Classification trees and logistic regression applied to prognostic studies: a comparison using meningococcal disease as an example.

The authors used logistic regression and classification trees to develop prediction models for fatal outcomes in meningococcal disease in a cohort of 829 children hospitalized for meningococcal disease during 1989-1990 in Rio de Janeiro. The area under the receiver operator characteristic (ROC) curve was 92 per cent for logistic regression and 88 per cent for classification trees. Logistic regression may be preferred when the main objective is to obtain explicit measures for statistical inference and measures of the force of the association between each variable and the outcome. However, estimation of the probability of dying for each patient involves manipulation of the logistic regression formula, which would not easily be done in an emergency room. Classification trees provided comparable discrimination between fatal and non-fatal outcomes, and yielded a graphical display of the results that is easier to understand and is straightforward to apply in clinical settings.

[Meningococcal disease: epidemiology and control of secondary cases]. / Doença meningocócica: epidemiologia e controle dos casos secundários.

Epidemiological features of meningococcal disease described as from the second half of the 80's inclusive, have motivated a revision of current guidelines for sporadic disease and outbreak control. The increase of disease among teenagers and linked cases involving schools are the two most significant aspects that have prompted the revision of control measures. Vaccination routines and advice for the disease management of clusters are also relevant features recently revised. This present paper describes the management and some epidemiological features of secondary cases.

Antibiotic regimen as an independent risk factor for disseminated fungal infections in neutropenic patients in Brazil.

In a cohort of 79 febrile episodes in 50 consecutive neutropenic patients seen at the University Hospital, Federal University of Rio de Janeiro, Brazil, between 1987 and 1991, it was observed that the cumulative incidence of disseminated fungal infections rose from 3% to 19% after the introduction of a new empirical antibiotic regimen. In order to identify risk factors, as well as to assess the impact of the new antibiotic regimen on the emergence of fungal infections, a nested case-control study was undertaken, in which 10 cases of disseminated fungal infections were compared with 30 randomly chosen controls, drawn from the same cohort. In a multiple logistic regression analysis, the predictive factors for disseminated fungal infection were younger age (odds ratio 0.85, 95% confidence interval 0.75-0.97) and use of the new antibiotic regimen (odds ratio 14.18, 95% confidence interval 1.05-191.80) The probable explanation for the emergence of fungal infections is that the new antibiotic regimen, by lowering the incidence of bacteraemia-related deaths, allowed patients to be at risk for the development of disseminated fungal infections.

[Prevalence of virus B infection in a hospital community]. / Prevalência da infecção pelo vírus B na comunidade hospitalar.

The authors studied the prevalence of HBV markers among health care personnel of the University Hospital of the Federal University of Rio de Janeiro. The aim of study was to identify the high risk groups, in order to counsel vaccination of those groups, in order to counsel vaccination of those groups at risk as a routine. As a control group, a group of office workers of the hospital were chosen. A significant difference of incidence of HBV markers, mainly in surgeons (40%) and hemodialysis unit personnel (36.4%) when compared to the control group, was observed (p less than 0.05). The incidence increased with age and the time spent in the profession. The authors concluded that vaccination is indicated in surgeons hemodialysis personnel, dentists and laboratory personnel.