Clinical outcomes and risk factors for immune recovery and all-cause mortality in Latin Americans living with HIV with virological success: a retrospective cohort study.

INTRODUCTION: Immune reconstitution following antiretroviral therapy (ART) initiation is crucial to prevent AIDS and non-AIDS-related comorbidities. Patients with suppressed viraemia who fail to restore cellular immunity are exposed to an increased risk of morbidity and mortality during long-term follow-up, although the underlying mechanisms remain poorly understood. We aim to describe clinical outcomes and factors associated with the worse immune recovery and all-cause mortality in people living with HIV (PLWH) from Latin America following ART initiation. METHODS: Retrospective cohort study using the CCASAnet database: PLWH ≥18 years of age at ART initiation using a three drug-based combination therapy and with medical follow-up for ≥24 months after ART initiation and undetectable viral load were included. Patients were divided into four immune recovery groups based on rounded quartiles of increase in CD4 T-cell count at 2 years of treatment (<150, [150, 250), [250, 350] and >350 cells/mm3 ). Primary outcomes included all-cause mortality, AIDS-defining events and non-communicable diseases that occurred >2 years after ART initiation. Factors associated with an increase in CD4 T-cell count at 2 years of treatment were evaluated using a cumulative probability model with a logit link. RESULTS: In our cohort of 4496 Latin American PLWH, we found that patients with the lowest CD4 increase (<150) had the lowest survival probability at 10 years of follow-up. Lower increase in CD4 count following therapy initiation (and remarkably not a lower baseline CD4 T-cell count) and older age were risk factors for all-cause mortality. We also found that older age, male sex and higher baseline CD4 T-cell count were associated with lower CD4 count increase following therapy initiation. CONCLUSIONS: Our study shows that PLWH with lower increases in CD4 count have lower survival probabilities. CD4 increase during follow-up might be a better predictor of mortality in undetectable PLWH than baseline CD4 count. Therefore, it should be included as a routine clinical variable to assess immune recovery and overall survival.

Rank intraclass correlation for clustered data.

Clustered data are common in biomedical research. Observations in the same cluster are often more similar to each other than to observations from other clusters. The intraclass correlation coefficient (ICC), first introduced by R. A. Fisher, is frequently used to measure this degree of similarity. However, the ICC is sensitive to extreme values and skewed distributions, and depends on the scale of the data. It is also not applicable to ordered categorical data. We define the rank ICC as a natural extension of Fisher's ICC to the rank scale, and describe its corresponding population parameter. The rank ICC is simply interpreted as the rank correlation between a random pair of observations from the same cluster. We also extend the definition when the underlying distribution has more than two hierarchies. We describe estimation and inference procedures, show the asymptotic properties of our estimator, conduct simulations to evaluate its performance, and illustrate our method in three real data examples with skewed data, count data, and three-level ordered categorical data.