Excellent vaccines have been available for the past decade, but the incidence of hepatitis B in the United States has increased by 37% [1]. Approximately 300 000 Americans become infected with hepatitis B virus (HBV) annually, and an estimated 5000 HBV-related deaths from fulminant hepatitis, cirrhosis, or hepatocellular cancer occur each year. Seroepidemiologic studies indicate that about 5% of the U.S. population has been infected with HBV [2].

The paradoxic increase of HBV since the introduction of an effective vaccine reflects the difficulties that stand between the scientific development of a powerful preventive measure and its effective use to achieve the desired individual and societal benefits. After the 1981 licensure of a hepatitis B vaccine with 80% to 95% protective efficacy, guidelines for its use were issued, specific to risk groups defined by occupation, medical conditions, and lifestyle. The implementation of these guidelines has been alarmingly poor for all target groups.

Many reasons have been cited for this slow acceptance of hepatitis B vaccine. Many physicians caring for adults have not become attuned to the provision of preventive services and have been unaware of the public health impact of hepatitis B. Furthermore, to select from among their patients those who should receive vaccine, physicians have had to ask about recreational risk factors, including intravenous drug use and sexuality (for example, whether patients have had multiple sex partners and whether men have had sex with other men). Physicians often do not explore such features of the medical history or are not sufficiently skilled to elicit complete responses; and patients may protect their privacy. With 13 different risk groups identified and each group requiring a different implementation strategy, vaccine delivery programs have been complex. Furthermore, 30% of patients with acute hepatitis B have no acknowledged risk factor for the disease and, therefore, fall outside the scope of a targeted vaccination program [3]. The high cost of vaccine and (unfounded) concerns that the plasma-derived vaccine was unsafe were further impediments to enthusiastic acceptance, both in the medical community and by the public.

The development of the current recombinant HBV vaccines, synthesized in yeast and therefore free of even the theoretic concern of vaccine risk (but still costly), still has not resulted in a dramatic increase in acceptance. Even among the target group most knowledgeable about the vaccine's benefits (health professionals), most persons did not receive immunization. In the meantime, HBV has increased dramatically among intravenous drug users and other young adults, whose focus on the present has made future-directed public health measures a low priority.

Progress is currently being made in two areas. Prenatal testing of pregnant women for hepatitis B surface antigen and consequent immunoprophylaxis of newborns at risk with hepatitis B immune globulin and vaccine are now routine obstetric care and promise to prevent approximately 6000 chronic hepatitis B infections per year [4]. In addition, the full implementation of the recent Occupational Safety and Health Administration regulations affecting health care workers, laboratory technologists, fire and police personnel, morticians, and others whose work might expose them to blood-borne pathogens has the potential to prevent about 5% of hepatitis B infections.

In response to the increasing lifetime risk for HBV infection for all Americans and recognizing the practical impediments to effectively identifying patients with risk factors, the long-term public health strategy has shifted to focus on a population more accessible for immunization, infants. The Immunization Practices Advisory Committee of the Centers for Disease Control [1], the American Academy of Pediatrics, and the American Academy of Family Practice now recommend the universal immunization of all infants against hepatitis B. Similar recommendations have been made in Canada. This recent addition to the routine "baby shots" has not been without pain. Indeed, doctors and nurses who provide well-baby care have not been happy at the prospect of three added intramuscular injections during the first 18 months of life—especially because the main risk for acquiring hepatitis B is many years in the baby's future. The anticipated inclusion of hepatitis B vaccine in the same syringe with diphtheria-pertussis-tetanus vaccine should alleviate this concern.

The thoughtful cost-effectiveness study of Bloom and colleagues [5] in this issue of the Annals further informs the debate regarding the various hepatitis B vaccination strategies. Public health policy makers are increasingly interested in using such economic analyses to guide their decisions. Thus, it was reassuring that the authors' analysis [5] indicated that hepatitis B vaccine now is even more cost-effective than it was a decade ago [6]. Bloom and colleagues conclude that there should be a two-pronged immunization strategy. Their recommendation of a selective infant immunization program has already been pre-empted by the universal standard that is now in place. They also support universal immunization of 10-year-olds just before they enter the high-risk teen and young adult years. Such a policy should be well received by pediatricians, family physicians, obstetricians, and internists who support greater preteen-physician interactions as an opportunity for immunization updating and lifestyle counseling.

However, the feasibility of reaching all 10-year-olds with an expensive vaccine requiring three intramuscular injections faces formidable logistic problems. To be successful, both the public and physicians must be convinced of the benefit by effective educational programs. This must be linked to expanded collaborative efforts of the public and private health care sectors to reach the majority of preteens who currently have no regular contact with the health care system. As with other universal immunization programs, persuasion alone is unlikely to suffice. New school immunization requirements presumably would provide the necessary regulatory impetus. Indeed, a few health departments already have begun to provide hepatitis B vaccine in junior high school, and some colleges are contemplating making it an admission requirement. The suggested implementation of a booster dose to all 20-year-olds will be even more difficult because no school registration regulation can be brought to bear. Although the duration of protection after hepatitis B vaccination is not yet known, protection against clinical disease and the acquisition of the hepatitis B surface antigen carrier state seem to extend substantially beyond the time when measurable antibody levels have waned [7]. Therefore, it would be pragmatic to await further epidemiologic evidence of need before committing to a universal booster program.

Although it is cost-effective in the long run, start-up costs of any universal immunization initiative will be large. High vaccine costs remain a substantial impediment. Vaccine prices have been reduced marginally in recent years (usually for large-volume purchasers); however, they remain substantially higher than in other parts of the world where population-based vaccination programs are underway. Both the study by Bloom and colleagues [5] and a similar analysis by Krahn and Detsky in Canada [8] indicate that the cost-effectiveness of universal immunization is sensitive to vaccine price.

Because hepatitis B virus infection is strongly associated with the occurrence of subsequent hepatocellular carcinoma [9, 10], many consider hepatitis B vaccine the first anti-cancer vaccine. In parts of the developing world with high rates of liver cancer, the vaccine is being applied with the intention of reducing the incidence of this cancer. Similarly, because almost half of hepatitis B infections in the United States are acquired sexually [3], hepatitis B vaccine is the only current immunization against a sexually transmitted infection.

Universal immunization of infants (or 10-year-olds) will not have a substantial impact on the hepatitis B epidemic for more than a decade. Therefore, such a program would not obviate the need for vigorously implementing the current program of immunizing adults in targeted risk groups, especially sexually active young adults.