METHODS: This cross-sectional seroprevalence study with a two-stage stratified random cluster sampling design included 5,131 representative community dwellers in Malaysia aged ≥1 year. Data collection lasted from 7 August to 11 October 2020 involving venous blood sampling and interviews for history of COVID-19 symptoms and diagnosis. Previous SARS-CoV-2 infection was defined as screened positive using the Wantai SARS-CoV-2 Total Antibody enzyme-linked immunosorbent assay and confirmed positive using the GenScript SARS-CoV-2 surrogate Virus Neutralization Test. We performed a complex sampling design analysis, calculating sample weights considering probabilities of selection, non-response rate and post-stratification weight.

RESULTS: The overall weighted prevalence of SARS-CoV-2 infection was 0.49% (95%CI 0.28-0.85) (N = 150,857). Among the estimated population with past infection, around 84.1% (95%CI 58.84-95.12) (N = 126 826) were asymptomatic, and 90.1% (95%CI 67.06-97.58) (N = 135 866) were undiagnosed.

Context: The survey is nationally representative and community based and is conducted by the Institute for Public Health, part of the National Institutes of Health, to generate health-related evidence and to support the Malaysian Ministry of Health in policy-making. Its planned scope for 2020 was the seroprevalence of communicable diseases such as hepatitis B and C.

Action: Additional components were added to the survey to increase its usefulness, including COVID-19 seroprevalence and facial anthropometric studies to ensure respirator fit. The survey's scale was reduced, and data collection was changed from including only face-to-face interviews to mainly self-administered and telephone interviews. The transmission risk to participants was reduced by screening data collectors before the survey and fortnightly thereafter, using standard droplet and contact precautions, ensuring proper training and monitoring of data collectors, and implementing other administrative infection prevention measures.

Outcome: Data were collected from 7 August to 11 October 2020, with 5957 participants recruited. Only 4 out of 12 components of the survey were conducted via face-to-face interview. No COVID-19 cases were reported among data collectors and participants. All participants were given their hepatitis and COVID-19 laboratory test results; 73 participants with hepatitis B and 14 with hepatitis C who had been previously undiagnosed were referred for further case management.

METHODS: The primary data sources were population-based serosurveys, claims and hospital discharges, cancer registries, vital registration systems, and published case series. We estimated chronic HBV infection and the burden of HBV-related diseases, defined as an aggregate of cirrhosis due to hepatitis B, liver cancer due to hepatitis B, and acute hepatitis B. We used DisMod-MR 2.1, a Bayesian mixed-effects meta-regression tool, to estimate the prevalence of chronic HBV infection, cirrhosis, and aetiological proportions of cirrhosis. We used mortality-to-incidence ratios modelled with spatiotemporal Gaussian process regression to estimate the incidence of liver cancer. We used the Cause of Death Ensemble modelling (CODEm) model, a tool that selects models and covariates on the basis of out-of-sample performance, to estimate mortality due to cirrhosis, liver cancer, and acute hepatitis B.

FINDINGS: In 2019, the estimated global, all-age prevalence of chronic HBV infection was 4·1% (95% uncertainty interval [UI] 3·7 to 4·5), corresponding to 316 million (284 to 351) infected people. There was a 31·3% (29·0 to 33·9) decline in all-age prevalence between 1990 and 2019, with a more marked decline of 76·8% (76·2 to 77·5) in prevalence in children younger than 5 years. HBV-related diseases resulted in 555 000 global deaths (487 000 to 630 000) in 2019. The number of HBV-related deaths increased between 1990 and 2019 (by 5·9% [-5·6 to 19·2]) and between 2015 and 2019 (by 2·9% [-5·9 to 11·3]). By contrast, all-age and age-standardised death rates due to HBV-related diseases decreased during these periods. We compared estimates for 2019 in 194 WHO locations to WHO-GHSS 2020 targets, and found that four countries achieved a 10% reduction in deaths, 15 countries achieved a 30% reduction in new cases, and 147 countries achieved a 1% prevalence in children younger than 5 years. As of 2019, 68 of 194 countries had already achieved the 2030 target proposed in WHO Interim Guidance of an all-age HBV-related death rate of four per 100 000.

INTERPRETATION: The prevalence of chronic HBV infection declined over time, particularly in children younger than 5 years, since the introduction of hepatitis B vaccination. HBV-related death rates also decreased, but HBV-related death counts increased as a result of population growth, ageing, and cohort effects. By 2019, many countries had met the interim seroprevalence target for children younger than 5 years, but few countries had met the WHO-GHSS interim targets for deaths and new cases. Progress according to all indicators must be accelerated to meet 2030 targets, and there are marked disparities in burden and progress across the world. HBV interventions, such as vaccination, testing, and treatment, must be strategically supported and scaled up to achieve elimination.