An EM-based Semi-Parametric Mixture Model Approach to the Regression Analysis of Competing-Risks Data

We consider a mixture model approach to the regression analysis of competing-risks data. Attention is focused on inference concerning the effects of factors on both the probability of occurrence and the hazard rate conditional on each of the failure types. These two quantities are specified in the mixture model using the logistic model and the proportional hazards model, respectively. We propose a semi-parametric mixture method to estimate the logistic and regression coefficients jointly, whereby the component-baseline hazard functions are completely unspecified. Estimation is based on maximum likelihood on the basis of the full likelihood, implemented via an expectation-conditional maximization (ECM) algorithm. Simulation studies are performed to compare the performance of the proposed semi-parametric method with a fully parametric mixture approach. The results show that when the component-baseline hazard is monotonic increasing, the semi-parametric and fully parametric mixture approaches are comparable for mildly and moderately censored samples. When the component-baseline hazard is not monotonic increasing, the semi-parametric method consistently provides less biased estimates than a fully parametric approach and is comparable in efficiency in the estimation of the parameters for all levels of censoring. The methods are illustrated using a real data set of prostate cancer patients treated with different dosages of the drug diethylstilbestrol. Copyright (C) 2003 John Wiley Sons, Ltd.

Effect of functional gain on satisfaction with medical rehabilitation after stroke

Objective: To examine the association between gain in motor and cognitive functional status with patient satisfaction 3-6 mo after rehabilitation discharge. Design: Patient satisfaction and changes in functional status were examined in 18,375 patients with stroke who received inpatient medical rehabilitation. Information was obtained from 144 hospitals and rehabilitation facilities contributing records to the Uniform Data System for Medical Rehabilitation and the National Follow-up Services. Results: Data analysis revealed significant (P < 0.05) differences in satisfaction responses based on whether information was collected from patient self-report or from a family member proxy, and the two subsets were analyzed separately. Logistic regression revealed the following significant predictors of satisfaction for data collected from stroke patients: cognitive and motor gain, rehospitalization, who the patient was living with at follow-up, age, and follow-up therapy. In the patient-reported data subset, compared with patients who showed improved cognitive or motor functional status, those with no change, respectively, had a 31% and 33% reduced risk of dissatisfaction. In addition, rehospitalized patients had a higher risk of dissatisfaction. For the proxy reported data subset, significant influences on satisfaction were health maintenance, rehospitalization, stroke type, ethnicity, cognitive FIM(TM) gain, length of stay, and follow-up therapy. Conclusions: Ratings of satisfaction with rehabilitation services were affected by change in functional status and whether the information was collected from patient rating or proxy response.

A reduced load approximation accounting for link interactions in a loss network

This paper is concerned with evaluating the performance of loss networks. Accurate determination of loss network performance can assist in the design and dimen- sioning of telecommunications networks. However, exact determination can be difficult and generally cannot be done in reasonable time. For these reasons there is much interest in developing fast and accurate approximations. We develop a reduced load approximation that improves on the famous Erlang fixed point approximation (EFPA) in a variety of circumstances. We illustrate our results with reference to a range of networks for which the EFPA may be expected to perform badly.

Poor correlation of mid-femoral measurements by CT and hip measurements by DXA in the elderly

Background and aims: Hip fracture is a devastating event in terms of outcome in the elderly, and the best predictor of hip fracture risk is hip bone density, usually measured by dual X-ray absorptiometry (DXA). However, bone density can also be ascertained from computerized tomography (CT) scans, and mid-thigh scans are frequently employed to assess the muscle and fat composition of the lower limb. Therefore, we examined if it was possible to predict hip bone density using mid-femoral bone density. Methods: Subjects were 803 ambulatory white and black women and men, aged 70-79 years, participating in the Health, Aging and Body Composition (Health ABC) Study. Bone mineral content (BMC, g) and volumetric bone mineral density (vBMD, mg/cm(3)) of the mid-femur were obtained by CT, whereas BMC and areal bone mineral density (aBMD, g/cm(2)) of the hip (femoral neck and trochanter) were derived from DXA. Results: In regression analyses stratified by race and sex, the coefficient of determination was low with mid-femoral BMC, explaining 6-27% of the variance in hip BMC, with a standard error of estimate (SEE) ranging from 16 to 22% of the mean. For mid-femur vBMD, the variance explained in hip aBMD was 2-17% with a SEE ranging from 15 to 18%. Adjusting aBMD to approximate volumetric density did not improve the relationships. In addition, the utility of fracture prediction was examined. Forty-eight subjects had one or more fractures (various sites) during a mean follow-up of 4.07 years. In logistic regression analysis, there was no association between mid-femoral vBMD and fracture (all fractures), whereas a 1 SD increase in hip BMD was associated with reduced odds for fracture of similar to60%. Conclusions: These results do not support the use of CT-derived mid-femoral vBMD or BMC to predict DXA-measured hip bone mineral status, irrespective of race or sex in older adults. Further, in contrast to femoral neck and trochanter BMD, mid-femur vBMD was not able to predict fracture (all fractures). (C) 2003, Editrice Kurtis.

Enhanced effector and memory CTL responses generated by incorporation of receptor activator of NF-kappa B(RANK)/RANK ligand costimulatory molecules into dendritic cell immunogens expressing a human tumor-specific antigen

The outcome of dendritic cell (DC) presentation of Ag to T cells via the TCR/MHC synapse is determined by second signaling through CD80/86 and, importantly, by ligation of costimulatory ligands and receptors located at the DC and T cell surfaces. Downstream signaling triggered by costimulatory molecule ligation results in reciprocal DC and T cell activation and survival, which predisposes to enhanced T cell-mediated immune responses. In this study, we used adenoviral vectors to express a model tumor Ag (the E7 oncoprotein of human papillomavirus 16) with or without coexpression of receptor activator of NF-kappaB (RANK)/RANK ligand (RANKL) or CD40/CD40L costimulatory molecules, and used these transgenic DCs to immunize mice for the generation of E7-directed CD8(+) T cell responses. We show that coexpression of RANK/RANKL, but not CD40/CD40L, in E7-expressing DCs augmented E7-specific IFN-gamma-secreting effector and memory T cells and E7-specific CTLs. These responses were also augmented by coexpression of T cell costimulatory molecules (RANKL and CD40L) or DC costimulatory molecules (RANK and CD40) in the E7-expressing DC immunogens. Augmentation of CTL responses correlated with up-regulation of CD80 and CD86 expression in DCs transduced with costimulatory molecules, suggesting a mechanism for enhanced T cell activation/survival. These results have generic implications for improved tumor Ag-expressing DC vaccines, and specific implications for a DC-based vaccine approach for human papillomavirus 16-associated cervical carcinoma.

Ventricular performance and Na+-K+ ATPase activity are reduced early and late after myocardial infarction in rats

Myocardial infarction leads to compensatory ventricular remodeling. Disturbances in myocardial contractility depend on the active transport of Ca2+ and Na+, which are regulated by Na+-K+ ATPase. Inappropriate regulation of Na+-K+ ATPase activity leads to excessive loss of K+ and gain of Na+ by the cell. We determined the participation of Na+-K+ ATPase in ventricular performance early and late after myocardial infarction. Wistar rats (8-10 per group) underwent left coronary artery ligation (infarcted, Inf) or sham-operation (Sham). Ventricular performance was measured at 3 and 30 days after surgery using the Langendorff technique. Left ventricular systolic pressure was obtained under different ventricular diastolic pressures and increased extracellular Ca2+ concentrations (Ca2+e) and after low and high ouabain concentrations. The baseline coronary perfusion pressure increased 3 days after myocardial infarction and normalized by 30 days (Sham 3 = 88 ± 6; Inf 3 = 130 ± 9; Inf 30 = 92 ± 7 mmHg; P < 0.05). The inotropic response to Ca2+e and ouabain was reduced at 3 and 30 days after myocardial infarction (Ca2+ = 1.25 mM; Sham 3 = 70 ± 3; Inf 3 = 45 ± 2; Inf 30 = 29 ± 3 mmHg; P < 0.05), while the Frank-Starling mechanism was preserved. At 3 and 30 days after myocardial infarction, ventricular Na+-K+ ATPase activity and contractility were reduced. This Na+-K+ ATPase hypoactivity may modify the Na+, K+ and Ca2+ transport across the sarcolemma resulting in ventricular dysfunction.

Os fundos de investimento imobiliário listados em bolsa e as variáveis que influenciam seus retornos

Os fundos de investimento imobiliário combinam características tanto do mercado imobiliário, fonte de seus rendimentos, quanto do mercado de capitais, ambiente em que são negociados. O impacto de cada um desses mercados subjacentes no comportamento, desempenho e risco dessa classe de ativos não é, no entanto, ainda claramente definida, sendo um dos grandes temas em análise, tanto na literatura acadêmica, quanto na indústria de fundos internacionais. Em face da significativa expansão dessa alternativa de investimento no mercado brasileiro, no presente estudo foram analisadas as variáveis que influenciam os retornos dos fundos imobiliários brasileiros para uma amostra de fundos listados em Bolsa de Valores de São Paulo, período de 2008-2013. Seguindo a metodologia de Clayton e Mackinnon (2003), os fatores explicativos dos retornos foram decompostos em quatro componentes principais, sendo três fatores de retorno de mercado (mercado de ações, mercado de renda fixa e mercado imobiliário) e risco idiossincrático. De acordo com a estatística descritiva, os fundos imobiliários da amostra apresentaram maior retorno em relação aos demais mercados, exceto em comparação com o mercado imobiliário, porém com menor risco. As análises de correlação, regressão e decomposição da variância indicam que o mercado de ações e o mercado imobiliário direto são, em geral, significativos no modelo, porém explicam apenas cerca de 15% da volatilidade dos retornos dos fundos da amostra. À luz da Moderna Teoria do Portfólio, esses resultados indicam que a inclusão de fundos imobiliários pode ter potencial diversificador numa carteira multi-ativo, seja aumentando o retorno total de uma carteira formada de ações e títulos de renda fixa, sem acréscimo em risco; ou mantendo o retorno dessa carteira, com diminuição da volatilidade, ampliando assim a fronteira eficiente da carteira. Esse efeito questiona o tradicional equilibrium fund de carteiras de investimentos formadas apenas de ações e renda fixa e aponta os fundos imobiliários como uma alternativa de investimento diversificadora, enquanto classe de ativo única. A análise de subamostras por tipologia indica, porém, que o papel diversificador dos fundos imobiliários está atrelado ao tipo de empreendimento que lastreia esse fundo, uma vez que os fatores explicativos e seus impactos nos retornos diferem de uma tipologia para outra. Esse resultado tem importantes implicações no critério de seleção a ser adotado tanto por investidores para seleção de ativos para uma carteira otimizada, como para gestores de fundos imobiliários na formatação e gestão desses produtos. Conclui-se também que os retornos dos fundos, de certo modo, refletem seu caráter híbrido, mas o modelo decomposto em 4 componentes não é suficiente para explicar os retornos dos fundos imobiliários, uma vez que o modelo estendido, demonstrou que outras variáveis, inclusive parâmetros desses próprios mercados, além de variáveis macroeconômicas e as características de cada fundo (eg. market-to-book, tamanho), podem ser responsáveis por explicar considerável parte da variância dos retornos dos FIIs.

Error bounds for low-rank approximations of the first exponential integral kernel

A hierarchical matrix is an efficient data-sparse representation of a matrix, especially useful for large dimensional problems. It consists of low-rank subblocks leading to low memory requirements as well as inexpensive computational costs. In this work, we discuss the use of the hierarchical matrix technique in the numerical solution of a large scale eigenvalue problem arising from a finite rank discretization of an integral operator. The operator is of convolution type, it is defined through the first exponential-integral function and, hence, it is weakly singular. We develop analytical expressions for the approximate degenerate kernels and deduce error upper bounds for these approximations. Some computational results illustrating the efficiency and robustness of the approach are presented.

Survival analysis: a tool in the study of post-harvest diseases in peaches

Survival analysis is applied when the time until the occurrence of an event is of interest. Such data are routinely collected in plant diseases, although applications of the method are uncommon. The objective of this study was to use two studies on post-harvest diseases of peaches, considering two harvests together and the existence of random effect shared by fruits of a same tree, in order to describe the main techniques in survival analysis. The nonparametric Kaplan-Meier method, the log-rank test and the semi-parametric Cox's proportional hazards model were used to estimate the effect of cultivars and the number of days after full bloom on the survival to the brown rot symptom and the instantaneous risk of expressing it in two consecutive harvests. The joint analysis with baseline effect, varying between harvests, and the confirmation of the tree effect as a grouping factor with random effect were appropriate to interpret the phenomenon (disease) evaluated and can be important tools to replace or complement the conventional analysis, respecting the nature of the variable and the phenomenon.

Susceptibility of asthmatic children to respiratory infection

OBJECTIVE: A case-control study of patients with pneumonia was conducted to investigate whether wheezing diseases could be a risk factor. METHODS: A random sample was taken from a general university hospital in S. Paulo City between March and August 1994 comprising 51 cases of pneumonia paired by age and sex to 51 non-respiratory controls and 51 healthy controls. Data collection was carried out by two senior paediatricians. Diagnoses of pneumonia and presence of wheezing disease were independently established by each paediatrician for both cases and controls. Pneumonia was radiologically confirmed and repeatability of information on wheezing diseases was measured. Logistic regression analysis was used to identify risk factors. RESULTS: Wheezing diseases, interpreted as proxies of asthma, were found to be an important risk factor for pneumonia with an odds ratio of 7.07 (95%CI= 2.34-21.36), when the effects of bedroom crowding (odds ratio = 1.49 per person, 95%CI= 0.95-2.32) and of low family income (odds ratio = 5.59 against high family income, 95%CI= 1.38-22.63) were controlled. The risk of pneumonia attributable to wheezing diseases is tentatively calculated at 51.42%. CONCLUSION: It is concluded that at practice level asthmatics should deserve proper surveillance for infection and that at public health level pneumonia incidence could be reduced if current World Health Organisation's guidelines were reviewed as to include comprehensive care for this illness.

Long-term price range forecast applied to risk management using regression models

Long-term contractual decisions are the basis of an efficient risk management. However those types of decisions have to be supported with a robust price forecast methodology. This paper reports a different approach for long-term price forecast which tries to give answers to that need. Making use of regression models, the proposed methodology has as main objective to find the maximum and a minimum Market Clearing Price (MCP) for a specific programming period, and with a desired confidence level α. Due to the problem complexity, the meta-heuristic Particle Swarm Optimization (PSO) was used to find the best regression parameters and the results compared with the obtained by using a Genetic Algorithm (GA). To validate these models, results from realistic data are presented and discussed in detail.

Job control, job demands, social support at work and health among adolescent workers

OBJECTIVE: To evaluate physical and psychological dimensions of adolescent labor (such as job demands, job control, and social support in the work environment), and their relation to reported body pain, work injuries, sleep duration and daily working hours. METHODS: A total of 354 adolescents attending evening classes at a public school in São Paulo, Brazil, answered questionnaires regarding their living and working conditions (Karasek's Job Content Questionnaire, 1998), and their health status. Data collection took place in April and May 2001. Multiple logistic regression analysis was used to determine relations among variables. RESULTS: Psychological job demands were related to body pain (OR=3.3), higher risk of work injuries (OR=3.0) and reduced sleep duration in weekdays (Monday to Thursday) (p<0.01). Lower decision authority in the workplace (p=0.03) and higher job security (p=0.02) were related to longer daily working hours. CONCLUSIONS: It was concluded that besides physical stressors, psychological factors are to be taken into account when studying adolescent working conditions, as they may be associated with negative job conditions and health effects.