Introduction Adherence of 95% or more to antiretroviral therapy is generally considered necessary for optimal virologic suppression in patients living with HIV/AIDS (PLWHA). In the African context of access followed by low viral load, understand the determinants of poor adherence is essential to improve compliance, optimise virologic suppression and reduce morbidity and mortality.

Method A transversal analytical study referred to was conducted on patients aged at least 18 years on antiretroviral treatment for at least three months. We randomly selected patients and 9 support structures for PLWHA from 63 following more than 100 patients in Kinshasa, from May 2015 to August 2015. The Case Adherence Index (subjective method) and renewal of order (method objective) were used as indicators of compliance. Khi-2 and Student tests were used for comparisons. The determinants of nonadherence were sought in multivariate logistic regression analysis.

Results On a global active file of 5724 patients, 400 were included. The median age was 43 years and the sex ratio in favour of the three women. The overall incidence of non compliance was 25%. The prevalence of non-objective compliance was higher than that of non-subjective compliance (29% vs 21%, p=0.01). In multivariate analysis, the payment of the consultation [adjusted OR 1.70 95% CI (1.020 to 2.813), p=0.042], side effects [adjusted OR 2.23 95% CI (1.32 to 3.74), p=0.002] and lack of perception that the missed dose can aggravate the disease [adjusted OR 4.15, 95% CI (1.03 to 16.67), p=0.045] have emerged as determinants of the overall non-adherence.

Conclusion The results are in accordance with what has already been shown. The discrepancy observed between the two methods highlights the importance of access to biological methods. The identified risk factors will through regular evaluation, better identify PLWHA at greater risk of non-compliance and to offer enhanced compliance support.