Cognitive change in adults ageing with well-controlled HIV was slow when it occurred at all over a decade of repeated testing, and the strongest baseline predictors of decline were cardiovascular, not viral, according to a prospective multicentre Canadian cohort study published online in the journal AIDS on 7 July 2026. Marie-Josee Brouillette of the Chronic Viral Illness Service at the McGill University Health Centre and colleagues report that group-based trajectory modelling identified two paths, one stable and one declining, with 47 per cent of participants falling into the declining group over up to 10 years. No decline was detected during the first three years of follow-up, and the estimated gap between the two trajectories was 0.35 to 0.5 of a standard deviation on the Brief Cognitive Ability Measure, the computerised battery used at every visit.
The participants come from Positive Brain Health Now, which enrolled 856 adults aged 35 or over living with HIV, and without dementia at entry, at five urban Canadian clinics, according to an earlier report from the cohort's inaugural visit. People were reassessed at intervals of nine to 12 months across two phases, 2013 to 2018 and 2019 to 2024, and the trajectory analysis was restricted to those with at least one visit in each phase. Baseline predictors of ending up in the declining group were tested with generalised estimating equations.
Two vascular measures stood out. A higher Framingham Risk Score, a cardiovascular score built from factors including age, blood pressure, cholesterol, smoking and diabetes, carried an odds ratio of 2.84 (95 per cent confidence interval 1.35 to 5.98), and peripheral artery disease an odds ratio of 3.07 (2.06 to 4.57). Older age, lower education, having needed academic support, cannabis use, anxiety, low motivation, detectable plasma HIV RNA and higher VACS Index 1.0 scores were also associated with decline. Nadir CD4 count and duration of HIV infection, the markers of historical disease severity that underpinned the older framing of HIV-associated neurocognitive disorder, were not independently associated with decline. The authors conclude that vascular comorbidities may matter more than traditional HIV-related factors in the current treatment era.
The counterpoint is in the design. This is an observational cohort, so the associations reported are not proof that vascular disease causes the decline, or that treating blood pressure or diabetes would prevent it. Higher baseline cognitive performance also predicted membership of the declining group, which the authors attribute to a floor effect in the measure, a reminder that part of the movement seen here is a property of the test as much as of the person. The cohort was drawn from specialist Canadian clinics, and the B-CAM measures cognitive ability on a scale, it does not diagnose dementia.