Comparison of two HIV testing strategies in primary care centres: indicator-condition-guided testing vs. testing of...

The aim of the study was to compare prospectively indicator-condition (IC)-guided testing versus testing of those with non-indicator conditions (NICs) in four primary care centres (PCCs) in Barcelona, Spain.

From October 2009 to February 2011, patients aged from 18 to 65 years old who attended a PCC for a new herpes zoster infection, seborrhoeic eczema, mononucleosis syndrome or leucopenia/thrombopenia were included in the IC group, and one in every 10 randomly selected patients consulting for other reasons were included in the NIC group. A proportion of patients in each group were offered an HIV test; those who agreed to be tested were given a rapid finger-stick HIV test (€6 per test). Epidemiological and clinical data were collected and analysed.

During the study period, 775 patients attended with one of the four selected ICs, while 66,043 patients presented with an NIC. HIV screening was offered to 89 patients with ICs (offer rate 11.5%), of whom 85 agreed to and completed testing (94.4 and 100% acceptance and completion rates, respectively). In the NIC group, an HIV test was offered to 344 persons (offer rate 5.2%), of whom 313 accepted (90.9%) and 304 completed (97.1%) testing. HIV tests were positive in four persons [prevalence 4.7%; 95% confidence interval (CI) 1.3-11.6%] in the IC group and in one person in the NIC group (prevalence 0.3%; 95% CI 0.01-1.82%; P < 0.009). If every eligible person had taken an HIV test, we would have spent €4650 in the IC group and €396,258 in the NIC group, and an estimated 36 (95% CI 25-49) and 198 persons (95% CI 171-227), respectively, would have been diagnosed with HIV infection. The estimated cost per new HIV diagnosis would have been €129 (95% CI €107-153) in the IC group and €2001 (95% CI €1913-2088) in the NIC group.

Although the number of patients included in the study was small and the results should be treated with caution, IC-guided HIV testing, based on four selected ICs, in PCCs seems to be a more feasible and less expensive strategy to improve diagnosis of HIV infection in Spain than a nontargeted HIV testing strategy.