Sage Journals: Discover world-class research

Abstract

Objectives:

The incidence of hepatitis A declined in the United States following the introduction of hepatitis A vaccines, before increasing in the setting of recent widespread outbreaks associated with person-to-person transmission. We describe the hepatitis A epidemiology in the United States, identify susceptible populations over time, and demonstrate the need for improved hepatitis A vaccination coverage, especially among adults at increased risk for hepatitis A.

Methods:

We calculated the hepatitis A incidence rates for sociodemographic characteristics and percentages for risk factors and clinical outcomes for hepatitis A cases reported to the National Notifiable Diseases Surveillance System during 1990-2020. We generated nationally representative estimates and 95% CIs of hepatitis A seroprevalence during 1976–March 2020 and self-reported hepatitis A vaccination coverage during 1999–March 2020 for the noninstitutionalized civilian US population using data from the National Health and Nutrition Examination Survey.

Results:

Overall, the rate per 100 000 population of reported cases of hepatitis A virus infection in the United States declined 17.3-fold, from 10.4 during 1990-1998 to 0.6 during 2007-2015, and then increased to 2.8 during 2016-2020. The overall hepatitis A seroprevalence in the United States increased from 38.2% (95% CI, 36.2%-40.1%) during 1976-1980 to 47.3% (95% CI, 45.4%-49.2%) during 2015–March 2020. The prevalence of self-reported hepatitis A vaccination coverage in the United States increased more than 2.5-fold, from 16.3% (95% CI, 15.0%-17.7%) during 1999-2006 to 41.9% (95% CI, 40.2%-43.7%) during 2015–March 2020.

Conclusions:

Hepatitis A epidemiology in the United States changed substantially during 1976-2020. Improved vaccination coverage, especially among adults recommended for vaccination by the Advisory Committee on Immunization Practices, is vital to stop current hepatitis A outbreaks associated with person-to-person transmission in the United States and prevent similar future recurrences.

Keywords

hepatitis A seroprevalence vaccine-preventable disease hepatitis A vaccine infectious disease outbreak

Hepatitis A is a vaccine-preventable liver infection acquired through fecal–oral transmission, either from direct person-to-person contact or consumption of contaminated food or water. Hepatitis A virus (HAV) infection is typically mild and self-limited; however, symptomatic illness, jaundice, hospitalization, and death occur more frequently among adults who are susceptible to HAV infection than among children.¹

During the latter half of the 20th century, hepatitis A was common in the United States, with large epidemics occurring approximately every 10 to 15 years (Figure 1).^2,3 Hepatitis A incidence rates per 100 000 population declined 96.7%, from 12.0 in 1995, when hepatitis A vaccines were first licensed in the United States, to 0.4 in 2015, when hepatitis A incidence rates reported to the National Notifiable Diseases Surveillance System (NNDSS) reached the nadir.⁴ Since 1996, the Advisory Committee on Immunization Practices (ACIP) has incrementally expanded hepatitis A vaccine recommendations for children and for adults who are at risk for HAV infection or for severe disease from HAV; universal childhood vaccination was recommended in 2006.^2,5-9

Figure 1.

Number and rate of reported hepatitis A virus infections, United States, 1966-2020. Data source: National Notifiable Diseases Surveillance System.¹⁰

We examined the totality of available data from NNDSS and the National Health and Nutrition Examination Survey (NHANES) to comprehensively describe the shifting hepatitis A epidemiology in the United States, identify susceptible populations over time, and demonstrate the need for improved surveillance and hepatitis A vaccination coverage, especially among adults at increased risk for HAV infection.

Methods

Data Sources

We obtained data analyzed in this article from NNDSS and NHANES. NNDSS is the national passive surveillance system through which health departments inform the Centers for Disease Control and Prevention (CDC) of nationally notifiable diseases and conditions.¹⁰ Hepatitis A has been nationally notifiable since 1966. During the analysis period, the surveillance definition used for hepatitis A was consistent (eg, clinical evidence of acute hepatitis and a positive test for immunoglobulin M antibody to HAV or clinical evidence of acute hepatitis in a person who had close contact with a laboratory-confirmed hepatitis A case during their exposure period).¹¹ NHANES collects nationally representative data on the health and nutritional status of the noninstitutionalized civilian US population. NHANES uses a stratified, multistage probability sampling design and collects information using standardized household interviews, physical examinations, and testing of biological specimens.¹² The analysis period was 1990-2020 for the NNDSS dataset and 1976–March 2020 for the NHANES dataset. We used deidentified NNDSS and publicly available NHANES datasets for this article, which do not require CDC institutional review board approval.

Serologic Testing

Eligibility of NHANES participants for hepatitis A serological testing varied across surveys: during NHANES II (1976-1980), people aged 6 months to 74 years were eligible; during NHANES III (1988-1994), people aged ≥6 years were eligible; and during 1999–March 2020, people aged ≥2 years were eligible. Serum specimens were tested for total HAV antibody (anti-HAV) using solid-phase competitive enzyme immunoassay (HAVAB-EIA solid-phase EIA Assay [Abbott Diagnostics]) for samples collected during 1976-1980, 1988-1994, and 1999-2006 and using a competitive immunoassay technique (Anti-HAV Total, VITROS Immunodiagnostic System [Ortho-Clinical Diagnostics, Inc]) for samples collected during 2007–March 2020. Tests initially considered indeterminate were repeated in duplicate for confirmation.¹³ For this study, we classified indeterminate results (88 of 113 917 total anti-HAV results) as negative. The serological assays used cannot differentiate between infection and vaccination; anti-HAV positivity reflects either immunity from prior infection or vaccination. We excluded people with missing anti-HAV results from the seroprevalence analysis.

Definitions

NNDSS

We assessed sociodemographic characteristics, risk factors, and clinical outcomes of hepatitis A cases reported to NNDSS: age, sex, race and ethnicity, urbanicity, and US Department of Health and Human Services (HHS) region. We defined urbanicity in accordance with the National Center for Health Statistics (NCHS) urban–rural classification scheme for counties; for 1990-2005, we applied the 1990 NCHS scheme; for 2006-2012, we applied the 2006 NCHS scheme; and for 2013-2020, we applied the 2013 NCHS scheme.¹⁴ We categorized HHS regions according to the grouping of states assigned under each of the 10 HHS regional offices.¹⁵ We assessed risk factors for acquiring hepatitis A based on their presence or absence during a patient’s exposure period (ie, the 15-50 days before symptom onset).

NHANES

We defined hepatitis A seroprevalence as the prevalence of positive total anti-HAV serology results. We considered survey participants vaccinated against hepatitis A if they self-reported receiving at least 1 dose of hepatitis A vaccine in response to the question, “Have you ever received hepatitis A vaccine?” We assessed sociodemographic characteristics: age, sex, race and ethnicity, country of birth, federal poverty level, and health insurance coverage. During NHANES II, ethnicity information was not collected; race options from NHANES II were recategorized for the analysis (“Black” was recategorized as non-Hispanic Black and “White” was recategorized as non-Hispanic White).

We categorized participants as having hepatitis B or hepatitis C if they were aged ≥2 years and tested positive for hepatitis B surface antigen (HBsAg), hepatitis C RNA, or both. We assessed men who have sex with men (MSM) status during 1999-2016 among male NHANES participants aged 20-59 years. We categorized males who met any of the following 3 scenarios as MSM for the purposes of the analysis: (1) provided a nonzero answer to, “In your lifetime, with how many men have you had sex?” or “In your lifetime, with how many males have you had anal or oral sex?”; (2) responded to sexual orientation/identity questions with “homosexual or gay,” “bisexual,” or “gay”; and (3) responded yes to the question, “Have you ever had any kind of sex with a man, including oral or anal?” We assessed drug use history during 1999-2018 among adult NHANES participants aged 20-59 years. We defined any drug use as participants responding yes to any of the following questions: “Have you ever used cocaine, including crack or freebase, or other street drugs? Do not include marijuana,” “Have you ever used a needle to take street drugs?,” “Have you ever, even once, used marijuana or hashish?,” “Have you ever used cocaine, crack cocaine, heroin, or methamphetamine?,” or “Have you ever, even once, used a needle to inject a drug not prescribed by a doctor?” We categorized a yes response to “Have you ever used a needle to take street drugs?” or “Have you ever, even once, used a needle to inject a drug not prescribed by a doctor?” as injection drug use. We categorized non–injection drug use according to 3 scenarios: (1) responded yes to “Have you ever used cocaine, including crack or freebase, or other street drugs? Do not include marijuana” and responded no to “Have you ever used a needle to take street drugs?”; (2) responded yes to “Have you ever, even once, used marijuana or hashish?”; and (3) responded yes to “Have you ever used cocaine, crack cocaine, heroin, or methamphetamine?” and responded no to “Have you ever, even once, used a needle to inject a drug not prescribed by a doctor?”

Statistical Analysis

For the NNDSS analysis, we calculated hepatitis A incidence rates by dividing the number of hepatitis A case reports annually during 1990-2020 by the US Census population for each year.¹⁶ We restricted percentage calculations for risk factors and clinical outcomes to those with available information. For the NHANES analyses, we used the NHANES survey sampling design and weighting variables to calculate nationally representative estimates and corresponding 95% CIs of the noninstitutionalized civilian US population; for these analyses, we used Stata version 14 (StataCorp LLC). We adhered to NCHS data presentation standards.¹⁷

Results

NNDSS

Overall, the rate per 100 000 population of reported cases of HAV infection in the United States declined 17.3-fold, from 10.4 during 1990-1998 to 0.6 during 2007-2015, and then increased to 2.8 during 2016-2020 (Table 1). During 1990-1998, the highest incidence rates were reported in people aged <40 years; however, by 2007-2015, rates of reported pediatric cases (aged <20 years) declined substantially, reaching the lowest age-specific rates of the analysis period during 2016-2020. During the analysis period, the highest age-specific reported HAV infection rates per 100 000 population occurred in increasingly older age groups; by 2016-2020, the highest rates (7.0 and 5.0) occurred among people aged 30-39 and 40-49 years, respectively. The race and ethnicity categories with reported HAV infection rates higher than the national average changed over time; during 1990-1998, American Indian/Alaska Native and Hispanic people had rates that were 6.5-fold and 1.8-fold higher than the national average, respectively. During subsequent periods, the highest race- and ethnicity-specific rates were reported among Hispanic (1999-2006), Asian/Pacific Islander (2007-2015), and non-Hispanic White (2016-2020) people (1.6-fold, 1.5-fold, and 1.2-fold higher, respectively, than the national average during the indicated period). Reported HAV infection rates were higher in urban areas than in rural areas during 1990-2015 but lower in urban areas than in rural areas during 2016-2020. The geographic distribution of reported cases shifted during the analysis period (Figure 2). During 1990-1998, the highest rates of reported HAV infections were in the western United States; by 2016-2020, the highest rates were in the eastern and midwestern United States.

Table 1.

Characteristics, risk behaviors or exposures, and outcomes among reported cases of hepatitis A virus infection, United States, 1990-2020^a

	1990-1998		1999-2006		2007-2015		2016-2020
Characteristic	No.	Rate^b	No.	Rate^b	No.	Rate^b	No.	Rate^b
Overall	244 868	10.4	71 257	3.1	16 591	0.6	46 645	2.8
Age, y
0-9	50 775	14.6	10 374	3.3	1011	0.3	307	0.1
10-19	40 540	12.2	9776	3.0	1848	0.5	769	0.4
20-29	55 241	15.8	13 016	4.2	3181	0.8	8872	3.9
30-39	47 465	12.0	13 054	3.8	2556	0.7	15 333	7.0
40-49	22 656	6.9	9942	2.9	2218	0.6	10 178	5.0
50-59	10 268	4.6	6246	2.4	2204	0.6	6492	3.1
≥60	13 426	3.4	8066	2.1	3423	0.7	4673	1.3
Sex
Female	101 939	8.4	29 071	2.5	7938	0.6	17 797	2.1
Male	135 158	11.7	41 721	3.7	8560	0.6	28 778	3.6
Race and ethnicity
American Indian/Alaska Native	11 868	68.1	525	3.0	86	0.4	147	1.1
Asian/Pacific Islander	3581	4.6	1982	2.1	1226	0.9	758	0.7
Non-Hispanic Black	21 732	7.6	6278	2.2	987	0.3	2713	1.3
Non-Hispanic White	116 293	6.7	29 427	1.9	7512	0.4	33 003	3.3
Hispanic	45 371	18.7	15 017	5.0	3093	0.7	2479	0.8
Urbanicity^c
Urban	200 468	10.6	61 801	3.4	13 010	0.6	35 090	2.5
Rural	42 424	8.9	9069	2.0	1913	0.5	8740	3.8
HHS region^d
Region 1	4302	3.4	3613	2.5	862	0.6	1439	2.1
Region 2	15 756	5.9	6149	2.6	1977	0.8	2110	1.7
Region 3	10 214	4.3	6004	3.0	1418	0.5	5413	6.6
Region 4	18 136	3.8	12 152	2.3	2640	0.4	19 596	7.1
Region 5	30 819	7.2	9386	2.1	2598	0.5	8421	3.1
Region 6	45 089	19.1	10 450	3.0	1859	0.4	3339	2.9
Region 7	14 116	10.6	2488	2.2	675	0.6	1449	1.6
Region 8	14 545	18.1	1356	1.2	500	0.5	966	1.2
Region 9	71 448	21.6	17 144	4.0	3443	0.8	3123	2.2
Region 10	20 443	23.9	2515	2.4	619	0.5	789	0.9
Risk behavior/exposure^e	n/N^f	%	n/N^f	%	n/N^f	%	n/N^f	%
Outbreak associated^g	11 336/62 548	18.1	5880/33 094	17.8	773/10 764	7.2	21 313/35 444	60.1
International travel	2500/39 442	6.3	3207/23 688	13.5	930/6591	14.1	362/24 173	1.5
Injection drug use	3355/31 872	10.5	1425/19 189	7.4	133/5861	2.3	11 971/26 290	45.5
Household contact (nonsexual)	5579/34 839	16.0	2526/20 678	12.2	261/6559	4.0	1144/18 364	6.2
Sexual contact	1273/34 839	3.7	854/20 678	4.1	124/6559	1.9	1446/18 364	7.9
Contact of daycare center child/employee	4396/37 941	11.6	1656/23 228	7.1	327/7395	4.4	351/20 633	1.7
Other contact^h	4831/34 839	13.9	1238/20 678	6.0	170/6559	2.6	1542/18 364	8.4
Child/employee in a daycare center	3530/40 779	8.7	1083/24 947	4.3	235/8403	2.8	193/23 272	0.8
Male-to-male sexual contactⁱ	2278/11 155	20.4	1912/6728	28.4	46/564	8.2	742/7299	10.2
Clinical outcomes
Hospitalized for hepatitis A^j	7881/43 493	18.1	6932/25 150	27.6	4085/9834	41.5	21 681/35 027	61.9
Died from hepatitis A^k	3388/41 478	8.2	927/27 001	3.4	65/9090	0.7	310/33 133	0.9

Abbreviation: HHS, US Department of Health and Human Services.

Data source: National Notifiable Diseases Surveillance System.¹⁰ Total case counts by characteristics may not sum to the overall sample size because of missing or unknown values.

Rate per 100 000 population.

Urbanicity was categorized according to the National Center for Health Statistics (NCHS) urban–rural classification scheme for counties and county-equivalent entities.¹⁴ The 1990 US Census–based scheme was applied to data from 1990-2005, the 2006 scheme to data from 2006-2012, and the 2013 scheme to data from 2013-2020. Large central metropolitan, large fringe metropolitan, medium metropolitan, and small metropolitan counties were grouped as urban. Micropolitan and noncore counties were grouped as rural.

HHS regions were categorized according to the grouping of states and US territories assigned under each of the 10 HHS regional offices.¹⁵ For the purposes of this article, regions with US territories (regions 2 and 9) contain data from states only.

Restricted to those with available risk behavior/exposure information. Reported cases may include more than 1 risk behavior/exposure; risk behavior/exposures are not mutually exclusive. Risk behavior/exposure data not available for 1990-1991.

Percentage calculations restricted to those with available data. n = number of cases with a response of yes, N = total number of cases with a response of yes or no.

Includes cases associated with common source (eg, foodborne, waterborne) and community (eg, person-to-person transmission) outbreaks.

For example, playmate, drug-sharing contact, or care provider.

Restricted to hepatitis A cases reported among males.

Restricted to those with available information on hospitalization status. Hospitalization data not available for 1990-1991.

Restricted to those with available information on death. Death data not available for 1990-1991. Hepatitis A deaths are not a reportable condition in the United States. Consequently, the accuracy of hepatitis A death rates is uncertain.

Figure 2.

Rate of reported hepatitis A virus infections per 100 000 population, by state, United States, 1990-2020. Data source: National Notifiable Diseases Surveillance System.¹⁰

The most frequently reported risk behaviors and exposures also changed over time. Among those with available information on risk factors, male-to-male sexual contact (20.4%), outbreak-associated (18.1%), and nonsexual household contact of a hepatitis A case (16.0%) were the most frequently reported risk factors during 1990-1998; outbreak-associated (60.1%), injection drug use (45.5%), and male-to-male sexual contact (10.2%) were the most frequently reported risk factors during 2016-2020 (Table 1).

Among those with available information, the percentage of reported hepatitis A cases that were hospitalized increased 3.4-fold during the analysis period, from 18.1% during 1990-1998 to 61.9% during 2016-2020 (Table 1). Conversely, the percentage of reported hepatitis A cases who died decreased from 8.2% during 1990-1998 to 0.7% during 2007-2015 and then increased to 0.9% during 2016-2020.

NHANES

Hepatitis A seroprevalence

NHANES response rates varied over time (Table 2). The overall hepatitis A seroprevalence in the United States increased from 38.2% (95% CI, 36.2%-40.1%) during 1976-1980 to 47.3% (95% CI, 45.4%-49.2%) during 2015–March 2020 (Table 3). Seroprevalence increased from 1976-1980 to 2015–March 2020 among people aged ≤9, 10-19, and 20-29 years but decreased during the same time frame among people aged ≥30 years, with larger decreases occurring among older age groups. Seroprevalence increased among males and females during the analysis period. By race and ethnicity, seroprevalence was highest among Hispanic people and lowest among non-Hispanic White people throughout the analysis period. During 2015–March 2020, hepatitis A seroprevalence among Hispanic people was 2.2-fold higher than among non-Hispanic White people, 1.5-fold higher than among non-Hispanic Black people, and 1.2-fold higher than among people in the “other” racial and ethnic category. In all time intervals examined, seroprevalence was higher among non–US-born than among US-born people and among those living at or below the federal poverty level than among those living above the federal poverty level. Seroprevalence more than doubled during the analysis period among those without health insurance (2.1-fold increase); those with any health insurance had a 1.3-fold increase in seroprevalence.

Table 2.

National Health and Nutrition Examination Survey response rates, United States, 1976–March 2020^a,b

Variable	1976-1980	1988-1994	1999-2006	2007-2014	2015–March 2020
Total no. sampled	27 801	30 930	47 281^c	50 693^c	40 080^c
No. (%) interviewed^d	25 286 (91.0)	25 733 (83.2)	38 124 (80.6)	37 790 (74.5)	23 862 (59.5)
No. (%) medically examined^e	20 322 (80.4)	23 527 (91.4)	36 126 (94.8)	36 447 (96.4)	22 302 (93.5)
No. (%) with serological specimen available for anti-HAV testing^f	9156 (45.1)	21 260 (90.4)	32 261 (89.3)	32 058 (88.0)	19 182 (86.0)

Abbreviations: anti-HAV, HAV antibody; HAV, hepatitis A virus.

Data source: National Health and Nutrition Examination Survey (NHANES).¹⁰

Restricted to participants who were eligible for anti-HAV serological testing during the respective time intervals. During 1976-1980, restricted to participants aged 6 months to 74 years. During 1988-1994, restricted to participants aged ≥6 years. During 1999–March 2020, restricted to participants aged ≥2 years.

Email communication, Deanna Kruszon-Moran, MS, and Te-Ching Chen, PhD, National Center for Health Statistics, February 28 and October 3, 2022.

Percentage calculated as number interviewed divided by total number sampled by NHANES.

Percentage calculated as number medically examined divided by number interviewed.

Percentage calculated as number with serological specimen available for anti-HAV testing divided by number medically examined.

Table 3.

Hepatitis A seroprevalence, United States, 1976–March 2020^a,b

	1976-1980		1988-1994		1999-2006		2007-2014		2015–March 2020
Characteristic	n/N	% (95% CI)	n/N	% (95% CI)	n/N	% (95% CI)	n/N	% (95% CI)	n/N	% (95% CI)
Total	4241/9516	38.2 (36.2-40.1)	9872/21 260	27.9 (23.8-32.0)	13 353/32 261	34.0 (32.4-35.7)	16 944/32 058	39.7 (37.8-41.5)	11 780/19 182	47.3 (45.4-49.2)
Age, y
≤9^c	39/391	10.5 (7.3-13.7)	—^d	—^d	1282/4771	22.4 (18.7-26.2)	3676/4941	70.2 (66.9-73.5)	2492/2747	90.7 (88.3-93.0)
10-19	277/1875	14.0 (11.4-16.5)	—^d	—^d	2921/9648	22.9 (19.1-26.6)	3395/6009	49.2 (45.6-52.7)	2816/3512	79.6 (76.0-83.2)
20-29	406/1595	23.4 (20.2-26.5)	1228/3275	19.2 (15.8-22.6)	1250/3398	28.0 (25.2-30.7)	1276/3503	32.1 (29.5-34.8)	993/1989	47.3 (43.5-51.1)
30-39	445/1122	37.1 (33.3-40.9)	1404/3124	24.9 (21.1-28.6)	1185/3045	30.0 (27.4-32.6)	1324/3621	31.8 (29.3-34.4)	840/2064	35.0 (31.6-38.4)
40-49	461/844	51.2 (47.1-55.3)	1327/2458	32.9 (29.1-36.6)	1399/3028	32.1 (29.5-34.6)	1510/3647	29.3 (26.8-31.9)	909/2108	33.6 (29.8-37.4)
50-59	561/827	65.1 (60.7-69.6)	1085/1765	47.3 (43.2-51.4)	1075/2298	36.4 (33.5-39.4)	1633/3358	31.8 (28.9-34.6)	1082/2195	31.6 (29.2-34.0)
≥60	2009/2408	81.8 (79.0-84.6)	3845/4958	66.2 (62.8-69.6)	4241/6073	59.4 (56.4-62.5)	4130/6979	44.7 (42.2-47.3)	2648/4567	40.2 (37.5-43.0)
Sex
Female	2348/5089	39.5 (37.2-41.7)	5152/11 173	28.2 (22.0-34.3)	6836/16 479	35.0 (33.3-36.7)	8560/16 139	39.6 (37.7-41.5)	5929/9739	46.4 (44.5-48.3)
Male	1893/4427	36.7 (34.6-38.7)	4720/10 087	27.7 (22.4-32.9)	6517/15 782	33.0 (31.2-34.8)	8384/15 919	39.7 (37.7-41.7)	5851/9443	48.2 (45.9-50.4)
Race and ethnicity^e
Non-Hispanic Black	534/1024	49.5 (43.2-55.9)	2227/6134	34.2 (28.5-40.0)	2461/7976	34.8 (32.6-36.9)	3406/6976	42.9 (40.1-45.7)	2529/4557	50.2 (46.3-54.0)
Non-Hispanic White	3604/8287	36.6 (34.7-38.6)	2919/7961	22.8 (18.4-27.3)	3810/12 760	26.6 (25.0-28.1)	3952/12 451	27.6 (26.0-29.3)	2685/6268	35.1 (33.1-37.0)
Hispanic	—^f	—^f	4198/6276	58.5 (50.7-66.4)	6542/10 236	64.5 (61.6-67.3)	7323/9318	75.3 (73.3-77.3)	4262/5247	76.6 (73.6-79.7)
Other	—^d	—^d	—^d	—^d	540/1289	49.6 (45.3-53.8)	2263/3313	63.4 (60.8-66.0)	2304/3110	65.5 (61.6-69.4)
Country of birth
United States	—^f	—^f	6633/16 685	24.1 (20.4-27.8)	8677/26 630	27.5 (26.0-28.9)	11 235/25 170	32.9 (31.4-34.5)	8023/14 711	41.2 (39.4-43.1)
Other	—^f	—^f	—^d	—^d	4670/5622	77.3 (75.3-79.2)	5694/6871	77.0 (74.8-79.2)	3753/4464	78.6 (76.4-80.8)
Federal poverty level
At or below	783/1459	50.6 (46.8-54.3)	2842/5360	40.4 (33.1-47.7)	3531/7705	42.4 (39.4-45.5)	4815/8077	51.0 (47.4-54.7)	2912/4119	61.2 (57.8-64.6)
Above	3458/8057	36.3 (34.5-38.2)	5889/13 978	24.8 (20.4-29.3)	8678/22 270	31.8 (30.1-33.5)	10 513/21 416	36.5 (34.8-38.2)	7386/12 847	43.8 (42.0-45.5)
Health insurance
Any	—^f	—^f	6884/13 301	34.5 (32.2-36.8)	9781/25 606	32.2 (30.7-33.7)	13 400/25 746	38.4 (36.8-40.1)	10 145/16 523	46.5 (44.6-48.3)
None	—^f	—^f	1971/4283	25.1 (16.4-33.8)	3429/6332	43.1 (40.0-46.2)	3515/6267	45.4 (41.8-48.9)	1604/2618	52.9 (47.8-58.0)
Risk groups
People with hepatitis B or hepatitis C^g	—^d	—^d	—^d	—^d	176/319	44.5 (37.9-51.0)	203/351	51.6 (42.8-60.4)	133/215	53.2 (42.9-63.5)
Men who had any sex with men^h	—^f	—^f	—^f	—^f	117/263	33.9 (25.7-42.1)	186/436	36.6 (30.1-43.0)	41/106	32.2 (18.4-45.9)
People who reported any drug useⁱ	—^f	—^f	—^f	—^f	556/1569	26.0 (22.9-29.1)	2257/7595	23.9 (22.1-25.6)	1314/3544	31.0 (28.9-33.1)
Non-IDUⁱ	—^f	—^f	—^f	—^f	495/1428	25.5 (22.0-29.0)	2223/7528	23.8 (22.1-25.6)	1289/3484	31.0 (28.8-33.2)
IDUⁱ	—^f	—^f	—^f	—^f	60/138	31.6 (21.8-41.3)	119/321	28.4 (21.5-35.2)	65/183	25.5 (19.8-31.2)

Abbreviations: anti-HAV, antibody to hepatitis A virus; IDU, injection drug use.

Data source: National Health and Nutrition Examination Survey (NHANES).¹⁰

Hepatitis A seroprevalence was defined as the prevalence of positive total anti-HAV serology results. Borderline total anti-HAV results were considered negative for the purposes of the analysis. The seroprevalence estimates are weighted; the sample sizes displayed are unweighted.

During 1976-1980, hepatitis A serology results were available only for those aged 6 months to 74 years. During 1988-1994, hepatitis A serology results were only available for those aged ≥6 years. During 1999–March 2020, hepatitis A serology results were available for those aged ≥2 years.

Suppressed in accordance with National Center for Health Statistics data presentation standards.

During 1976-1980, data on ethnicity were not collected, and race options included “Black” (presented in the row labeled “Non-Hispanic Black”), “White” (presented in the row labeled “Non-Hispanic White”), and “Other.”

Data not available during this time interval.

Participants aged ≥2 years who tested positive for hepatitis B surface antigen, hepatitis C RNA, or both. Hepatitis B surface antigen results were available for those aged ≥2 years. Hepatitis C RNA results were available for those aged ≥6 years.

Restricted to male NHANES participants aged 20-59 years during 1999-2016. During 2017–March 2020, sexual behavior questions about men who have sex with men were not available in NHANES public use datasets and were not included in the analysis.

Restricted to NHANES participants aged 20-59 years during 1999-2018. During 2019–March 2020, drug use questions were not available in NHANES public use datasets and were not included in the analysis.

Hepatitis A seroprevalence changed among groups at risk for HAV infection from 1999-2006 to 2015–March 2020. People with hepatitis B or hepatitis C, people who reported any drug use, and people who reported non–injection drug use had 1.2-fold increases in seroprevalence, while MSM and people who reported injection drug use had 0.9-fold and 0.8-fold decreases in seroprevalence, respectively (Table 3).

Self-reported hepatitis A vaccination coverage

Overall, self-reported hepatitis A vaccination coverage in the United States increased more than 2.5-fold, from 16.3% (95% CI, 15.0%-17.7%) during 1999-2006 to 41.9% (95% CI, 40.2%-43.6%) during 2015–March 2020 (Table 4). We noted increases across all groups examined. Self-reported hepatitis A vaccination coverage rates were inversely related to age; across the analysis period, the highest rates were among the youngest age group, while decreasing rates were among increasingly older age groups. During 2015–March 2020, self-reported hepatitis A vaccination coverage was 76.8% among those aged 2-9 years, 70.5% among those aged 10-19 years, 60.0% among those aged 20-29 years, and 43.6% among those aged 30-39 years; self-reported hepatitis A vaccination coverage was <33% for all people aged ≥40 years. Throughout the analysis period, the self-reported vaccination coverage rate was higher among non-Hispanic Black, Hispanic, and non-Hispanic Other people than the overall national estimate, while coverage was lower among non-Hispanic White people than the overall national estimate.

Table 4.

Prevalence of self-reported hepatitis A vaccination,^a United States, 1999–March 2020^b

Characteristic	1999-2006		2007-2014		2015–March 2020
Characteristic	n/N	% (95% CI)	n/N	% (95% CI)	n/N	% (95% CI)
Total	6757/35 305	16.3 (15.0-17.7)	14 965/33 124	37.7 (36.3-39.2)	9401/19 805	41.9 (40.2-43.6)
Age, y
2-9	1724/6217	24.7 (21.6-27.9)	4958/6540	70.9 (68.5-73.4)	2870/3729	76.8 (74.8-78.8)
10-19	2602/10 155	24.5 (21.3-27.6)	4318/6210	62.9 (60.2-65.6)	2668/3646	70.5 (67.4-73.5)
20-29	644/3464	18.4 (16.3-20.5)	1616/3187	48.9 (46.3-51.4)	1028/1723	60.0 (55.8-64.3)
30-39	487/3164	14.9 (13.3-16.4)	1219/3334	35.6 (33.0-38.1)	793/1820	43.6 (40.3-46.9)
40-49	419/3121	12.6 (11.0-14.3)	1018/3475	26.7 (24.5-28.8)	683/1958	32.5 (29.4-35.6)
50-59	300/2399	12.4 (10.9-13.9)	739/3311	20.6 (18.7-22.5)	540/2168	23.4 (20.3-26.4)
≥60	581/6785	9.9 (8.5-11.2)	1097/7067	15.4 (13.7-17.1)	819/4761	17.8 (15.8-19.8)
Sex
Female	3464/18 201	15.7 (14.3-17.1)	7549/16 705	37.5 (35.8-39.1)	4835/10 025	41.5 (39.5-43.4)
Male	3293/17 104	17.0 (15.7-18.4)	7416/16 419	38.0 (36.6-39.5)	4566/9780	42.4 (40.2-44.6)
Race and ethnicity^c
Non-Hispanic Black	1654/8933	17.0 (15.1-18.9)	3709/7779	44.5 (41.3-47.7)	2402/5165	44.3 (41.5-47.2)
Non-Hispanic White	2041/14 084	14.3 (12.8-15.9)	4133/12 505	31.6 (29.8-33.4)	2566/6439	36.9 (34.6-39.2)
Hispanic	2703/10 850	22.2 (19.6-24.7)	5152/9316	52.6 (50.6-54.6)	2546/4915	52.1 (49.3-54.9)
Other	359/1438	26.5 (22.6-30.4)	1971/3524	48.4 (45.9-51.0)	1887/3286	52.2 (49.1-55.2)
Country of birth
United States	5534/29 516	15.7 (14.3-17.1)	12 403/26 636	37.5 (35.9-39.2)	7634/15 619	41.9 (40.1-43.7)
Other	1217/5768	20.8 (18.9-22.7)	2558/6473	39.0 (37.3-40.8)	1762/4179	42.0 (39.3-44.6)
Federal poverty level
At or below	1994/8185	20.7 (19.1-22.4)	4658/8498	48.8 (46.0-51.6)	2205/4223	47.5 (44.5-50.4)
Above	4328/24 239	15.7 (14.3-17.2)	9017/21 856	35.1 (33.7-36.6)	6096/13 106	41.2 (39.3-43.1)
Health insurance
Any	5357/28 267	16.4 (14.9-17.8)	12 561/26 967	38.1 (36.7-39.6)	8431/17 289	42.6 (40.8-44.5)
None	1319/6559	16.3 (14.8-17.8)	2379/6115	35.7 (33.5-38.0)	960/2484	36.2 (33.3-39.2)
Risk groups
People with hepatitis B or hepatitis C^d	43/299	15.4 (10.0-20.7)	104/368	29.4 (23.3-35.5)	48/189	32.8 (21.8-43.8)
Men who had any sex with men^e	66/256	26.1 (19.3-32.9)	159/406	39.2 (34.1-44.2)	28/90	27.3 (16.0-38.6)
People who reported any drug use^f	198/1559	12.1 (9.8-14.4)	2440/6859	32.7 (31.0-34.4)	1120/3071	35.6 (32.7-38.6)
Non-IDU^f	172/1406	11.7 (9.4-14.1)	2416/6784	32.8 (31.1-34.5)	1110/3010	35.9 (32.9-38.8)
IDU^f	26/150	16.1 (9.1-23.1)	109/339	30.7 (24.4-37.0)	39/172	22.4 (14.5-30.4)

Abbreviations: anti-HAV, antibody to hepatitis A virus; IDU, injection drug use.

For the purposes of the analysis, survey participants were considered vaccinated against hepatitis A if they self-reported receiving at least 1 dose of hepatitis A vaccine in response to the question, “Have you ever received hepatitis A vaccine?” The prevalence estimates are weighted; the sample sizes displayed are unweighted.

Data source: National Health and Nutrition Examination Survey (NHANES).¹⁰

Self-reported hepatitis A vaccination coverage increased from 1999-2006 to 2015–March 2020 among all risk groups examined. However, self-reported vaccination coverage remained <36% for all risk groups during 2015–March 2020. Throughout the analysis period, the largest changes in self-reported vaccination coverage from 1999-2006 to 2015–March 2020 were among people who reported non–injection drug use, people who reported any drug use, and people with hepatitis B or hepatitis C (3.1-fold, 2.9-fold, and 2.1-fold increases, respectively), while smaller increases occurred among people who reported injection drug use (1.4-fold) and MSM (1.0-fold) (Table 4).

Discussion

Hepatitis A epidemiology in the United States has changed substantially since the condition first became nationally notifiable in 1966, owing largely to the 1996 introduction of hepatitis A vaccines and implementation of the 2006 ACIP recommendation for universal childhood vaccination. From 1990 to 2020, hepatitis A transitioned from a disease primarily affecting people aged <30 years, living in the western United States, and of Hispanic ethnicity, to a disease affecting people aged 30-49 years, living in the eastern and midwestern United States, and of non-Hispanic White race and ethnicity. Hepatitis A seroprevalence increased over time among people aged <30 years but decreased among those aged ≥30 years. Seroprevalence among adults at risk for HAV infection or serious consequences of HAV infection was lower than the national average among all risk groups examined in the analysis except for people with hepatitis B or hepatitis C. Similarly, self-reported vaccination coverage was lower than the national average among all risk groups examined, except MSM during 1999-2014.

The US populations that are most susceptible to HAV infection have shifted over time in response to the changing epidemiology and the introduction of hepatitis A vaccines. Prior to the introduction of hepatitis A vaccines, people aged ≤19 years were the most susceptible to HAV infection; however, during 2015–March 2020, they were the most highly protected against infection, and adults aged 30-59 years were the most susceptible to HAV infection. Among the race and ethnicity categories examined, non-Hispanic White people had the highest HAV infection susceptibility throughout the entire analysis period. In addition, during 1999–March 2020, all adult risk groups examined (except for people with hepatitis B or hepatitis C) had higher susceptibility to HAV infection than the overall national average.

During 2007-2014, the high levels of hepatitis A susceptibility among adults aged 20-59 years (>66%), non-Hispanic White people (>72%), and people who reported any drug use, injection drug use, or non–injection drug use (>71%) set the stage for the hepatitis A outbreaks associated with person-to-person transmission that have been widespread across the United States since 2016. These outbreaks have been unprecedented in the era of the hepatitis A vaccine; 37 states have been affected, and these states have reported >44 850 outbreak-associated cases, >27 400 hospitalizations, and approximately 421 deaths as of April 7, 2023.^18,19 The US hepatitis A incidence rate increased 850% from the beginning of the outbreaks in 2016 to their peak in 2019.⁴ As the epidemiology of susceptible populations shifted, people more likely to have severe outcomes (eg, hospitalization) represented an increasingly higher proportion of cases reported nationally. Infections have spread primarily through close contact among people who use drugs—a group recommended for hepatitis A vaccination by ACIP for more than 25 years—and people experiencing homelessness.^20,21

Limitations

Our analysis had several limitations. First, risk factors and self-reported vaccination status are subject to social desirability and recall bias, which might have resulted in underestimates of the prevalence of risk factors and self-reported vaccination. Second, NNDSS is a passive surveillance system, and the capacity for case ascertainment and reporting varies considerably across states; case counts and incidence rates might underestimate the true prevalence of hepatitis A in the United States. Third, substantial data (approximately 19% of data on race and ethnicity, 34% of data on risk factors, and 70% of data on clinical outcomes) were missing from NNDSS hepatitis A reports during the analysis period (CDC, unpublished data), which might have introduced bias in our analysis. Fourth, NNDSS and NHANES lack data to describe hepatitis A incidence among people experiencing homelessness or incarceration, 2 groups that have increased risk for HAV infection; consequently, we were unable to calculate incidence rates, seroprevalence, and self-reported vaccination coverage for these groups. Fifth, NHANES used different anti-HAV assays during the analysis period, which might have resulted in varying prevalence estimates. However, previous research indicates that the impact would be minimal.²²

Conclusions

HHS and CDC have called for a reduction in new hepatitis A infections by 40% by 2025 and 65% by 2030.^23,24 To reach the goal of eliminating hepatitis A as a public health threat, improved universal and catchup childhood vaccination and increased vaccination coverage among adult populations most affected by recent outbreaks are essential. Enhancements to core case surveillance are critical to support vaccination outreach and other preventive efforts. Improvements to variables collected in NNDSS are underway to better characterize the risk behaviors and exposures of people infected with HAV. In addition, investments in workforce development for disease intervention specialists will support jurisdictions in case investigation, linkage to preventive services, and outbreak control.²⁵ These investments are particularly important given the challenges often encountered during case investigations of people who use drugs, people experiencing homelessness, and people in correctional facilities, where loss to follow-up, mistrust of government, and stigma are often prevalent. Lastly, the call for a reduction in new hepatitis A infections is an important opportunity to leverage investments made to immunization information systems and vaccine infrastructure during the COVID-19 pandemic. Improved data quality and better linkage of surveillance systems and immunization registries with other data sources, such as Homeless Management Information Systems, Department of Corrections, and Medicaid databases, can provide a better understanding of the local epidemiology of hepatitis A and inform interventions for targeted and equitable vaccine delivery.

The epidemiology of hepatitis A has changed substantially during the past 4 decades. The public health gains we observed after the introduction of hepatitis A vaccines, however, have been interrupted by widespread outbreaks among people who use drugs and people experiencing homelessness. In the United States, hepatitis A is now a disease that primarily affects medically underserved people with inadequate access to preventive and community resources, and many people who experience increased risk for hepatitis A remain susceptible to infection and severe outcomes. Improvements to viral hepatitis surveillance, immunization information systems, vaccine delivery models, and services for disproportionately affected populations (eg, access to substance use treatment and sanitation) are needed to reduce health disparities and eliminate hepatitis A as a public health threat.

Footnotes

Disclaimer

The findings and conclusions in this article are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Megan G. Hofmeister, MD, MS, MPH

Shaoman Yin, PhD

References

Kemmer

Miskovsky

EP.

Hepatitis A. Infect Dis Clin North Am. 2000;14(3):605-615. doi:10.1016/s0891-5520(05)70123-9

Centers for Disease Control and Prevention. Prevention of hepatitis A through active or passive immunization: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Recomm Rep. 1999;48(RR-12):1-37.

Murphy

Denniston

Hill

, et al. Progress toward eliminating hepatitis A disease in the United States. 2022. MMWR Suppl. 2016;65(1):29-41. doi:10.15585/mmwr.su6501a6

Centers for Disease Control and Prevention. Viral hepatitis surveillance—United States. 2022. Accessed February 12, 2022. https://www.cdc.gov/hepatitis/statistics/SurveillanceRpts.htm

Prevention of hepatitis A through active or passive immunization: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Recomm Rep. 1996;45(RR-15):1-30.

Fiore

Wasley

Bell

BP.

Prevention of hepatitis A through active or passive immunization: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Recomm Rep. 2006;55(RR-7):1-23.

Centers for Disease Control and Prevention; Advisory Committee on Immunization Practices. Updated recommendations from the Advisory Committee on Immunization Practices (ACIP) for use of hepatitis A vaccine in close contacts of newly arriving international adoptees. MMWR Morb Mortal Wkly Rep. 2009;58(36):1006-1007.

Doshani

Weng

Moore

Romero

Nelson

NP.

Recommendations of the Advisory Committee on Immunization Practices for use of hepatitis A vaccine for persons experiencing homelessness. MMWR Morb Mortal Wkly Rep. 2019;68(6):153-156. doi:10.15585/mmwr.mm6806a6

Nelson

Weng

Hofmeister

, et al. Prevention of hepatitis A virus infection in the United States: recommendations of the Advisory Committee on Immunization Practices, 2020. MMWR Recomm Rep. 2020;69(5):1-38. doi:10.15585/mmwr.rr6905a1

10.

Centers for Disease Control and Prevention. National Notifiable Diseases Surveillance System (NNDSS). 2022. Accessed April 24, 2023. https://www.cdc.gov/nndss/index.html

11.

Centers for Disease Control and Prevention. National Notifiable Diseases Surveillance System (NNDSS): hepatitis A, acute. 2021. Accessed October 5, 2022. https://ndc.services.cdc.gov/conditions/hepatitis-a-acute

12.

National Center for Health Statistics. National Health and Nutrition Examination Survey: NHANES questionnaires, datasets, and related documentation. 2022. Accessed March 18, 2022. https://wwwn.cdc.gov/nchs/nhanes

13.

Centers for Disease Control and Prevention. Laboratory procedure manual: hepatitis A antibody. September 2013. Accessed January 31, 2022. https://wwwn.cdc.gov/nchs/data/nhanes/2007-2008/labmethods/hepa_e_hepatitis-a-antibody_met.pdf

14.

Centers for Disease Control and Prevention. NCHS urban–rural classification scheme for counties. 2017. Accessed January 31, 2022. https://www.cdc.gov/nchs/data_access/urban_rural.htm

15.

US Department of Health and Human Services. HHS regional offices. 2021. Accessed October 5, 2022. https://www.hhs.gov/about/agencies/iea/regional-offices/index.html

16.

US Census Bureau. Population data. 2021. Accessed October 5, 2022. https://www.census.gov/topics/population/data.html

17.

Parker

Talih

Malec

, et al. National Center for Health Statistics data presentation standards for proportions. Vital Health Stat 2. 2017;175:1-22.

18.

Centers for Disease Control and Prevention. Person-to-person outbreaks of hepatitis A across the United States. 2022. Accessed January 9, 2023. https://www.cdc.gov/hepatitis/outbreaks/2017March-HepatitisA.htm

19.

Foster

Hofmeister

Kupronis

, et al. Increase in hepatitis A virus infections—United States, 2013-2018. MMWR Morb Mortal Wkly Rep. 2019;68(18):413-415. doi:10.15585/mmwr.mm6818a2

20.

Foster

Ramachandran

Myatt

, et al. Hepatitis A virus outbreaks associated with drug use and homelessness—California, Kentucky, Michigan, and Utah, 2017. MMWR Morb Mortal Wkly Rep. 2018;67(43):1208-1210. doi:10.15585/mmwr.mm6743a3

21.

Hofmeister

Xing

Foster

, et al. Hepatitis A person-to-person outbreaks: epidemiology, morbidity burden, and factors associated with hospitalization—multiple states, 2016-2019. J Infect Dis. 2021;223(3):426-434. doi:10.1093/infdis/jiaa636

22.

Klevens

Denniston

Jiles-Chapman

Murphy

TV.

Decreasing immunity to hepatitis A virus infection among US adults: findings from the National Health and Nutrition Examination Survey (NHANES), 1999-2012. Vaccine. 2015;33(46):6192-6198. doi:10.1016/j.vaccine.2015.10.009

23.

US Department of Health and Human Services. Viral hepatitis national strategic plan for the United States: a roadmap to elimination (2021-2025). 2020. Accessed February 18, 2022. https://www.hhs.gov/sites/default/files/Viral-Hepatitis-National-Strategic-Plan-2021-2025.pdf

24.

Centers for Disease Control and Prevention. Division of Viral Hepatitis: 2025 strategic plan. 2020. Accessed March 18, 2022. https://www.cdc.gov/hepatitis/pdfs/DVH-StrategicPlan2020-2025.pdf

25.

Centers for Disease Control and Prevention. CDC DIS workforce development funding. 2023. Accessed March 18, 2022. https://www.cdc.gov/std/funding/pchd/development-funding.html

Trends and Opportunities: Hepatitis A Virus Infection,Seroprevalence,and Vaccination Coverage—United States,1976-2020

Abstract

Objectives:

Methods:

Results:

Conclusions:

Keywords

Methods

Data Sources

Serologic Testing

Definitions

NNDSS

NHANES

Statistical Analysis

Results

NNDSS

NHANES

Hepatitis A seroprevalence

Self-reported hepatitis A vaccination coverage

Discussion

Limitations

Conclusions

Footnotes

Disclaimer

Declaration of Conflicting Interests

Funding

ORCID iDs

References