Cibele Comini César

Universidade Federal da Bahia, Bahia, Estado de Bahía, Brazil

Are you Cibele Comini César?

Claim your profile

Publications (80)86.74 Total impact

  • Source
    [show abstract] [hide abstract]
    ABSTRACT: The measurement of CD4+ T-cell (CD4) counts is a strong predictor of progression to AIDS and a means of monitoring antiviral therapy (ART). The success or failure of controlling virus levels in untreated patients or those taking ART may be associated with treatment adherence, habits, correlated infections unrelated to HIV, cancer, immunosuppressive drugs; as well as socio-economic and psychosocial aspects and access to healthcare. The aim of the present study was to identify, using a multilevel model, the factors related to the variations of CD4 counts over time, in patients living with HIV. A cohort study was conducted with patients living with HIV, selected from July 2007 to December 2010. Patients were monitored from records of their first CD4 count after being diagnosed with HIV. A multilevel model with 3 levels of aggregation was applied to analyze the associations of predictor variables and the behavior of CD4 over time. A total of 1870 patients were enrolled. The mean number of CD4 at the beginning of the cohort was 393.1 cells/mm(3), and there was a mean increase of 1.529 cells/mm(3) per month. Patient's age, smoking, use of illicit drugs, hospital treatment, changing doctors and the use of ART, were factors that affected the kinetics of the CD4 count during the follow-up period. The results of this study indicated increased levels of CD4 over time in a cohort of patients living with HIV/AIDS and identified factors that may influence this increase and are liable to intervention.
    PLoS ONE 01/2014; 9(2):e84276. · 3.73 Impact Factor
  • [show abstract] [hide abstract]
    ABSTRACT: Work ability depends on a balance between individual resources and work demands. This study evaluated factors that are associated with inadequate work ability among workers in the clothing industry. We conducted a cross-sectional observational study of 306 workers in 40 small and medium-sized enterprises. We assessed work ability, individual resources, physical and psychosocial demands, and aspects of life outside work using a binary logistic regression model with hierarchical data entry. The mean work ability was 42.5 (SD=3.5); when adjusted for age, only 11% of the workers showed inadequate work ability. The final model revealed that smoking, high isometric physical load, and poor physical environmental conditions were the most significant predictors of inadequate work ability. Good working conditions and worker education must be implemented to eliminate factors that can be changed and that have a negative impact on work ability. These initiatives include anti-smoking measures, improved postures at work, and better physical environmental conditions.
    Work 11/2013; · 0.52 Impact Factor
  • [show abstract] [hide abstract]
    ABSTRACT: The study aimed to estimate the prevalence of hypertension in members of quilombos (slavedescendant African-Brazilian communities) and to evaluate possible associated factors. A population-based survey included 797 individuals 18 years or older. Hypertension was defined as systolic blood pressure > 140mmHg or diastolic blood pressure > 90mmHg or reported use of antihypertensive medication. We used Poisson regression with robust variance and hierarchical entry of variables. Prevalence of hypertension was 45.4% (95%CI: 41.89-48.85). Factors associated with hypertension were: neighborhood security (distal), age, economic class, education, and physical inactivity (intermediate), and body mass index (proximal). Health promotion is needed in the quilombo communities through action at the individual and population levels. The high prevalence of hypertension highlights the need for broad access to health services for prevention, early diagnosis, and guidelines for adequate management.
    Cadernos de saúde pública / Ministério da Saúde, Fundação Oswaldo Cruz, Escola Nacional de Saúde Pública 09/2013; 29(9):1889-902. · 0.83 Impact Factor
  • [show abstract] [hide abstract]
    ABSTRACT: The objective of this study was to analyze factors associated with access to kidney transplants from living and cadaver donors in Belo Horizonte, Minas Gerais State, Brazil. The authors conducted a non-concurrent cohort study of patients on the waiting list for kidney transplants from 2000 to 2004 and followed until transplantation, death, exclusion, or continued presence on the line at the end of the study on December 31, 2005. The Cox model was used for competing risks. Of the 835 patients, 22.7% were transplanted. Lower risk of transplantation from living donors and cadavers was observed in patients with more time on dialysis and blood type O. Lower risk of transplantation from living donors was observed in residents in a high health risk area and in recipients with diabetes. The greatest disparity in access was observed in transplants from living donors, since there were no significant socio-demographic differences in transplants from cadaver donors. One can infer that the organ allocation system contributed to mitigating socio-demographic inequalities, and that clinical issues were more relevant in access to transplants from cadaver donors.
    Cadernos de saúde pública / Ministério da Saúde, Fundação Oswaldo Cruz, Escola Nacional de Saúde Pública 12/2012; 28(12):2315-26. · 0.83 Impact Factor
  • [show abstract] [hide abstract]
    ABSTRACT: PURPOSE: The aim of this study was to assess the agreement of four renowned interaction lists on potentially severe warfarin drug interactions (DI) in outpatients at a university hospital in Brazil, specifically in subgroups of Trypanosoma cruzi-infected and non-infected patients and those with previous bleeding episodes. METHODS: This was a cross-sectional study in which adult outpatients with heart disease and indications for chronic warfarin use were enrolled. The occurrence of potentially severe warfarin DI was evaluated based on the lists provided by three compendia, i.e., Drug Interaction Facts (DIF), Drug Interactions: Analysis and Management (DIAM) and DRUG-REAX, and by the World Health Organization (WHO) Model Formulary. A kappa coefficient was used to calculate the agreement among the sources. RESULTS: A total of 280 patients were studied. Most patients were female (54.6 %) with an average age of 56.8 (standard deviation 13.1) years. The agreement among the four sources was fair (Fleiss' kappa coefficient = 0.295). T. cruzi-infected individuals were less likely to have severe warfarin DI than non-infected patients (p < 0.05 for DIAM, DRUG-REAX and the WHO Model Formulary). Potentially severe DI were more frequent in patients with previous bleeding episodes, based on the DIF compendia (p = 0.007). CONCLUSIONS: This evaluation of warfarin DI revealed that the disagreement between compendia is also observed in clinical practice. T. cruzi infection is associated with a lower prevalence of potentially severe warfarin DI, but with a wider variation in its detection. Our results suggest a wide spectrum of discrepancies in detecting heart disease patients at higher risk for severe warfarin DI and a possible heterogeneity in clinical guidance.
    European Journal of Clinical Pharmacology 09/2012; · 2.74 Impact Factor
  • [show abstract] [hide abstract]
    ABSTRACT: There is mixed evidence that socioeconomic status (SES) affects the predictive power of self-rated health (SRH) for mortality. We sought to compare the predictive value of SRH for 6-year mortality in English and Brazilian older adults, and to assess whether this association varies by SES in these populations. Data came from the English and the Bambui (Brazil) cohort studies of aging. Potential confounding variables included sociodemographic characteristics, lifestyle, self-reported diseases, physical functioning, mental symptoms, and selected biomarker measures. Participants were 5183 English and 1499 Brazilians aged 60 years and over. Low health ratings were independently associated with subsequent mortality in both populations. However, the predictive power of poor SRH for death was much higher for English (a population with higher SES level) than for Brazilians (adjusted hazard ratios 4.45 [95% confidence interval, 3.04-6.51] and 1.88 [1.25-2.81], respectively). In both populations, the predictive value of SRH for mortality was higher among those in the highest income tertile. Our results suggest that the association between SRH and mortality is underestimated in populations and in subgroups of population with low SES level. Further international research is needed to examine the generalizability of this pattern.
    Annals of epidemiology 07/2012; 22(9):644-8. · 2.95 Impact Factor
  • [show abstract] [hide abstract]
    ABSTRACT: Although specific measurement instruments are necessary to better understand the relationship between features of neighborhoods and health, very few studies have developed instruments to measure neighborhood features in developing countries. The objective of the study was to develop valid and reliable measures of neighborhood context useful in a Latin American urban context, assess their psychometric and ecometric properties, and examine individual and neighborhood-level predictors of these measures. We analyzed data from a multistage household survey (2008-2009) conducted in Belo Horizonte City by the Observatory for Urban Health. One adult in each household was selected to answer a questionnaire that included scales to measure neighborhood domains. Census tracts were used to proxy neighborhoods. Internal consistency was evaluated by Cronbach's alpha, and multilevel models were used to estimate ecometric properties and to estimate associations of neighborhood measures with socioeconomic indicators. The final sample comprised 4048 survey respondents representing 149 census tracts. We assessed ten neighborhood environment dimensions: public services, aesthetic quality, walking environment, safety, violence, social cohesion, neighborhood participation, neighborhood physical disorder, neighborhood social disorder, and neighborhood problems. Cronbach's alpha coefficients ranged from 0.53 to 0.83; intraneighborhood correlations ranged from 0.02 to 0.53, and neighborhood reliability varied from 0.76 to 0.99. Most scales were associated with individual and neighborhood socioeconomic predictors. Questionnaires can be used to reliably measure neighborhood contexts in developing countries.
    Journal of Urban Health 06/2012; · 1.89 Impact Factor
  • [show abstract] [hide abstract]
    ABSTRACT: To assess individual and/or health service factors associated with patients returning for results of HIV or sexually transmitted infection (STI) tests in mental health centers. Cross-sectional national multicenter study among 2,080 patients randomly selected from 26 Brazilian mental health centers in 2007. Multilevel logistic regression was used to assess the effect of individual (level 1) and mental health service characteristics (level 2) on receipt of test results. The rate of returning HIV/STI test results was 79.6%. Among health service characteristics examined, only condom distribution was associated with receiving HIV/STI test results, whereas several individual characteristics were independently associated including living in the same city where treatment centers are; being single; not having heard of AIDS; and not having been previously HIV tested. It is urgent to expand HIV/STI testing in health services which provide care for patients with potentially increased vulnerability to these conditions, and to promote better integration between mental health and health services.
    Revista de saude publica 02/2012; 46(2):290-9. · 1.01 Impact Factor
  • [show abstract] [hide abstract]
    ABSTRACT: Interest in self-rated health (SRH) as a tool for use in disease and mortality risk screening is increasing. The authors assessed the discriminatory ability of baseline SRH to predict 10-year mortality rates compared with objectively measured health status. Principal component analysis was used to create a health score that included systolic blood pressure, presence of diabetes mellitus, body mass index, electrocardiographic parameters, B-type natriuretic peptide, and other biochemical and hematologic measures. From 1997 to 2007, a total of 474 of the 1,388 baseline participants died and 81 were lost to follow-up, yielding 11,833 person-years of observation. The adjusted hazard ratio for death was 1.74 (95% confidence interval (CI): 1.32, 2.29) for persons reporting poor health versus those reporting good health. When combined with age and sex, SRH had a C statistic to predict death equal to 0.69 (95% CI: 0.67, 0.71), which was comparable to that of the inclusive health score (C = 0.69, 95% CI: 0.67, 0.72). The addition of other parameters, such as lifestyle, physical functioning, mental symptoms, and physical symptoms, had little effect on these 2 predictive models (C = 0.71 (95% CI: 0.69, 0.73) and C = 0.71 (95% CI: 0.69, 0.74), respectively). The abilities of the SRH and the health score models to predict death decreased in parallel fashion over time. These results suggest that older adults who report poor health warrant particular attention as persons who have accumulated biologic markers of disease.
    American journal of epidemiology 12/2011; 175(3):228-35. · 5.59 Impact Factor
  • [show abstract] [hide abstract]
    ABSTRACT: Researchers in the health field often deal with the problem of incomplete databases. Complete Case Analysis (CCA), which restricts the analysis to subjects with complete data, reduces the sample size and may result in biased estimates. Based on statistical grounds, Multiple Imputation (MI) uses all collected data and is recommended as an alternative to CCA. Data from the study Saúde em Beagá, attended by 4,048 adults from two of nine health districts in the city of Belo Horizonte, Minas Gerais State, Brazil, in 2008-2009, were used to evaluate CCA and different MI approaches in the context of logistic models with incomplete covariate data. Peculiarities in some variables in this study allowed analyzing a situation in which the missing covariate data are recovered and thus the results before and after recovery are compared. Based on the analysis, even the more simplistic MI approach performed better than CCA, since it was closer to the post-recovery results.
    Cadernos de saúde pública / Ministério da Saúde, Fundação Oswaldo Cruz, Escola Nacional de Saúde Pública 12/2011; 27(12):2299-313. · 0.83 Impact Factor
  • [show abstract] [hide abstract]
    ABSTRACT: This study aimed to analyze and compare the quality of life of renal replacement therapy patients undergoing hemodialysis, peritoneal dialysis and those with renal transplantation in Brazil. In addition, we aimed to verify factors associated with patients' quality of life and the relationship between quality of life and treatment modality, socioeconomic and demographic conditions as well as aspects related to the disease and health services. A representative sample of the dialysis units and transplant centers was obtained. Structured questionnaires were used to interview 3,036 patients in one of three treatment modalities: hemodialysis, peritoneal dialysis and renal transplant. Information was collected about socioeconomic and demographic characteristics and quality of life measures. There were significant differences between renal transplants and both forms of dialysis for all dimensions of the SF-36. Hemodialysis patients showed better results in the dimensions of functional capacity, physical aspects and social aspects, compared to peritoneal dialysis patients. Renal transplant patients had the best mean score in the physical component of quality of life. There were no significant differences among treatment groups regarding the mental component of quality of life. The physical and mental components were associated with comorbidities and age; however, older patients had better mental quality of life but worse physical quality of life. Patients in a higher socioeconomic class and patients that were not hospitalized also reported better quality of life. Unmarried and male patients presented better physical quality of life. The dialysis units and transplant centers influenced the patients' quality of life. Renal transplant patients have the best quality of life of the three treatment modalities. It is necessary to increase access to renal transplants.
    Quality of Life Research 09/2011; 21(6):983-91. · 2.41 Impact Factor
  • [show abstract] [hide abstract]
    ABSTRACT: To analyze the maximum axial force, the mean axial force, the amount of energy accumulated by the tongue, and the time taken to reach the maximum axial force, in different age ranges. The records of 92 individuals - students, staff and visitors at an university -, 29 (32.6%) men and 63 (67.4%) women, with ages between 14 and 53 years old, were analyzed. Subjects were divided into four age groups: 14 to 18 years, 19 to 23 years, 24 to 28 years, and 29 to 53 years. Each subject underwent clinical and instrumental assessment of the tongue. Instrumental assessment used FORLING. Data were statistically analyzed. Regarding the maximum force, the mean force and the tongue's accumulated energy, no differences were observed between groups. Regarding the time taken to reach the maximum force, the greatest values were obtained at the age range from 14 to 18 years (4.5 s), and the shortest values, at the age range from 19 to 23 years (3.1 s), with significant difference between the groups (p=0.001). Only the time taken to reach the tongue's maximum force is influenced by age range, indicating that teenagers are not able to reach the maximum lingual force as fast as young adults.
    Jornal da Sociedade Brasileira de Fonoaudiologia. 09/2011; 23(3):201-5.
  • [show abstract] [hide abstract]
    ABSTRACT: Objective To evaluate the existence of an association between the use of medication by nursing women and duration of breastfeeding in Itaúna, MG, Brazil.Methods Longitudinal cohort study was carried out with 12-month follow-up, including 246 women admitted in the only Maternity Hospital of the city of Itaúna, MG, Brazil. The effect of medications on duration of breastfeeding was analyzed with the Cox regression model, with time-dependent covariables. Drugs were classified for safety during breastfeeding according to criteria established by the American Academy of Pediatrics (2001) and Hale (2004).ResultsNinety-eight percent of nursing women used medications after hospital discharge. The duration of breastfeeding was longer for mothers who did not use medications or who used drugs considering safety during lactation (P<0.05).Conclusions Health professionals should prescribe medications with well established safety for both infants and for the lactation process, so that maternal medication is compatible with breastfeeding.
    Asian Pacific Journal of Tropical Disease. 09/2011; 1(3):216–221.
  • [show abstract] [hide abstract]
    ABSTRACT: To assess the validity of the estimates obtained through telephone survey and to measure the impact of the post-stratification weighting factor to adjust estimates. The same questionnaire was completed by two independent samples of the population living in the municipality of Belo Horizonte city (Barreiro and West regions). One sample (n=440) completed the questionnaire of VIGITEL 2008 (telephone survey), and the other (n=4,048) of Saúde em Beagá (face to face household interview). The results of the two samples for 18 health-related variables were compared by means of test statistics. At first, residents who had a landline telephone line were compared to those who reported not having a telephone line; then, VIGITEL estimates, with and without post-stratification weight, were compared with Saúde em Beagá estimates. Subjects who owned a landline telephone line had indicators for better economic conditions (housing, schooling, and skin color); higher prevalence of chronic diseases; lower exposure to risk factors for chronic diseases; and improved access to health services, compared to the those who reported not having a telephone line. Most VIGITEL estimates (without post-stratification weight) were similar to the sample of Saúde em Beagá that reported owning a residential landline, showing no major impact of the methodology to obtain this data (lower information bias). Even without post-stratification weight, VIGITEL estimates were similar to those of Saúde em Beagá. With post-stratification weight, the estimates of "number of residents", "skin color" and "physical activity" did not differ from those obtained by the face to face survey. The results of both surveys were very similar. Because of the lower cost, the telephone interview is a good option in public health for the behavioral risk-factor surveillance system.
    Revista Brasileira de Epidemiologia 09/2011; 14 Suppl 1:16-30.
  • Source
    Cibele C César, Marilia S Carvalho
    [show abstract] [hide abstract]
    ABSTRACT: Longitudinal studies often employ complex sample designs to optimize sample size, over-representing population groups of interest. The effect of sample design on parameter estimates is quite often ignored, particularly when fitting survival models. Another major problem in long-term cohort studies is the potential bias due to loss to follow-up. In this paper we simulated a dataset with approximately 50,000 individuals as the target population and 15,000 participants to be followed up for 40 years, both based on real cohort studies of cardiovascular diseases. Two sample strategies--simple random (our golden standard) and Stratified by professional group, with non-proportional allocation--and two loss to follow-up scenarios--non-informative censoring and losses related to the professional group--were analyzed. Two modeling approaches were evaluated: weighted and non-weighted fit. Our results indicate that under the correctly specified model, ignoring the sample weights does not affect the results. However, the model ignoring the interaction of sample strata with the variable of interest and the crude estimates were highly biased. In epidemiological studies misspecification should always be considered, as different sources of variability, related to the individuals and not captured by the covariates, are always present. Therefore, allowance must be made for the possibility of unknown confounders and interactions with the main variable of interest in our data. It is strongly recommended always to correct by sample weights.
    BMC Medical Research Methodology 06/2011; 11:99. · 2.21 Impact Factor
  • [show abstract] [hide abstract]
    ABSTRACT: Detecting potential drug interactions can lead to early interventions that protect patients from serious drug-related problems. The aim of this study was to evaluate the agreement among the lists of warfarin interactions provided by five information sources. The lists of warfarin interactions and the corresponding severity ratings and documentation levels presented by the three compendia and by the World Health Organization (WHO) Model Formulary were all compared, and each list was compared to that provided on the package insert of Marevan, a brand of warfarin. The compendia used were: Drug Interaction Facts, Drug Interactions: Analysis and Management and DRUG-REAX. A kappa coefficient was used to calculate the agreement among the sources. A total of 537 interactions were listed. Only 13 (2.4%) were common to the five sources. The global Fleiss' kappa coefficient was -0.0080, which indicated poor agreement. Eleven warfarin interactions appeared only in the Marevan package insert. Importantly, 243 interactions (45.3% of the total) were deemed significant in at least one compendium. Only two warfarin interactions were reported as critical by all three compendia and by WHO. The most critical interactions cited by the compendia were missing from the package insert. Poor agreement was found among five sources listing warfarin interactions. Potentially severe clinical consequences might occur due to these discrepant recommendations. Finally, the lack of standard terminology and clinical guidance, as well as the possible inaccuracy of severity ratings and documentation might contribute to heterogeneous procedures in clinical practice.
    European Journal of Clinical Pharmacology 06/2011; 67(12):1301-8. · 2.74 Impact Factor
  • [show abstract] [hide abstract]
    ABSTRACT: The objective of this present study is to analyze individual and contextual factors associated with access to renal transplant in Brazil. An observational, prospective and non-concurrent study was carried out, based on data from the National Database on renal replacement therapies in Brazil. Patients undergoing dialysis between 01/Jan/2000 and 31/Dec/2000 were included and monitored up to the point of transplant, death or until the end of the study period. Variables that were analyzed included: individual variables (age, sex, region of residence, primary renal disease, hospitalizations); and context variables concerning both the dialysis unit (level of complexity, juridical nature, hemodialysis machines and location) and the city (geographic region, location and HDI). Proportional hazard models were adjusted with hierarchical entry to identify factors associated with the risk of transplant. The results point to differentials in access according to socio-demographic, clinical, geographic and social factors, indicating that the organ allocation system has not eliminated avoidable disparities for those who compete for an organ in the nationwide waiting list.
    Cadernos de saúde pública / Ministério da Saúde, Fundação Oswaldo Cruz, Escola Nacional de Saúde Pública 01/2011; 27 Suppl 2:S284-97. · 0.83 Impact Factor
  • [show abstract] [hide abstract]
    ABSTRACT: In order to investigate the association between satisfaction with the neighborhood environment and self-rated health among older elderly, data from 814 participants of the eleventh wave of the Bambuí Cohort Study of Aging were analyzed using robust Poisson regression analyses. Those elderly with higher satisfaction with their neighborhoods (PR = 0.75; 95%CI: 0.63-0.87) were less likely to report worse self-rated health. The number of chronic diseases (two, PR = 1.69; 95%CI: 1.05-2.70, three or more, PR = 1.99, 95%CI: 1.27-3.13), difficulty in performing daily activities (PR = 1.51; 95%CI: 1.28-1.78), presence of depressive symptoms (PR = 1.68; 95%CI: 1.44-1.95) and frequency of leisure-time exercise in previous 90 days (less than once a week, PR =1.24; 95%CI: 1.03-1.50) were all positively and significantly associated with poor self-rated health. This study provided empirical evidence that satisfaction with the neighborhood environment was associated with the health of the older elderly. The findings further suggest the potential importance of including this indicator in analyses of place and health among the elderly.
    Cadernos de saúde pública / Ministério da Saúde, Fundação Oswaldo Cruz, Escola Nacional de Saúde Pública 01/2011; 27 Suppl 3:S390-8. · 0.83 Impact Factor
  • [show abstract] [hide abstract]
    ABSTRACT: The aim of this study was to evaluate the effect of birth cohort on nutritional status among older elderly (71-81 years old) from the Bambuí Cohort Study of Aging, conducted in Brazil in 1997 and 2008. We compared the two birth cohorts--1916-1926 (older cohort) and 1927-1937 (recent cohort)--considering body mass index (BMI = weight/height²), waist circumference (WC) and prevalence of overweight (BMI ³ 27 kg/m²). BMI (β = 0.09, 95%CI: 0.04, 0.15) and prevalence of overweight (PR = 1.02; 95%CI: 1.01; 1.03) were higher in the recent cohort than the earlier cohort, regardless of sex and schooling. No difference was observed in WC. Stratified by sex, similar overall trends were observed for men, and WC was higher in the recent cohort. Among women there was no difference in BMI and overweight, but WC was lower in the recent cohort. The cohort effect was greater among older men and, in the near future, may result in greater prevalence of overweight in this group.
    Cadernos de saúde pública / Ministério da Saúde, Fundação Oswaldo Cruz, Escola Nacional de Saúde Pública 01/2011; 27 Suppl 3:S418-26. · 0.83 Impact Factor
  • Source
    [show abstract] [hide abstract]
    ABSTRACT: A major goal of health systems is to reduce inequities in access to services, that is, to ensure that health care is provided based on health needs rather than social or economic factors. This study aims to identify the determinants of health services utilization among adults in a large Brazilian city and intraurban disparities in health care use. We combine household survey data with census-derived classification of social vulnerability of each household's census tract. The dependent variable was utilization of physician services in the prior 12 months, and the independent variables included predisposing factors, health needs, enabling factors, and context. Prevalence ratios and 95% confidence intervals were estimated by the Hurdle regression model, which combined Poisson regression analysis of factors associated with any doctor visits (dichotomous variable) and zero-truncated negative binomial regression for the analysis of factors associated with the number of visits among those who had at least one. Results indicate that the use of health services was greater among women and increased with age, and was determined primarily by health needs and whether the individual had a regular doctor, even among those living in areas of the city with the worst socio-environmental indicators. The experience of Belo Horizonte may have implications for other world cities, particularly in the development and use of a comprehensive index to identify populations at risk and in order to guide expansion of primary health care services as a means of enhancing equity in health.
    Journal of Urban Health 12/2010; 87(6):994-1006. · 1.89 Impact Factor

Publication Stats

211 Citations
86.74 Total Impact Points


  • 2013
    • Universidade Federal da Bahia
      Bahia, Estado de Bahía, Brazil
  • 2004–2013
    • Federal University of Minas Gerais
      • • Faculty of Medicine
      • • Departamento de Medicina Preventiva e Social
      Cidade de Minas, Minas Gerais, Brazil
  • 2012
    • Federal University of São João del-Rei
      São José del Rey, Minas Gerais, Brazil
  • 2010
    • Fundação Oswaldo Cruz
      • Centro de Pesquisas René Rachou - Fiocruz Minas
      Rio de Janeiro, Rio de Janeiro, Brazil
  • 2008
    • Unimed Belo Horizonte
      Cidade de Minas, Minas Gerais, Brazil
  • 2007
    • Universidade Federal de Ouro Preto
      • School of Nutrition
      Ouro Preto, Minas Gerais, Brazil