ABSTRACT:
Introduction:
Tuberculosis and diabetes comorbidity remains a challenge for global public health.
Objective:
To analyze the sociodemographic profile and the diagnostic and treatment characteristics of tuberculosis cases with and without diabetes in Brazil.
Methods:
This is a cross-sectional study with data from the Notifiable Diseases Information System and the Hypertension and Diabetes Mellitus Primary Care Clinical Management System, from 2007 to 2011. We adopted a Poisson regression model with robust variance to estimate the prevalence ratios (PR) and their respective confidence intervals.
Results:
We found the studied comorbidity in 7.2% of cases. The hierarchical model showed a higher PR among women (PR=1.31; 95% confidence interval - 95%CI 1.27-1.35); a greater association in the age groups 40-59 years and ≥ 60 years (PR=11.70; 95%CI 10.21-13.39, and PR=17.49; 95%CI 15.26-20.05), and in those with positive sputum smear microscopy results - 1st sample (PR=1.40; 95%CI 1.35-1.47). Return after treatment discontinuation and treatment discontinuation were inversely associated with comorbidity (PR=0.66; 95%CI 0.57-0.76 and PR=0.79; 95%CI 0.72-0.87).
Conclusion:
The findings, such as the inverse relationship with tuberculosis treatment discontinuation in the group of people with comorbidity, reinforce the importance of integrated actions in health services to change the scenario of this challenging comorbidity.
Keywords:
Tuberculosis; Diabetes mellitus; Comorbidity; Information management; Regression analysis