEARCH · PubMed Health

Results for “Logistic Model”

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 469 records · Page 26Linked to original sources

PubMedDate not supplied

Assessing the adequacy of the logistic regression model for matched case-control studies.

We apply results for the detection of outliers in logistic regression to the analysis of matched pairs data and illustrate the techniques with data from a matched pairs study of the relationship between iron-binding proteins and mortality in the Solomon Islands. Although the study includes 90 matched pairs, the conclusions change substantially with deletion of one or two pairs. Thus, even with reasonably large data sets, single observations may have a great impact on the results of an analysis. The identification of influential observations should constitute part of the analysis of matched case-control data.

Clinical Trials as Topic↗

PubMed1997

A logistic regression model when some events precede treatment: the effect of thrombolytic therapy for acute myocardial infarction on the risk of cardiac arrest.

When outcomes occur in clinical trials before treatment can be given, neither intent-to-treat nor according-to-protocol analyses give optimal estimates of the treatment effect. A better approach employs a time-dependent variable for treatment. Intent-to-treat analyses are conservative, biasing against treatment; according-to-protocol analyses bias in favor of treatment. We show how to measure the effect of a time-dependent variable in a logistic regression using person-time intervals as units of measurement and describe appropriate methods for reporting model performance. The method is applied to develop a model to predict the probability that a patient with a myocardial infarction will have a sudden cardiac arrest within 48 hours of presentation to emergency medical services both when treated with thrombolysis and when not treated. We use a time-dependent treatment variable because many patients went into cardiac arrest while awaiting treatment. This technique has been programmed into an electrocardiograph for real-time use in an emergency department.

Aged↗

PubMed2001

Modelling the effects of genetic and environmental factors on the risk of autoimmune disease.

The presence or absence of autoimmune disease is a qualitative outcome resulting from the combined influences of genetic, environmental and stochastic factors that interact in complex ways. Because of the nature of these interactions, the effects of risk factors that contribute to the aetiology of disease cannot be adequately assessed individually. Here, I propose an approach to multiparameter modelling of autoimmune disease that incorporates Falconer's concept of liability in a generalized linear model. Specifically, a logistic model, allied to that used to assess the influence of genetic risk factors, is suggested. The limits of modelling autoimmune disease and the requirements for data collection are also briefly discussed.

Autoimmune Diseases↗

PubMed1994

Tumor growth in vivo and as multicellular spheroids compared by mathematical models.

In vivo volume growth of two murine tumor cell lines was compared by mathematical modeling to their volume growth as multicellular spheroids. Fourteen deterministic mathematical models were studied. For one cell line, spheroid growth could be described by a model simpler than needed for description of growth in vivo. A model that explicitly included the stimulatory role for cell-cell interactions in regulation of growth was always superior to a model that did not include such a role. The von Bertalanffy model and the logistic model could not fit the data; this result contradicted some previous literature and was found to depend on the applied least squares fitting method. By the use of a particularly designed mathematical method, qualitative differences were discriminated from quantitative differences in growth dynamics of the same cells cultivated in two different three-dimensional systems.

Animals↗

PubMedDate not supplied

Bacterial growth measurement using an automated system: mathematical modelling and analysis of growth kinetics.

The reliability of automated systems was assessed by simultaneously using an MS2 and a methodology of data interpretation/mathematical modelling. Two experimental situations were analysed. Bacterial growth both in synthetic and in complex media could be well described by Monod's model and the logistic model. In the two cases, the fit of the models to the data was satisfactory.

Bacteria↗

PubMed2006

Extensive multiallelic analysis of the relationship between HLA-DRB1 and rheumatoid arthritis using a Bayesian partition model.

To analyse the association between individual HLA-DRB1 locus genotypes and rheumatoid arthritis (RA) susceptibility, taking in account the multiallelic nature of the shared epitope (SE). In total, 538 patients and 536 controls were genotyped for 12 alleles of the HLA-DRB1 locus. A Bayesian partition model and multivariate logistic models were used to assess the role of the SE and of its individual components. The SE was associated with RA susceptibility (odds ratio (OR) 2 versus 0 SE copy=9.99 (95 CI 4.69-15.30) and OR 1 versus 0 SE copy=3.16 (95% CI 2.42-4.12)). The Bayesian partition model supplied a permutation of the HLA-DRBA locus alleles ordered by increasing disease risk. Alleles associated with highest risks are those that code for the SE. The individual OR estimations for the HLA-DRB1 locus genotypes went from OR=1.00 (95% CI 1.00-1.25) for the less associated genotype to OR=21.40 (95% CI 8.02-65.79) for the most associated one. In conclusion, the allele order risk and the OR estimations for individual genotypes of the HLA-DRB1 locus were consistent with the SE theory. Using an exploratory statistical method without a priori hypothesis, our study allowed a detailed analysis of the multiallelic nature of the SE.

Adult↗

PubMed2005

Participant retention in clinical trials of candidate HIV vaccines.

OBJECTIVE: : To determine predictors of loss to follow-up (LTFU) in trials of candidate HIV vaccines. METHODS: : Data were obtained from trials of candidate preventive HIV vaccines conducted by the AIDS Vaccine Evaluation Group (AVEG) and HIV Network for Prevention Trials (HIVNET) that enrolled HIV-negative volunteers. Analytic models included multiple logistic regression and generalized estimating equations. RESULTS: : Of 3033 volunteers enrolled in 48 trials, 282 (9.3%) persons did not complete follow-up. In univariate analyses, age, trial duration, and number of immunizations were associated with LTFU. In a multivariate logistic model, age (per year) (adjusted odds ratio [AOR] = 0.96, 95% confidence interval [CI]: 0.95, 0.98) and study duration (per month) (AOR = 1.04, 95% CI: 1.01, 1.08) remained significantly associated with LTFU. CONCLUSIONS: : Younger age and increasing trial duration predicted LTFU. Limiting enrollment in trials of novel products to those less than 40 years of age may exclude participants shown to have improved retention. Trials should be designed to last only as long as required to address the scientific question. Retention efforts in future trials should especially address younger persons.

AIDS Vaccines↗

PubMed1990

Use of survey data and small area statistics to assess the link between individual morbidity and neighbourhood deprivation.

STUDY OBJECTIVE: The aim of the study was to examine the relationship between sociogeographic factors and health, using a particular social indicator of neighbourhood deprivation. DESIGN: The study analysed the relationship between health problems (reported by randomly selected respondents to a household survey) and an area social indicator for the neighbourhoods where the respondents lived (based on census data). The area social indicator tested was the underprivileged areas indicator developed by the St Mary's Hospital Department of General Practice, London. Generalised linear interactive modelling with a logistic model was used to test the strength of the relationship between social indicators and morbidity, and to calculate the probability of reporting illness or consultations for survey respondents living in different types of area. SETTING: The study population was derived from three London health districts and their corresponding census enumeration districts. PARTICIPANTS: Responses were obtained from 738 households drawn from the local taxation evaluation list (66% of those sampled), and 1384 persons over 16 participated in the survey (94% of eligible adults in households surveyed). Of these, 1221 provided complete data on health problems. The survey population was considered representative of the general population in the areas studied since its characteristics were similar to those reported for the population as a whole in the 1981 census. RESULTS: Within different age and sex groups, those living in very deprived areas, with high underprivileged area scores, were more likely to consult their doctor and to report various indicators of poor health than those living in privileged areas, with low underprivileged area scores. CONCLUSIONS: The underprivileged areas index may provide a useful surrogate indicator to estimate morbidity and use of general practitioner services in small areas. It is likely to be most effective in areas where sociodemographic profiles of the local population are highly contrasting.

Adolescent↗

PubMed1993

Mortality and incidence of cancer among sewage workers: a retrospective cohort study.

To study the incidence of and mortality from cancer among sewage workers a retrospective analysis was performed on a cohort of 656 men employed for at least one year at any one of 17 Swedish sewage plants during the years 1965-86. Assessment of exposure was done by classification of work tasks. Lower than expected total mortality (standardised mortality ratio (SMR) = 0.75, 95% confidence interval (95% CI) 0.58-0.97) and cardiovascular mortality (SMR = 0.61, 95% CI 0.39-0.91) was found. This was interpreted as a result of the healthy worker effect. For all cancers combined the mortality (SMR = 1.08, 95% CI 0.68-1.67) and morbidity (SMR = 1.02, 95% CI 0.72-1.38) were comparable with those of the general population. There were increased incidences for brain tumours (SMR = 2.19, 95% CI 0.45-6.39), gastric cancers (SMR = 2.73, 95% CI, 1.00-5.94), and renal cancers (SMR = 1.68, 95% CI = 0.35-4.90). For lung cancer the risk was reduced (SMR = 0.70, 95% CI 0.15-2.05). Allowance for a latency period of 10 years from the start of exposure did not change the pattern. Logistic modelling was used to search for exposure-response relations. In a logistic model with the confounder age forced in, renal cancer had a significant positive relation with a weighted sum of employment times, where the weights describe the classification of exposure. No exposure-response relations were found for brain tumors or gastric cancers. The increased risks are based on small numbers of cases. A future follow up will add more conclusive power to the study. Specific exposures need to be identified to allow for a better dose-response analysis.

Cause of Death↗

PubMed1988

Exponential models for analyses of time-related factors, illustrated with asbestos textile worker mortality data.

In any study based on an occupational cohort, it is important to consider the variation in risk factors over time. Cumulative exposure is the most important time-related factor for exposure-response analyses, whereas other time-related factors such as age at risk, year at risk, and length of follow-up may be confounders and effect modifiers. This paper examines the family of exponential models which can be used for time-related analyses of studies based on an occupational cohort. Analyses using Poisson regression, the proportional hazards model, and the logistic model are presented, and their interrelationships explored. These models are illustrated with data from a cohort study of lung cancer mortality among asbestos textile plant workers. All three approaches yielded similar effect estimates. In particular, Poisson regression and the proportional hazards model yielded very similar findings, but Poisson regression has some conceptual and computational advantages.

Adult↗

PubMed1990

Early physiologic predictors of injury severity and death in blunt multiple trauma.

The importance of admission physiological and biochemical variables was modeled on data from 185 patients with blunt liver trauma with regard to their significance in prediction of mortality. The variables used were admission Glasgow Coma Score, base excess (or deficit), arterial lactate, Injury Severity Score, and initial 24-hour volume of blood required for replacement. Each variable was modeled as a predictor of survival alone and in combination, using a linear logistic model. In any two-variable combination, Glasgow Coma Score had a high likelihood ratio for prediction representing the influence of brain injury. But as a single variable reflecting the probability of death, both base excess (LD50 = -11.8 mmol/L) and initial 24-hour volume of blood (LD50 = 5.4 L) were highly significant. A combined logistic model of admission Glasgow Coma Score and base excess had the greatest likelihood of accurate prediction of outcome: P death = e lambda/l + e lambda; where lambda = -0.21(Glasgow Coma Score) -0.147(base excess) + 0.285. Testing of this predictive model on data from 323 additional patients with multiple trauma who had pelvic fracture as their index injury also showed it to be a highly significant early predictor of outcome.

Fluid Therapy↗

PubMed2005

[Hospital admissions forms versus medical records to assess risk of in-hospital death from acute myocardial infarction in Rio de Janeiro, Brazil].

The objective of this study was to analyze the usefulness of the Brazilian Hospital Information System (SIH) in comparison to medical records to study factors associated with in-hospital mortality due to acute myocardial infarction (AMI). We evaluated a stratified random sample of 391 medical records (out of 1,936 hospital admissions forms) with AMI as the primary diagnosis in the city of Rio de Janeiro. Factors associated with in-hospital death were studied through logistic modeling. Models were developed directly from the SIH and from medical records. ROC curves were constructed to allow comparison of the different models. We found an AMI diagnostic confirmation = 91.7% and hospital mortality = 20.6%. The logistic model derived from medical records produced the best fit (concordance = 90.1%). Although the SIH model had a worse fit (concordance = 70.6%), the correction of keying-in and information errors using data from medical records did not significantly modify its performance. Under-recording of secondary diagnosis was high in the SIH forms and was the main limiting factor.

Acute Disease↗

PubMed2006

[Factors affecting breastfeeding rate and duration].

OBJECTIVES: The purpose of this study is to examine the factors affecting the rate and duration of breastfeeding. METHODS: We analyzed the data from the year 2000 Korea National Fertility Survey that was collected through direct interviews. In particular, the mothers who delivered their last child and the child was under 1 year of age from January 1998 to June 2000 (N=1,066) were analyzed via a logistic model to assess the factors affecting the breastfeeding rate. Among the study subjects, those who had initiated breastfeeding (N=740) were analyzed through Cox's proportional hazard model to evaluate the factors affecting the duration of breastfeeding. RESULTS: The multivariate logistic model showed that the delivery type and the baby's birth-weight have a statistically significant influence on the breastfeeding rate. Women who delivered their babies through Cesarean section were less likely than others to breastfeed. In contrast, the women whose babies weighed 2.5 Kg or more were more likely than others to breastfeed. The results obtained from the survival analysis are as follows : the higher the mother's education level, the shorter is the breastfeeding duration. The mother's work status played a significant role in the early termination of breastfeeding. Women aged 35 or older showed a longer breastfeeding duration than the younger age groups, whereas the maternal age was not a significant factor in affecting whether or not a mother would breastfeed. CONCLUSIONS: Reducing the cases of operative delivery (Cesarean section) and low weight births, enlightening young and highly educated women on breastfeeding and improving the environment for breastfeeding on the job are important strategies to encourage women to breastfeed.

Adult↗

PubMed2000

Coronary heart disease incidence in northern and southern European populations: a reanalysis of the seven countries study for a European coronary risk chart.

OBJECTIVE: A systematic reanalysis of 10 year coronary heart disease incidence data from the northern and the southern European cohorts of the seven countries study, to contribute indirectly to the production of a European coronary risk chart. DESIGN AND SETTING: Men aged 40-59 years at entry were studied in three northern European cohorts based in Finland and Netherlands (n = 2213); and in 10 southern European cohorts based in Italy, former Yugoslavia, and Greece (n = 5897). Multiple logistic models for the prediction of coronary deaths, coronary incidence (hard criteria), and coronary incidence (any criterion) were solved for the two geographical groups and their pool. Risk factors fed into the models were age, systolic blood pressure, serum total cholesterol, and cigarette smoking. RESULTS: 10 year coronary heart disease mortality and incidence were higher in northern than in southern Europe, with ratios around 2.65. Ratios among the three coronary heart disease manifestations were identical in the two cultural groupings. Coefficients of the multiple logistic models were similar and not significantly different between the two groupings. When applying the coefficients back to the same or the opposite population, the relative risk was large and similar in the different cultures. Relative risk was larger for more severe coronary heart disease manifestations. The absolute risk was overestimated when applying the northern European model to southern European populations and vice versa, with ratios of about 1.5 and 0.5, respectively. Coronary risk charts created to reproduce the shape of those incorporated in recent European guidelines confirmed the excess of absolute risk in the northern compared with the southern European cohorts, all else being equal. CONCLUSIONS: In theory, a more appropriate European coronary risk chart could be produced by adopting coefficients to correct for different background incidence rates in different cultures. Other coefficients could appropriately be used to transform mortality risk into incidence risk.

Adult↗

PubMed2006

Challenging the validity of the EPO index.

BACKGROUND: With use of recombinant erythropoietin (EPO) and intravenous iron, the majority of hemodialysis patients can achieve target hemoglobin concentrations. EPO resistance arises as a consequence of inflammation and other processes that can adversely affect survival. We hypothesized that the EPO dose-hematocrit (EPO/Hct) ratio, also known as the EPO index, may be a surrogate for inflammation and that greater EPO/Hct ratios would be associated with decreased survival. METHODS: We used proportional hazards regression models and time-varying logistic models to analyze the association between EPO index and survival in US hemodialysis patients initiating hemodialysis therapy between January 1, 1999, and December 31, 2000, and followed up for up to 3 years until December 31, 2001. RESULTS: We found an unexpected and consistent association between greater EPO index and survival in all models. The associations of EPO/Hct ratio were most prominent at intermediate Hct values and with longer dialysis vintage. Iron administration was associated with a lower risk for death independent of Hct. Conversely, greater average prior EPO dose was associated with a greater risk for death. CONCLUSION: EPO resistance may be reflected better by total cumulative EPO dose than the EPO/Hct ratio. The mechanism(s) responsible for the association between a greater EPO/Hct ratio and survival remains to be established, but may be a result of nonerythrogenic effects of EPO.

Adult↗

PubMed2002

A biomechanical analysis of four different methods of harvesting bone-patellar tendon-bone graft in porcine knees.

BACKGROUND: Anterior cruciate ligament reconstruction with use of bone-patellar tendon-bone graft is commonly performed to treat functional instability of the knee after injury of the anterior cruciate ligament. Patellar fracture is an underreported complication of this procedure, yet it may be a career-ending injury for a professional athlete. The purpose of this study was to design a new technique to procure the patellar bone plug when reconstructing the cruciate ligament with use of patellar tendon. METHODS: Fifty-eight knees were harvested from eighteen to twenty-four-month-old pigs. Bone plugs of four geometrical shapes were cored out of the patellae from forty knees (four plug shapes for each of ten knees). Ten knees had no graft removed. Fifty knees were tested for the effects of a single impact. Eight knees (two plug shapes for each of four knees) were tested for the effects of repeated impacts. Data were analyzed statistically with use of an ordinal logistic model. The probability of impact failure of the patella at different energy levels was calculated. RESULTS: Impact energy has been shown to have a significant effect on the outcome (p < 0.001). When controlling for energy level, there was a significant difference between the sharp and round corners (p < 0.01). The ordinal logistic model was found to be very good for predicting the outcome of impact testing. The impact energy required to create a 1% probability of complete fracture was 7 J for a patella with a sharp-cornered plug defect, 17 J for a patella with a trapezoidal plug, 22 J for a patella with a sharp-cornered defect with a drill-hole at the corner, 40 J for a patella with a round-cornered defect, and about 49 J for a normal patella. Repeat impact testing substantially damaged the patellae with sharp-cornered defects, as they failed earlier at a lower energy and after a lower number of impacts than did the patellae with round-cornered defects. CONCLUSIONS: We propose the use of the round-cornered patellar bone plug when bone-patellar tendon-bone graft is harvested for anterior cruciate ligament reconstruction, to further reduce the small likelihood of a patellar fracture occurring intraoperatively or in the early postoperative rehabilitation phase.

Animals↗

PubMed2001

Departmental consumption of antibiotic drugs and subsequent resistance: a quantitative link.

OBJECTIVE: To look for a quantitative model linking departmental consumption of antibiotic drugs to the subsequent isolation of resistant hospital-acquired coliform pathogens. MATERIALS AND METHODS: Included in the study were all patients with hospital-acquired bloodstream infections caused by a coliform pathogen, detected in six departments of internal medicine of one university hospital during the period 1991-1996, who had not been hospitalized in the month before the infection (n = 394). Departmental consumption of antibiotics in the year before the infection [expressed as defined daily dosages (DDD)/100 patient days], antibiotic treatment given to the individual patient before the infection, the day of hospital stay on which the infection occurred, and the department and the calendar year were all included in a logistic model to predict the isolation of a resistant pathogen. We looked at five drugs: gentamicin, amikacin, cefuroxime, ceftazidime and ciprofloxacin. RESULTS: Five logistic models were fitted for the resistance to each of the antibiotic drugs. The multivariable-adjusted odds ratios for a pathogen resistant to the specific antibiotic were 1.03 [95% confidence interval (CI) 0.70-1.50] for gentamicin, 1.80 (95% CI 1.00-3.24) for amikacin, 1.12 (95% CI 1.02-1.23) for cefuroxime, 1.45 (95% CI 1.19-1.76) for ceftazidime and 1.06 (95% CI 0.57-1.97) for ciprofloxacin, per 1 DDD/100 patient days. CONCLUSIONS: The departmental consumption of cephalosporin drugs and amikacin in six autonomous departments of medicine in the same hospital was associated with a measurable and statistically significant increase in the probability of infection caused by a resistant pathogen.

Amikacin↗

PubMed2006

Predictors for response to rehabilitation in patients with hip or knee osteoarthritis: a comparison of logistic regression models with three different definitions of responder.

OBJECTIVE: To identify pre-treatment predictors of who will benefit from a 3-4-week comprehensive rehabilitation intervention in patients with osteoarthritis (OA) of the knee or hip. METHODS: A prospective cohort study with assessments at admission to the clinic and after 6 months was conducted. Two hundred and fifty patients from the rehabilitation clinic Rehaclinic Zurzach, Switzerland, were included. Three different measures of response to a 3-4-week comprehensive rehabilitation intervention were used: one indirect measure (minimal clinically important difference (MCID) in the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) global score=18% improvement), one direct measure (transition question) and a combination of both criteria. Responders were predicted by a sequential logistic regression analysis with nine personal variables, five lifestyle risk factors, seven psychological status variables and the WOMAC global baseline score. RESULTS: The set of statistically significant predictors was dependent on the definition of response. The comparison of predictors that were statistically significant in any of the prediction models showed similar odds ratios (ORs) for the majority of predictors across three regression models with the different response definitions as dependent variable. Female gender, absence of depressive symptoms (dep), history of complementary medicine (cm) and low comorbidity (com) were the most stable predictors and had ORs above 2.0 (female) and above 1.5 (dep, cm, com) across the three regression models with different response definitions. CONCLUSION: A set of predictors for the outcome of rehabilitation in patients with OA was identified. If these predictors could be confirmed in future research, this knowledge might help to adopt and individualize the treatment of patients who are, at present, less likely to respond.

Aged↗