Patients who are retained in HIV care have a higher likelihood of viral suppression and increased survival. Lab markers have been used as surrogate markers for clinical visits to estimate retention, but the accuracy of these markers at predicting retention in care has not been validated. A retrospective cohort study was conducted using patients newly diagnosed with HIV in the Emergency Department of Interim Louisiana Public Hospital (ILPH). Retention in care was defined as two clinical visits to an HIV provider separated by at least three months within a one-year period as per the Health Resources and Services Administration (HRSA) definition. Retention by lab markers was defined as two documented labs, either a CD4 count or an HIV viral load, separated by at least three months within the same one-year period. Ninety-nine patients were newly diagnosed with HIV; 36 patients (36%) were retained at 1 year using the HRSA definition and 40 patients (40%) using lab markers. The sensitivity and specificity of using lab markers among the newly diagnosed were 100% and 93.7%, respectively. The positive predictive value (PPV) and negative predictive value (NPV) were 90% and 100%, respectively. Among the 99 patients, 56 were linked to the HIV clinic associated with our hospital, of which 63% (36) were retained at year 1 using the HRSA definition and 70% (39) using lab markers. The sensitivity and specificity of using lab markers among linked patients were 100% and 85%, respectively. The PPV and NPV were 92% and 100%, respectively. Lab markers slightly overestimate currently accepted definitions of retention. While lab markers may be the easiest way to estimate retention at the population level, further study should be done before lab markers are accepted as the gold standard surrogate measure for retention.
Keywords: CD4 counts; HIV viral loads; Linkage to care; engagement in care; retention in care.