Factors associated with directly observed treatment in tuberculosis/HIV coinfection cases in Porto Alegre, 2009-2013: a retrospective cohort

Abstract

Background TB/HIV coinfection is a serious public health issue in Brazil, and patients with coinfection have difficulty adhering to treatments. Directly observed treatment (DOT) has been recommended by the World Health Organization, considering the vulnerability of those affected. The purpose is to investigate the occurrence of DOT and associated factors compared to conventional treatment in Porto Alegre, Brazil. Methods A retrospective cohort study was carried out with all patients with coinfection from 2009 to 2013 in the city of Porto Alegre, Brazil, the state capital with the highest rate of coinfection in Brazil. The data came from national health information systems. The dependent variable was the performance of DOT. Bivariate and multivariable models were used to determine factors associated with DOT. The percentage of cure and death was verified in a period of two years, comparing patients who received and did not receive DOT. Results 2,400 cases of coinfection were reported, with 1,574 males and 826 females and a mean age of 38 years ± 9.91 years. The occurrence of DOT was 16.9%. In the multivariable analysis, factors independently associated to DOT were the year (with greater chances of beingreceived in 2012 and 2013), place of origin, non-white race (OR = 1.29, 95% CI = 1.08– 1.54), cases of relapse (OR = 1.33; 95% CI = 1.03–1.73), readmission after abandonment (OR = 1.48, 95% CI = 1.20–1.83), transfer (OR = 2.04; 95% CI = 1.40–2.98), acid-fast bacilli (AFB) test with positive result in first sample (OR = 1.73, 95% CI = 1.24–2.42), alcohol abuse (OR = 1.39; 95% CI = 1.16–1.67), and mental disorders (OR = 1.83; 95% CI = 1.38– 2.44.) Of the 532 cases of death, occurring in two years, 10.2% were in patients who underwent DOT and 89.8% in patients who did not undergo DOT (p<0.001). O percentual de o´bitos em pessoas que receberam DOT foi de 13% e o percentual de o´bitos para pessoas que receberam tratamento convencional foi de 24%. Conclusions There was an increase in the percentage of DOT over the years in the scenario studied, and the predictors for DOT were related to social vulnerability. In relation to death within two years, a lower proportion was found in patients who underwent DOT, suggesting a protective effect of the strategy.