Development and evaluation of a simple treatment eligibility score (HEPSANET) to decentralise hepatitis B care in Africa: a cross-sectional study
Nicolas Minier,Alice Nanelin Guingané,Edith Okeke,Edford Sinkala,Asgeir Johannessen,Monique I Andersson,Pantong Davwar,Hailemichael Desalegn,Mary Duguru,Fatou Fall,Souleyman Mboup,Tongai Maponga,Philippa C Matthews,Adrià Ramírez Mena,Gibril Ndow,Stian M S Orlien,Nicholas Riches,Moussa Seydi,Mark Sonderup,C Wendy Spearman,Alexander J Stockdale,Jantjie Taljaard,Michael Vinikoor,Gilles Wandeler,Maud Lemoine,Yusuke Shimakawa,Roger Sombié
DOI: https://doi.org/10.1016/s2468-1253(23)00449-1
IF: 45.042
2024-02-01
The Lancet Gastroenterology & Hepatology
Abstract:BACKGROUND: Hepatitis B virus (HBV) elimination requires expanding and decentralising HBV care services. However, peripheral health facilities lack access to diagnostic tools to assess eligibility for antiviral therapy. Through the Hepatitis B in Africa Collaborative Network (HEPSANET), we aimed to develop and evaluate a score using tests generally available at lower-level facilities, to simplify the evaluation of antiviral therapy eligibility in people living with HBV.METHODS: We surveyed the availability of clinical and laboratory parameters across different health-care levels in sub-Saharan Africa. We used data from the HEPSANET dataset, the largest cross-sectional dataset of treatment-naive people living with HBV in sub-Saharan Africa, to derive and validate the score. Participants from this dataset were included in the analysis if they were aged 18 years or older and had liver fibrosis stages determined by a liver stiffness measurement or liver histopathology. Participants with co-infections or metabolic disorders were excluded. We allocated participants to the derivation and validation sets by geographical site. In the derivation set, we used stepwise logistic regression to identify the best performing parameters for identifying participants that met the 2017 European Association for the Study of the Liver (EASL) criteria. Regression coefficients were converted into integer points to construct simplified algorithms for different health-care levels. In the validation set, we estimated the area under the receiver operating characteristic, sensitivity, and specificity of the simplified algorithm for identifying antiviral therapy eligibility defined by the 2017 EASL criteria.FINDINGS: At 11 sites from eight countries that returned surveys, aspartate aminotransferase (AST), alanine aminotransferase (ALT), and platelet count were generally available at district hospital levels, and hepatitis B e antigen and point-of-care HBV DNA tests were available only at regional and provincial hospital levels or above. Among 2895 participants included from the HEPSANET database (1740 [60·1%] male, 1155 [39·9%] female), 409 (14·1%) met EASL antiviral therapy eligibility criteria. In the derivation set, the optimal district-level hospital score was: ALT (IU/L), less than 40 (0 points), 40-79 (+1), 80 or greater (+2); AST (IU/L), less than 40 (0), 40-79 (+1), 80 or greater (+2); and platelet counts (10<sup>9</sup>/L), less than 100 (+2), 100-149 (+1), 150 or greater (0). When combined with family history and clinical data for decompensated cirrhosis that do not require any biological tests, a cut-off of 2 points or more had a sensitivity and specificity of 82% (95% CI 76-86) and 95% (93-96) to identify treatment-eligible individuals in the derivation set, and 78% (71-85) and 87% (86-89) in the validation set, respectively.INTERPRETATION: Using a score incorporating platelet counts, AST, and ALT, the majority of people living with HBV requiring antiviral therapy can be identified. Our findings suggest that clinical staging can be decentralised down to district hospital level in sub-Saharan Africa.FUNDING: European Association for the Study of the Liver Foundation, John C Martin Foundation.TRANSLATION: For the French translation of the abstract see Supplementary Materials section.
gastroenterology & hepatology