Fewer than four in 10 US adults had serologic immunity to hepatitis A virus (HAV) and fewer than three in 10 had immunity to hepatitis B virus (HBV), with substantial susceptibility also observed in high-risk clinical groups, according to a cross-sectional study published in JAMA Internal Medicine.
The study analyzed National Health and Nutrition Examination Survey (NHANES) data collected from January 2017 through August 2023. Investigators included adults aged 20 years or older with available serologic results for HAV and HBV. Data were weighted to generate nationally representative estimates of the US adult population.
Among 13,514 participants representing an estimated 226.1 million US adults, 39.5% had serologic immunity to HAV, corresponding to approximately 89.4 million adults. About 27% had serologic immunity to HBV, representing approximately 61.3 million adults, while 25% had vaccine-derived HBV immunity, representing approximately 54.2 million adults.
Investigators defined HAV immunity as positivity for total HAV antibody and HBV immunity as positivity for hepatitis B surface antibody. Vaccine-derived HBV immunity was defined by surface antibody positivity in the absence of hepatitis B core antibodies. Multivariable survey-weighted logistic regression was used to identify factors independently associated with viral hepatitis immunity.
The analysis was published as the Centers for Disease Control and Prevention and the Advisory Committee on Immunization Practices moved in December 2025 to recommend individual-based decision-making on birth-dose hepatitis B vaccination for infants born to mothers without HBV infection, replacing the universal birth-dose policy adopted in 1991. The authors wrote that their findings establish a baseline for evaluating that change.
The authors also placed the findings in historical context. HAV immunity, which reflects both vaccination introduced in 1995 and prior natural exposure, was 38% in 1976 to 1980, 31% in 2005 to 2008, and 47% in 2015 to 2016, according to the paper. Vaccine-derived HBV immunity rose from approximately 11% in 1999 and 2000 to 23% in 2015 and 2016.
Immunity by demographic characteristics
Among adults aged 65 years or older, 39% had HAV immunity and 14% had HBV immunity. Among adults born outside the US, 77% had HAV immunity and 31% had HBV immunity. Race and ethnicity were self-reported using NHANES categories. HAV immunity was highest among non-Hispanic Asian participants at 79%, Mexican American participants at 75%, and other Hispanic participants at 65%. HBV immunity was highest among non-Hispanic Asian participants at 46%, non-Hispanic Black participants at 32%, and other Hispanic participants at 27%.
In adjusted analyses, HAV immunity was associated with younger age, lower educational attainment, non-Hispanic Asian, non-Hispanic Black, Mexican American, other Hispanic, and multiracial or other race and ethnicity, birth outside the US, and awareness of liver disease. Obesity was associated with lower odds of HAV immunity. HBV immunity was associated with younger age, female sex, non-Hispanic Asian or non-Hispanic Black race and ethnicity, and higher educational attainment, while obesity was associated with lower odds of immunity. Birth outside the US was not independently associated with HBV immunity after adjustment.
For vaccine-derived HBV immunity, younger age, female sex, non-Hispanic Asian race and ethnicity, and higher educational attainment were associated with immunity in adjusted analyses, with the association for non-Hispanic Black participants of borderline statistical significance. Obesity was associated with lower odds of vaccine-derived immunity, while place of birth and insurance status were not included in the adjusted model.
Immunity across high-risk groups
Among all US adults, 67%, corresponding to 151 million adults, were categorized as being in high-risk subgroups. This included 89.9 million adults with chronic liver disease (CLD), 8 million with chronic kidney disease (CKD), 114.4 million with dysglycemia or diabetes (defined by self-reported diabetes, use of antidiabetic medications, fasting glucose of 100 mg/dL or greater, or hemoglobin A1c of 5.7% or higher), 4.8 million receiving immunosuppressive therapy, and 2.1 million pregnant adults. Categories overlapped, and adults could be counted in more than one.
Among adults with CLD, HAV serologic immunity was 38% among those without advanced fibrosis and 37% among those with advanced liver fibrosis. HBV serologic immunity was 23% among those with CLD without advanced liver fibrosis and 18% among those with advanced liver fibrosis. When CLD was stratified by diagnosis, HAV serologic immunity ranged from 27% among adults with metabolic dysfunction–associated alcohol-related liver disease (MetALD) to 64% among those with chronic HBV infection. HBV serologic immunity ranged from 15% among adults with MetALD to 29% among those with alcohol-related liver disease.
Among adults receiving immunosuppressive therapy, 36% had HAV serologic immunity and 20% had HBV serologic immunity. Among all adults with CKD, 41% had HAV immunity and 14% had HBV immunity. Adults with advanced CKD were the only high-risk group with higher HAV and HBV immunity than the general population, at 54% and 32%, respectively. Among pregnant adults, 47% had HAV serologic immunity and 39% had HBV serologic immunity. Across high-risk groups, vaccine-derived HBV immunity ranged from 14% among adults with CKD to 39% among pregnant adults. Overall, depending on the high-risk subgroup, 36% to 73% of adults remained susceptible to HAV, while 61% to 86% lacked vaccine-derived immunity to HBV.
The investigators noted that advanced CKD was an exception to the generally low immunity observed in high-risk groups. They said this pattern likely reflects protocol-driven vaccination practices, including postvaccination titer monitoring and revaccination.
Study limitations
The authors noted several limitations. NHANES samples the noninstitutionalized US population and therefore excludes some groups at high risk for viral hepatitis, including incarcerated individuals and people experiencing houselessness, potentially underestimating population susceptibility. Behavioral data, including alcohol use, were self-reported and subject to recall bias, which may have led to misclassification, and residual nonresponse bias could not be excluded. Medication data were available only for the 2017 to 2020 NHANES cycle, limiting the generalizability of findings on immunosuppression across the full study period.
The authors also cautioned that relying on serologic markers may underestimate true immunologic protection against HBV because anti-HBs titers often decline over time, resulting in apparent serologic susceptibility, particularly in older cohorts vaccinated decades ago. High-risk populations are also more likely to have vaccine nonresponse, further complicating interpretation of serologic immunity.
First and corresponding author Giovanni A. Roldan, MD, of the Division of Gastroenterology, Hepatology and Nutrition at the University of Minnesota, Minneapolis, told GI & Hepatology News that measurable serologic protection against HAV and HBV remains suboptimal among US adults, despite decades of vaccine availability and long-standing vaccination recommendations. “This gap is particularly important because the lowest levels of protection were seen among groups most vulnerable to severe outcomes, including patients with chronic liver disease, advanced fibrosis, chronic kidney disease, immunosuppression and pregnancy,” he said.
Implications for practice and public health
Although the study has clinical implications, its broader importance is in public health, Dr. Roldan noted. “Our findings suggest that current vaccination strategies have not translated into adequate population-level serologic protection, even under decades of stronger HBV vaccine policy,” he said. “This should prompt renewed attention to systematic vaccine assessment, targeted outreach and implementation strategies for both the general population and high-risk clinical groups.”
For clinicians, the practical message is to avoid assuming that patients are protected simply because vaccines are widely recommended, Dr. Roldan explained. “In high-risk settings, especially hepatology, nephrology, transplant, pregnancy care and primary care, clinicians should consider routine assessment of HAV and HBV immunity and use every eligible encounter as an opportunity to vaccinate susceptible patients,” he said. “For patients with chronic liver disease or advanced fibrosis, this is especially important because viral hepatitis can lead to severe complications.”
Cautions and next steps for research
Dr. Roldan cautioned that these findings should be interpreted as serologic immunity estimates, not direct measures of vaccine coverage or complete lifelong immune protection. “Antibody levels may wane over time, and our study cannot determine causality or predict the future impact of recent HBV policy changes. However, the study provides an important national baseline for evaluating how changes in vaccine policy may affect population-level immunity over time,” he concluded. “Our group is currently working on additional analyses evaluating the impact of vaccine policies across different age cohorts, including work that will be presented at The Liver Meeting 2026.”
Study coauthor John R. Lake, MD, reported receiving salary support from Miromatrix. No other authors reported disclosures.