PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “Multivariate response regression”

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 73 records · Page 4Linked to original sources
PubMed1991

[Left ventricular filling studied by echo-Doppler in hypertensive and normotensive subjects].

Purpose of the study was to evaluate the influence of different clinical and echocardiographic parameters on left ventricular diastolic filling in a group of 66 mild to moderate hypertensive subjects (mean age = 45.9 +/- 13.9 yrs) and in 49 normotensive controls (mean age = 44.2 +/- 15.9 yrs). All subjects underwent an echocardiographic study with pulsed Doppler evaluation of left ventricular filling. In the hypertensives, a noninvasive 24-hour blood pressure monitoring was performed. The ratio of early to atrial peak diastolic filling velocity (VmaxE/VmaxA) and the ratio between the respective velocity flow integral (Earea/Aarea) were significantly lower in the hypertensives compared to the normotensives (p less than 0.001). As within the hypertensives, office systolic blood pressure, average 24-hour mean blood pressure (24MBP), left atrial dimensions, and left ventricular mass index (MAXIND) were related both to age and diastolic filling. To study the relation between these parameters and diastolic filling a partial correlation test was used. The only variable related to diastolic filling, independently of age, was 24MBP (VmaxE/VmaxA vs 24MBP: r = -0.307, p less than 0.05, Earea/Aarea vs 24MBP: r = -0.261, p less than 0.05). Linear multivariate analysis using the method of stepwise regression established that age, 24MBP and heart rate were responsible for 75% of the VmaxE/VmaxA variance in the hypertensive group. To evaluate if the difference between the indexes of diastolic filling in the normotensives and the hypertensives varied according to age, we divided each group into two classes of subjects older and younger than 40 yrs. Under the age of 40 only VmaxE/VmaxA was significantly different in the two groups (p less than 0.05), while in the subjects older than 40 yrs both VmaxE/VmaxA and Earea/Aarea differed to a great level of statistical significance (p less than 0.001). The results of the present study allow the following conclusions: 1) age has the greatest influence on diastolic filling in both hypertensive and normotensive individuals. Diastolic filling is impaired in mild to moderate hypertension, when systolic function is still normal; 2) the worsening of diastolic filling determined by hypertension is more evident in the subjects greater than 40 yrs; 3) the only clinical variable that influences diastolic filling, independently of age, is 24MBP, while office blood pressure, MAXIND and left ventricular ejection fraction do not seem to be related to this clinical entity.

Adolescent↗
PubMed1984

Prognostic factors in surgically resected limited-stage, nonsmall cell carcinoma of the lung.

The gross pathological, microscopic, and clinical features of 173 Stage I and Stage II primary nonsmall cell carcinomas resected for cure by segmental resection, lobectomy, or pneumonectomy at the Johns Hopkins Hospital were analyzed to determine which provided useful independent prognostic information. The tumors studied included 79 squamous carcinomas (56%); 74 adenocarcinomas (44%), including 15 undifferentiated tumors which contained intracellular mucin; 18 large-cell undifferentiated carcinomas (10%); one giant cell carcinoma (less than 1%); and one adenosquamous carcinoma (less than 1%). Clinical features evaluated for each case included age, sex, race, and history of previous or subsequent malignancy; and pathologic features evaluated included tumor size, lymph node metastases, tumor location, cellular anaplasia, desmoplastic response, inflammatory response, preexisting scar, tumor necrosis, and degree of tumor differentiation. Multivariate analysis using the Cox life table regression model indicated that five features had a significant (p less than 0.05) independent association with subsequent death due to the tumor, and that the final set was highly significant (p = 0.001). These features were the following: large-cell undifferentiated histology, lymph node metastases expressed as N classification, tumor size expressed as T classification, tumor giant cells in any histologic type, and absent or minimal plasma cell infiltration. No additional prognostic information was obtained from any of the other features analyzed.

Adenocarcinoma↗
PubMed1992

Quantitative correlation of mutagenic and carcinogenic potencies for heterocyclic amines from cooked foods and additional aromatic amines.

Aromatic amines have long been recognized as animal and human carcinogens. Recently heterocyclic aromatic amines (thermic amines) have been found in small amounts in cooked foods, primarily meats, and have proven to be potent mutagens and rodent carcinogens. Availability of quantitative databases for mutagenic potency in Salmonella and for carcinogenic potency in rodents has made possible a study of ten heterocyclic thermic amines and 24 aromatic amines. Potencies on mutagenic and carcinogenic scales were significantly correlated. By multiple linear regression analysis and multivariate analysis of variance, two descriptive structural factors were found to modulate the two modes of biological response. These factors were number of rings and methyl substitution at carbon atoms. The quantitative correlation between mutagenic and carcinogenic potencies and the modulating structural factors suggest a significant similarity of molecular mechanisms and support the utility of the short-term bacterial assay in evaluating hazard levels.

Amines↗
PubMed1991

The influence of preoperative patient characteristics on early and late survival following cardiac transplantation.

In an attempt to identify the preoperative factors that influence survival following cardiac transplantation, we retrospectively analyzed multiple demographic, clinical, and hemodynamic data of 301 consecutive patients who underwent cardiac transplantation between December 1980 and July 1988 (cyclosporine era). Univariate and multivariate regression analyses revealed that the two most deleterious risk factors for premature death following transplantation were pulmonary hypertension not responsive to vasodilator challenge and preoperative requirement of hemodynamic support (intravenous inotropes or mechanical assistance). The combination of these two independent risk factors was an even stronger predictor of mortality (relative risk, 6:1; p less than 0.0001): the 3-month actuarial survival rate of the 20 patients with this combination was 30.3% versus 78.4% of the 47 patients with only pulmonary hypertension, 87.9% of the 42 patients with only the requirement of hemodynamic support, and 95.3% of the 172 patients with neither of these two risk factors. The difference in postoperative mortality between patients with versus those without these two risk factors is due to a higher rate of fatal infectious complications: six of 20 patients (30%) with both risk factors died from infection within 3 months after transplant compared with three deaths among 172 patients (1.7%) with neither of these risk factors. The majority of fatal infections were pulmonary. There were significantly more fatal and nonfatal infectious episodes in patients with one or both of these risk factors.(ABSTRACT TRUNCATED AT 250 WORDS)

Adult↗
PubMed1991

Age, cardiovascular risk factors and coronary heart disease as determinants of platelet function in men. A multivariate approach.

Older age, the cardiovascular risk factors and arteriosclerosis have been reported to be associated with stimulated platelet function. To evaluate the relative importance of these factors in determining platelet function, a cross-sectional multivariate study in 191 men, 113 healthy subjects and 78 patients with angiographically documented coronary heart disease, was performed. In healthy subjects, stepwise multiple linear regression identified age to be a major determinant of platelet aggregability. After induction with both ADP and collagen the platelet aggregatory response markedly increased with age. In the patients, platelet function was not age dependent. In multivariate analysis of variance, neither smoking status nor hypercholesterolemia (greater than or equal to 240 mg/dl) were determinants of platelet function in either group. An increase in systolic blood pressure was associated with slightly more inhibited ADP induced aggregation in both healthy subjects and patients with coronary heart disease. In patients compared to healthy subjects, aggregation after induction with ADP and collagen was markedly enhanced and the in vitro formation of thromboxane after collagen stimulation increased. Thus, by multivariate analysis, age and the presence or absence of coronary heart disease were found to be major determinants of platelet function. In contrast, the cardiovascular risk factors smoking, hypercholesterolemia and hypertension were associated with only minor or no alterations of platelet function.

Adult↗
PubMed1991

The prognostic importance of site and type of radiation-induced bowel injury in patients requiring surgical management.

A multivariate analysis was performed to determine the outcome, and factors prognostic for outcome, in 57 patients requiring surgical intervention for radiation bowel injury. The actuarial 2- and 5-year cause-specific survival (CSS) was 76 and 74%, respectively, with a median follow-up of 62 months for the survivors. The median time from surgery to death from complications was 4 months. Identified sites of injury were both large and small bowel. The types of injury were defined as stricture, perforation, inflammation, and fistula. At surgery 9 patients had more than one site, and 15 patients had more than one type of injury. Cox proportional hazards regression models relating survival to individual patient characteristics were constructed using surgical procedure, radiation-surgery interval, age, stage, radiotherapy technique and dose, and the individual sites and types of injuries. Only the site of injury was found to be of prognostic significance for CSS (P less than 0.03). However, when the site and type of injury were recoded as single or multiple, Cox regression analysis found both the site (P = 0.008) and the type (P = 0.02) of injury to be statistically significant for CSS (favoring single sites and types). Stepwise multivariate regression analysis found type of injury to be insignificant when site of injury was already in the model. Ileal damage was associated with the lowest CSS of any single site of injury (56%) and also appeared to be responsible for the poor CSS of those with multiple sites of injury (46%).

Adult↗
PubMed1983

Incremental value of clinical assessment, supine exercise electrocardiography, and biplane exercise radionuclide ventriculography in the prediction of coronary artery disease in men with chest pain.

The incremental value of clinical assessment, exercise electrocardiography (ECG) and biplane radionuclide ventriculography (RVG) in the prediction of coronary artery disease (CAD) was assessed in 105 men without myocardial infarction who were undergoing coronary angiography for investigation of chest pain. Independent clinical assessment of chest pain was made prospectively by 2 physicians. Graded supine bicycle exercise testing was symptom-limited. Right anterior oblique ECG-gated first-pass RVG and left anterior oblique ECG-gated equilibrium RVG were performed at rest and exercise. Regional wall motion abnormalities were defined by agreement of 2 of 3 blinded observers. A combined strongly positive exercise ECG response was defined as greater than or equal to 2 mm ST depression or 1.0 to 1.9 mm ST depression with exercise-induced chest pain. A multivariate logistic regression model for the preexercise prediction of CAD was derived from the clinical data and selected 2 variables: chest pain class and cholesterol level. A second model assessed the incremental value of the exercise test in prediction of CAD and found 2 exercise variables that improved prediction: RVG wall motion abnormalities, and a combined strongly positive ECG response. Applying the derived predictive models, 37 of the 58 patients (64%) with preexercise probabilities of 10 to 90% crossed either below the 10% probability threshold or above the 90% threshold and 28 (48%) also moved across the 5 and 95% thresholds. Supine exercise testing with ECG and biplane RVG together, but neither test alone, effectively adds to clinical prediction of CAD. It is most useful in men with atypical chest pain and when the ECG and RVG results are concordant.

Adult↗
PubMed1991

A multivariate analysis of risk factors for anterior atlantoaxial subluxation and an evaluation of the effect of glucocorticoid treatment on the upper rheumatoid cervical spine.

A cross-sectional, retrospective computerized analysis of risk factors for anterior atlantoaxial subluxation (AAS) was performed. Logistic regression performed on the clinical variables involved in 145 cases of rheumatoid arthritis (RA) disclosed a high joint score index and a low blood haemoglobin level as significant independent risk factors. This means that the development of anterior AAS is connected with widespread RA. Linear multiple regression analysis showed an association between the extent of anterior AAS in millimetres and the spread of erosions of the dens of axis and negative correlation with the severity of vertical atlantoaxial dislocation (VD). This suggests that whereas the presence of anterior AAS is connected with the severity of the systemic disease, its actual extent is associated with signs of local involvement, other than severe VD. The duration or cumulative dosage of glucocorticoids were not associated with the development or extent of anterior AAS, nor with the severity of vertical dislocation. This suggests that low dose glucocorticoid treatment is not involved with the development of rheumatoid changes in the upper cervical spine. It should be borne in mind, of course, that although no correlation was found, a causal relation cannot be excluded.

Arthritis, Rheumatoid↗
PubMed1992

Development and validation of prognostic models in metastatic breast cancer: a GOCS study.

The significance of several prognostic factors and the magnitude of their influence on response rate and survival were assessed by means of uni- and multivariate analyses in 362 patients with stage IV (UICC) breast carcinoma receiving combination chemotherapy as first systemic treatment over an 8-year period. Univariate analyses identified performance status and prior adjuvant radiotherapy as predictors of objective regression (OR), whereas the performance status, prior chemotherapy and radiotherapy (adjuvants), white blood cells count, SGOT and SGPT levels, and metastatic pattern were significantly correlated to survival. In multivariate analyses favorable characteristics associated to OR were prior adjuvant radiotherapy, no prior chemotherapy and postmenopausal status. Regarding survival, the performance status and visceral involvement were selected by the Cox model. The predictive accuracy of the logistic and the proportional hazards models was retrospectively tested in the training sample, and prospectively in a new population of 126 patients also receiving combined chemotherapy as first treatment for metastatic breast cancer. A certain overfitting to data in the training sample was observed with the regression model for response. However, the discriminative ability of the Cox model for survival was clearly confirmed.

Analysis of Variance↗
PubMed1991

Hormonal content of combined oral contraceptives in relation to the reduced risk of endometrial carcinoma. The WHO Collaborative Study of Neoplasia and Steroid Contraceptives.

The relationship between the strength of the estrogenic and progestational components of combined oral contraceptives and risk of endometrial carcinoma was examined in a multinational hospital-based case-control study comparing 220 cases from 7 countries with 1,537 age- and hospital-matched controls. Oral contraceptives were classified according to their relative content of estrogen and progestin. The risk was not altered in women who used preparations containing high-dose estrogen/low-dose progestin compounds [odds ratio (OR) = 1.10, 95% CI = 0.13-9.96]. In contrast, the risk among users of low-dose estrogen/high-dose progestin oral contraceptives was greatly decreased (OR = 0, 95% CI = 0-1.08). Risks observed for users of high-dose estrogen/high-dose progestin (OR = 0.15, 95% CI = 0.045-0.50) and low-dose estrogen/low-dose progesterone (OR = 0.59, 95% CI = 0.26-1.30) preparations were intermediate between those for users of the 2 other types of preparation. A significantly lower risk was observed for high-dose progestin users than for low-dose users [ratio of odds ratios (ROR) = 0.21, 95% CI = 0.05-0.84]. Although the numbers are small, findings from this study suggest that combined oral contraceptives with varying strengths of estrogen and progestin have different effects on the risk of endometrial carcinoma.

Case-Control Studies↗
PubMed1992

Relationships between carboplatin exposure and tumor response and toxicity in patients with ovarian cancer.

PURPOSE: The study was undertaken to define the relationship between tumor response and carboplatin area under the curve (AUC) in patients with ovarian cancer; to study the relationship between carboplatin AUC and myelosuppression in the same population; to establish the true impact of carboplatin AUC, prior therapy, and pretreatment platelet and WBC counts on toxicity; and to define an optimal carboplatin exposure for treating patients with ovarian cancer. METHODS: With the equation AUC = dose/(glomerular filtration rate [GFR]+25), carboplatin AUC (course 1) was calculated for 1,028 patients (450 previously untreated) who received single-agent carboplatin (40 to 1,000 mg/m2) for advanced ovarian cancer. GFR was measured (chromium-51-edathamil [51Cr-EDTA] or creatinine clearance) in all patients. RESULTS: Regression analysis showed that carboplatin AUC, prior treatment, and Eastern Cooperative Oncology Group grade performance status (PS) are predictors of tumor response, thrombocytopenia, and leukopenia. Pretreatment platelet and WBC counts are additional predictors of thrombocytopenia and leukopenia, respectively. Although the likelihood of tumor response increased with increasing carboplatin AUC, this relationship was nonlinear. In all patient subsets, the likelihood of complete response (CR) or overall response did not increase significantly above a carboplatin AUC of 5 to 7 mg/mL x minutes. At any given carboplatin AUC, thrombocytopenia occurred more frequently than leukopenia, although both approached 100% as carboplatin AUC increased. Both thrombocytopenia and leukopenia were more frequent in pretreated than in untreated patients regardless of pretreatment count. At any carboplatin AUC, the influence of PS on likelihood of response and toxicity was profound. CONCLUSION: Carboplatin dosing by AUC will lead to more predictable toxicity, and increasing carboplatin AUC above 5 to 7 mg/mL x minutes does not improve the likelihood of response but does increase myelotoxicity. Therefore, careful evaluation of high-dose carboplatin therapy in a prospective, randomized trial is needed before such treatment becomes accepted practice.

Aged↗
PubMed1990

[Measuring the vascular caliber in diabetic retinopathy before and following photocoagulation].

The calibers of 4 retinal arterioles and 4 venoles of each of 58 patients with diabetic retinopathy (DR) were measured on fundus photographs before and after photocoagulation (PC). The measurements were repeated at approximately one-year intervals following photocoagulation (PC) of one eye of each patient. The mean caliber of arterioles and venoles of DR patients was 15.4% greater than in nondiabetic subjects. This finding may be interpreted as an autoregulatory response to relative tissue hypoxia. Following PC there was a significant reduction in both arterial and venous vessel caliber. In a multivariate regression analysis, the reduction in venous caliber correlated significantly with the number of coagulation spots. A weak correlation was found with arterial hypertension, change in visual acuity, and change in a calculated DR score. The reduction in pathologic vessel calibers following PC is indicative of an improvement in retinal hemodynamics. Photographic quantification of retinal vessel changes may be of prognostic value, i.e., for assessing the course of DR before and after treatment by PC.

Diabetic Retinopathy↗
PubMed1991

Multivariate analysis of metastasis risk in laryngeal carcinoma. II. Immune response.

The presence of inflammatory reaction, plasma cells, and eosinophils in peritumoral connective tissue and in neoplastic stroma was evaluated with morphometrical method in 181 patients affected by laryngeal carcinoma. A logistic multiple regression model was applied making it with the use of an independent variable represented by the "infiltrating" or "expansive" types of tumor growth, in order to evaluate the probability of nodal metastatsis of each parameter. The results suggest an inverse correlationship between plasma cells and inflammatory infiltration and incidence of nodal metastatsis only in the comparison of the extreme conditions: those with scarce infiltration versus the ones with large infiltration. Inflammatory or plasmacellular infiltration may represent both a defense mechanism against cancer and an aspecific or allergic reaction. The eosinophilic infiltration shows no value in the prevention of nodal involvement.

Eosinophils↗
PubMed1991

Estimating equations for parameters in means and covariances of multivariate discrete and continuous responses.

Generalized estimating equations are introduced in an ad hoc fashion for the covariance matrix of a multivariate response. These equations are to be solved jointly with score equations from a generalized linear model for mean parameters. A class of quadratic exponential models is used to develop joint estimating equations for mean and covariance parameters in a more systematic fashion, and proposals for the use of such equations are developed. Comments on the relative merits of the ad hoc and model-based approaches to estimation are given and a regression illustration with a bivariate response is provided.

Anticonvulsants↗
PubMed1992

The use of cusums and other techniques in modelling continuous covariates in logistic regression.

The assessment of continuous covariates singly as possible predictors in a multivariable logistic regression model is an important first step in the analysis. An approach to plotting which uses a cusum (cumulative sum) of the binary response variable is described. Extreme-deviation statistics associated with the cusum may be used to detect monotonic and non-monotonic trends. Probability plots of the covariate in the two groups defined by the response variable may help to determine the appropriate scale (transformation) of the covariate and to anticipate possible problems with the logistic fit. The ratio of the variances in the response/non-response groups is informative about the need for a quadratic term in the logistic model. Smoothed scatterplots of the response are valuable in displaying the observed and fitted values. The techniques are illustrated with two data sets.

Binomial Distribution↗
PubMed1992

Blocked arteries and multivariate regression.

Ultrasound blood flow waveforms may be used in the diagnosis of arterial occlusive disease in human legs. We develop a statistical model to predict disease severity, conditional on the ultrasound data and some training data. It belongs to the class of models known as seemingly unrelated regressions, for which the Bayesian predictive density function cannot be evaluated analytically. Allowing for missing components of response vectors in the training data, we describe a first-order approximation to the predictive density, based on a Bayes estimate of the precision matrix. This approximation is then used to generate cross-validated predictions of disease severity in a set of 31 patients. We conclude with a discussion of the results.

Arterial Occlusive Diseases↗
PubMed1987

Predicting the pharmacodynamics of heparin: a clinical evaluation of the Hepcon System 4.

The magnitude of the anticoagulation response to heparin (heparin responsiveness) varies substantially from patient to patient. Identifying extremes of sensitivity and resistance prior to intravenous administration of heparin would facilitate anticoagulation for cardiopulmonary bypass (CPB). The performance of the Hepcon System 4 (HemoTec, Inc, Englewood, CO), an instrument designed for that purpose, was tested. Using nonheparinized blood samples from 157 patients scheduled for surgery requiring CPB, this device performed activated coagulation times (ACT) with three different concentrations of in vitro heparin. After determining each patient's in vitro heparin response, the heparin dose predicted to produce ACT values of 480 seconds (group 1, N = 77) or 600 seconds (group 2, N = 80) was administered. Five minutes later each patient's ACT was determined with the Hemochron method (International Technidyne, Inc, Edison, NJ). Simultaneously, several other variables that might predict heparin responsiveness were investigated. When compared with the observed ACT, the Hepcon System 4 inadequately predicted the response. There was considerable scatter in this comparison, but most frequently the in vitro method substantially underestimated the in vivo heparin dose requirement. Heparin responsiveness decreased significantly with high platelet counts and advanced age, but was unaffected by the initial hematocrit, ACT, partial thromboplastin time, or preoperative heparin therapy. Previous investigations have not identified a relationship between advanced age and reduced heparin responsiveness. Combining the Hepcon heparin dose-response in vitro method with the other parameters evaluated, stepwise regression could only account for 39% of the observed variability in heparin responsiveness.(ABSTRACT TRUNCATED AT 250 WORDS)

Age Factors↗