Editorial

Fernando A. Guerra, MD, MPh, Director of Health, San Antonio Metropolitan Health District, San Antonio, Texas

The results of two recently published studies have prompted renewed dialogue concerning the implementation of routine hepatitis A vaccination in children. These findings are critically important because the data not only support the recommended vaccination of hepatitis A in children known to be at high risk but also illustrate the existence and magnitude of the underreporting of hepatitis A infection in the remaining population. Complete epidemiological data is vital in evaluating and quantifying the potential impact of proposed hepatitis A vaccination programs throughout the United States (U.S.). Reported cases from local and state level health departments, when placed within a historical and geographic context that emphasizes epidemiological measures, are extremely valuable for guiding population based programs.

This report will summarize the data presented in each of those major studies. In the first study, published in the May 2002 issue of Pediatrics, Armstrong and Bell present a modeling analysis that delineates congruence between the reported incidence of hepatitis A and the observed prevalence of antibody to hepatitis A virus and thus allows an estimate of the true incidence of hepatitis A infection in the U.S.¹ The modeling analysis estimates an annual average of 271,000 hepatitis A infections from 1980 to 1999, representing approximately 10 times as many infections of hepatitis A as is reported to public health departments (see Table 1). Moreover, more than half of these infections occur in children less than 10 years of age.

The outcomes of a successful hepatitis A vaccination program are demonstrated in the second published study. Published in the December 2001 issue of JAMA, Averhoff et al determine the effect of routine vaccination of children in a community with recurrent hepatitis A epidemics.² This study has important implications in that it is one of the first studies to examine the impact of routine hepatitis A vaccination of children living in a large community but not necessarily considered high risk, vis-а-vis a small community.

The highest incidence of hepatitis A infection occurs in children between the ages of 5 and 14 years.³ In communities with the highest rates of hepatitis A, approximately 33% of the children acquire the hepatitis A infection before the age of 5 years, and almost 100% are infected before reaching adulthood.³ Clearly, children play a role in hepatitis A transmission, since infected children are typically asymptomatic and essentially serve as a vector to older children and adults.³ The public health, pediatric and primary care responsibilities for recognizing children in out-of-home placements, either child care or school settings, warrant particular consideration.

One benefit of routine hepatitis A vaccination of children that should not be overlooked may be the reduction in hepatitis A infections in adults.¹ Morbidity and mortality associated with hepatitis A is significant. Approximately one fifth of adult hepatitis A patients require hospitalization and the case-fatality rate among persons over 50 years of age is approximately 2%.³ This is especially important in those communities and populations where the prevalence of hepatitis C infection is increasing and its comorbidity with hepatitis A often times leads to serious consequences. The concept of strategic vaccination is not without precedence. Hepatitis A may be analogous to influenza in that children may harbor much of the reservoir of infection while adults incur most of the disease burden. Although not demonstrated conclusively, there are data to suggest that immunizing children against influenza may result in substantial reductions in influenza morbidity and mortality among adults.⁴ It is of interest to note that serious consideration is being given to immunize healthy children 6 months of age and older against influenza because of these observed benefits, not only to adults but also to the child. Such can be the benefit with the broader use of hepatitis A vaccine in children 2 years of age and older.

Until recently, immunoglobulin and improved hygiene were the only measures available to prevent and control hepatitis A. Highly effective vaccines against hepatitis A (hepatitis A vaccine, inactivated; Havrix®, Vaqta®) have been available in the U.S. since the mid-1990s.² Currently, hepatitis A vaccination is recommended for people with certain risk factors and for children residing in communities where the reported annual incidence exceeds twice the average annual U.S. rate of approximately 10 per 100,000.³ The dynamic principles of "herd immunity" suggest that routine hepatitis A vaccination of children could significantly lower the incidence of hepatitis A infection in nonimmunized children and adults. The impact of hepatitis A vaccination is made more dramatic with the evidence that when routine vaccination is instituted for children in high-prevalence communities, outbreaks of hepatitis A are not only interrupted, but subsequent outbreaks are also prevented.⁵

Determining the True Incidence of Hepatitis A

Dynamic modeling of disease transmission is useful for evaluating the potential impact of proposed immunization programs.¹ However, in terms of hepatitis A, application of these models requires knowledge of the incidence of hepatitis A infection, which is unknown in the U.S. In order to circumvent difficulties associated with underestimated hepatitis A disease incidence and eliminate the discordance between high prevalence of anti-hepatitis A virus antibodies in the U.S. and reported low incidence of infection, Armstrong and Bell have devised a model that allows concordance between the observed prevalence and reported incidence of hepatitis A.¹

A "catalytic modeling" method was utilized to estimate past incidence rates of hepatitis A based on the current prevalence of antibody to hepatitis A virus in the U.S. The model assumed that each hepatitis A antibody-positive person in the U.S. symbolized one hepatitis A infection in the past. The National Health and Nutrition Examination Survey (NHANES III, 1988-1994) provided hepatitis A antibody prevalence data.¹ Estimates of 47 million past hepatitis A infections were derived from these analyses. Additional modeling was then used to determine the age groups of infected people and the dates the infections occurred.

To account for anicteric infections assumed not to have been recognized or reported, the modeling estimates were divided by the age-specific probability that infection would result in jaundice. To model the age-specific probability of developing jaundice, data was used from studies of acutely infected individuals.¹ Each of the studies involved community-wide hepatitis A outbreaks and included serological surveys of symptomatic and asymptomatic children or adults.

The model estimated "force of infection," which is approximately equivalent to incidence in the susceptible (ie, hepatitis A-antibody-negative) population, from various sources of raw data including the reported number of hepatitis A cases from 1980 to 1999 from the National Notifiable Diseases Surveillance System.¹ There were on average 26,000 cases of acute hepatitis A reported to public health agencies annually between 1980 and 1999.¹

Modeling Results

The average reported age-specific incidence of hepatitis A from 1980 to 1999 was highest in children aged 5 to 9 years and adults aged 20 to 29 years. Incidence remained highest in these age groups after adjustment for the reported incidence in the prevalence of hepatitis A antibody to estimate the incidence in the susceptible population. Further, adjustment for anicteric infections using the probability jaundice model greatly increased the incidence in younger age groups. The final model observed the force of infection declining 4.5% per annum in the incidence of hepatitis A infection in the susceptible population resulting in increases in both the average age at infection (3.4 years in 1920 to 15.7 years in 2000) and the average age at icteric infection (11.5 years to 28.3 years).

From 1980 to 1999, the model estimated an average of 271,000 infections annually, representing 10.4 times the actual number of reported hepatitis A cases during that period (see Table 1). Because a majority of these infections occurred in children less than 10 years of age, fewer than half (111,800) resulted in icteric infection. As a result, only one of every 4.3 cases of icteric hepatitis A infection was actually reported. This modeling analysis illustrates the magnitude of the underreporting that has occurred in relation to the incidence of hepatitis A infection.

The model also predicted a logarithmic decrease in the incidence of hepatitis A infection as a function of time and an increasing prevalence of hepatitis A-antibody with age, attributable to cohort effects rather than to increasing rates of infections in older individuals. Similar cohort effects have been observed in Germany and Australia as well. Furthermore, the annual rate of decline (4.2% to 4.8%) in hepatitis A infection incidence is in agreement with the observed logarithmic decrease in infectious disease mortality rates in the U.S. throughout the 20th century.

The predicted increase in incidence of clinical hepatitis A during a time of declining incidence of hepatitis A infection is not without precedence. For example, Israel observed an increase in hepatitis A cases coincident with improvements in sanitary conditions.¹

Because the force of infection declines dramatically with age, Armstrong's model would predict that routine vaccination of hepatitis A in children should have particularly strong dynamic effects ("herd immunity")¹ and could significantly lower the incidence of hepatitis A infection in nonimmunized children and adults. Supplementary evidence exists in support of this proposition. Coincident with the implementation of recommended hepatitis A vaccination in selected U.S. states, the number of hepatitis A cases fell from 30,021 in 1997 to fewer than 12,000 in 2001¹, the lowest incidence since the Centers for Disease Control and Prevention began collecting this type of data in the 1950s.

Effective Vaccination Program Controls Hepatitis A

Routine hepatitis A vaccination of children living in small communities that had experienced recurrent hepatitis A epidemics and in which most adults are immune has been shown to be effective in helping to interrupt disease transmission. However, the impact of routine hepatitis A vaccination of children living in large communities with elevated hepatitis A infection rates has not been evaluated. Averhoff et al examined the results of routine hepatitis A vaccination of children in one such community.²

Butte County, California has experienced recurrent hepatitis A outbreaks from the mid 1980s to the late 1990s. During 1990-1997, annual hepatitis A incidence in Butte County was at least equal to, and up to 5.7 times higher than, the overall California rate. From January 1995 through December 2000, hepatitis A vaccine was offered free of charge to children residing in the county via a demonstration project conducted in two phases. The first phase (1995) focused primarily on controlling the ongoing outbreak of hepatitis A by vaccination of children. From 1996 to 2000, the second phase evaluated the effectiveness of routine vaccination of children. By the end of the project, children aged 2 to 17 years were eligible to receive vaccine and were included in the analysis.

During the 6-year study period, 29,789 (66.2%) of the eligible children received at least 1 dose of vaccine and 17,681 (39.3%) received 2 doses. First-dose vaccination coverage increased annually from 35.2% to 66.2%, and second-dose coverage increased from 14.5% to 39.3%.

The average annual incidence of hepatitis A in Butte County during the 5 years before the project (1990-1994) was 47.9 per 100,000 population and declined by 56.8% to 20.7 per 100,000 population during 1995-2000. The age-specific incidence decreased among all groups during 1995-2000 compared with that in 1990-1994, but the effect was most pronounced among the younger age groups (see Figure 1).

Among children ≤17 years of age, incidence decreased by 78.9% (67.3 to 13.5 per 100,000 population) compared with 44.3% among persons older than 17 years (41.3 to 23.0 per 100,000 population). Out of 245 reported cases (Butte County Health Department) of disease during the study period, 40 (16.3%) cases occurred in children ≤17 years of age, of which 16 cases were reported in 1995 and only one case in 2000. Dramatic decreases in hepatitis A cases were observed among the whole county population, from 57 in 1995 to 4 in 2000, representing a 93.5% decrease.

During the period of this project, the vaccine was demonstrated to be highly effective in preventing hepatitis A infection; the estimated protective efficacy of 1 or more doses of vaccine was 98% (95% Confidence Interval, 86%-100%). Further, no serious adverse events were reported although mild injection site reactions, fever, or rash were noted.

Conclusion

Hepatitis A continues to be one of the most frequently reported vaccine-preventable diseases in the U.S. The morbidity and mortality associated with hepatitis A infection is significant and the economic implications dramatic. These studies establish the underreported incidence of hepatitis A infection and document the achievability and efficacy of routine childhood hepatitis A vaccination and its impact on reduction of community-wide rates of disease. Furthermore, these findings suggest that sustained routine hepatitis A vaccination of children results in the reduction of overall disease rates of hepatitis A in the community in children as well as adults. As stated in a NEJM editorial by Dr. Raymond S. Koff⁶, "until such a policy is implemented, the eradication of hepatitis A virus by vaccination will remain unachievable."

References

1. Armstrong GL, Bell BP. Hepatitis A virus infections in the United States: model-based estimates and implications for childhood immunization. Pediatrics. 2002; 109:839-845.
2. Averhoff F, Shapiro CN, Bell BP, et al. Control of hepatitis A through routine vaccination of children. JAMA. 2001;286:2968-2973.
3. Prevention of hepatitis A through active and passive immunization: recommendations of the Advisory Committee on Immunization Practices. MMWR Morb Mortal Wkly Rep. 1999;48(RR-12):1-37.
4. Reichert TA, Sugaya N, Fedson DS, Glezen WP, Simonsen L, Tashiro M. The Japanese experience with vaccinating schoolchildren against influenza. N Engl J Med. 2001;344:889-896.
5. Centers for Disease Control and Prevention. Hepatitis A vaccination programs in communities with high rates of hepatitis A. MMWR Morb Mortal Wkly Rep. 1997;46:600-603.
6. Koff RS. The case for routine childhood vaccination against hepatitis A. N Engl J Med. 1999;340:644-645.