 |
|
|
| |
Less Education Linked to Later HIV Diagnosis, Higher Mortality in 13-Cohort European Study
|
| |
| |
7th IAS Conference on HIV Pathogenesis, Treatment and Prevention, June 30-July 3, 2013, Kuala Lumpur
Mark Mascolini
People with less education got diagnosed with HIV infection later in the course of their illness and, largely as a result, started antiretroviral therapy (ART) later than did people with more education [1]. Probably because of both delays, people with less education had a higher death rate in this analysis of 13 European cohorts.
Access to HIV testing and care is universal in Europe. Plentiful evidence from general-population studies show that people of lower socioeconomic status have worse health outcomes than people higher up the socioeconomic ladder. Because few large studies have examined the potential role of socioeconomic status on HIV diagnosis, ART initiation, and mortality, COHERE in EuroCoord investigators mounted this large analysis (www.cohere.org and www.EuroCoord.net).
COHERE investigators used UNESCO educational attainment classifications as a proxy for socioeconomic status: noncompleted basic education, basic, secondary, and tertiary. They used logistic regression analysis to explore the impact of education level on delayed HIV diagnosis, defined as advanced HIV disease (CD4 count under 200 or AIDS within 6 months of HIV diagnosis) or late HIV disease (CD4 count below 350 or AIDS within 6 months of HIV diagnosis). The investigators used median regression models to estimate the impact of education on CD4 count distribution at ART initiation in (1) all patients who started treatment and (2) those who started without advanced HIV disease. Kaplan-Meier and Cox regression models explored the impact of education on all-cause mortality in (1) all patients who began combination ART when antiretroviral naive and (2) all who began ART without advanced HIV disease. Models adjusted for gender, HIV transmission group, European or non-European origin, year before 2001 or after, and age at HIV diagnosis.
The delayed diagnosis analysis involved 15,414 people, 40% with advanced HIV disease and 60% with late HIV disease. Three quarters of these people (76%) were men, and the largest proportion (42%) acquired HIV through sex between men. Median age at HIV diagnosis stood at 35 years (interquartile range 29 to 43).
Proportions of cohort members with advanced HIV disease and late HIV disease fell significantly with each higher education level (P < 0.001 for trend). For men, proportions with advanced HIV disease were 54% for those with noncompleted basic education, 48% with basic education, 36% with secondary education, and 30% with tertiary education. The downtrend proved less steep for women: 49%, 40%, 39%, and 41%.
Overall advanced HIV disease rates for the four education categories were 52%, 45%, 37%, and 31%. Advanced HIV disease rates were greater by education category before 2001 (55%, 43%, 43%, and 44%) than in 2001 or later (50%, 46%, 34%, and 28%). But the differential between education levels was steeper (worse) after 2001 than before.
The analysis of CD4 count when ART began involved 11,035 people. Compared with people who had tertiary education, those with secondary education had a 19-cell lower adjusted median CD4 count when they started ART, people with basic education had a 27-cell lower adjusted count, and people with less than basic education had a 49-cell lower adjusted count (P < 0.001). Among people who began ART with more than 200 CD4s, adjusted median CD4 count at ART initiation was 4 cells lower in those with secondary versus tertiary education, 9 cells lower in those with basic education, and 18 cells lower in those with less than basic education; this worsening trend was not statistically significant (P = 0.375).
During 55,501 person-years of follow-up, 478 people died. Adjusted analysis determined that, compared with people who had tertiary-level education, those with secondary education had a 60% higher risk of all-cause death (adjusted hazard ratio [aHR] 1.60, 95% confidence interval [CI] 1.06 to 2.41), those with basic education had more than a doubled risk of dying (aHR 2.27, 95% CI 1.49 to 3.47), and those with less than basic education had a tripled risk of dying (aHR 3.00, 95% CI 1.89 to 4.79) (P < 0.001). When COHERE investigators limited the analysis to people without delayed ART initiation, hazard ratios compared with the tertiary group held true, though the comparison with people who had secondary education was no longer statistically significant (aHR 1.54, 95% CI 0.75 to 3.18).
Although access to HIV care is universal in Europe, the COHERE team concluded, education level has an impact on timing of HIV diagnosis and ART initiation, and less education raises the risk of all-cause mortality. They suggested that the weaker impact of less education on delayed HIV diagnosis in women than men may reflect the universal opt-out HIV testing offered to all pregnant women in Europe.
The researchers noted that delayed ART initiation with less education can be largely attributed to later diagnosis with less education. "It is reassuring," they add, "that earlier knowledge of . . . HIV-positive status leads to timely treatment regardless of educational level."
Notably, less education was more strongly associated with delayed HIV diagnosis after 2001 than before. The researchers called for further investigation "to confirm or refute this temporal trend of increasing socioeconomic inequities."
Finally, the COHERE team stressed that delayed HIV diagnosis does not fully explain higher on-treatment mortality in people with less education. They encouraged further research to determine whether higher mortality in more poorly educated people reflects differences in adherence and response to treatment or social disadvantages.
Reference
1. Lodi S, COHERE in EuroCoord. Late HIV diagnosis, late initiation of combination ART and mortality in Europe: variation by educational level. 7th IAS Conference on HIV Pathogenesis, Treatment and Prevention, June 30-July 3, 2013, Kuala Lumpur. Abstract TUPE257.
|
| |
|
|
|
|
|
