Acceptability and usability of oral fluid HCV self-testing for hepatitis C diagnosis: A systematic review and meta-analysis

Data on the acceptability and usability of hepatitis C virus self-testing (HCVST) remain scarce. We estimated the pooled rates of acceptability/feasibility and re-reading/re-testing agreement of HCVST using oral fluid tests (PROSPERO-CRD42022349874). We searched online databases for studies that evaluated acceptability, usability and inter-reader/operator variability for HCVST using oral fluid tests. Pooled estimates of feasibility, agreement and post-testing perspectives were analysed. Sensitivity analyses were performed in men who have sex with men (MSM) and people who inject drugs (PWID). Heterogeneity was assessed using the I-2 statistics. A total of six studies comprising 870 participants were identified: USA (n = 95 with liver disease), Kenya (n = 150 PWID), Egypt (n = 116 from the general population), Vietnam (n = 104 MSM and n = 105 PWID), China (n = 100 MSM) and Georgia (n = 100 MSM and n = 100 PWID)]. All studies used OraQuick & REG; HCV Rapid Antibody Test. The pooled overall estimates for correct sample collection and for people who performed HCVST without needing assistance in any step (95% confidence interval [CI]) were 87.2% [76.0-95.3] (n = 755; I-2 = 93.7%) and 62.6% [37.2-84.8] (n = 755; I-2 = 98.0%), respectively. The pooled estimate of agreement for re-reading was 95.0% [95% CI 91.5-97.6] (n = 831; I-2 = 74.0%) and for re-testing was 94.4% [90.3-97.5] (n = 726; I-2 = 77.1%). The pooled estimate of those who would recommend HCVST was 94.4% [84.7-99.6] (n = 625; I-2 = 93.7%). Pooled estimates (95% CI) of correct sample collection (72.8% [63.3-81.5] vs. 90.8% [85.9-94.8]) and performance of HCVST without needing assistance (44.1% [14.1-76.7] vs. 78.1% [53.4-95.3]) was lower in PWID compared to MSM. In summary, HCV testing with oral fluid HCVST was feasible and well-accepted. Oral fluid HCVST should be considered in key populations for uptake HCV testing.