Modified score function for monotone likelihood in the semiparametric mixture cure model.

The cure fraction models are intended to analyze lifetime data from populations where some individuals are immune to the event under study, and allow a joint estimation of the distribution related to the cured and susceptible subjects, as opposed to the usual approach ignoring the cure rate. In situations involving small sample sizes with many censored times, the detection of nonfinite coefficients may arise via maximum likelihood. This phenomenon is commonly known as monotone likelihood (ML), occurring in the Cox and logistic regression models when many categorical and unbalanced covariates are present. An existing solution to prevent the issue is based on the Firth correction, originally developed to reduce the estimation bias. The method ensures finite estimates by penalizing the likelihood function. In the context of mixture cure models, the ML issue is rarely discussed in the literature; therefore, this topic can be seen as the first contribution of our paper. The second major contribution, not well addressed elsewhere, is the study of the ML issue in cure mixture modeling under the flexibility of a semiparametric framework to handle the baseline hazard. We derive the modified score function based on the Firth approach and explore finite sample size properties of the estimators via a Monte Carlo scheme. The simulation results indicate that the performance of coefficients related to the binary covariates are strongly affected to the imbalance degree. A real illustration, in the melanoma dataset, is discussed using a relatively novel data set collected in a Brazilian university hospital.

Clinical characteristics and risk factors for death among hospitalised children and adolescents with COVID-19 in Brazil: an analysis of a nationwide database.

BACKGROUND: COVID-19 is usually less severe and has lower case fatality in children than in adults. We aimed to characterise the clinical features of children and adolescents hospitalised with laboratory-confirmed SARS-CoV-2 infection and to evaluate the risk factors for COVID-19-related death in this population. METHODS: We did an analysis of all patients younger than 20 years who had quantitative RT-PCR-confirmed COVID-19 and were registered in the Influenza Epidemiological Surveillance Information System (SIVEP-Gripe, a nationwide surveillance database of patients admitted to hospital with severe acute respiratory disease in Brazil), between Feb 16, 2020, and Jan 9, 2021. The primary outcome was time to recovery (discharge) or in-hospital death, evaluated by competing risks analysis using the cumulative incidence function. FINDINGS: Of the 82 055 patients younger than 20 years reported to SIVEP-Gripe during the study period, 11 613 (14·2%) had available data showing laboratory-confirmed SARS-CoV-2 infection and were included in the sample. Among these patients, 886 (7·6%) died in hospital (at a median 6 days [IQR 3-15] after hospital admission), 10 041 (86·5%) patients were discharged from the hospital, 369 (3·2%) were in hospital at the time of analysis, and 317 (2·7%) were missing information on outcome. The estimated probability of death was 4·8% during the first 10 days after hospital admission, 6·7% during the first 20 days, and 8·1% at the end of follow-up. Probability of discharge was 54·1% during the first 10 days, 78·4% during the first 20 days, and 92·0% at the end of follow-up. Our competing risks multivariate survival analysis showed that risk of death was increased in infants younger than 2 years (hazard ratio 2·36 [95% CI 1·94-2·88]) or adolescents aged 12-19 years (2·23 [1·84-2·71]) relative to children aged 2-11 years; those of Indigenous ethnicity (3·36 [2·15-5·24]) relative to those of White ethnicity; those living in the Northeast region (2·06 [1·68-2·52]) or North region (1·55 [1·22-1·98]) relative to those in the Southeast region; and those with one (2·96 [2·52-3·47]), two (4·96 [3·80-6·48]), or three or more (7·28 [4·56-11·6]) pre-existing medical conditions relative to those with none. INTERPRETATION: Death from COVID-19 was associated with age, Indigenous ethnicity, poor geopolitical region, and pre-existing medical conditions. Disparities in health care, poverty, and comorbidities can contribute to magnifying the burden of COVID-19 in more vulnerable and socioeconomically disadvantaged children and adolescents in Brazil. FUNDING: National Council for Scientific and Technological Development, Research Support Foundation of Minas Gerais.

Estimating the grid of time-points for the piecewise exponential model.

One of the greatest challenges related to the use of piecewise exponential models (PEMs) is to find an adequate grid of time-points needed in its construction. In general, the number of intervals in such a grid and the position of their endpoints are ad-hoc choices. We extend previous works by introducing a full Bayesian approach for the piecewise exponential model in which the grid of time-points (and, consequently, the endpoints and the number of intervals) is random. We estimate the failure rates using the proposed procedure and compare the results with the non-parametric piecewise exponential estimates. Estimates for the survival function using the most probable partition are compared with the Kaplan-Meier estimators (KMEs). A sensitivity analysis for the proposed model is provided considering different prior specifications for the failure rates and for the grid. We also evaluate the effect of different percentage of censoring observations in the estimates. An application to a real data set is also provided. We notice that the posteriors are strongly influenced by prior specifications, mainly for the failure rates parameters. Thus, the priors must be fairly built, say, really disclosing the expert prior opinion.

Setting priorities for environmental sanitation interventions based on epidemiological criteria: a Brazilian study.

The present study addresses the use of analytical epidemiologic approaches to subsidize the establishment of priorities in environmental sanitation interventions. An epidemiological investigation was carried out in 1993 in the urban area of Betim, a southeast Brazilian City of 160,000 inhabitants. The case-control 'inclusive' (or case-cohort) design, with a sample of 997 cases and 999 controls, was employed. Cases were defined as children of less than five years of age presenting diarrhoea episodes, while controls were randomly selected among children of the same age, living in the study area. After logistic regression adjustment, 11 of several exposure variables analysed were significantly associated with diarrhoea. Four different criteria, using as risk measures the relative risk, the attributable risk, the standardized coefficient of the logistic regression and the cost standardized coefficient, were used in order to define intervention priorities.

Indicadores sanitários sentinelas da mortalidade infantil: um estudo ecológico de base municipal em quatro estados brasileiros / Sentinel sanitary indicators for infant mortality: a municipality-based ecological study in four brazilian states

É relativamente recente no país a prática da vigilância ambiental em saúde, área que se ocupa do monitoramento e controle dos fatores ambientais de risco à saúde humana. A vigilância da qualidade da água para consumo humano, parte integrante da vigilância ambiental em saúde, ressente de estudos e pesquisas que forneçam suporte científico à formulação de instrumentos para a prática de tal vigilância. Nesse campo, a formulação e seleção de indicadores epidemiológicos e sanitários constituem atividade essencial para representar os efeitos da insuficiência das ações de saneamento sobre a saúde humana e, portanto, como ferramenta para a vigilância e a orientação de programas e planos de alocação de recursos em saneamento. O presente artigo é parte de um trabalho mais abrangente, elaborado para selecionar indicadores sanitários como sentinelas na prevenção e controle de doenças e agravos relacionados ao saneamento, a partir da análise estatística entre um banco de dados, de base municipal, de indicadores sanitários sociais e epidemiológicos. Neste artigo são enfocados os indicadores sanitários que melhor relação apresentaram com a mortalidade infantil e que possam ser considerados sentinelas na detecção precoce daquele indicador. Destacam-se, dentre os indicadores sanitários avaliados, a proporção de turbidez fora do padrão e o percentual da população coberta por coleta de lixo.