The association between viral load suppression and AIDS at diagnosis probably relates to the fact that these patients are monitored more closely and frequently (or even hospitalized for opportunistic infections), thereby facilitating optimal antiretroviral adherence and subsequent virological suppression. However, analyses examining whether stage of infection predicts MLN0128 nmr antiretroviral adherence remain inconclusive [25]. Baseline CD4 cell count may predict eventual long-term outcomes of antiretroviral therapy [26,27]. However, our work demonstrates that baseline viral load is a more important predictor of time to virological suppression, which supports findings

from past studies [28–30]. Furthermore, our subanalysis exploring whether baseline viral load remains an important predictor of suppression later in follow-up indicates that, after 18 months of therapy, baseline viral load is no longer significantly associated with suppression. This finding supports those of past studies in which it was concluded that time to suppression is a mathematical function corresponding to baseline viral load [28,29]. In our cohort, women were less likely than men to achieve virological suppression. This is in contrast to other evaluations that have

found similar [31,32] or improved [33] virological suppression compared with men. These differing results may be a consequence of the specific characteristics of our population. In our cohort, a large BGB324 concentration proportion of our female population faced barriers to successful treatment, including IDU (IDU in 26% of women compared with 16% of men; P<0.001). This is well established to negatively influence virological suppression [34]. We speculate that other socioeconomic and mental health issues not controlled for in our models may explain our findings. Unfortunately, this information is not currently captured in the CANOC database. It is important to note that our data were obtained from only three provinces, and thus may not be generalizable to the entire Canadian HIV-positive Galeterone population.

However, the majority of HIV-positive individuals in Canada receive care in these three regions. In fact, CANOC contains approximately one-quarter of all patients on therapy and a much larger proportion of those who initiated since 2000 [35]. As with other cohort analyses, there is the potential for selection bias as a result of the differential losses to follow-up at the various clinic sites of those individuals who did not achieve suppression. As reported, loss to follow-up differed significantly among the provinces. Also, there is a clinic-based selection bias, which may explain the difference among provinces in viral load suppression, as British Columbia represents the entire sample of people on antiretroviral therapy in the province while data from the other provinces are based on a selection of clinics.