RESUMO
INTRODUCTION: Tuberculin skin testing (TST) for tuberculosis (TB) is recommended for all patients with HIV infection because of the known relationship between these two conditions. In this report we analyze the incidence and variables associated with non-prescription of TST in a cohort of HIV-infected people. PATIENTS AND METHODS: Longitudinal study conducted between 2000 and 2002 at 10 HIV hospital-based clinics. All HIV-infected patients who had not been regularly followed-up previously in dedicated clinics were identified. Data about TST and other variables related to TB were obtained from the clinical records. We calculated the percentage of patients who did not undergo TST and the associated factors, using odds ratios (ORs) and the 95% CI to investigate associations. A multivariate logistic regression analysis was performed. RESULTS: A total of 1242 patients met the inclusion criteria. TST was not performed in 185 patients (17.6% of those eligible). The fact of being an intravenous drug abuser was associated with a higher probability of TST non-prescription (OR: 2.6, 95% CI 1.1-6.5), whereas being unemployed (OR: 0.6, 95% CI 0.3-1.0), having a CD4 cell count >200 (CD4 200-499: OR 0.5, 95% CI 0.3-0.9. CD4> or =500: OR 0.3, 95% CI 0.2-0.6), and contact with persons with TB (OR 0.2, 95% CI 0.1-0.5) were associated with a lower probability. CONCLUSIONS: In this study, the percentage of TST non-prescription was quite high. The results suggest that TST non-prescription in this population is related to the clinicians' expectations regarding the results of the test and the patients' adherence to treatment for latent TB infection.
