Educational gradient in HIV diagnosis delay, mortality, antiretroviral treatment initiation and response in a country with universal health care

Abstract

Background: The aim of this study was to analyse associations between educational level and delayed HIV diagnosis (DD), late initiation of combined antiretroviral therapy (cART), overall and in subjects with timely HIV diagnosis, virological and immunological responses to cART, and mortality from HIV diagnosis and cART initiation.

Methods: This was a multicentre cohort study of HIV-positive treatment-naive subjects in Spain between 2004-2009. Logistic and Cox regression analyses were used.

Results: Of 4,549 subjects, 44.5% had low education level (LOW), 34.4% medium education level (MED) and 21.1% high education level (HIG). In men, DD was more common in MED (OR 1.3 [95% CI 1.0, 1.7]) or LOW [OR 1.8 (95% CI 1.4, 2.3)] compared to HIG. In women, the opposite was observed; women with HIG were 40% more likely to have DD than those with LOW (OR 1.4 [95% CI 0.8, 2.5]). In individuals with timely HIV diagnoses, percentages of late cART initiators were similar (LOW 9.5%, MED 11.4% and HIG 7.0%; P=0.114). Immunological (LOW 68%, MED 76% and HIG 84%) and virological (LOW 76%, MED 83% and HIG 86%) responses to cART increased significantly with educational level; these increases remained significant in multivariate analyses. Mortality for LOW subjects was higher than for HIG, from HIV diagnosis (hazard ratio [HR] 2.3 [95% CI 1.1, 4.9]) and from cART initiation (HR 1.8 [95% CI 0.8, 3.9]).

Conclusions: We found important differences by educational level in diagnosis delay, virological and immunological responses to cART and mortality in a country with universal health care. Women with high educational level are at higher risk of having delayed HIV diagnoses. Educational level should be taken into account when designing HIV testing and clinical management strategies.