RESUMO
We describe a microcomputer program (COXSURV) for proportional hazards multiple regression analysis of survival and other failure-time data generated in clinical trials and in retrospective clinical epidemiology studies. COXSURV is menu-driven and has powerful variable factoring and data exploratory capabilities for multivariate modeling. A batch mode allows automatic uni- or multivariate analyses for confounder summarization. Model selection for predictive purposes is possible through a step-up algorithm. The partial likelihood method used in the program allows the use of either discrete or continuous time scales by treating tied uncensored observations by either the exact method or by a robust approximation method. The program calculates most standard model fitting statistics for either overall or stratified analyses and uses data layout files compatible with those of other related epidemiologic analysis software.
Assuntos
Análise Multivariada , Software , Análise de Sobrevida , Algoritmos , Neoplasias da Mama/mortalidade , Métodos Epidemiológicos , Feminino , Humanos , Leucemia/mortalidade , Computação Matemática , Microcomputadores , Modelos de Riscos Proporcionais , Estudos RetrospectivosRESUMO
A frequent procedure in matched case-control studies is to report results from the multivariate unmatched analyses if they do not differ substantially from the ones obtained after conditioning on the matching variables. Although conceptually simple, this rule requires that an extensive series of logistic regression models be evaluated by both the conditional and unconditional maximum likelihood methods. Most computer programs for logistic regression employ only one maximum likelihood method, which requires that the analyses be performed in separate steps. This paper describes a Pascal microcomputer (IBM PC) program that performs multiple logistic regression by both maximum likelihood estimation methods, which obviates the need for switching between programs to obtain relative risk estimates from both matched and unmatched analyses. The program calculates most standard statistics and allows factoring of categorical or continuous variables by two distinct methods of contrast. A built-in, descriptive statistics option allows the user to inspect the distribution of cases and controls across categories of any given variable.
Assuntos
Métodos Epidemiológicos , Microcomputadores , Probabilidade , Análise de Regressão , Software/métodos , Algoritmos , Modelos EstatísticosRESUMO
We describe a microcomputer program (KMSURV) for exploratory univariate statistical analysis of survival data which is directly applicable to the evaluation of clinical trials and to retrospective epidemiological studies of hospital registry-based data. The program calculates life-table-like information based on Kaplan-Meier's product-limit estimators of the survivorship function S(t) and provides summary measures of average survival times. In addition, two non-parametric tests for the comparison of survival distributions are performed. A report-quality, high resolution plot of the S(t) estimates for all groups being compared complements each set of analyses. KMSURV is not a simple adaptation of a mainframe statistical analysis package and, thus, it utilizes efficiently the interactive environment which is inherent in microcomputing.
Assuntos
Análise Atuarial/métodos , Interpretação Estatística de Dados , Computação Matemática , Microcomputadores , Análise Numérica Assistida por Computador , Humanos , Leucemia/mortalidade , Linguagens de ProgramaçãoRESUMO
Ninety-four soft-tissue sarcomas of head and neck occurring among 10,700 malignancies, seen during the period 1953-1980 in the head-and-neck surgical department, were studied with reference to age, sex, ethnic origin, site, histological type and survival. The objective was to review our own experience with sarcomas in these sites seen in the Department of Pathology, A.C. Camargo Hospital. The predominant site was the head. Fibrosarcoma was the most frequent type in adults and rhabdomyosarcoma, located exclusively in the head, was the most frequent in children.
Assuntos
Neoplasias de Cabeça e Pescoço/epidemiologia , Sarcoma/epidemiologia , Adolescente , Adulto , Idoso , Criança , Pré-Escolar , Feminino , Neoplasias de Cabeça e Pescoço/mortalidade , Neoplasias de Cabeça e Pescoço/patologia , Humanos , Lactente , Recém-Nascido , Masculino , Pessoa de Meia-Idade , Sarcoma/mortalidade , Sarcoma/patologiaRESUMO
Sex-specific relative frequencies (RF) of oral, esophageal, stomach, colon, rectal, laryngeal, lung, female breast, cervical and penile cancers obtained from a government-sponsored, nation-wide data base of histopathological diagnoses were evaluated with respect to all possible inter-site correlations and with 12 socioeconomic and demographic variables for 23 States in Brazil. Use of bivariate and multivariate methods detected a high positive intercorrelation among RFs of lung, laryngeal and colon cancers regardless of sex. RFs for these 3 sites were also positively correlated with many markers of State development and affluence. Cervical and penile cancers emerged as a distinct subset with respect to their correlation patterns. RFs for these neoplasms were highly (positively) correlated (r = 0.8606, p less than 0.001) with each other and exhibited intense negative associations with many of the affluence markers and the former cancer sites. Bivariate correlations generally exhibited a better fit with female-specific RFs than with data from males which was reflected in the number of strong correlations detected by sex.