Development of CART model for prediction of tuberculosis treatment loss to follow up in the state of São Paulo, Brazil: A case–control study

Yamaguti, Verena Hokino; Alves, Domingos; Rijo, Rui; Miyoshi, Newton Shydeo Brandão; Ruffino-Netto, Antônio

doi:10.1016/j.ijmedinf.2020.104198

Cited by 5 publications

(5 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Several predictive models for TB treatment LTFU are available from other countries; for example, in Spain, those who live alone or in an institution, are immigrants, intravenous drug user (IVDU), patients with poor knowledge of TB and previously treated TB cases were identified as predictors for LTFU [ 33 ]. Another study from Brazil that used the Classification and Regression Trees (CART) prediction model ranked several important variables that contribute to LTFU, such as the number of doses taken during treatment, age, total number of people in the household, non-IVDU and HIV-negative patients [ 34 ]. In comparison with other prognostic models for loss to follow-up among general TB patients [ 33 , 34 ], our model identifies several predictive factors that are significant for TB smokers, such as young adults (age <50 years old) and HIV-positive individuals.…”

Section: Discussionmentioning

confidence: 99%

“…Another study from Brazil that used the Classification and Regression Trees (CART) prediction model ranked several important variables that contribute to LTFU, such as the number of doses taken during treatment, age, total number of people in the household, non-IVDU and HIV-negative patients [ 34 ]. In comparison with other prognostic models for loss to follow-up among general TB patients [ 33 , 34 ], our model identifies several predictive factors that are significant for TB smokers, such as young adults (age <50 years old) and HIV-positive individuals. Young adult who are in their working age population are mostly associated with smoking behavior and are at risk for LTFU from TB treatment [ 8 , 14 , 35 ].…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

T-BACCO SCORE: A predictive scoring tool for tuberculosis (TB) loss to follow-up among TB smokers

et al. 2023

View full text Add to dashboard Cite

Introduction Loss to follow-up (LTFU) and smoking during TB treatment are major challenges for TB control programs. Smoking increases the severity and prolongs TB treatment duration, which lead to a higher rate of LTFU. We aim to develop a prognostic scoring tool to predict LTFU among TB patients who smoke to improve successful TB treatment outcomes. Materials and methods The development of the prognostic model utilized prospectively collected longitudinal data of adult TB patients who smoked in the state of Selangor between 2013 until 2017, which were obtained from the Malaysian Tuberculosis Information System (MyTB) database. Data were randomly split into development and internal validation cohorts. A simple prognostic score (T-BACCO SCORE) was constructed based on the regression coefficients of predictors in the final logistic model of the development cohort. Estimated missing data was 2.8% from the development cohort and was completely at random. Model discrimination was determined using c-statistics (AUCs), and calibration was based on the Hosmer and Lemeshow goodness of fit test and calibration plot. Results The model highlights several variables with different T-BACCO SCORE values as predictors for LTFU among TB patients who smoke (e.g., age group, ethnicity, locality, nationality, educational level, monthly income level, employment status, TB case category, TB detection methods, X-ray categories, HIV status, and sputum status). The prognostic scores were categorized into three groups that predict the risk for LTFU: low-risk (<15 points), medium-risk (15 to 25 points) and high-risk (> 25 points). The model exhibited fair discrimination with a c-statistic of 0.681 (95% CI 0.627–0.710) and good calibration with a nonsignificant chi-square Hosmer‒Lemeshow’s goodness of fit test χ2 = 4.893 and accompanying p value of 0.769. Conclusion Predicting LTFU among TB patients who smoke in the early phase of TB treatment is achievable using this simple T-BACCO SCORE. The applicability of the tool in clinical settings helps health care professionals manage TB smokers based on their risk scores. Further external validation should be carried out prior to use.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

T-BACCO SCORE: A predictive scoring tool for tuberculosis (TB) loss to follow-up among TB smokers

et al. 2023

View full text Add to dashboard Cite

show abstract

“…Methodologically, our work was related to other studies predicting outcomes by inferring a model being trained by a set of historical data [15,22,25,40,[54][55][56]. Given appropriate assumptions, such techniques allow for valid predictions about the counterfactual outcomes under different settings for determining interventions.…”

Section: Plos Global Public Healthmentioning

confidence: 99%

“…These high treatment non-adherence rates possibly imply an existence of factors intrinsic to the patients or the treatment strategies amongst others. Indeed, several studies attempting to identify these patient factors aiming to avert high treatment non-adherence rates have been conducted [7,[13][14][15][16][17][18]. These studies employed both traditional statistics and epidemiological approaches utilizing logistic regression and other generalized linear models.…”

Section: Introductionmentioning

confidence: 99%

A machine learning approach to explore individual risk factors for tuberculosis treatment non-adherence in Mukono district

Gichuhi

Magumba

Kumar

et al. 2023

PLOS Glob Public Health

View full text Add to dashboard Cite

Despite the availability and implementation of well-known efficacious interventions for tuberculosis treatment by the Ministry of Health, Uganda (MoH), treatment non-adherence persists. Moreover, identifying a specific tuberculosis patient at risk of treatment non-adherence is still a challenge. Thus, this retrospective study, based on a record review of 838 tuberculosis patients enrolled in six health facilities, presents, and discusses a machine learning approach to explore the individual risk factors predictive of tuberculosis treatment non-adherence in the Mukono district, Uganda. Five classification machine learning algorithms, logistic regression (LR), artificial neural networks (ANN), support vector machines (SVM), random forest (RF), and AdaBoost were trained, and evaluated by computing their accuracy, F1 score, precision, recall, and the area under the receiver operating curve (AUC) through the aid of a confusion matrix. Of the five developed and evaluated algorithms, SVM (91.28%) had the highest accuracy (AdaBoost, 91.05% performed better than SVM when AUC is considered as evaluation parameter). Looking at all five evaluation parameters globally, AdaBoost is quite on par with SVM. Individual risk factors predictive of non-adherence included tuberculosis type, GeneXpert results, sub-country, antiretroviral status, contacts below 5 years, health facility ownership, sputum test results at 2 months, treatment supporter, cotrimoxazole preventive therapy (CPT) dapsone status, risk group, patient age, gender, middle and upper arm circumference, referral, positive sputum test at 5 and 6 months. Therefore, machine learning techniques, specifically classification types, can identify patient factors predictive of treatment non-adherence and accurately differentiate between adherent and non-adherent patients. Thus, tuberculosis program management should consider adopting the classification machine learning techniques evaluated in this study as a screening tool for identifying and targeting suited interventions to these patients.

show abstract

“…Todos os campos mapeados são importados, os quais devem ser posteriormente revisados. O breve texto apresentado em (LIMA et al, 2021d), em formato de Carta ao Editor, buscou apresentar as ações dos pesquisadores do LIS para oferecer suporte ao manejo e atenção aos pacientes de TB nos serviços públicos do Estado de São Paulo, como a plataforma para TDO remoto (ALBANO DOS SANTOS et al, 2019;CREPALDI et al, 2019), modelos de predição de abandono de tratamento (HOKINO YAMAGUTI et al, 2020) e um sistema capaz de monitorar e predizer a evolução dos casos de COVID-19 em uma região, de forma a emitir alertas antecipados aos serviços de TB. Um DSS deve oferecer funcionalidades de análise inteligente, compilação, processamento e visualização de um grande volume de dados obtidos a partir de diversas fontes, automatizando a integração de bases de dados heterogêneas, como bases de informações administrativas, de dados clínicos, de sistemas epidemiológicos, da literatura, entre outros.…”

Section: Considerações Finaisunclassified

Arquitetura e recursos computacionais para coleta, gestão e interoperabilidade de dados de tuberculose

Lima¹

View full text Add to dashboard Cite

À minha esposa Raquel, pelos insights, pelas sábias orientações sobre a prática da pesquisa científica e todo o apoio incondicional; Aos meus pais, pela educação proporcionada, da qual me orgulho fortemente, e pelo sempre efetivo apoio à continuidade dos meus estudos.Aos meus grandes amigos Felipe Pellison, Filipe Bernardi e Francisco Júnior, que caminharam comigo durante todo o tempo e que, sem eles, os conceitos de colaboração e trabalho em equipe teriam de ser reinventados.Aos profissionais Rafael Galliez e Karla Gram, pelas dicas, ensinamentos e trocas de experiências que ajudaram a nortear as atividades desenvolvidas.Ao meu caríssimo coorientador, Prof. Rui Rijo, pelo acolhimento em todos os momentos e por estar sempre de braços abertos.Ao meu orientador, Prof. Domingos Alves, pela disponibilidade, pela supervisão recebida e por, principalmente, acreditar no potencial de seus alunos.Ao Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), pela atribuição da bolsa de Desenvolvimento Tecnológico e Industrial (DTI), Nível A, que permitiu a execução de atividades relacionadas a este projeto de pesquisa. Eu não tenho ídolos.Tenho admiração por trabalho, dedicação e competência.

show abstract

Development of CART model for prediction of tuberculosis treatment loss to follow up in the state of São Paulo, Brazil: A case–control study

Cited by 5 publications

References 9 publications

T-BACCO SCORE: A predictive scoring tool for tuberculosis (TB) loss to follow-up among TB smokers

T-BACCO SCORE: A predictive scoring tool for tuberculosis (TB) loss to follow-up among TB smokers

A machine learning approach to explore individual risk factors for tuberculosis treatment non-adherence in Mukono district

Arquitetura e recursos computacionais para coleta, gestão e interoperabilidade de dados de tuberculose

Contact Info

Product

Resources

About