Legend of hepatitis B vaccination: The Taiwan experience



    1. Division of Gastroenterology, Department of Medicine, Taipei Veterans General Hospital and National Yang-Ming University School of Medicine, Taipei, Taiwan
    Search for more papers by this author

    Corresponding author
    1. Division of Gastroenterology, Department of Medicine, Taipei Veterans General Hospital and National Yang-Ming University School of Medicine, Taipei, Taiwan
    Search for more papers by this author

    1. Division of Gastroenterology, Department of Medicine, Taipei Veterans General Hospital and National Yang-Ming University School of Medicine, Taipei, Taiwan
    Search for more papers by this author

Professor Shou-Dong Lee, Department of Medicine, Taipei Veterans General Hospital, 201 Shih-Pai Road, Sec 2, Taipei 112, Taiwan. Email: sdlee@vghtpe.gov.tw


Hepatitis B, a disease entity currently affecting more than 350 million persons worldwide, is also a serious health problem in Taiwan. Liver cirrhosis and hepatoma, which are both closely correlated with hepatitis B, are among the 10 leading causes of death in Taiwan. A mass hepatitis B vaccination program, conducted by the government of Taiwan, was started in 1984. Prior to this vaccination program, a series of viral epidemiological surveys, transmission pattern studies, and pilot immunization trials proved the clinical, economic, and strategic benefits of mass immunization, thus providing the impetus for the implementation of this mass vaccination program. The success of this program has led to a decline in hepatitis B carrier rates among children in Taiwan from 10% to <1%. Furthermore, the mortality rate of fulminant hepatitis in infants and the annual incidence of childhood hepatoma have also decreased significantly in recent years. This is one of the most remarkable success stories in the field of public health.


Hepatitis B virus (HBV) infection has been one of the world's  greatest  health  problems,  not  only  because  of its wide-spreading nature, but also because of its severe complications and sequelae. According to a World Health Organization (WHO) report,1 there were more than 350 million carriers of HBV in the world, over 75% of whom were Chinese. Each year, approximately 2 million people worldwide die of HBV-related diseases, including hepatocellular carcinoma (HCC), cirrhosis, chronic hepatitis, and acute hepatitis. Again, more than three-quarters of these cases were Chinese. Therefore, some people refer to hepatitis B and its related diseases as ‘the Chinese diseases’. In 1981, when the first vaccine against viral hepatitis was invented, a legendary struggle against HBV infection began in Taiwan, a country where more than 98% of the residents are Chinese.


Before the hepatitis B vaccine was introduced, a series of epidemiological studies had disclosed the harsh situation of HBV infection in Taiwan. A survey in 1980 revealed that over 90% of the general population under the age of 40 years had been infected by HBV, and 15–20% of these were chronic carriers.2 In 1981, Beasley et al. reported the results of a large-scale follow-up study.3 The incidence of HCC among 22 707 male government employees who worked in Taipei City and received regular health checkups from 1978 to 1980 was monitored. Over 3000 of these men were found to have hepatitis B surface antigen (HBsAg), 49 of whom eventually developed HCC. In contrast, only one of the remaining non-carriers (about 20 000 in number) developed liver cancer during the follow-up period. That study indicated that HBsAg carriers might have a 273-fold higher risk of HCC compared with non-carriers.

The annual incidence of HBV infection in Taiwanese adults has been found to be 2.7%.4 Investigators began to realize that perinatal infection may lead to chronic liver diseases in 90% of those affected, while adults who contracted the infection rarely developed chronic hepatitis.5 Early in 1975, Stevens et al. confirmed that the mother-to-infant transmission was the main route of HBV transmission in Taiwan.6 Their work included monitoring 158 HBsAg-positive mothers and their infants at the Taipei Veterans General Hospital. At 6 months of age, 40% of the infants were found to be positive for HBsAg. Among the 20 control infants born to HBsAg-negative mothers, not one was positive for HBsAg. In addition, they found that maternal hepatitis B e antigenemia correlated with a high HBsAg titer, and that both parameters were good predictors of perinatal transmission.7

Further studies also revealed that over 90% of infants were born to hepatitis B e antigen (HBeAg)-positive carrier mothers; the infection rate was nearly 100% when the mothers were positive for HBV DNA.8–11 In contrast, the infection rate was only 10% in babies whose mothers were HBeAg-negative carriers.12 Beasley et al. reported that hepatitis B immune globulin (HBIG) could only provide partial protection to these infants.13,14 These studies suggested that infants born to HBeAg-positive carrier mothers were actually the high-risk group with the most urgent need for vaccination. Therefore, early administration of the vaccine at the newborn stage is essential to a successful vaccination program.

In 1982, 2 years prior to the launch of the vaccination program, the population of Taiwan was 18 million persons, and new cases of viral hepatitis reached 20 000. There were 3200 deaths from cirrhosis (the sixth leading cause of death in Taiwan) and 3700 deaths from HCC (the leading cancer among males). Related medical costs reached approximately US$60 million per year. The authors presented cost–benefit data to the government and urged officials to begin a vaccination program, in order to prevent infants from becoming carriers, from developing liver diseases, and from burdening the national treasury.4,15,16 The economics of the viral hepatitis problem were first calculated in Taiwan, based on an annual birth rate of 300 000. Every year, at a perinatal transmission rate of 6%, there would be 20 000 infants who would become HBV carriers. If the vaccination program were to be postponed, 1500–1750 new cases would develop every month, or 50–58 per day.15 These infants would grow up and develop chronic hepatitis, liver cirrhosis, or HCC.

As the vaccines were fairly new, hostility to them was strong. The general public was against the use of the Chinese people as ‘guinea pigs’, alleging that the plasma-derived vaccine might cause infectious diseases such as AIDS or other conditions whose course cannot be predicted conditions. To disprove the myths regarding the vaccine, a group of investigators from Taipei Veterans General Hospital and their preschool-age children became the first volunteers to receive the hepatitis B vaccine before the clinical trial of hepatitis B vaccination in Taiwan was formally conducted.


Since 1981, the authors have conducted three pilot vaccine trials, using a plasma-derived hepatitis B vaccine as the test agent, at the Taipei Veterans General Hospital.17,18 Initially, the authors performed a serological-marker screening of nearly 10 000 pregnant women. Of these, 16% were found to be HBsAg-positive and 41% of these HBV carriers were also HBeAg-positive.17 Neonates born to HBsAg-positive mothers were included as candidates for the vaccination trials.

The first vaccine trial was conducted in order to identify the most economic and efficient way for immunoprophylaxis of HBV infection in high-risk neonates (born to HBeAg-positive carrier mothers).8,17 Neonates were randomly assigned into three study groups: to receive the hepatitis B vaccine alone (Group 1), in combination with one dose of HBIG at birth (Group 2), or vaccine plus two doses of HBIG administered at birth and at 1 month after (Group 3). The control group included the high-risk infants whose parents refused vaccination, but who were willing to receive regular follow up. At 6 months of age, the HBsAg-positive rate was 24% in Group 1, 11% in Group 2, and 5% in Group 3. In contrast, the HBV carrier rate was up to 90% in the control group. Therefore, the efficacy of HBV vaccination in preventing HBV infection among high-risk infants was 74% in Group 1, 88% in Group 2, and 94% in Group 3. The prophylactic efficacy in Groups 2 and 3 was significantly superior to that of Group 1, but there was no significant difference between vaccinees who received either one dose or two doses of HBIG.

During the same period, another study group, Beasley et al. at Taipei Municipal Women and Children Hospital, Mackay Memorial Hospital and the Washington Medical Research Unit in Taipei, also conducted a series of similar studies.18–21 The results from these studies showed that, with HBIG coverage from birth, the timing of the start of vaccination does not seem to be of importance within the first month of life. However, they also suggested that hepatitis B vaccination should be initiated during childbirth in order to maximize compliance and to minimize costs.18 Based on these results, the hepatitis B vaccine plus one dose of HBIG at birth became the protocol-of-choice for high-risk neonates in the nationwide vaccination program.

After the protocol for high-risk neonates had been established, the second vaccine trial at Taipei Veterans General Hospital was conducted in neonates with relatively lower risk, that is, infants born to HBeAg-negative carrier mothers.12 These neonates were randomly divided into two groups to receive either the hepatitis B vaccine alone (Group 1) or the vaccine plus one dose of HBIG (Group 2) at birth. As in the first trial, neonates (born to HBeAg-negative carrier mothers) whose parents refused vaccination trial, but who were willing to be followed up were enrolled as controls. At 6 months of age, all of the neonates in Group 1 developed antibodies to HBsAg (anti-HBs), while 96% of the neonates in Group 2 were positive for anti-HBs. None of the infants who received vaccination became HBV carriers. These results indicated that injection of HBIG at birth would not provide an additional benefit to neonates born to HBeAg-negative carrier mothers. Therefore, the government accepted that the hepatitis B vaccine alone was sufficient for the prevention of HBV infection in neonates born to HBeAg-negative carrier mothers, provided that it was administered at the earliest possible time after birth.

Although a nationwide vaccination program could be highly cost-beneficial, there might also be an immense national financial burden at the beginning of the project. Therefore, the aim of the authors’ third vaccine trial was to find an appropriate method to reduce the costs of vaccination.22 To do this, the authors evaluated the immunogenicity of two lower doses (1 µg and 2 µg) of the hepatitis B vaccine. The results showed that lower doses of hepatitis B vaccine produced significantly lower antibody titers compared with the standard dose (5 µg). As a result, 5 µg of plasma-derived hepatitis B vaccine was considered as the standard dose for all of the vaccines until the 1990s, when the recombinant hepatitis B vaccine became available.

As the hepatitis B vaccines available in the 1980s were plasma-derived, that is, they were prepared from the serum of HBV carriers after the virus was inactivated,23,24 it was important to evaluate the safety of these vaccines. To allay anxieties about the side-effects of the vaccine in neonates, the authors compared the hepatitis B vaccine with the diphtheria–pertussis–tetanus (DPT) and measles vaccines.8 Fever occurred in only 2.8% of neonates vaccinated against HBV, compared with 38.4% and 28% in neonates given the DPT and measles vaccines, respectively. Other side-effects, such as diarrhea and local skin redness, were also significantly reduced in the hepatitis-B-vaccinated group.

After the hepatitis B vaccine was proven to be highly effective and safe in the prevention of perinatal transmission, the study of its long-term efficacy gave rise to a new concern.25–27 The infant vaccinees had long been followed up for evaluation of the long-term immunogenicity and efficacy of the hepatitis B vaccine. Among children born to HBeAg-positive mothers, 92% still had high anti-HBs titers, and only 5% lost their protective antibodies by the age of 7 years. In contrast, 10% of the children of non-carrier mothers, who received the standard dose of the vaccine as neonates, had no detectable antibodies by the age of 9 years.26,27 In some cases, abrupt elevations of anti-HBs titers were noted in the follow-up period, but antibody to hepatitis B core (anti-HBc) had not been developed and the serum aminotransferase levels of these vaccinees were kept within normal limits. In these cases, it was considered that the HBV entered the body and that, instead of producing the disease, acted as a natural booster to enhance the titer levels. This phenomenon could explain why most of the high-risk infant vaccinees still had high anti-HBs titers 6 years after the last dose of hepatitis B vaccination was administered. Another important finding was that none of the children who were regularly followed up became positive for HBsAg, despite the disappearance of protective antibodies in some of them.

The recombinant HBV vaccine became available in 1992 and replaced the plasma-derived vaccine because of its improved safety profile. The authors then compared the immunogenicity of the new vaccine with the traditional plasma-derived vaccine by administering it as a booster in primary-school children who had received the original plasma-derived vaccine.28,29 Having been monitored until 14 years of age, children in both groups were found to have similar antibody levels. Approximately 80% of the children, regardless of whether they had the recombinant vaccine or the plasma-derived vaccine, still had serum antibodies. None became positive for HBsAg, including even those who had lost their anti-HBs during the follow-up period. There seemed to be no need for a second booster, a finding compatible with the current proposal from WHO that booster vaccination is not recommended for children under 15 years of age.


The government of Taiwan finally launched the nationwide vaccination program in 1984. The program, for the first 2 years, covered only neonates born to HBsAg-carrier mothers, but was extended to cover all neonates by 1986. Those who missed the scheduled vaccination were encouraged to receive the vaccine on a fee-for-service basis. Based on the results of the authors’ trials, the government accepted 5 µg of the plasma-derived vaccine as the standard dose. Hepatitis B vaccination plus one dose of HBIG at birth was regarded as the method-of-choice for the prevention of perinatal transmission of HBV in high-risk neonates. For neonates born to HBeAg-negative carrier mothers, hepatitis B vaccine alone was enough.

In the first 15 months, more than 350 000 pregnant women were screened for HBsAg; 62 359 (18%) of these were HBV carriers, and half were highly infectious (HBeAg-positive). In total, infants of 55 620 carrier mothers received the vaccination. The coverage rate of the HBIG for infants born to highly infectious mothers was 77%, and for the first, second, third, and fourth doses of vaccine was 88%, 86%, 84%, and 71%, respectively. The reported untoward reactions to the immunization were negligible.30

Four years into the program, the government carried out an evaluation to compare the results of the mass vaccination with those of the pilot study.31 Sera from children born to 786 carrier mothers, positive for HBeAg, were tested. Positivity for HBsAg was found to be 11%, and the protective efficacy 85%, exactly the same as in the pilot study findings.


The efficacy of the nationwide mass vaccination programs has been closely monitored by a study group from the National Taiwan University since the program was put in place. In 1989, 5 years after the mass vaccination project was launched, a follow-up seroepidemiological study was carried out in the Taipei district where prevaccination seroepidemiology had been previously investigated in 1984. Hepatitis B virus markers were studied in 1134 apparently healthy children under 13 years of age. The prevalence of HBsAg in children under 5 years of age decreased from 9.3% in 1984 to approximately 2% in 1989. Children between 5 and 8 years of age who were not immunized against HBV showed a significant decrease in the prevalence of HBsAg and anti-HBc. This finding suggested that the nationwide vaccination program not only protected vaccinated subjects, but also decreased the horizontal infection among children who were not vaccinated. Hepatitis B vaccination is obviously effective in protecting the majority of children in hyperendemic areas from HBV infection and from becoming chronic carriers.32

Ten years after the launch of the nationwide vaccination program, the HBsAg-carrier rates in children decreased from 9.8% in 1984 to 1.3% in 1994. The overall prevalence rate of anti-HBc was 26% in 1984, 15% in 1989, and 4.0% in 1994.33

Fifteen years after the vaccination program was implemented, the prevalence of HBsAg among persons <15 years of age decreased from 9.8% in 1984 to 0.7% in 1999, and among persons 15–20 years of age, the prevalence of HBsAg was 7% (P < 0.001) in 1999. Seropositivity for anti-HBc, which represents the prevalence of HBV infection, was found in 2.9% of persons <15 years of age and in 20.6% of persons aged 15–20 years (P < 0.001). In the same age groups, the rate of anti-HBs seropositivity was 75.8% and 70.7%, respectively (P = 0.02).34

In addition to the declined HBsAg-carrier rates, the mortality rate from fulminant hepatitis in infants was also decreased.35 Data from the National Mortality Registry System of Taiwan showed that the ratio of yearly mortality from 1975 to 1998 was 1.10 (P < 0.001), representing a progressive decrease in the number of cases. The average mortality associated with fulminant hepatitis in infants, measured from 1975 to 1984 and from 1985 to 1998, was 5.36 and 1.71 per 100 000 infants, respectively. The ratio of the average mortality in the period from 1985 to 1998 to that in the period from 1975 to 1984 was 0.32 (P < 0.001).

The study group from the National Taiwan University also reported a series of studies on the rates of HCC in children. In 1997, Chang et al. reported that the average annual incidence of HCC in children aged 6–14 years declined from 0.70 per 100 000 children between 1981 and 1986 to 0.57 between 1986 and 1990, and to 0.36 between 1990 and 1994 (P < 0.01).36 The corresponding rates of mortality from HCC also decreased. The incidence of HCC in children aged 6–9 years declined from 0.52 for those born between 1974 and 1984 to 0.13 for those born between 1984 and 1986 (P < 0.001). An interesting finding was also noted by Chang et al. that the boy–girl incidence ratio decreased steadily from 4.5 in 1981–1984 (before the implementation of the program) to 1.9 in 1990–1996 (6–12 years after the vaccination program was launched).36 The incidence of HCC in boys born after 1984 was significantly reduced in comparison with those born before 1978. No significant decrease in HCC incidence was observed in girls born in the same periods. These findings suggest that boys may benefit more from the HBV vaccination than girls in terms of the prevention of HCC.


Although effective vaccines have been available for almost 20 years, hepatitis B remains a major public health problem. In 1992, government delegates to the World Health Assembly agreed that countries with an HBV carrier prevalence of 8% or higher should have hepatitis B vaccine integrated into their national immunization programs by 1995, and that all countries should have such immunization in place by 1997. Since 1997, global eradication of hepatitis B has become one of the most important goals of the WHO.37

Besides Taiwan,  at  least  four  other  countries  have had universal immunization programs against hepatitis B implemented for more than 10 years, including Bulgaria (1991),38 Italy (1991),39 Israel (1989),40 and Spain (1991).41 Because the rewards of effective implementation of the programs in Taiwan and these countries are becoming apparent, the success of these vaccination programs offers an exemplary model for other countries.

In 2001, 9 years after the Global Advisory Group of the Expanded Programme on Immunization (WHO) set 1997 as the target year for integrating hepatitis B vaccination into national immunization programs worldwide, 129 countries have included hepatitis B vaccine as part of their routine immunization programs. In addition, the universal hepatitis B immunization program will be implemented in 41 out of the 51 countries of the WHO European Region by the end of 2002.

However, some countries outside Europe still have difficulties in incorporating the hepatitis B vaccine into universal childhood immunization programs. The most important reasons for this are economic constraints and the inability to have a constant vaccine supply. In the next decade, the main tasks will be to expand the use of hepatitis B vaccines and to sustain vaccination programs for those countries and regions that cannot afford it. As work still remains to be done to support and implement interventions, the WHO goal to eradicate hepatitis B will be achieved within the first half of this century.42


The study was supported by a grant (VGH-TPE 91–254) from Taipei Veterans General Hospital, Taipei, Taiwan.