Elisabeth Carmen Duarte
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Elisabeth Carmen Duarte has a total of 30 co-authors. Here are the top co-authors by number of shared publications:
- Leila Posenato Garcia
Fundação Jorge Duprat Figueiredo de Segurança e Medicina do Trabalho
5 shared publications Pedro Luiz Tauil
Faculdade de Medicina, Universidade de Brasília, Brasilia, DF
3 shared publications- Nereu Henrique Mansano
University of Brasilia
2 shared publications - Rui Moreira Braz
Secretaria de Vigilância em Saúde, Departamento de Vigilância de Doenças Transmissíveis, Brasília-DF, Brasil
2 shared publications - Walter Massa Ramalho
Universidade de Brasília, Brasil
2 shared publications - Ana Luiza Bierrenbach
Hospital Sírio-Libanês, Brazil
1 shared publication
Latest Publications
- Open Access
- 12 May 2025
- Published by MDPI AG in Tropical Medicine and Infectious Disease
- Vol. 10 (5) , 131
Evaluating Machine Learning Models for Predicting Late Leprosy Diagnosis by Physical Disability Grade in Brazil (2018–2022)
The severity of physical disability at leprosy diagnosis reflects the timeliness of case detection and the effectiveness of disease surveillance. This study evaluates machine learning models to predict factors associated with late leprosy diagnosis—defined as grade 2 physical disability (G2D)—in Brazil from 2018 to 2022. Using an observational cross-sectional design, we analyzed data from the Notifiable Diseases Information System and trained four machine learning models: Random Forest, LightGBM, CatBoost, XGBoost, and an Ensemble model. Model performance was assessed through accuracy, area under the receiver operating characteristic curve (AUC-ROC), recall, precision, F1 score, specificity, and Matthew’s correlation coefficient (MCC). An increasing trend in G2D prevalence was observed, averaging 11.6% over the study period and rising to 13.1% in 2022. The Ensemble model and LightGBM demonstrated the highest predictive performance, particularly in the north and northeast regions (accuracy: 0.85, AUC-ROC: 0.93, recall: 0.90, F1 score: 0.83, MCC: 0.70), with similar results in other regions. Key predictors of G2D included the number of nerves affected, clinical form, education level, and case detection mode. These findings underscore the potential of machine learning to enhance early detection strategies and reduce the burden of disability in leprosy, particularly in regions with persistent health disparities.
- 12 May 2025
- Published by MDPI AG in Tropical Medicine and Infectious Disease
- Vol. 10 (5) , 131
Open AccessEvaluating Machine Learning Models for Predicting Late Leprosy Diagnosis by Physical Disability Grade in Brazil (2018–2022)
The severity of physical disability at leprosy diagnosis reflects the timeliness of case detection and the effectiveness of disease surveillance. This study evaluates machine learning models to predict factors associated with late leprosy diagnosis—defined as grade 2 physical disability (G2D)—in Brazil from 2018 to 2022. Using an observational cross-sectional design, we analyzed data from the Notifiable Diseases Information System and trained four machine learning models: Random Forest, LightGBM, CatBoost, XGBoost, and an Ensemble model. Model performance was assessed through accuracy, area under the receiver operating characteristic curve (AUC-ROC), recall, precision, F1 score, specificity, and Matthew’s correlation coefficient (MCC). An increasing trend in G2D prevalence was observed, averaging 11.6% over the study period and rising to 13.1% in 2022. The Ensemble model and LightGBM demonstrated the highest predictive performance, particularly in the north and northeast regions (accuracy: 0.85, AUC-ROC: 0.93, recall: 0.90, F1 score: 0.83, MCC: 0.70), with similar results in other regions. Key predictors of G2D included the number of nerves affected, clinical form, education level, and case detection mode. These findings underscore the potential of machine learning to enhance early detection strategies and reduce the burden of disability in leprosy, particularly in regions with persistent health disparities.
- Open Access
- 2 December 2017
- Published by Springer Nature in BMC Public Health
- Vol. 17 (1) , 924
Outcomes of Zika virus infection during pregnancy: contributions to the debate on the efficiency of cohort studies
Zika infection during pregnancy (ZIKVP) is known to be associated with adverse outcomes. Studies on this matter involve both rare outcomes and rare exposures and methodological choices are not straightforward. Cohort studies will surely offer more robust evidences, but their efficiency must be enhanced. We aim to contribute to the debate on sample selection strategies in cohort studies to assess outcomes associated with ZKVP. A study can be statistically more efficient than another if its estimates are more accurate (precise and valid), even if the studies involve the same number of subjects. Sample size and specific design strategies can enhance or impair the statistical efficiency of a study, depending on how the subjects are distributed in subgroups pertinent to the analysis. In most ZIKVP cohort studies to date there is an a priori identification of the source population (pregnant women, regardless of their exposure status) which is then sampled or included in its entirety (census). Subsequently, the group of pregnant women is classified according to exposure (presence or absence of ZIKVP), respecting the exposed:unexposed ratio in the source population. We propose that the sample selection be done from the a priori identification of groups of pregnant women exposed and unexposed to ZIKVP. This method will allow for an oversampling (even 100%) of the pregnant women with ZKVP and a optimized sampling from the general population of pregnant women unexposed to ZIKVP, saving resources in the unexposed group and improving the expected number of incident cases (outcomes) overall. We hope that this proposal will broaden the methodological debate on the improvement of statistical power and protocol harmonization of cohort studies that aim to evaluate the association between Zika infection during pregnancy and outcomes for the offspring, as well as those with similar objectives.
- 2 December 2017
- Published by Springer Nature in BMC Public Health
- Vol. 17 (1) , 924
Open AccessOutcomes of Zika virus infection during pregnancy: contributions to the debate on the efficiency of cohort studies
Zika infection during pregnancy (ZIKVP) is known to be associated with adverse outcomes. Studies on this matter involve both rare outcomes and rare exposures and methodological choices are not straightforward. Cohort studies will surely offer more robust evidences, but their efficiency must be enhanced. We aim to contribute to the debate on sample selection strategies in cohort studies to assess outcomes associated with ZKVP. A study can be statistically more efficient than another if its estimates are more accurate (precise and valid), even if the studies involve the same number of subjects. Sample size and specific design strategies can enhance or impair the statistical efficiency of a study, depending on how the subjects are distributed in subgroups pertinent to the analysis. In most ZIKVP cohort studies to date there is an a priori identification of the source population (pregnant women, regardless of their exposure status) which is then sampled or included in its entirety (census). Subsequently, the group of pregnant women is classified according to exposure (presence or absence of ZIKVP), respecting the exposed:unexposed ratio in the source population. We propose that the sample selection be done from the a priori identification of groups of pregnant women exposed and unexposed to ZIKVP. This method will allow for an oversampling (even 100%) of the pregnant women with ZKVP and a optimized sampling from the general population of pregnant women unexposed to ZIKVP, saving resources in the unexposed group and improving the expected number of incident cases (outcomes) overall. We hope that this proposal will broaden the methodological debate on the improvement of statistical power and protocol harmonization of cohort studies that aim to evaluate the association between Zika infection during pregnancy and outcomes for the offspring, as well as those with similar objectives.
- Open Access
- 5 September 2016
- Published by Springer Nature in BMC Infectious Diseases
- Vol. 16 (1) , 1-10
Trends of main indicators of leprosy in Brazilian municipalities with high risk of leprosy transmission, 2001–2012
Leprosy incidence has reduced in recent years in Brazil, although the disease still persists as a public health problem in some regions. To investigate the trends of selected leprosy indicators in Brazilian municipalities with high risk of transmission is essential to provide effective control of the disease, yet this area has not been investigated. This is an ecological time-series study with multiple groups using Notifiable Diseases Information System (SINAN) data. All 692 municipalities of the states of Mato Grosso, Tocantins, Rondônia, Pará and Maranhão were included. The incidence rates of leprosy were calculated, as well as incidence rates in children under 15 years per 100,000 inhabitants and rates of new cases presenting grade-2 disabilities per 100,000 inhabitants. Joinpoint Regression was used to analyse the time trends of the different indicators studied. The spatial distribution of temporal variations of the indicators in the period was presented. Between 2001 and 2012, 176,929 leprosy cases were notified in the area studied, this being equivalent to 34.6 % of total cases in Brazil. In the aggregate of municipalities, there was a reduction in incidence rate of leprosy from 89.10 to 56.98 new cases per 100,000 inhabitants between 2001 and 2012, with a significant reduction between 2003 and 2012 (APC: − 6.2 %, 95 % CI: −7.2 % to −5.2 %). The incidence rate in <15 years also reduced significantly between 2003 and 2012 (APC: −5.6 %; 95 % CI: −7.2 % to −4.1 %). The rate of new cases with grade 2 disability remained stable between 2001 and 2012 (APC: −1.3 %; 95 % CI: −2.6 % to 0.1 %). Despite the reduction in the leprosy incidence rate, strategies for controlling this disease need to be enhanced to enable early case detection, especially in hyperendemic municipalities, in order to prevent disability.
- 5 September 2016
- Published by Springer Nature in BMC Infectious Diseases
- Vol. 16 (1) , 1-10
Open AccessTrends of main indicators of leprosy in Brazilian municipalities with high risk of leprosy transmission, 2001–2012
Leprosy incidence has reduced in recent years in Brazil, although the disease still persists as a public health problem in some regions. To investigate the trends of selected leprosy indicators in Brazilian municipalities with high risk of transmission is essential to provide effective control of the disease, yet this area has not been investigated. This is an ecological time-series study with multiple groups using Notifiable Diseases Information System (SINAN) data. All 692 municipalities of the states of Mato Grosso, Tocantins, Rondônia, Pará and Maranhão were included. The incidence rates of leprosy were calculated, as well as incidence rates in children under 15 years per 100,000 inhabitants and rates of new cases presenting grade-2 disabilities per 100,000 inhabitants. Joinpoint Regression was used to analyse the time trends of the different indicators studied. The spatial distribution of temporal variations of the indicators in the period was presented. Between 2001 and 2012, 176,929 leprosy cases were notified in the area studied, this being equivalent to 34.6 % of total cases in Brazil. In the aggregate of municipalities, there was a reduction in incidence rate of leprosy from 89.10 to 56.98 new cases per 100,000 inhabitants between 2001 and 2012, with a significant reduction between 2003 and 2012 (APC: − 6.2 %, 95 % CI: −7.2 % to −5.2 %). The incidence rate in <15 years also reduced significantly between 2003 and 2012 (APC: −5.6 %; 95 % CI: −7.2 % to −4.1 %). The rate of new cases with grade 2 disability remained stable between 2001 and 2012 (APC: −1.3 %; 95 % CI: −2.6 % to 0.1 %). Despite the reduction in the leprosy incidence rate, strategies for controlling this disease need to be enhanced to enable early case detection, especially in hyperendemic municipalities, in order to prevent disability.
- 1 November 2015
- Vol. 38 (5) , 418-24
[Alcohol use as a necessary cause of death in Brazil, 2010-2012].
To describe mortality from diseases, conditions, and injuries for which alcohol consumption is a necessary cause during the 2010-2012 triennium in Brazil. A descriptive study was conducted with data from the Brazilian Ministry of Health's Mortality Information System (SIM). The analysis included deaths whose primary cause was classified as any of the 78 codes of the International Statistical Classification of Diseases and Related Health Problems (ICD-10) corresponding to the diseases, conditions, and injuries for which the use of alcohol is a necessary cause. Deaths with alcohol consumption as a necessary cause totaled 55 380 (88.5% in men). The crude mortality rate for the triennium was 9.6/100 000 people in the overall population, 17.35/100 000 men in males, and 2.15/100 000 women in females. Higher mortality rates were observed in the 50-59 year (28.45) and 60-69 year (27.23) age groups and among people with black or brown skin color (10.15). The Northeast (11.70) and Midwest (11.04) regions exhibited higher age-adjusted mortality rates. Liver diseases were the leading cause of death (55.3%). Mortality from causes related to alcohol consumption is high in Brazil, especially among men, people aged 50-69 years, and residents in the Northeast and Midwest regions.
- 1 November 2015
- Vol. 38 (5) , 418-24
[Alcohol use as a necessary cause of death in Brazil, 2010-2012].
To describe mortality from diseases, conditions, and injuries for which alcohol consumption is a necessary cause during the 2010-2012 triennium in Brazil. A descriptive study was conducted with data from the Brazilian Ministry of Health's Mortality Information System (SIM). The analysis included deaths whose primary cause was classified as any of the 78 codes of the International Statistical Classification of Diseases and Related Health Problems (ICD-10) corresponding to the diseases, conditions, and injuries for which the use of alcohol is a necessary cause. Deaths with alcohol consumption as a necessary cause totaled 55 380 (88.5% in men). The crude mortality rate for the triennium was 9.6/100 000 people in the overall population, 17.35/100 000 men in males, and 2.15/100 000 women in females. Higher mortality rates were observed in the 50-59 year (28.45) and 60-69 year (27.23) age groups and among people with black or brown skin color (10.15). The Northeast (11.70) and Midwest (11.04) regions exhibited higher age-adjusted mortality rates. Liver diseases were the leading cause of death (55.3%). Mortality from causes related to alcohol consumption is high in Brazil, especially among men, people aged 50-69 years, and residents in the Northeast and Midwest regions.
- Open Access
- 1 October 2015
- Published by FapUNIFESP (SciELO) in Epidemiologia e Serviços de Saúde
- Vol. 24 (4) , 803-804
Resenha do livro Saúde Brasil 2014: uma análise da situação de saúde e das causas externas
- 1 October 2015
- Published by FapUNIFESP (SciELO) in Epidemiologia e Serviços de Saúde
- Vol. 24 (4) , 803-804
Open AccessResenha do livro Saúde Brasil 2014: uma análise da situação de saúde e das causas externas
- Open Access
- 1 December 2014
- Published by FapUNIFESP (SciELO) in Revista da Sociedade Brasileira de Medicina Tropical
- Vol. 47 (6) , 763-769
The changing distribution of malaria in the Brazilian Amazon, 2003-2004 and 2008-2009
Key words: Malaria; Risk ratio; Gini coefficient; Lorenz curve; Brazilian Amazon
- 1 December 2014
- Published by FapUNIFESP (SciELO) in Revista da Sociedade Brasileira de Medicina Tropical
- Vol. 47 (6) , 763-769
Open AccessThe changing distribution of malaria in the Brazilian Amazon, 2003-2004 and 2008-2009
Key words: Malaria; Risk ratio; Gini coefficient; Lorenz curve; Brazilian Amazon
- 1 March 2014
- Vol. 35 (3) , 186-92
[An algorithm to monitor the incidence of malaria in the Brazilian Amazon, 2003 to 2010].
To evaluate an algorithm developed for detecting variations in the incidence of malaria in the Brazilian Amazon. An evaluation was conducted of an automated monitoring system based on an algorithm that had been previously developed by the authors. The algorithm employs quartile diagrams to classify municipalities according to variations in the incidence of malaria: group 1 (reduced incidence)-municipalities with below the expected incidence rates; group 2 (expected incidence)- within the expected incidence rates; group 3 (epidemics)-higher than the expected incidence rates; and group 4 (sporadic case)-a single case during a year. The period from 2003 to 2010 was analyzed. All the municipalities in the nine states that make up the Brazilian Amazon were studied (805 municipalities in 2003 and 807 starting in 2004). Based on this method, Amazonian municipalities were classified as follows: group 1, 152 (18.8%) municipalities in 2003 and 109 (13.5%) in 2010; group 2, 206 (25.6%) municipalities in 2003 and 331 (41.0%) in 2010; group 3, 391 (48.6%) municipalities in 2003 and 308 (38.2%) in 2010; and group 4, 56 (7.0%) municipalities in 2003 and 59 (7.3%) in 2010. The use of the algorithm revealed that in 2010, the number of Amazonian municipalities in group 3 (epidemics) decreased when compared to 2003, while the number of municipalities in group 2 (expected incidence) increased. In the same period, there was no significant variation in the number of municipalities in group 1 (reduced incidence) and in group 4 (sporadic case).
- 1 March 2014
- Vol. 35 (3) , 186-92
[An algorithm to monitor the incidence of malaria in the Brazilian Amazon, 2003 to 2010].
To evaluate an algorithm developed for detecting variations in the incidence of malaria in the Brazilian Amazon. An evaluation was conducted of an automated monitoring system based on an algorithm that had been previously developed by the authors. The algorithm employs quartile diagrams to classify municipalities according to variations in the incidence of malaria: group 1 (reduced incidence)-municipalities with below the expected incidence rates; group 2 (expected incidence)- within the expected incidence rates; group 3 (epidemics)-higher than the expected incidence rates; and group 4 (sporadic case)-a single case during a year. The period from 2003 to 2010 was analyzed. All the municipalities in the nine states that make up the Brazilian Amazon were studied (805 municipalities in 2003 and 807 starting in 2004). Based on this method, Amazonian municipalities were classified as follows: group 1, 152 (18.8%) municipalities in 2003 and 109 (13.5%) in 2010; group 2, 206 (25.6%) municipalities in 2003 and 331 (41.0%) in 2010; group 3, 391 (48.6%) municipalities in 2003 and 308 (38.2%) in 2010; and group 4, 56 (7.0%) municipalities in 2003 and 59 (7.3%) in 2010. The use of the algorithm revealed that in 2010, the number of Amazonian municipalities in group 3 (epidemics) decreased when compared to 2003, while the number of municipalities in group 2 (expected incidence) increased. In the same period, there was no significant variation in the number of municipalities in group 1 (reduced incidence) and in group 4 (sporadic case).
- Open Access
- 1 January 2014
- Published by FapUNIFESP (SciELO) in Revista Brasileira de Epidemiologia
- Vol. 17 (suppl 1) , 3-16
Adolescent drivers in Brazil: prevalence and associated factors estimated from the National Adolescent School-based Health Survey (PeNSE 2012).
There is evidence that adolescent drivers can represent risks to themselves, to passengers and to society in general.
- 1 January 2014
- Published by FapUNIFESP (SciELO) in Revista Brasileira de Epidemiologia
- Vol. 17 (suppl 1) , 3-16
Open AccessAdolescent drivers in Brazil: prevalence and associated factors estimated from the National Adolescent School-based Health Survey (PeNSE 2012).
There is evidence that adolescent drivers can represent risks to themselves, to passengers and to society in general.
- Open Access
- 1 June 2013
- Published by FapUNIFESP (SciELO) in Epidemiologia e Serviços de Saúde
- Vol. 22 (2) , 203-214
Homicídios em homens jovens de 10 a 24 anos e condições sociais em municípios do Paraná e Santa Catarina, Brasil, 2001 - 2010
- 1 June 2013
- Published by FapUNIFESP (SciELO) in Epidemiologia e Serviços de Saúde
- Vol. 22 (2) , 203-214
Open AccessHomicídios em homens jovens de 10 a 24 anos e condições sociais em municípios do Paraná e Santa Catarina, Brasil, 2001 - 2010
- 1 May 2013
- Vol. 29 (5) , 935-44
[Characteristics of malaria epidemics in the municipalities of the Brazilian Amazon, 2010].
Malaria epidemics occur annually in various municipalities (counties) in the Brazilian Amazon. However, health services do not systematically adopt tools to detect and promptly control these events. This article aimed to characterize malaria epidemics in the Brazilian Amazon Region based on their duration, the Plasmodium species involved, and the population's degree of vulnerability. An automatic malaria incidence monitoring system based on quartiles was assessed for prompt identification of malaria epidemics. In 2010, epidemics were identified in 338 (41.9%) of the counties in the Brazilian Amazon. P. falciparum and P. vivax epidemics were detected, both singly and in combination. Epidemics lasted from 1 to 4 months in 58.3% of the counties, 5 to 8 months in 34.5%, and 9 to 12 months in 17.4%. Systematic monitoring of malaria incidence could contribute to early detection of epidemics and improve the effectiveness of control measures.
- 1 May 2013
- Vol. 29 (5) , 935-44
[Characteristics of malaria epidemics in the municipalities of the Brazilian Amazon, 2010].
Malaria epidemics occur annually in various municipalities (counties) in the Brazilian Amazon. However, health services do not systematically adopt tools to detect and promptly control these events. This article aimed to characterize malaria epidemics in the Brazilian Amazon Region based on their duration, the Plasmodium species involved, and the population's degree of vulnerability. An automatic malaria incidence monitoring system based on quartiles was assessed for prompt identification of malaria epidemics. In 2010, epidemics were identified in 338 (41.9%) of the counties in the Brazilian Amazon. P. falciparum and P. vivax epidemics were detected, both singly and in combination. Epidemics lasted from 1 to 4 months in 58.3% of the counties, 5 to 8 months in 34.5%, and 9 to 12 months in 17.4%. Systematic monitoring of malaria incidence could contribute to early detection of epidemics and improve the effectiveness of control measures.