Late entry into HIV care: lessons from Brazil, 2003 to 2006

BMC Infectious Diseases

Table 5 Proportion of late entry into HIV care, as defined by the extended and stricter criteria, with adjusted risk ratios. Brazil (2003–2006)

Characteristic		Extended Criterion				Stricter Criterion
		Adjusted risk ratio	95% Confidence interval		p	Adjusted risk ratio	95% Confidence interval		p
			Min	Max			Min	Max
Geopolitical Region	South	1	-	-	-	1	-	-	-
	North	1.26	1.23	1.28	< 0.001	1.41	1.35	1.47	< 0.001
	Northeast	1.10	1.08	1.11	< 0.001	1.30	1.26	1.34	< 0.001
	Central-West	1.11	1.08	1.13	< 0.001	1.20	1.14	1.24	< 0.001
	Southeast	1.00	0.98	1.01	0.498	1.11	1.08	1.13	< 0.001
Size of municipality (number of inhabitants)	≥ 500,000	1	-	-	-	1	-	-
	100,000-499,999	1.01	1.00	1.02	0.194	1.07	1.05	1.09	< 0.001
	< 100,000	1.02	1.01	1.04	0.005	1.10	1.07	1.13	< 0.001
HIV testing rate^†	High	1	-	-	-	1	-	-	-
	Medium	1.05	1.03	1.08	< 0.001	1.11	1.06	1.16	<0.001
	Low	1.07	1.05	1.09	< 0.001	1.11	1.08	1.15	< 0.001
Sex	Female	1	-	-	-	1	-	-	-
Sex	Male	1.20	1.19	1.21	< 0.001	1.40	1.37	1.42	< 0.001
Age	15-29 years	1	-	-	-	1	-	-	-
	30-59 years	1.39	1.37	1.40	< 0.001	1.70	1.66	1.74	< 0.001
	≥ 60 years	1.56	1.52	1.60	< 0.001	2.10	2.00	2.20	< 0.001
HIV Category of transmission^§	Heterosexual women	1	-	-	-	1	-	-	-
	Injection drug user	1.07	1.05	1.09	< 0.001	1.15	1.09	1.21	< 0.001
	Heterosexual man	1.08	1.07	1.09	< 0.001	1.15	1.12	1.19	< 0.001
	Blood transfusion	1.11	1.03	1.19	0.004	0.96	0.73	1.27	0.793
	Homosexual/bisexual male	1.01	0.99	1.02	< 0.203	1.00	0.96	1.04	< 0.001

^†High = > 10 tests/1,000 population; Medium = from 1.01 to 10 tests/1,000 population; Low = ≤ 1 test/1,000 population. ^§The categories of transmission were analyzed with a separate Poisson regression model, including the 37,391 individuals for whom information on transmission category was available. Control of confounding variables = region, HIV testing rate, and age. Population size was disregarded as a misleading variable as it did not show statistical significance in this regression model.

ISSN: 1471-2334