SEARCH · PubMed Health

Results for “Multivariate Analysis”

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

PubMedDate not supplied

Papillary thyroid carcinoma: a multivariate analysis of prognostic factors including an evaluation of the p-TNM staging system.

OBJECTIVE: To analyse the prognostic factors in papillary thyroid carcinoma, and in particular to evaluate the accuracy of the pathological tumour, nodes, metastases (p-TNM) staging. DESIGN: Retrospective univariate and multivariate analysis. SETTING: University hospital in Norway. SUBJECTS: 167 patients who were operated on for papillary thyroid carcinoma between 1971 and 1985. Main outcome measures--Death of papillary thyroid carcinoma, and length of recurrence free survival. RESULTS: Male sex, increasing age, larger tumours, and spread of growth beyond the thyroid all independently increased the risk of dying of papillary thyroid carcinoma, whereas the period of recurrence free survival was influenced only by the presence of regional metastases and the patient's age. The age related p-TNM staging is suitable for predicting the likelihood of death, but is less accurate in the prediction of recurrence free survival. The age of 45 years is too low to be useful in predicting survival, especially in women. CONCLUSION: The identification of sex in the multivariate analysis as a strong independent predictor of death of papillary thyroid cancer suggests that the prognostic value of the age related p-TNM staging system could be improved if sex was adjusted for, and if a different age was used for men and women.

Adult↗

PubMed1991

Radiotherapy in T1 laryngeal cancer: prognostic factors for locoregional control and survival, uni- and multivariate analysis.

From 1975 through 1985, 194 patients with T1 glottic, 37 patients with T1 supraglottic, and 3 patients with T1 subglottic cancer were treated with radiotherapy. Local control and ultimate locoregional control (after salvage surgery) was 91% and 97% for T1 glottic, 84% and 81% for T1 supraglottic, and 2/3 and 3/3, respectively for subglottic tumors. In uni- and multivariate analysis local control for glottic tumors was associated with extension of the tumor on the vocal cord (entire length of vocal cord vs others, p = 0.01) and continuation of smoking after therapy (yes/no, p = 0.03). No prognostic factor for local control was found in supraglottic tumors. However, regional control and survival were impaired by N stage (N0 vs N+, p less than 0.0005), local recurrence (yes/no, p less than 0.0005), and extension of the tumor (one supraglottic subsite vs more than one, p less than 0.05). Mild late complications were seen in 13% of patients without salvage therapy. Following univariate analysis, field size, fraction size (greater than 2 Gy), maximum tumor dose (greater than 70 Gy), age, post-treatment biopsy, and tumor site were associated with complication rate. Following multivariate analysis, site, fraction size, maximum tumor dose, and continuation of smoking after therapy were independent prognostic factors for mild late complications (mostly arytenoid edema).

Adult↗

PubMedDate not supplied

Impact of neck dissection on survival in well-differentiated thyroid cancer: a multivariate analysis of 218 cases.

The management of cervical lymph node metastases in well-differentiated carcinoma of the thyroid has been highly controversial. In the Department of Surgery (II), Kanazawa University Hospital, the surgical management of cervical lymph node metastases in curable well-differentiated carcinoma of the thyroid has been changed from a conservative approach to an aggressive one since 1973. In order to determine whether an aggressive approach for cervical lymph node metastases is adequate, a retrospective multivariate analysis was carried out of 218 cases of well-differentiated thyroid cancer. The patients have been followed up from 5 to 30 years. Multivariate analysis was conducted following Cox's model. As for the results, the aggressive management of cervical lymph node metastases appeared to have an impact on survival. Furthermore, age and sex were confirmed to be important prognostic factors and a partial lobectomy was confirmed to be inadequate as a type of thyroidectomy.

Adenocarcinoma↗

PubMed1989

[Effective evaluation of combination chemotherapy in stage I-II of non-Hodgkin's lymphoma--a comparison of treatment with multivariate analysis].

Sixty-one patients with non-Hodgkin's lymphoma (Stage I and II) were treated solely with radiotherapy or in combination with chemotherapy. A multivariate analysis was used to identify the prognostic factors that significantly affected the treatment outcome. The variables examined included: stage of disease, according to the Ann Arbor classification, sex, age, pathological criteria, according to the LSG, the presence or absence of systemic symptoms, the sites of disease involvement, the LDH, the bulk of the disease, and the combination chemotherapy. Variables achieving a p-value in the 0 to 0.10 range with a multivariate analysis for the overall survival were the stage of disease, the sites of the disease involvement, the bulk of the disease, and the combination chemotherapy. Finally, the overall survival was affected by bulk of the disease (p less than 0.01) and the combination chemotherapy (p less than 0.01). The 5-year survival rates were 84.4% for stage I and 50.7% for stage II. The survival rates were 47.5% for radiotherapy treatment alone and 75.3% for combination chemotherapy. It thus was concluded that combination chemotherapy was effective for treatment of stages I and II of non-Hodgkin's lymphoma.

Adolescent↗

PubMed1992

Selection of patients for programmed ventricular stimulation: a clinical decision-making model based on multivariate analysis of clinical variables.

OBJECTIVE: This study was conducted to assess the utility of clinical variables in predicting the inducibility of sustained ventricular arrhythmias in a heterogeneous group of patients undergoing programmed ventricular stimulation. METHODS: Variables were considered in a simulated chronologic order to determine the incremental information added by the signal-averaged electrocardiogram (ECG) and left ventricular ejection fraction. All patients undergoing baseline programmed ventricular stimulation for induction of ventricular tachyarrhythmia during a 30-month period were included in the study. Fourteen historical, ECG, signal-averaged ECG and left ventricular wall motion variables were evaluated for their ability in predicting inducibility of a sustained ventricular arrhythmia, a "positive" event, at programmed ventricular stimulation. RESULTS: On univariate analysis of the clinical variables, comparison between patients with positive or negative results showed significant differences in 10 of the 14 clinical variables: major cardiac diagnosis, history of ventricular tachycardia, myocardial infarction by history or ECG, all five signal-averaged ECG variables, left ventricular ejection fraction and presence of left ventricular aneurysm. On multivariate analysis, five independent variables were determined to be important: history of ventricular tachycardia, historical or ECG evidence of myocardial infarction, history of loss of consciousness, filtered QRS duration on the signal-averaged ECG and left ventricular ejection fraction. However, with sequential multivariate analysis, a model based only on historical and conventional ECG data was found to do as well as a model that included signal-averaged ECG and left ventricular ejection fraction data. CONCLUSIONS: Routinely available noninvasive historical, ECG, signal-averaged ECG and left ventricular wall motion variables can be used to accurately predict the outcome of programmed ventricular stimulation. The majority of the predictive power was obtained with the routine model, using only historical and ECG data. The signal-averaged ECG and left ventricular wall motion analysis added no significant incremental information.

Aged↗

PubMed1988

[Gastric adenocarcinoma surgically treated. A multivariate analysis of prognostic factors].

One hundred and twelve curatively resected gastric adenocarcinomas were studied retrospectively to appreciate the survival factors. Twenty different criteria (clinic, histologic and therapeutic parameters) were assessed using univariate and then multivariate analysis (semi parametric regression (COX's) model). Only three criteria were very important according to the multivariate analysis: 1) invasion of neighboring organs (p less than 0.006) with a relative risk score (RRS) of 4.26; 2) intravascular or intralymphatic tumor embols outside the tumor (p less than 0.004; RRS = 2.11); 3) invaded distal nodes (located at the origin of the vessels (p less than 0.04; RRS = 1.88). A prognosis index was described according to these results. A repartition of the patients in three prognostic groups according to these 3 criteria was proposed. Future, trials should consider these three different prognostic groups.

Adenocarcinoma↗

PubMed1992

[A multivariate analysis of prognostic factors of spinocerebellar degenerations].

To clarify the factors relating to the prognosis of spinocerebellar degenerations (SCD), we performed a follow-up study on the survival of patients with Friedreich disease, familial spastic paraparasis, sporadic olivopontocerebellar atrophy (OPCA), hereditary OPCA of Menzel type, sporadic late cortical cerebellar atrophy (LCCA), cerebellar atrophy of Holmes type, Shy-Drager syndrome, striatonigral degeneration, dentatorubropallidoluysian atrophy, or Joseph disease. One hundred and forty-eight patients admitted to the Nagoya University Hospital during the period of 1976 to 1990 were dealt with. They had been followed-up for 15 years at longest through the medical records, or through the direct informations from the patients or their family members or both. In order to find factors influencing the survival of the patients, Kaplan-Meier method was used as the 1st step to construct the survival curve for each factor. These were calculated by generalized Wilcoxon test. Employing Cox's proportional hazard model, a multivariate analysis was then performed based on the factors which were shown significant in the 1st step. The multivariate analysis showed that the following four factors are related to the prognosis of SCD patients, i.e., lack of hereditary trait, existence of orthostatic hypotension, lack of walking disorder at onset, and lack of nystagmus on admission. Of these factors, the lack of hereditary trait had the most intimate relation to their poor prognosis. Therefore, these four factors are statistically informative to predict prognosis of each patient with SCD.

Adolescent↗

PubMed1992

Multivariate analysis of the histopathologic prognostic factors of cervical cancer in patients undergoing radical hysterectomy.

Three hundred forty-five patients with invasive carcinoma of the uterine cervix, Stages Ib (211 patients) and II (134 patients), underwent radical hysterectomy and pelvic lymphadenectomy. The influence of histologic factors including histologic subtype, maximum depth of cervical stromal invasion, degree of stromal invasion, longitudinal tumor diameter, lymph-vascular space invasion, corpus invasion, parametrial invasion, vaginal invasion, and pelvic lymph node (PLN) metastases on survival were examined by multivariate analysis. Univariate analysis revealed that all the variables except corpus invasion and vaginal invasion were significant in survival (P less than 0.05). Among these variables, however, PLN metastases, histologic subtype, and longitudinal tumor diameter were identified as independent and significant prognostic factors by multivariate analysis using Cox regression models. The prognostic index (PI), defined by the model (an indicator of the patient's place in the prognostic spectrum), was able to divide the patients into three prognostic groups. The key factors in the definition of these groups were (1) squamous cell carcinoma, small tumor diameter, and no PLN metastases for the good prognostic group and (2) PLN metastasis in two or more node groups, adenocarcinoma with one positive PLN group, or squamous cell carcinoma with one PLN group and large diameter for the poor prognostic group. These prognostic findings could predict the prognosis more precisely than that of clinical staging.

Adult↗

PubMed1987

Multivariate analysis of lipoprotein cholesterol fractions.

Intervention and prevention of multifactorial diseases such as coronary heart disease can be effective only when the joint effects of multiple risk factors are known. This process is facilitated by multivariate analysis of correlated risk factors, such as the serum cholesterol fractions, high density lipoprotein (HDL), low density lipoprotein (LDL), and very low density lipoprotein (VLDL). Whereas evidence for genetic covariation provides focus for further refined biochemical analysis, covariation among environmental factors can point to efficacious intervention strategies. To assess sources of variation and covariation among HDL, LDL, and VLDL, a multivariate path model was developed and applied to family data. Phenotypic variance is due primarily to specific environmental influences with substantial genetic influences, with the common family environment contributing less than 10% of the variance. There are genetic correlations of -0.22 for HDL-VLDL and 0.35 for VLDL-LDL, consistent with the known inverse associations of HDL and VLDL and the precursor-product relationship between VLDL and LDL, whereas there is no evidence for a direct HDL-LDL genetic relationship. Strong specific environmental correlations are found between HDL and VLDL (-0.35 in children and -0.50 in adults). Thus, intervention focused primarily on one fraction (e.g., triglycerides and VLDL) might beneficially affect levels of both lipoproteins (e.g., lowering VLDL cholesterol and elevating HDL cholesterol). Multivariate analysis can facilitate understanding of the linked effects of intervention on lipoprotein cholesterols, and, hence, should benefit approaches to maximize the effects of lipoprotein cholesterol intervention on coronary heart disease morbidity and mortality.

Analysis of Variance↗

PubMed1989

Craniometrical estimation of the native Japanese Mishima cattle, using multivariate analysis.

The present study on measurement of the skull of Mishima cattle, which has been postulated as the only pure representative breed of native Japanese cattle, was performed using craniometrical multivariate analysis. The data of the skull of Mishima cattle was compared with 17 breeds of cattle, i.e. Korean cattle (Hamhung, Pyongyang, Chinju Suwon, and Kwangju), Mongolian cattle, Hainan Tao cattle, northeastern Chinese cattle (Shuangliao, Shenyang, Tongliao, Lüta, and Chilin), Astatic Water Buffalo, Yak, Bos Banteng, American Bison, and Holstein-Friesian. The Mishima cattle was included in the group of Korean breeds, especially it was closed on the group of Pyongyang and Chinju breeds. The distance on the craniometrical multivariate analyzing co-ordinate between Mishima cattle and Hainan Tao breed of Zebu cattle was larger than the distance between Mishima cattle and Korean breeds. While result, as a above the present study was very important for the origin of "Wagyu" (native Japanese cattle). Since the northern route theory of the origin of Mishima cattle has been reported on the type of serum enzymes and hemotypes. It was suggested that the craniometrical multivariate analysis supported to the northern route theory of the origin of Mishima cattle.

Animals↗

PubMed1990

[Therapeutic factors related to irradiation in primary and metastatic liver cancer using multivariate analysis].

Between December 1973 and August 1987, 21 patients with primary liver cancer and 41 patients with metastatic liver cancer were treated with external irradiation, intra-arterial infusion chemotherapy and/or trans-arterial embolization (TAE) at the National Medical Center Hospital, the National South Kyushu Central Hospital and the National Kure Hospital. They were all inoperable cases. We diagnosed the tumor site and the involved area with many imagings and we decided the target volume. 1. Primary liver cancer: the average survival period was 10.9 months, the 1-year survival rate was 28.6%, the 2-year survival rate was 14.3% and the 3-year survival rate was 4.7%. Using multivariate analysis, stage, cancer reduction rate, Child classification and field size were valuable factors of the prognosis in the arterial infusion group. In the TAE group, cancer reduction rate was the most valuable factor. 2. Metastatic liver cancer: The average survival period was 8.0 months, the 1-year survival rate was 19.5%, the 5-year survival rate was 2.4%. Using multivariate analysis, pre-treatment K.P.S, distant metastasis, H, Child classification were valuable factors and pre-treatment K.P.S. was most valuable factor. Although the indication of hepatic irradiation were generally thought to limit those cases which were Child A or B, H1 or H2 and the cases which have no distant metastasis, the case which have their main portal tumor thrombus were also the indication of this therapy.

Aged↗

PubMed1986

Is the EORTC prognostic index of thyroid cancer valid in differentiated thyroid carcinoma? Retrospective multivariate analysis of differentiated thyroid carcinoma with long follow-up.

The European Organization for Research on Treatment of Cancer (EORTC) Thyroid Cancer Cooperative Group presented a prognostic index in 1979 that included all histologic groups of thyroid carcinomas, and was based on a multivariate analysis of 507 patients with a median follow-up of 40 months. The current report not only uses a multivariate analysis to study the clinical validity and reproducibility of this index on case material consisting of 226 differentiated thyroid carcinomas with a considerably long follow-up (11 years), but also it investigates possible prognostic factors, other than those proposed by the EORTC. Three EORTC variables could be reproduced as important: age at diagnosis, locally advanced disease, and distant metastases. Of the additional histopathologic variables tested, microscopic tumor invasion beyond thyroid capsule (pT4) and marked cellular atypia (MCA) proved to be significant. When the effect of the age-correlated tumor factors pT4 and MCA of differentiated thyroid carcinomas were taken into consideration, age alone lost its prognostic importance as a guide for treatment.

Adult↗

PubMed1992

A population-based, multivariate analysis of the association between 911 access and per-capita county trauma death rates.

STUDY OBJECTIVE: Decreased response times for injury should decrease the morbidity and mortality of trauma. The 911 telephone access is designed to improve the response time for emergencies. The purpose of this study was to analyze the association between county 911 access and per-capita county trauma death rates. METHODS: Data on all trauma deaths from 1986 through 1988 were obtained from the North Carolina Medical Examiner's data base. Counties were divided into those that had 911 access during the entire study period (15), those that never had 911 access (62), and those that installed 911 during 1987 (ten). Counties obtaining 911 access in 1986 or 1988 were excluded (13). RESULTS: The per-capita trauma death rate in counties that had 911 access throughout the study was 4.3 +/- 0.8 versus 5.0 +/- 1.1 per 10,000 population in counties that never had 911 access (P less than .01). Compared with counties with 911 access, counties without 911 were more rural, were less likely to have a trauma center, and were less likely to have advanced life support certification (P less than .03 for all). Controlling for these other factors, multivariate analysis demonstrated that 911 access had no significant independent association with per-capita county trauma death rates. In the ten counties that implemented 911 access in 1987, no significant change occurred in per-capita county trauma death rates after implementation of 911. CONCLUSION: Although counties with 911 access had lower trauma death rates by t-test, multivariate analysis showed no significant independent association of 911 access with per-capita county trauma death rates. In the ten counties that implemented 911 access in 1987, no significant changes in trauma death rates occurred after implementation. Although other factors may explain these findings, this study showed no significant independent impact of 911 access on per-capita county trauma death rates.

Emergency Medical Service Communication Systems↗

PubMed1987

Multivariate analysis of karyotypic abnormality in leukemia facilitated by numerical encoding of cytogenetic data.

A system is proposed whereby human karyotype data is expressed in quantitative terms rather than in the ISCN (1985) terminology used at present. This recoding facilitates application of multivariate analysis using standard statistical packages. As an example, karyotypes of 714 cases from 11 leukemias (Mitelman, 1983) are here recorded and subjected to discriminant analysis (SPSSx, 1983). Significant karyotypic specificity is apparent in six of the 11 FAB leukemia types. Four others show insignificant levels of specificity, while the last is equivocal. These results merely confirm present views. However, their generation by means of computerized multivariate analysis is novel, and confirms the feasibility of the approach. In this quantitative form, karyotypic data may be combined with any other data of diagnostic or prognostic value. Given such a consolidated data set, desired information concerning any aspect of neoplasia could be extracted via a single procedure.

Acute Disease↗

PubMed1989

Chronic myelomonocytic leukemia--clinicobiological characteristics: a multivariate analysis in a series of 70 cases.

In a series of 70 patients diagnosed according to the FAB criteria, 42 clinical and biological disease characteristics were analyzed in order to identify significant prognostic factors by means of univariate and multivariate analysis. The univariate analysis identified ten parameters associated with poor prognosis: Symptoms of anemia, WBC over 10 x 10(9)/l, presence of blast cells, myeloid precursors or erythroblasts in peripheral blood (PB), high bone marrow (BM) cellularity, severe dysthrombopoiesis, percent of blast cells in BM and high serum levels of bilirubin and LDH. The Cox proportional hazards regression method revealed that the combination of high leukocyte counts and BM percentage of blast cells had the strongest predictive relation to survival length (p = 0.002 and p = 0.060 respectively). A new multivariate analysis treating the presence of myeloid and erythroid precursors in PB as a single variable selected only this as the most significant prognostic factor (p = 0.001). Both regression models allowed us to discriminate two significantly different risk groups of patients.

Actuarial Analysis↗

PubMed1989

Non-seminomatous testicular cancer clinical stage I: prediction of outcome by histopathological parameters. A multivariate analysis.

The clinical course of 186 orchiectomized patients with testicular non-seminoma clinical stage I (CSI) was reviewed together with a reevaluation of the haematoxylin-eosin-stained histological sections of the primary tumour. Treatment (1970-1980) was as follows: abdominal radiotherapy (group 1): 132 patients; observation only (group 2): 23 patients; retroperitoneal lymph node dissection (group 3): 31 patients. Failure of the management (diagnostic error/relapse after treatment) was defined as demonstration of retroperitoneal lymph node metastases (group 3) or relapse during follow-up (group 2 and 3). Ten of the 31 primarily operated patients had retroperitoneal lymph node metastases. (These patients received 3-4 cycles adjuvant cisplatin based on chemotherapy.) Forty-six patients relapsed after a median time of 6 months (range: 2-113). The 10-year cancer-related survival rates for group 1 and 2 were 85 and 73%, respectively. No cancer-related death occurred in group 3 within the first 5 years. In a univariate analysis the following parameters were significantly (p less than 0.05) correlated with management failure: vascular invasion (blood, lymphatic), the presence of syncytiotrophoblasts, the demonstration of the histological subtype MTU. In a multivariate analysis both lymphatic and blood vessel invasion significantly predicted management failure. In 80% of the non-seminoma patients with CSI and both lymphatic and blood vessels invasion in the primary tumour failure of the primary management must be expected. In the multivariate analysis the demonstration of MTU was a predictive factor only if no differentiation was made between lymphatic and blood vessel invasion. The presence or absence of endodermal sinus tumour was not correlated with the failure rate in non-seminoma CSI. Based on easily assessable histopathological parameters of the primary tumour (small vessel invasion in particular) it is possible to define subgroups of patients with non-seminoma CSI who have a high or a low risk of relapse or who are likely to have microscopic retroperitoneal lymph node metastases. Such a differentiation should have therapeutic consequences.

Humans↗

PubMed1988

Prognostic value of CEA and ferritin assay in breast cancer: a multivariate analysis.

The prognostic significance of preoperative serum carcinoembryonic antigen (CEA) and ferritin levels was evaluated in 191 women operated for breast cancer. The influence of CEA, ferritin and another 11 clinical and pathological features on the disease-free survival was investigated in a multivariate analysis, using Cox's proportional hazard model. Axillary node status (P = 0.004), CEA level (P = 0.011), and the histological grade of the tumor (P = 0.029) emerged as independent prognostic factors. By contrast, no significant relationship was found between ferritin and disease-free survival. These three parameters were used to derive a prognostic index (I) for each patient. Multivariate analysis showed that its prognostic value was better than the value of any single factor (P less than 0.0001). The I score was used to divide patients into groups at different risk of recurrence: low, moderate and high (97.5%, 45% and 22.5% of recurrence-free patients at 3 years respectively). The data showed that the prognosis of patients with different combinations of node status and tumor grade was related to the level of CEA. Only women with very good (node-negative with well-differentiated tumors) or very bad prognosis (node-positive with four or more metastatic nodes and poorly differentiated tumors) had a disease-free survival independent of CEA values. These findings suggest that the preoperative measurement of CEA enhances the possibility of correctly predicting outcome and hence could be of assistance in the planning of adjuvant therapies.

Adult↗

PubMed1991

Factors determining the occurrence of late potentials on the signal-averaged electrocardiogram after a first myocardial infarction: a multivariate analysis.

To determine the natural history of late potentials on the signal-averaged electrocardiogram (ECG), multivariate analysis was performed in 167 patients (138 men, 29 women) with a first anterior or inferior acute myocardial infarction. Seventy-four patients received thrombolytic therapy; the remaining 93 patients were treated conventionally. All patients underwent coronary angiography, left ventricular ejection fraction determination and signal-averaged ECG recording. Eight variables thought to be correlated with the presence of late potentials were studied; that is, age, infarct location, number of diseased coronary vessels, left ventricular ejection fraction, infarct-related coronary artery patency, treatment received, delay between admission and signal-averaged recording and delay between admission and coronary angiography. Statistical analysis showed that two independent factors (coronary artery occlusion and impaired left ventricular ejection fraction) were highly correlated with the incidence of late potentials. The occurrence of late potentials was multiplied by 5 in case of an occluded infarct-related vessel and by 1.75 each time the left ventricular ejection fraction value decreased by 0.10. This study suggests that coronary artery patency is the most important factor that decrease the rate of late potentials after a first acute myocardial infarction and it occurs independently of infarct location and left ventricular function.

Coronary Angiography↗