A previous anonymized HIV prevalence survey in the same clinic demonstrated that the prevalence of HIV infection was particularly high among patients born in sub-Saharan Africa [15, 16]. Over one in ten patients attending the open-access returning traveller clinic fall into this category. Our study demonstrates

that approximately half (49.4%) of Black Africans consented to have an HIV test; compared with the average acceptance rate in our clinic of 42.5%. Patients travelling to areas of lower prevalence such as Europe were less likely to accept a test, which may reflect the patients’ perception learn more of risk. Evidence shows that patients with high-risk behaviours report to be more likely to accept POCT compared with standard laboratory testing [13]. Of the seven individuals with new diagnoses in phases DAPT 1 and 2, three were Black African, two of White ethnicity and two from other ethnic backgrounds. Two of the patients required direct admission from the clinic for investigation of suspected

opportunistic infection. The others, who did not require admission and had no clinical evidence of immunosuppression, were counselled in the clinic, had confirmatory laboratory tests dispatched and were referred directly to the health advisors in the genitourinary medicine clinic. The distribution of CD4 cell counts for all these patients illustrates that universal testing identifies people with and without advanced immunosuppression. In a clinical study evaluating the performance of the INSTI Rapid POCT compared with ‘gold standard’ laboratory tests in known and unknown HIV-1-infected patients, a specificity of 99% (95% CI 96.3–99.7) was calculated [23]. In our setting, the estimated positive predictive value (PPV) is 0.89 and the estimated negative predictive value (NPV) is 1 using HIV-1 prevalence data from an anonymized study carried out in 1992 [15]. With our data, the false reactive rate was 0.002 (two of 1261). Both patients with a false reactive INSTI Rapid POCT had

confirmed P. falciparum malaria. It has been demonstrated elsewhere Dapagliflozin that co-existent malarial infections may give rise to false reactive rapid antigen tests for HIV [24]. To explore this further, we tested by INSTI 19 consecutive stored plasma samples from patients with confirmed malaria and documented negative 4th generation HIV enzyme immunoassay (EIA) results, and identified three that demonstrated a weakly reactive spot (indeterminate result). Clearly caution is required in communicating reactive results to patients in relatively low-prevalence settings where alternative diagnoses such as malaria are prevalent. The positive predictive value of a reactive POCT is not as high as in high HIV prevalence settings and therefore patients and staff should to be counselled accordingly.