Valeska Andreozzi

Análise de sobrevivência: teoria e aplicações em saúde

O objetivo apoiar o desenvolvimento metodologico em analise de dados longitudinais, atraves da aplicacao, teste e disseminacao de metodos estatisticos inovadores no contexto da saude, particularmente os voltados para a analise de sobrevida

Hospital Resource Utilization and Treatment Cost of Skeletal-Related Events in Patients with Metastatic Breast or Prostate Cancer: Estimation for the Portuguese National Health System

BACKGROUND Skeletal-related events (SREs) occur frequently in patients with bone metastases as a result of breast (BC) and prostate (PC) cancers. They increase both morbidity and mortality and lead to extensive health-care resource utilization. METHODS Health care resource utilization by BC/PC patients with at least one SRE during the preceding 12 months was assessed through retrospective chart review. SRE-treatment costs were estimated using the Portuguese Ministry of Health cost database and analyzed using generalized linear models. RESULTS This study included 152 patients from nine hospitals. The mean (SD) annual SRE-treatment cost per patient was €5963 (€3646) and €5711 (€4347), for BC (n=121) and PC (n=31) patients, respectively. Mean cost per single episode ranged between €1485 (radiotherapy) and €13,203 (spinal cord compression). Early onset of bone metastasis (P = 0.03) and diagnosis of bone metastases at or after the occurrence of the first SRE (P < 0.001) were associated with higher SRE-treatment costs. CONCLUSION These results reveal the high hospital SRE-treatment costs, highlighting the need for early diagnosis and treatment, and identify key factors determining the economic value of therapies for patients with skeletal metastases.

Maternal child-feeding practices and dietary inadequacy of 4-year-old children

This study aimed to evaluate the association between maternal perceived responsibility and child-feeding practices and dietary inadequacy of 4-year-old children. We studied 4122 mothers and children enrolled in the population-based birth cohort - Generation XXI (Porto, Portugal). Mothers self-completed the Child Feeding Questionnaire and a scale on covert and overt control, and answered to a food frequency questionnaire in face-to-face interviews. Using dietary guidelines for preschool children, adequacy intervals were defined: fruit and vegetables (F&V) 4-7 times/day; dairy 3-5 times/day; meat and eggs 5-10 times/week; fish 2-4 times/week. Inadequacy was considered as below or above these cut-points. For energy-dense micronutrient-poor foods and beverages (EDF), a tolerable limit was defined (<6 times/week). Associations between maternal perceived responsibility and child-feeding practices (restriction, monitoring, pressure to eat, overt and covert control) and childrens diet were examined by logistic regression models. After adjustment for maternal BMI, education, and diet, and childrens characteristics (sex, BMI z-scores), restriction, monitoring, overt and covert control were associated with 11-18% lower odds of F&V consumption below the interval defined as adequate. Overt control was also associated with 24% higher odds of their consumption above it. Higher perceived responsibility was associated with higher odds of children consuming F&V and dairy above recommendations. Pressure to eat was positively associated with consumption of dairy above the adequate interval. Except for pressure to eat, maternal practices were associated with 14-27% lower odds of inadequate consumption of EDF. In conclusion, children whose mothers had higher levels of covert control, monitoring, and restriction were less likely to consume F&V below recommendations and EDF above tolerable limits. Higher overt control and pressure to eat were associated, respectively, with higher possibility of children consuming F&V and dairy above recommendations.

Fatores associados à morbidade respiratória entre 12 e 36 meses de vida de crianças nascidas de muito baixo peso oriundas de uma UTI neonatal pública

The aim of this paper was to estimate respiratory morbidity and its determinants for premature infants aged 12 to 36 months. The population comprised 84 infants from a cohort of very low birth weight premature infants. The outcome was the respiratory morbidity incidence rate. The relationship between the independent variables and respiratory morbidity was estimated using a Poisson regression model. From 12 to 24 months of age, 56.3% of children had experienced at least one episode of respiratory disease. >From 24 to 36 months, 38.1% of children were affected. Variables significantly associated with respiratory morbidity were bronchopulmonary dysplasia (RR = 1.9; 95%CI: 1.2-2.9), abnormal lung compliance (RR = 1.6; 95%CI: 1.1-2.3), neonatal pneumonia (RR = 2.8; 95%CI: 1.9-4.0), patent ductus arteriosus (RR = 1.6; 95%CI: 1.1-2.5), and respiratory morbidity in the first year of life (RR = 1.7; 95%CI: 1.2-2.5). The incidence of respiratory morbidity remains high in this group of high-risk infants, which calls for regular follow-up and effective interventions to prevent respiratory disease and to improve the quality of life of these children and their families.

Bayesian joint modeling of longitudinal and spatial survival AIDS data

Joint analysis of longitudinal and survival data has received increasing attention in the recent years, especially for analyzing cancer and AIDS data. As both repeated measurements (longitudinal) and time-to-event (survival) outcomes are observed in an individual, a joint modeling is more appropriate because it takes into account the dependence between the two types of responses, which are often analyzed separately. We propose a Bayesian hierarchical model for jointly modeling longitudinal and survival data considering functional time and spatial frailty effects, respectively. That is, the proposed model deals with non-linear longitudinal effects and spatial survival effects accounting for the unobserved heterogeneity among individuals living in the same region. This joint approach is applied to a cohort study of patients with HIV/AIDS in Brazil during the years 2002-2006. Our Bayesian joint model presents considerable improvements in the estimation of survival times of the Brazilian HIV/AIDS patients when compared with those obtained through a separate survival model and shows that the spatial risk of death is the same across the different Brazilian states. Copyright

Social inequalities in BMI trajectories: 8-year follow-up of the Pró-Saúde study in Rio de Janeiro, Brazil.

Objective In a cohort of government employees in Rio de Janeiro, Brazil, we investigated prospectively, sex-specific associations between education and BMI trajectories and their potential effect modification by race. Design Of the 4030 participants in Phase 1 (1999), 3253 (81 %) participated in Phase 2 (2003) and 3058 (76 %) participated in Phase 3 (2006). Education was categorized as elementary, high school or college graduate. Study participants self-identified as White, Black or Pardo. BMI was calculated from measured weight and height. BMI trajectories were modelled using a generalized additive regression model with mixed effects (GAMM). Setting The Pro-Saúde Study, a longitudinal investigation of social determinants of health. Subjects Women (n 1441) and men (n 1127) who participated in the three phases of data collection and had complete information for all study variables. Results Women and men with less than high school, or only a high school education, gained approximately 1 kg/m2 more than college graduates (women: 1·06 kg/m2 (P<0·001) and 1·06 kg/m2 (P<0·001), respectively; men: 1·04 kg/m2 (P=0·013) and 1·01 kg/m2 (P=0·277), respectively). For women only, race was independently associated with weight gain. Women identifying as Pardo or Black gained 1·03 kg/m2 (P=0·01) and 1·02 kg/m2 (P=0·10), respectively, more than Whites. No effect modification by race was observed for either men or women. Conclusions While both lower education and darker race were associated with greater weight gain, gender similarities and differences were observed in these associations. The relationship between weight gain and different indicators of social status are therefore complex and require careful consideration when addressing the obesity epidemic.

Joint analysis of longitudinal and survival AIDS data with a spatial fraction of long‐term survivors: A Bayesian approach

A typical survival analysis with time-dependent covariates usually does not take into account the possible random fluctuations or the contamination by measurement errors of the variables. Ignoring these sources of randomness may cause bias in the estimates of the model parameters. One possible way for overcoming that limitation is to consider a longitudinal model for the time-varying covariates jointly with a survival model for the time to the event of interest, thereby taking advantage of the complementary information flowing between these two-model outcomes. We employ here a Bayesian hierarchical approach to jointly model spatial-clustered survival data with a fraction of long-term survivors along with the repeated measurements of CD4+ T lymphocyte counts for a random sample of 500 HIV/AIDS individuals collected in all the 27 states of Brazil during the period 2002-2006. The proposed Bayesian joint model comprises two parts: on the one hand, a flexible model using Penalized Splines to better capture the nonlinear behavior of the different CD4 profiles over time; on the other hand, a spatial cure model to cope with the set of long-term survivor individuals. Our findings show that joint models considering this set of patients were the ones with the best performance comparatively to the more traditional survival approach. Moreover, the use of spatial frailties allowed us to map the heterogeneity in the disease risk among the Brazilian states.

Hierarchical Normal Mixture Model to Analyse HIV/AIDS LOS

Inpatient length of stay (LOS) is an important measure of hospital activity and is often considered as a proxy of hospital resource consumption. In Portugal, hospitalizations related to HIV infection are some of the most expensive and the second major diagnosis category with greatest average LOS. This paper investigates factors associated with HIV/AIDS LOS. A hierarchical finite normal mixture model was fitted to the logarithm of LOS, to account for the inherent correlation of patients clustered in hospitals. We found that the model with two components had the best fit. In addition associated risk factors were identified for each component and the random effects make possible a comparison of relative efficiencies among hospitals.

Analysis of HIV/AIDS DRG in Portugal: a hierarchical finite mixture model

Inpatient length of stay (LOS) is an important measure of hospital activity, but its empirical distribution is often positively skewed, representing a challenge for statistical analysis. Taking this feature into account, we seek to identify factors that are associated with HIV/AIDS through a hierarchical finite mixture model. A mixture of normal components is applied to adult HIV/AIDS diagnosis-related group data (DRG) from 2008. The model accounts for the demographic and clinical characteristics of the patients, as well the inherent correlation of patients clustered within hospitals. In the present research, a normal mixture distribution was fitted to the logarithm of LOS and it was found that a model with two-components had the best fit, resulting in two subgroups of LOS: a short-stay subgroup and a long-stay subgroup. Associated risk factors for both groups were identified as well as some statistical differences in the hospitals. Our findings provide important information for policy makers in terms of discharge planning and the efficient management of LOS. The presence of “atypical” hospitals also suggests that hospitals should not be viewed or treated as homogenous bodies.