| |
An Updated Assessment of Chronic Hepatitis B Prevalence Among Foreign-Born Persons Living in the United States
|
| |
| |
Download the PDF here
August 2021 Hepatology - Robert J. Wong ,1,2 Carol L. Brosgart,3 Sue Welch,4 Tim Block,5,6 Mark Chen,4 Chari Cohen,5,6 W. Ray Kim ,1 Kris V. Kowdley,7 Anna S. Lok,8 Naoky Tsai,9 John Ward,10 Steven S. Wong,4 and Robert G. Gish
Abstract
Background and Aims
Although prevalence of chronic hepatitis B (CHB) in the USA includes 0.42 million (range, 0.28-0.67) U.S.-born persons, foreign-born (FB) persons contribute a substantially larger number to the burden of CHB in the USA. Over the past decade, patterns of U.S. immigration have changed and many countries have implemented HBV prevention programs. This study aims to estimate the number of FB persons with CHB in the USA by country of origin, updating our 2011 study.
Approach and Results
We performed systematic searches for articles published in 2009–2019 reporting HBsAg seroprevalence in emigrants and in-country populations of 117 countries. Data meeting inclusion criteria were combined with data from our 2011 study to calculate pooled prevalence estimates for 99 countries using meta-analyses (total 2,800 surveys involving 112 million subjects). Combining country-specific CHB rate estimates with the number of FB in the USA in 2018, by country of origin from the U.S. Census Bureau, we estimate that the number of FB with CHB in the USA in 2018 was 1.47 million (95% CI, 1.21-1.73), substantially higher than previously reported. The weighted average CHB prevalence for all FB in the USA in 2018 was 3.07%. Approximately 59% of FB with CHB in the USA in 2018 emigrated from Asia, 19% from the Americas, and 15% from Africa. Subgroup analyses found that for many countries, CHB rates are higher in males than females and have declined over the past three decades, but no consistent pattern is observed between emigrant and in-country rates.
Conclusions
Including FB and U.S.-born persons, the total prevalence of CHB in the USA may be as high as 2.4 million.
Our updated systematic review and meta-analysis from 1980 to 2019 incorporated changes that have occurred over the past decade, including an increase in the FB population in the USA, increased immigration from HBV endemic regions, and changes in country-specific CHB prevalence rates resulting from HBV prevention programs. We estimate that the number of FB with CHB living in the U.S. in 2018 was 1.47M (95% CI, 1.21-1.73), an increase of 11% from 1.32M (95% CI, 1.04-1.61) in 2009.(5) Assuming 0.42M (95% CI, 0.28-0.67) U.S.-born persons with CHB in 2018, as many as 1.89M (range, 1.49-2.40) persons with CHB (FB and U.S.-born) may be living in the USA in 2018-80% of whom acquired their infection in their country of birth.
The updated estimates of CHB prevalence among FB in the USA provide information to guide public health policies, set research priorities, and help raise greater awareness of the clinical burden of CHB among FB in the USA. The finding that as many as 1.7M FB with CHB and 2.4M total persons with CHB may be living in the USA highlights the importance of HBV screening, particularly given that >80% of CHB adults in the USA remain undiagnosed and less than half of those diagnosed receive appropriate care.(21, 22). Recent updated recommendations by the U.S. Preventative Services Task Force reaffirm the importance of screening adults and adolescents at increased risk for HBV, including persons born in countries with high prevalence of HBV infection ≥2%.(23) However, risk-based testing for HBV requires accurate assessment of "risk," contributing to suboptimal screening rates. Similar to the evolution of testing for HIV and HCV from risk-based to universal testing, our findings and others emphasize the case for universal HBV testing to close the gap in missed diagnoses and opportunities for early linkage to care and treatment. We advocate for universal screening with HBsAg, hepatitis B surface antibody, and hepatitis B core antibody, given the importance of understanding past HBV exposure and potential risk of reactivation, identifying occult HBV infections, linking those with chronic infection to care and treatment for HBV, and assessing need for HBV vaccination, given that only ∼25% of U.S. adults aged ≥19 years are vaccinated against hepatitis B and remain at risk for acute HBV infection.(24) Universal screening and universal adult immunization for HBV are recommendations currently undergoing consideration by the U.S. Centers for Disease Control and Prevention and Advisory Committee on Immunization Practices.
Discussion
Our updated systematic review and meta-analysis from 1980 to 2019 incorporated changes that have occurred over the past decade, including an increase in the FB population in the USA, increased immigration from HBV endemic regions, and changes in country-specific CHB prevalence rates resulting from HBV prevention programs. We estimate that the number of FB with CHB living in the U.S. in 2018 was 1.47M (95% CI, 1.21-1.73), an increase of 11% from 1.32M (95% CI, 1.04-1.61) in 2009.(5) Assuming 0.42M (95% CI, 0.28-0.67) U.S.-born persons with CHB in 2018, as many as 1.89M (range, 1.49-2.40) persons with CHB (FB and U.S.-born) may be living in the USA in 2018-80% of whom acquired their infection in their country of birth.
The increase in FB with CHB in the USA in 2018 can be attributed to a 25% increase in the FB population over the past decade and a greater proportion of FB from regions with high HBV endemicity.(1) Offsetting the population increase, CHB rates likely declined in many countries. Comparisons of country-specific pooled CHB rates by survey decade showed that rates from surveys conducted in 2010-2019 are significantly lower than rates from surveys conducted before 1990 for many countries in Asia and for a few countries in Africa.
Although limited by lack of data for many countries, these results are generally consistent with studies using more robust methods to measure time changes in CHB rates (e.g., modeling rate trends).(16) For the FB population in the USA as a whole, a nonsignificant trend toward lower prevalence over time was observed. The weighted average CHB prevalence for all FB from all countries living in the USA in 2018 of 3.07% (95% CI, 2.52-3.61) is similar to what we found for 2009 (3.45%; 95% CI, 2.72-4.19),(5) which was not unexpected given the large number of surveys common to both analyses.
Subgroup analyses comparing country-specific, sex-specific CHB prevalence rates in males and females support previous reports that rates are higher in males than in females in many countries, although this analysis was limited by the small number of surveys for many countries. Comparison of rates from surveys in emigrants versus in-country populations did not show a consistent pattern. For most countries (38 of 64 for which comparison was possible), CHB rates in emigrants were similar to rates in in-country populations; for 17 of 64, in-country prevalence was higher, and for 9 of 64, rates in emigrants were higher. This is not unexpected: Different countries may have differing proportions of emigrants with higher CHB rates (e.g., because they spent time in refugee camps) or with lower CHB rates because they had higher socioeconomic status, better health care, and resources to emigrate.
Our current estimate of 1.47M FB with CHB in the USA is substantially higher than estimates from other sources-2.7 times the estimate of 0.55M (95% CI, 0.40-0.74) FB with CHB based on NHANES surveys from 2011 to 2016 (Fig. 3).(2) Although NHANES methods were modified in 2011 to oversample Asians, these surveys may still under-represent FB populations.(15, 17, 18) The CLDF recently adjusted the NHANES estimate for FB with CHB, which is calculated by race and ethnicity, by replacing the NHANES CHB rate for FB Asians (3.5%) with a higher CHB rate (7.2%) from our 2011 study.(5) Using this method, the CLDF estimated 1.08M to 1.62M FB with CHB in the USA in 2018.(15)
Few other prevalence estimates of FB with CHB in the USA have been reported. A study estimating the burden of CHB in the USA attributable to immigration multiplied country-specific prevalence rates by the number of FB persons immigrating to the USA each year. They determined that 1.3M FB persons with CHB entered the USA between 1974 and 2008-4.6% of all FB entering during that time. Data include lawful permanent residents only and were not adjusted for mortality to provide a prevalence estimate. The study estimated that FB persons comprise >90% of all new cases of CHB and 96% of deaths from CHB in the USA.(3, 19)
Our current estimate of 1.89M total persons (FB and U.S.-born) living with CHB in the USA in 2018 is higher than estimates from other sources, with the difference attributable to a larger FB population. NHANES surveys indicate that the total number of persons living with CHB in the USA remained at ∼0.85M (range, 0.6-1.1) from 1999 to 2016.(2, 17) This is almost certainly an underestimate because populations at high risk for HBV are excluded and FB under-represented.(18) By adjusting NHANES data to include institutionalized, incarcerated, homeless, and military populations and revising the estimate for FB with CHB using a higher CHB rate for FB Asians, the CLDF estimated 1.59M (range, 1.25-2.49) total persons with CHB (FB and U.S.-born) in the USA in 2018, ∼16% lower than our estimate with the difference attributable to our larger number of FB.(15)
Some country-specific CHB rates calculated in this analysis are higher than rates reported elsewhere. To compare our pooled CHB rates (in aggregate) with country-specific CHB rates from two recent global HBV reviews, we calculated the number of FB with CHB in the USA using ACS FB populations combined with CHB rates from the reviews.(1, 9, 20) Replacing our pooled CHB rates with rates from a review that were calculated from surveys published 1965-2013 (from a literature review) yields an estimate of 1.25M (95% CI, 1.11-1.43) FB with CHB in the USA in 2018, ∼15% lower than our mid-range RE estimate of 1.47M (95% CI, 1.21-1.73). The weighted average CHB rate among all FB in the USA in 2018 using available rates from this review is 2.60% compared to 3.07% from our analysis. A second review identified surveys published 1960-2016 from a literature review and selected a rate from a single representative survey for each country based on quality assessment and expert opinion; 83% of the surveys selected for country rates were conducted after 2000 and 50% after 2010.(20) Combining available country-specific CHB rates from this review with FB populations by country of origin yields 1.56M (95% CI, 1.17-1.94) FB with CHB in the USA in 2018, which encompasses our estimate of 1.47M (95% CI, 1.21-1.73). CHB rates for individual countries from these reviews and our work show wide variability, highlighting difficulties inherent in estimating country-specific prevalence rates based on limited data available. Our analysis identified the three countries from which the largest number of FB with CHB in the USA originate as China, Vietnam, and the Philippines. Our CHB rate estimate for mainland China was 8.5% (95% CI, 8.3-8.6), compared to 5.5% (95% CI, 5.47-5.50) for China (excluding Hong Kong) and 7.2% (range, 6.7-7.7) for China reported in the first and second reviews, respectively. Rates for Vietnam were 11.7% (95% CI, 10.9-12.5), 10.8% (95% CI, 10.3-11.3), and 10.0% (range, 7.8-12.5), and for the Philippines, 7.0% (95% CI, 6.1-7.9), 4.6% (95% CI, 4.5-4.7), and 16.7% (range, 14.3-19.1).
Differences between our rates and others could result from differences in the primary HBsAg seroprevalence data included in the analyses (i.e., survey selection) and/or methods used to calculate country-specific pooled rates. Unlike the first review, our analysis did not include surveys of children or blood donors, groups that generally have lower seroprevalence rates than adults in the general population.(9) The second review excluded surveys in blood donors, but allowed surveys among persons of all ages. Our higher rates may result from the RE meta-analysis model we used to calculate pooled country-specific CHB rates. Other meta-analytical methods would likely produce different estimates. Using rates from an FE model, we calculated a substantially lower estimate of FB in the USA in 2018 (0.92M vs. 1.47M with an RE model). An RE model was selected based on the heterogeneous nature of the survey data-heterogeneity was expected, observed, and supported by formal heterogeneity assessment of country-specific meta-analyses (i.e., I2 values). Other variations of RE models (e.g., using different weighting or data transformations) may produce different results and should be considered for future analyses.
A major limitation to this approach is the scarcity of HBsAg seroprevalence studies in many countries. Data are especially scarce for Central America and the Caribbean, regions from which large numbers of FB migrate to the USA. Literature searches were limited to PubMed; additional, potentially relevant articles may have been found in other databases. In addition, coverage of potentially relevant surveys reported in languages other than English was limited. Country-specific pooled CHB rates were calculated using all eligible surveys identified for the country; data were not sufficient to calculate rates balanced by sex, age, and other variables to be representative of a country's demographics. Data were not available to account for CHB-related deaths among FB who arrived in early decades. Nationally representative surveys were included, but were unavailable for most countries. A crude adjustment for age was made by excluding recent surveys in children from the 2019 update.
Surveys used in the analysis may not be strictly representative of populations likely to immigrate to the USA. Although we included surveys of "general populations," the demographics (age, sex, education, and economic status) of persons likely to emigrate to the USA vary by country. The limited surveys available precluded development of inclusion criteria based on specific demographics for emigrants from each country. Biases will vary from country to country and depend on the available surveys. Inclusion criteria allowed a broad range of surveys in different populations, settings, and locations. We expected prevalence rates to differ among surveys as a consequence of differences in demographics, locally dominant routes of HBV transmission, history and reach of HBV prevention programs, and other factors. For this reason, RE meta-analysis was used to calculate a pooled weighted mean seroprevalence for the distribution of seroprevalences reported within each country.
The updated estimates of CHB prevalence among FB in the USA provide information to guide public health policies, set research priorities, and help raise greater awareness of the clinical burden of CHB among FB in the USA. The finding that as many as 1.7M FB with CHB and 2.4M total persons with CHB may be living in the USA highlights the importance of HBV screening, particularly given that >80% of CHB adults in the USA remain undiagnosed and less than half of those diagnosed receive appropriate care.(21, 22). Recent updated recommendations by the U.S. Preventative Services Task Force reaffirm the importance of screening adults and adolescents at increased risk for HBV, including persons born in countries with high prevalence of HBV infection ≥2%.(23) However, risk-based testing for HBV requires accurate assessment of "risk," contributing to suboptimal screening rates. Similar to the evolution of testing for HIV and HCV from risk-based to universal testing, our findings and others emphasize the case for universal HBV testing to close the gap in missed diagnoses and opportunities for early linkage to care and treatment. We advocate for universal screening with HBsAg, hepatitis B surface antibody, and hepatitis B core antibody, given the importance of understanding past HBV exposure and potential risk of reactivation, identifying occult HBV infections, linking those with chronic infection to care and treatment for HBV, and assessing need for HBV vaccination, given that only ∼25% of U.S. adults aged ≥19 years are vaccinated against hepatitis B and remain at risk for acute HBV infection.(24) Universal screening and universal adult immunization for HBV are recommendations currently undergoing consideration by the U.S. Centers for Disease Control and Prevention and Advisory Committee on Immunization Practices.
|
|
| |
| |
|
|
|
