Educational attainment associated with health outcomes after starting HIV therapy

People with HIV who have lower educational attainment have poorer outcomes after starting combination antiretroviral therapy (cART), according to data from a large European cohort collaboration published in the online edition of AIDS. Rates of mortality and AIDS decreased with increasing education level, and education was also associated with virological suppression and CD4 count at the time of cART initiation.

“The striking differences in mortality and clinical responses to cART could not be explained entirely by delayed HIV diagnosis or late cART initiation, since differences largely remained after we restricted analyses to those initiating cART with CD4 > 350 cells/mm³ and without previous AIDS diagnoses,” comment the authors. “The associations we found between educational level and clinical outcomes are probably mediated by material and psychosocial paths.”

Even in countries with free access to healthcare, lower socioeconomic group is associated with poorer health outcomes. Investigators from a large collaboration of HIV cohort studies – the Collaboration of Observational HIV Epidemiological Research in Europe (COHERE) – have already shown that late diagnosis of HIV and starting cART with a low CD4 count are associated with lower levels of education. The researchers now wanted to see if disparities in health outcomes by education persisted after cART initiation.

They therefore designed a study examining differences by educational level in mortality and new AIDS after starting cART and also viral suppression and CD4 count changes.

Data from 15 cohort studies in eight countries were available for analysis. Adult patients starting cART for the first time between 1996 and 2013 were eligible for inclusion. Educational level was standardised across the cohorts.

The main outcomes were:

• All cause mortality.
• New AIDS event or death.
• Virological suppression – two successive viral load measurements below 400 copies/ml.
• CD4 cell increases during the first six years of cART.

Results were adjusted to take account of sex, age at cART initiation, calendar period of cART initiation (before 2001; 2001-04; 2005-08; 2009-13), transmission category, pre-cART viral load and CD4 count; pre-cART AIDS and type of cART.

A total of 24,069 people were eligible for inclusion. Overall, 9% had not completed primary education; 32% had only a primary education; 44% had completed secondary education and 14% had a college education or equivalent.

Individuals with a secondary and college education were more likely to be male and in the men-who-have-sex-with-men transmission category. A fifth of women were in the lower educational strata.

There was an association between education and mortality. During 132,507 person-years of analysis, 1081 people died. Mortality decreased as educational level increased (p < 0.001) and these differences persisted in models that adjusted for potential confounders. People with a college education had a lower risk of death than all other educational groups. A similar mortality pattern was present in an analysis controlling for CD4 count at the time of cART initiation, previous AIDS and age at the time treatment was started.

A significant association was also present between education and mortality and AIDS. Over 122,765 person-years, there were 2598 new AIDS events or deaths. Differences in incidence of AIDS/death by educational level were even more marked than for death alone (p < 0.001).

One year after starting HIV therapy, 77% of patients had attained viral suppression. This was achieved by 67% of people with incomplete primary education, 85% of individuals with a primary education, 82% of individuals with a secondary education and 87% of individuals who had attended college (p < 0.001). Differences in outcomes between the groups became less pronounced over time and ceased to be significant after ten years of therapy. After adjusting for possible confounders, those with an incomplete primary education or a primary education only had 20% and 7% lower chances, respectively, of attaining a virological response than patients who attended college.

Turning to immunological outcomes, the data showed that the higher the educational level achieved, the higher the CD4 count at cART initiation. However, there was no evidence that educational level was associated with CD4 count recovery six months after starting therapy.

“HIV-positive patients on combination antiretroviral therapy who had less education had higher mortality, higher rates of new AIDS, and worse virological responses than patients under care who had more education,” write the authors. “We observed such health differentials for an eighteen-year period in eight European countries where access to health care and cART is universal.”