RESUMEN
OBJECTIVE: To analyse the HIV care cascade at the site level at the University Hospital of the West Indies,Kingston, Jamaica. SUBJECTS AND METHODS: This was a retrospective analysis of the national HIV treatment database for all entries between the years 2010 and 2013 identifying the key stepsin the HIV care cascade with an increased focus on parameters influencing viral load suppression. Proportions by classic cascade denominator and new denominators were calculated. Equality of proportions, with corresponding p-values and confidence intervals, was used to assess cascade step differences. RESULTS: Total enrolment in the clinic increased over the period (from 1206 to 1472 patients), as did number of patients receiving antiretroviral therapy [ART] (from 879to 1199). Using the classic denominator approach, significant increases were seen for patients on ART (8.4%,95% CI 0.053, 0.127) and significant declines were seen in viral load samples collected (-5.1%, 95% CI -0.099,-0.001) and viral load results returned (-12.1%, 95% CI-0.164, -0.056). No significant changes were identified in patients engaged in care (p = 0.885) or viral load suppression(p = 0.124). New denominator analysis showed significant changes in patients on ART engaged in care,-9.4% (95% CI -0.116, -0.072), as well as viral load suppression for patients on ART, 7.9% (95% CI 0.009,0.148). CONCLUSIONS: Site-level data can increase the detail included in the HIV care cascade and avoid limitations of the classic cascade. Site-level analysis can provide greater insight into the factors affecting patient outcomes and furnish the required data
