Table 2. Descriptive analysis and association of the proximal variables with the detection rate of tuberculosis^a. Brazil, 2012 to 2014.

Proximal variables (every 10%)	Median (IQR)	Increase in IRR (95%CI)
% of teams that conduct TB active case finding	70.6 (40–100)	3.15 (2.46–3.84)d
% of teams that request TB X-rayb	100 (100–100)	2.43 (1.18–3.69)d
% of teams that request smear microscopy for TBb	100 (100–100)	3.8 (2.26–5.36)d
% of teams that request culture for mycobacteriab	100 (71.4–100)	2.36 (1.46–3.27)d
% of teams that have the annual estimate of the number of TB cases and symptomatic respiratory cases in their territory	93.8 (66.7–100)	2.19 (1.4–2.99)d
% of teams whose first sputum sample for diagnosis of TB is collected at the first visit	50 (22.2–94.4)	0.81 (0.15–1.46)d
% of teams that have a registration of the number of TB cases identified in the last yearc	100 (100–100)	0.8 (-0.62–2.23)
% of teams that conduct household contact investigation of new TB cases	100 (0–100)	3.56 (3.02–4.11)d

IQR: interquartile range; IRR: incidence rate ratio; TB: tuberculosis

^aThe measure of association represents the increase in the incidence rate ratio (IRR-1) expressed as a percentage every 10% of the independent variable along with the 95% confidence interval (95%CI). All measures of association are adjusted by States.

^bPerformed by the network of health services.

^cWith document that proves it.

^dp < 0.05