14, 95% CI 1.04–1.25) were more likely to achieve suppression than individuals residing in British Columbia. Individuals with a history of IDU were less likely to achieve suppression (HR 0.58, 95% CI 0.53–0.64).

Patients on initial antiretroviral regimens containing efavirenz (HR 1.30, 95% CI 1.16–1.47), lopinavir (HR 1.19, 95% CI 1.06–1.34) and atazanavir (HR 1.29, 95% CI 1.14–1.46) were more likely to achieve suppression GSI-IX in vivo than those whose first regimen contained nevirapine. Patients who initiated nelfinavir were less likely to achieve suppression (HR 0.66, 95% CI 0.56–0.78). Finally, patients with low baseline viral load measurements were more likely to achieve suppression (<4 log10 copies/mL, HR 1.49, 95% CI 1.29–1.65; 4–5 log10 copies/mL, HR 1.27, 95% CI 1.17–1.37) than patients with baseline viral load measures ≥5 log10 copies/mL. A life table was used to further explore the association of baseline viral load with suppression during follow-up. In Table 3, the probabilities of suppression at 6, 12, 18 and 24 months are listed by baseline viral load measure. Using a Bonferroni correction

for multiple comparisons Galunisertib datasheet (statistical significance level of P<0.0125, which is 0.05/4), it was found that, while baseline viral load was significantly associated with suppression at both 6 and 12 months of follow-up (P<0.001), by 18 and 24 months, baseline viral load was no longer a significant factor (P=0.050 and 0.223, respectively). In order to ascertain what effect baseline viral load had beyond 12 months, a subset of the data was analysed (n=832), which excluded very patients who achieved suppression earlier than 12 months as well as those with a follow-up time of less than 12 months. A Kaplan–Meier analysis showed that baseline viral load was not significantly associated with suppression for those followed for more than 12 months (log-rank P=0.118) (data not shown). Kaplan–Meier curves exploring provincial differences in time to suppression for subset populations indicated that provincial differences in suppression still existed when men, women, injecting drug users, non-injecting drug users

and those testing positive for hepatitis C were examined exclusively (Fig. 2). There were no provincial differences in suppression for those testing negative for hepatitis C (P=0.115). In this large multi-site Canadian cohort study we found that increased age, lower baseline viral load, having an AIDS diagnosis at baseline, male sex, non-IDU history and treatment in Ontario rather than British Columbia predicted increased likelihood of suppression. We also found that suppression was more likely with currently preferred regimens that include two NRTIs plus either an NNRTI or a ritonavir-boosted PI. Our finding of a 57% probability of suppression after 6 months of therapy is consistent with findings from other cohorts [17,18].