ANÁLISE ESPACIAL DO ABANDONO DO TRATAMENTO DA TUBERCULOSE E DA DUPLA CARGA TUBERCULOSE-DIABETES.

• Autor(a) principal: SOEIRO, Vanessa Moreira da Silva
• Data de Publicação: 2021
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações da UFMA
• Texto Completo: https://tedebc.ufma.br/jspui/handle/tede/tede/3713
• topic: Tuberculose;; Estudos de séries temporais;; Análise espacial;; Comorbidade;; Diabetes;; Regressão Espacial; Tuberculosis;; Time series studies;; Spatial analysis;; Comorbidity;; Diabetes;; Spatial regression; Saúde Pública

The abandonment of tuberculosis (TB) treatment and the tuberculosis-diabetes mellitus comorbidity (TB-DM) are part of the set of challenges to TB control as a public health problem. We aimed to analyze the spatio-temporal distribution of TB treatment abandonment and the tuberculosis-diabetes double burden in Brazil. An ecological study was conducted on the abandonment of treatment of new cases of TB and cases of tuberculosis-diabetes comorbidity in Brazil, in the period from 2012 to 2018, notified in the Sistema de Informação de Agravos de Notificação - Sinan, with municipalities as units of analysis. The study population consisted of all new cases of TB, of all clinical forms, whose outcome was the abandonment of TB treatment, and also all cases of tuberculosis, in all clinical forms, with the comorbidity Diabetes Mellitus. The Prais-Winsten model was used for the trend analysis and the Moran's Global and Local indices for the spatial analysis. It was found that the mean proportion of TB treatment abandonment in the country was 10.44% (SD:±1.36) with a trend considered stable. The highest averages were observed in the Southeast (10.80±1.50), South (10.67±3.17), and Midwest (10.12±0.86). The states with the highest rates were Rondônia (13.94±1.65), Rio Grande do Sul (13.73±4.56), and Rio de Janeiro (12.36±1.68). There was stability in the proportion of TB treatment abandonment in the country, and the North and South regions showed a decreasing trend, with variation rates of -5.45% and -16.62%, respectively. There was heterogeneous and non-random distribution, with the existence of high-risk areas concentrated, above all, in the Southeast region. As for the prevalence of TB DM comorbidity, in the period under study this indicator was 7.02%. In the annual and period distribution, the municipalities with prevalences above 6% are concentrated in the Southeast, Northeast, and South regions. The proportion of comorbidity TB-DM was heterogeneously distributed in Brazilian municipalities and did not occur randomly, with statistically significant positive spatial autocorrelation in the analysis of the period aggregate (Moran I= 0.193; p<0.0001). Spatial autocorrelation was verified by means of the Moran indices, with prevalence in high-risk areas equal to 28.04% and in low-risk areas equal to 5.86%. The high risk cluster was composed of municipalities belonging mainly to the Northeast, Southeast, and South regions. In the classic multivariate regression model, the coverage of Primary Care, percentage of the population living in households with a density greater than two people per room, percentage of unemployment of people over 18 years of age, per capita GDP, and per capita income fitted better. A negative association was observed between the dependent variable and the variables 'Primary Care coverage' and 'GDP per capita', which expresses in numerical terms that, for each increase in these markers, there is a decrease in the log of the number of MD-TB cases. The other variables were positively associated with the outcome. These variables were entered into the Spatial Lag and Spatial Error models and the results compared, with the latter showing the best parameters: R2=0.421, Log of the Likelihood=- 4661.03, AIC=9334.07 and SBC=9371.00. The Spatial Error residuals showed normal distribution and the overall Moran's index was -0.022 (p<0.001), indicating that the inclusion of the spatial component eliminated spatial autocorrelation from the residuals (value closer to zero), especially when compared to the residuals from the classical regression (OLS). The findings of this paper suggest that both treatment abandonment and TB-DM double burden are influenced by space for their occurrence and that socioeconomic and health factors explained the occurrence of TB-DM comorbidity in Brazil. The study of the spatial analysis of TB treatment abandonment and TB-DM comorbidity allowed us to highlight the heterogeneity and the pattern of geographical distribution of these two challenges to TB control as a public health problem in Brazil. The results presented here reinforce the need to recognize the abandonment of TB treatment and the TB-DM comorbidity as barriers to combating the disease in Brazil. This recognition is indispensable to the construction and implementation of public policies. It is hoped that the results of this study can contribute to the improvement of health actions, providing elements for the development of strategies aimed at reducing cases of treatment abandonment and for confronting the synergy of TB and DM, through the targeting of interventions in areas of greater risk, thus supporting the actions of public health management.