SEARCH · PubMed Health

Results for “Proportional Hazards 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 19 recordsLinked to original sources

PubMed1992

Flexible covariate effects in the proportional hazards model.

The proportional hazards model is frequently used in analyzing the results of clinical trials, when it is often the case that the outcomes are right-censored. This model allows one to measure treatment effects and simultaneously identify and adjust for prognostic factors that might influence the outcome. In this paper, we outline a class of semiparametric models that allows one to model prognostic factors nonlinearly, and have the data suggest the form of their effect. The methods are illustrated in an analysis of data from a breast cancer clinical trial.

Adult↗

PubMedDate not supplied

Using the proportional hazards model to study heart valve replacement data.

The proportional hazards model is used to study the effect of various concomitant variables on the time to valve failure, mortality, or other complications, for patients who have had artificial heart valves inserted. The data are from a database, which is still being assembled as more information is acquired, at Killingbeck Hospital. A suite of computer programs, not specifically developed with this application in mind, has been used to carry out the exploratory data analysis, the estimation of parameters and the validation of the model. These three elements of the analysis are all illustrated. The present report is seen as a preliminary study to assess the usefulness of the proportional hazards model in this area. Follow-up work as more data are accumulated is intended.

Bioprosthesis↗

PubMed1992

Proportional hazards model with covariates subject to measurement error.

When covariates of a proportional hazards model are subject to measurement error, the maximum likelihood estimates of regression coefficients based on the partial likelihood are asymptotically biased. Prentice (1982, Biometrika 69, 331-342) presents an example of such bias and suggests a modified partial likelihood. This paper applies the corrected score function method (Nakamura, 1990, Biometrika 77, 127-137) to the proportional hazards model when measurement errors are additive and normally distributed. The result allows a simple correction to the ordinary partial likelihood that yields asymptotically unbiased estimates; the validity of the correction is confirmed via a limited simulation study.

Adult↗

PubMed1992

GOFCOX: a computer program for the goodness-of-fit analysis of the Cox proportional hazards model.

GOFCOX is a user-friendly FORTRAN program for assessing the adequacy of the Cox proportional hazards model. The underlying methodology is based on the comparison of the maximum partial likelihood estimator and a weighted parameter estimator. The latter is the root to an estimation equation that assigns varying weights to the individual contributions to the partial likelihood score function. The weighted and unweighted parameter estimators have the same expectation under the Cox model, but tend to differ when the model is inappropriate. The GOFCOX program computes a rich class of weighted parameter estimators and corresponding goodness-of-fit test statistics. The program runs on both mainframe computers and microcomputers. The running time is minimal even for large data sets. A simple example is provided to illustrate the features of the program.

Computers, Mainframe↗

PubMed1992

Using Cox's proportional hazards model for prognostication in carcinoma of the upper aero-digestive tract.

One of the major short comings of the traditional TNM system is its limited potential for prognostication. With the development of multifactorial analysis techniques, such as Cox's proportional hazards model, it has become possible to simultaneously evaluate a large number of prognostic variables. Cox's model allows both the identification of prognostically relevant variables and the quantification of their prognostic influence. These characteristics make it a helpful tool for analysis as well as for prognostication. The goal of the present study was to develop a prognostic index for patients with carcinoma of the upper aero-digestive tract which makes use of all prognostically relevant variables. To accomplish this, the survival data of 800 patients with squamous cell carcinoma of the oral cavity, oropharynx, hypopharynx or larynx were analyzed. Sixty-one variables were screened for prognostic significance; of these only 19 variables (including age, tumor location, T, N and M stages, resection margins, capsular invasion of nodal metastases, and treatment modality) were found to significantly correlate with prognosis. With the help of Cox's equation, a prognostic index (PI) was computed for every combination of prognostic factors. To test the proposed model, the prognostic index was applied to 120 patients with carcinoma of the oral cavity or oropharynx. A comparison of predicted and observed survival showed good overall correlation, although actual survival tended to be better than predicted.

Adult↗

PubMed1992

Methods for comparing cumulative hazard functions in a semi-proportional hazard model.

Graphical methods based on the analysis of differences between log cumulative hazard functions are considered for a two-group semi-proportional hazard model which allows for interaction between treatments and covariates. Confidence procedures and test statistics that can be used to test for interaction and for main effects are developed. Their use is illustrated by applying them to the analysis of kidney transplant data from the University of California, San Francisco.

Azathioprine↗

PubMed1991

Effect of repeated transcatheter arterial embolization on the survival time in patients with hepatocellular carcinoma. An analysis by the Cox proportional hazard model.

One hundred fifty-eight patients with hepatocellular carcinoma (HCC) were treated by transcatheter arterial embolization (TAE) as repeatedly as possible. Survival rates at the end of the first, second, and third year were 76.5%, 54.5%, and 41.1%, respectively. In 142 patients with repeated TAE, a significantly increased number of patients with complete necrosis of tumor was observed after repetition of the therapy. Adjusting the imbalance in prognostic factors among patients by using Cox proportional hazard model, it proved that the best response during the repeated therapy, rather than the first response, was significantly associated with survival period of the patients. Aside from the factor of response to the treatment, tumor size was the worst prognostic factor at the time when diagnosis was made. Other significant factors were portal vein invasion by HCC and bilirubin. The survival period of patients with HCC treated by repeated TAE was, therefore, affected by cancer factors, liver cirrhosis factors, and therapy-responsiveness factors. It is concluded that even if complete necrosis of tumor is not obtained after the first trial, repetition of TAE is an effective measure for prolonging of survival time in patients with HCC.

Adult↗

PubMed1992

Effects of antiviral agents on chronic hepatitis B. Analysis using Cox proportional hazard model.

Two hundred fifteen courses of antiviral treatment including interferon with or without steroid withdrawal, adenine arabinoside with steroid withdrawal, and steroid withdrawal alone were given to 175 patients with HBe-antigen positive chronic hepatitis B. The effectiveness was judged on loss of HBe antigen and formation of anti-HBe, and was compared with 80 controls. According to cumulative HBe seronegative and seroconversion rates as analyzed by the Kaplan-Meier method, interferon with steroid withdrawal increased both the cumulative HBe seronegative and seroconversion rates significantly (P < 0.0001). Adenine arabinoside with steroid withdrawal and interferon alone increased the cumulative HBe seronegative rate only (P < 0.001). The Cox proportional hazard regression model was fitted to the data of 188 cases whose pretreatment liver biopsy specimens were obtained. Among treatment protocols, interferon with steroid withdrawal shortened both the HBeAg-positive period and the duration until anti-HBe becomes reactive significantly (P < 0.0001). Interferon without steroid withdrawal and adenine arabinoside with steroid withdrawal shortened the HBeAg positive interval only (P < 0.05). Among patients' characteristics, female and advanced liver histology were favorable factors. Effects of treatment protocols were analyzed after averaging each parameter of the patients' characteristics. Interferon and adenine arabinoside with steroid withdrawal shortened the HBeAg-positive interval significantly (P < 0.0001 and P < 0.05, respectively), and interferon alone showed a tendency to shorten the interval. In particular, interferon with steroid withdrawal increased the chance of losing HBeAg 7.3 times more than control. The effectiveness of antiviral treatment on chronic hepatitis B, especially the priority of interferon with steroid withdrawal, was thought to be established through this study.

Antiviral Agents↗

PubMed1992

Covariance adjustment of survival curves based on Cox's proportional hazards regression model.

Cox's proportional hazards regression model is a useful statistical tool for the analysis of 'survival data' from longitudinal studies. This multivariate method compares the 'survival experience' between two or more exposure groups while allowing for simultaneous adjustment of confounding due to one or more covariates. In addition to the summary regression statistics, further insight on the exposure--response relationship can be gained by visually examining the covariates-adjusted survival curves in the respective comparison groups. Covariates-adjusted survival curves are usually computed by the 'average covariate method'. This method is, however, subject to potential drawbacks. A method that avoids these drawbacks is to estimate adjusted survival curves by the corrected group prognostic curves approach. We have written a computer program to construct survival curves by the latter method. The program is coded in the Interactive Matrix Language of SAS.

Humans↗

PubMed1992

An analysis of excess mortality rates for persons with non-insulin-dependent diabetes mellitus in Western Australia using the Cox proportional hazards regression model.

A cohort of 888 rural, nonaboriginal persons with non-insulin-dependent diabetes mellitus identified in Western Australia through surveys in 1978-1982 were followed for death until the end of 1986. A total of 257 deaths were observed. Excess mortality in this cohort as compared with the general Australian population was investigated by calculating standardized mortality ratios and using the Cox proportional hazards regression model with hazard rates for the general population as the baseline. The overall standardized mortality ratio was 1.83 (95% confidence interval 1.51-2.16) for women and 1.43 (95% confidence interval 1.18-1.67) for men. Cause-specific comparisons with the general population showed that the majority of excess deaths could be attributed to diseases of the circulatory system. Factors assessed at the baseline survey that were independently prognostic of shorter survival were early onset of diabetes (for females only), high plasma glucose level, retinopathy, macrovascular disease, albuminuria (for females only), and elevated plasma creatinine level. Reductions in life expectancy at 60 years of age as compared with the general population averaged about 5 years but could be as much as 16 years for female diabetics with early onset of diabetes, high plasma glucose levels, and several complications.

Aged↗

PubMed1992

An exploratory analysis of survival with AIDS using a nonparametric tree-structured approach.

We illustrate an analysis with classification and regression trees applied to survival data. Through this application, we provide a description of the opportunistic diseases and sociodemographic factors that contribute to survival among people with human immunodeficiency virus disease. The analyses are based on 43,795 cases reported to the Centers for Disease Control between January 1, 1984, and December 31, 1987. We used vital status as of December 31, 1989, to estimate mortality rates. We identified Kaposi's sarcoma and opportunistic diseases causing central nervous system damage (cryptococcosis, primary lymphoma of the brain, cytomegalovirus disease, and progressive multifocal leukoencephalopathy) as important predictors of death. In addition, advanced age at diagnosis (50+), race (white/other), and history of illicit drug use were found to be important determinants. Estimates of the cumulative probability of survival for subgroups of individuals defined by the tree structure illustrate the effect of these determinants on mortality. For the purpose of comparison, two proportional hazards models were also fit to the data using factors identified in the tree structure as the determinants of interest. This application illustrates the utility and limitations of both this new technique and proportional hazards models for epidemiologic research.

Acquired Immunodeficiency Syndrome↗

PubMed1992

[Long-term observation after percutaneous transluminal angioplasty in the pelvic and leg areas].

In this study 163 patients had long term follow ups after PTA; the value of a test which had previously hardly been used, the "Cox proportional hazards model", for prognostic purposes is stressed. In this test account is taken of the severity of the stenosis before PTA, the extent of re-stenosis following PTA and the distance of the occlusion from the aorta and these are related to the likelihood of success. The use of the Cox proportional hazards model with an accurate description of the clinical material and the use of the Kaplan-Meier method is recommended in order to obtain comparable figures for the results obtained in various centres. In this study patency after six months was 77%, after a year 67%, after three years 46% and after five years 37%.

Angioplasty, Balloon↗

PubMed1992

National outcomes of cataract extraction. Increased risk of retinal complications associated with Nd:YAG laser capsulotomy. The Cataract Patient Outcomes Research Team.

PURPOSE: The authors studied 57,103 randomly selected Medicare beneficiaries who underwent extracapsular cataract extraction in 1986 or 1987 to determine the possible association between performance of neodymium (Nd):YAG laser capsulotomy and the risk of subsequent retinal break or detachment. METHODS: Cases of cataract surgery were identified from Medicare claims submitted in 1986 and 1987 and were followed through the end of 1988. Episodes of cataract surgery, posterior capsulotomy, and retinal complications were ascertained based on procedure and diagnosis codes listed in physician bills and hospital discharge records. Lifetable and Cox's proportional hazards models were used to analyze the risk of retinal detachment or break in patients undergoing and not undergoing capsulotomy during the period of observation. RESULTS: Of the 57,103 persons identified as having undergone extracapsular cataract extraction in 1986 or 1987, 13,709 subsequently underwent Nd:YAG laser capsulotomy between 1986 and 1988. A total of 337 persons had aphakic or pseudophakic retinal detachments between 1986 and 1988 and an additional 194 underwent repair of a retinal break. Proportional hazards modeling shows a 3.9-fold increase in the risk of retinal break or detachment among those who underwent capsulotomy (95% confidence interval: 2.89 to 5.25). Younger patient age, male sex, and white race also were associated with increased risk of retinal complications after extracapsular cataract extraction. CONCLUSION: The authors conclude that there is a statistically significant increase in the risk of retinal detachment or break in those patients who undergo capsulotomy after cataract extraction. Therefore, capsulotomy should be deferred until the patient's impairment caused by capsular opacification warrants the increased risk of retinal complications associated with performance of capsulotomy.

Aged↗

PubMed1992

Nine years of maintenance treatment with ranitidine for patients with duodenal ulcer disease.

Four hundred and sixty-four patients with duodenal ulcer disease received continuous maintenance treatment with ranitidine for up to 9 years. Treatment failure was defined as either the first symptomatic recurrence of ulcer or the first ulcer recurrence accompanied by haemorrhage. Life tables were constructed using the Kaplan-Meier product-limit method; comparisons of survival curves were performed using the log-rank test; and multivariate analysis was carried out using the Cox proportional hazards model. Results. The proportion of patients remaining free from symptomatic recurrence of ulcer during maintenance treatment with either 150 mg/day or 300 mg/day ranitidine was: 95% at 1 year; 88% at 3 years; 86% at 5 years; and 81% at 7 and 9 years (95% C.I. 76-86%). Young age and the absence of exposure to non-steroidal anti-inflammatory drugs (NSAIDs) at the time of diagnosis of ulcer were shown to increase the likelihood of developing ulcer recurrence. The rate of ulcer recurrence after 9 years of maintenance treatment (17%) was significantly less than that after 5 years without active anti-ulcer therapy (80%), P = 0.0001. The proportion of patients who remained free from ulcer haemorrhage during maintenance treatment was: 99.1% at 1 year and 98.1% at 5 and 9 years (95% C.I. 96.7-99.6%). The risk of haemorrhage was significantly greater in patients who were not receiving active treatment (12.4% at 5 years) than in patients receiving maintenance treatment (less than 2% at 9 years), P = 0.0001 (log-rank test). Patients who had originally presented with haemorrhage had an increased risk of further ulcer bleeding compared with patients who presented with pain (P = 0.0013). A significantly greater proportion of patients with NSAID-associated duodenal ulcers remained free from ulcer recurrence during maintenance treatment compared with patients suffering from 'idiopathic' ulcers (P = 0.0238), although there was no difference between the two groups in respect of haemorrhage during maintenance treatment. Conclusions. (a) Continuous maintenance treatment with ranitidine for up to nine years successfully prevents ulcer recurrence in more than 80% of patients with duodenal ulcer disease. (b) Young age increases the risk of ulcer recurrence during maintenance treatment with ranitidine. Ulcers associated with NSAID-intake at the time of diagnosis have a lesser risk of recurrence during maintenance treatment. (c) The risk of haemorrhage in patients with ulcer disease receiving maintenance treatment with ranitidine for nine years was less than 2% compared with greater than 12% in untreated patients observed for 5 years.(ABSTRACT TRUNCATED AT 400 WORDS)

Adult↗

PubMed1992

Schizophrenia: gender and familial risk.

The morbid risks for schizophrenia and any nonaffective psychosis in the first degree relatives of male and female schizophrenic probands were compared utilizing Cox proportional hazards models. The schizophrenic probands (275 male; 106 female) were drawn from a larger sample of hospitalized patients obtained by systematically screening all psychiatric admissions to 15 facilities over a six-year period. Proband diagnoses (DSM-III) were based on a direct assessment of the patient and a review of medical records. The family history method was used to obtain information about the first degree relatives of the probands. Cox proportional hazards models were adjusted for duration of illness of the proband and gender of the relatives. First degree relatives of female probands had significantly higher morbid risks for schizophrenia and nonaffective psychosis than relatives of male probands. The differential risk for schizophrenia in the relatives of male and female probands demonstrated in this study, as well as others, suggests that males and females may be at different risk for subtypes of the disorder.

Adult↗

PubMed2024

Metagenomic Study of the MESA: Detection of Gemella Morbillorum and Association With Coronary Heart Disease.

BACKGROUND: Inflammation is a feature of coronary heart disease (CHD), but the role of proinflammatory microbial infection in CHD remains understudied. METHODS AND RESULTS: CHD was defined in the MESA (Multi-Ethnic Study of Atherosclerosis) as myocardial infarction (251 participants), resuscitated arrest (2 participants), and CHD death (80 participants). We analyzed sequencing reads from 4421 MESA participants in the National Heart, Lung, and Blood Institute Trans-Omics for Precision Medicine program using the PathSeq workflow of the Genome Analysis Tool Kit and a 65-gigabase microbial reference. Paired reads aligning to 840 microbes were detected in >1% of participants. The association of the presence of microbe reads with incident CHD (follow-up, ~18 years) was examined. First, important variables were ascertained using a single regularized Cox proportional hazard model, examining change of risk as a function of presence of microbe with age, sex, education level, Life's Simple 7, and inflammation. For variables of importance, the hazard ratio (HR) was estimated in separate (unregularized) Cox proportional hazard models including the same covariates (significance threshold Bonferroni corrected P<6×10-5, 0.05/840). Reads from 2 microbes were significantly associated with CHD: Gemella morbillorum (HR, 3.14 [95% CI, 1.92-5.12]; P=4.86×10-6) and Pseudomonas species NFACC19-2 (HR, 3.22 [95% CI, 2.03-5.41]; P=1.58×10-6). CONCLUSIONS: Metagenomics of whole-genome sequence reads opens a possible frontier for detection of pathogens for chronic diseases. The association of G morbillorum and Pseudomonas species reads with CHD raises the possibilities that microbes may drive atherosclerotic inflammation and that treatments for specific pathogens may provide clinical utility for CHD reduction.

Humans↗

PubMedDate not supplied

The accelerated failure time model: a useful alternative to the Cox regression model in survival analysis.

For the past two decades the Cox proportional hazards model has been used extensively to examine the covariate effects on the hazard function for the failure time variable. On the other hand, the accelerated failure time model, which simply regresses the logarithm of the survival time over the covariates, has seldom been utilized in the analysis of censored survival data. In this article, we review some newly developed linear regression methods for analysing failure time observations. These procedures have sound theoretical justification and can be implemented with an efficient numerical method. The accelerated failure time model has an intuitive physical interpretation and would be a useful alternative to the Cox model in survival analysis.

Proportional Hazards Models↗

PubMed1992

Effectiveness of Lipiodol in transcatheter arterial embolization of hepatocellular carcinoma.

The effectiveness of Lipiodol (iodized oil) in transcatheter arterial embolization (TAE) of hepatocellular carcinoma (HCC) was retrospectively evaluated using statistical analysis. A total of 343 HCC patients who underwent TAE at 5 institutions between 1984 and 1989 were divided into 2 groups: the GS-TAE group underwent TAE with Gelfoam sponge alone, whereas the LP-TAE group was given Lipiodol (LP) immediately before GS-TAE. The statistical T value calculated for the LP-TAE group showed that the administration of LP, the tumor size, intrahepatic metastasis, portal vein infiltration, and serum total bilirubin and alpha-fetoprotein levels significantly (P < 0.01) affected the patients' survival. Both the cumulative survival determined using the Kaplan-Meier model and the cumulative hazard calculated using Cox's proportional hazard model differed significantly (P < 0.01) between the GS-TAE group and the LP-TAE group (log-rank test). These results confirmed the effectiveness of LP used in combination with Gelfoam sponge for TAE of HCC.

Aged↗