High burden of Hepatitis B virus infection among chronic liver disease patients in Northwest Ethiopia: a cross-sectional study (2023–2024)

Abstract

Hepatitis B virus (HBV) remains a major driver of chronic liver disease (CLD), cirrhosis, and hepatocellular carcinoma (HCC) in sub-Saharan Africa. Patients with established CLD represent a high-risk group; acute or superimposed HBV infection in these individuals can trigger rapid hepatic decompensation, accelerate structural progression to cirrhosis, and elevate the lifetime risk of neoplastic transformation. Despite this vulnerability, clinical insights into its burden and specific risk profiles in public healthcare delivery networks remain poorly defined. To assess the prevalence and associated factors of HBV infection among chronic liver disease patients attending public referral hospitals in Bahir Dar, Ethiopia. A facility-based, cross-sectional study was conducted from October 2023 to January 2024 among 320 CLD patients across three public referral hospitals in Bahir Dar. Data were gathered through structured questionnaires, physical evaluations, and chart reviews. Bivariable and multivariable logistic regression analyses were utilized to compute adjusted odds ratios (AORs). To complement the odds ratio estimates in this high-prevalence setting (41.9%), a modified Poisson regression model with robust variance was additionally applied to calculate adjusted prevalence ratios (aPRs) to provide prevalence-based effect estimates. The prevalence of HBV infection among the surveyed CLD patients was 41.9% (134/320; 95% CI: 36.5–47.5). In multivariable binary logistic regression, several conventional exposures appeared strongly associated with HBsAg positivity: HIV-positive status (AOR = 3.00; 95% CI: 1.40–6.80), family history of liver disease (AOR = 4.70; 95% CI: 1.90–11.00), history of dental procedures (AOR = 4.50; 95% CI: 2.00–10.00), and contact with a jaundiced person (AOR = 2.40; 95% CI: 1.20–4.80). In the parallel modified Poisson model, prevalence ratio estimates provided conservative effect measurements: HIV-positive status (aPR = 1.68; 95% CI: 1.20–2.35), family history of liver disease (aPR = 1.76; 95% CI: 1.29–2.40), contact with a jaundiced person (aPR = 1.48; 95% CI: 1.14–1.92), and history of dental procedures (aPR = 1.62; 95% CI: 1.21–2.17). Post-hoc power calculations confirmed >85% statistical power for primary multivariable predictors. HBV infection is highly prevalent among patients with CLD in Northwest Ethiopia. The logistic and modified Poisson models provide complementary perspectives on risk estimation, confirming consistent positive associations across parenteral and familial exposure pathways. These findings suggest that integrating HBV screening into liver clinics, evaluating targeted adult vaccination strategies, and reinforcing standard infection control measures in health facilities warrant consideration in regional public health planning. Not applicable.