Assessing the Likely Utilization of New Vaccines
The expected health benefits from a vaccine development project depend not only on the number of potential recipients of the vaccine (the target population), but also on the proportion of the target population that will actually receive it.* However, predicting future utilization rates of vaccine candidates is difficult, even for a relatively homogeneous population, because vaccine utilization depends on many interrelated and complex factors. These factors include:
the capacity of the health care system to deliver the vaccine to potential recipients
the effects of statutory interventions, such as the U.S. school immunization requirements (including the extent to which they are enforced)
characteristics of the target population, including its access to health care providers and the ease with which its members can be identified by the health care system (in turn, dependent on size, composition, age, and socioeconomic status of the target population)
characteristics of the vaccine affecting provider utilization (route of administration, storage conditions, shelf-life, cost of delivery, and special procedures required prior to or during vaccination)
characteristics of the vaccine that affect patient acceptance (number of doses, route of administration, adverse reactions, and cost)
provider attitudes toward the vaccine, which are affected by perceptions of efficacy, safety, liability, ease of administration, profitability, professional consensus, and patient need or demand
target population attitudes toward the vaccine, which are influenced by perceptions of the likelihood of contracting the disease, its severity if contracted, and the vaccine’s efficacy and safety, and
by the influence of media coverage (e.g., pertussis vaccine acceptance in the United Kingdom)
These factors affect each other in ways that are not fully understood or quantifiable. In its report on vaccine priorities for important diseases in the United States, this committee considered the issues of vaccine availability and utilization separately (Institute of Medicine, 1985a). A similar approach has been adopted for this project because factors affecting availability and utilization in the developing world are different. In addition, for some vaccines considered in this assessment, there may not be a need for universal availability or utilization.
Availability of a licensed vaccine depends on the willingness of a pharmaceutical company or other entity to manufacture it and on the company’s ability to produce the vaccine in sufficient quantities to meet demand. Factors influencing a pharmaceutical company’s willingness to manufacture a vaccine include:
profitability in public or private sale, which is affected by market size and composition (e.g., ability to pay), public health initiatives, patentability or status as sole supplier, and provider and lay acceptance
legal concerns, particularly costs associated with vaccine injury compensation liability, which is a major issue in the United States
technical difficulty of production and of distribution (in a manner that ensures potency) in target areas
humanitarian and public relations issues
Probably some of the vaccine candidates considered in this analysis will not be commercially attractive to manufacturers in the United States or other industrialized countries, and they eventually will be made locally in developing countries. The committee had insufficient information to predict the availability of new vaccines,* so it assumed that any vaccine that met reasonable safety and efficacy standards (e.g., U.S. licensing standards or World Health Organization [WHO] guidelines) would be available in quantities sufficient to meet the demand. Such a scenario would probably require the transfer of vaccine
production technology to developing countries, most likely under the guidance and assistance of such international agencies as the WHO.
Prediction of Utilization in Defined Populations
To develop a theoretical basis for predicting vaccine utilization in the United States, the committee adapted basic concepts of the “health belief model” (HBM), a social-psychological model of health-related decision making developed by U.S. Public Health Service psychologists in the early 1950s (Becker, 1974; Janz and Becker, 1984). Questions exist concerning the validity of the HBM as a predictive tool, but it offers an appropriate conceptual framework for studying public immunization behavior within a defined, relatively homogeneous population. A brief review of the health belief model and a description of the methods used by the committee to predict vaccine utilization in the United States appear in Volume I of the committee’s report (Institute of Medicine, 1985a).
To estimate probable utilization, each vaccine candidate was assigned scores to reflect lay (target population) and provider perceptions of risk of illness, severity of disease, vaccine benefits, and barriers to vaccination. These scores were then combined into overall vaccine acceptance scores for each candidate. The combined scores were used to estimate vaccine utilization in situations in which individuals could freely choose their health behavior, and in which other factors, for example, the efficiency of the health care system, did not otherwise limit utilization. (Mandatory immunization requirements, e.g., for school admission, if enforced, tend to override the influence of these other factors in determining acceptance.)
Little research has been conducted to determine whether the health belief model as formulated for the U.S. population is applicable to other societies. Several studies of voluntary vaccine utilization in developing countries suggest that the factors determining relevant health decisions in the developing world are similar to those postulated in the HBM (Adeniyi, 1972; Azurin and Alvero, 1971; Hingson, 1974; Hingson and Lin, 1972; Hsu, 1955; Lin et al., 1971; Ogionowo, 1973; Ristori, 1969). It is important to recognize, however, that local perceptions of disease and health care are molded by social, educational, and economic conditions radically different from those in the United States. Accurate assessment of these perceptions requires input from individuals familiar with the cultures involved.
The committee believes that its general approach to estimate U.S. vaccine utilization probably could be adapted to any defined population, for example, a specific country, where the range of HBM perceptions can be estimated with reasonable confidence.
Utilization of Vaccines in the Developing World
A committee subgroup evaluated the feasibility of making utilization predictions for vaccine candidates in the entire developing world. The subgroup determined that because of the diversity of health care systems and populations involved, it would be unrealistic to attempt to combine predictions for countries or regions. Also, in many areas, health care systems cannot routinely reach large portions of the target population. In some countries, successful periodic mass immunization campaigns have been conducted, utilizing the mass media and personnel with minimal health training. For most of the developing world, however, the only consistent method for delivering vaccines is through the WHO-supported Expanded Program on Immunization (EPI), which generally provides DTP (diphtheria and tetanus toxoids and pertussis vaccine), polio, BCG (Bacillus Calmette-Guerin), and measles vaccines to infants and children, and tetanus toxoid to susceptible women (including those who possibly are pregnant). All of the vaccine candidates considered in this analysis could be adapted to conform to the current EPI schedule of administration (see Appendixes D-1 through D-19). Its vaccination schedule recommendations are intended to be adapted to local conditions and constraints. WHO-EPI flexibility in this regard (World Health Organization, 1985) may establish new opportunities for optimizing the delivery of vaccines discussed in this report.*
Because of these constraints, the committee assumed that utilization rates of vaccine candidates would be uniform in target populations in the developing world, and that these rates would be identical to those achieved through EPI. Thus, potential utilization is not used to differentiate among vaccine candidates in this analysis.
This assumption is not a recommendation that all or any of these vaccine candidates should be incorporated into EPI worldwide or in a particular country. Such decisions should be based on local assessments of disease burdens, resources, and other considerations. The committee
In some cases, delivery of a vaccine considered in this assessment at the ages when EPI vaccines are presently administered may not provide optimal protection against that disease. In these cases, predictions of vaccine efficacy and estimates of the proportion of the disease burdens that are vaccine preventable have been adjusted accordingly (see Chapter 5 and specific appendixes). While rabies immunization could be added to the EPI, this strategy is probably not practical for some vaccine candidates, and an alternative scenario has been analyzed for them, namely post-exposure prophylaxis. Although delivery in accordance with EPI schedules is assumed in this analysis, the epidemiologic features of certain diseases suggest that strategies for vaccine delivery shortly after birth (for hepatitis B vaccine) or before school entry (for epidemic meningococcal meningitis) deserve consideration in some regions. WHO-EPI presently encourages giving oral poliomyelitis vaccine beginning at birth in countries where polio remains endemic (Henderson, personal communication, 1985) .
also recognizes that various technical problems could interfere with the addition of these vaccines to the EPI. For example, antagonisms could exist between potential vaccine candidates and those already included in the WHO program. Moreover, changes may be made in the basic EPI as new and better vaccines are developed or as information on current vaccine use is refined. Further research will be needed to investigate the feasibility and desirability of substantially increasing the number of antigens administered at any one time.
The expected health benefits from a vaccine depend, in part, on the proportion of the target population that actually receives it (which, in turn, affects the probability of acquiring the disease). Predicting utilization rates is complicated by various factors, including availability, cost, the health system capacity to deliver vaccine, statutory interventions, the target population, and provider and recipient acceptance.
Because delivery of vaccines would probably be achieved through the WHO-EPI, the committee assumed for the analysis that utilization rates of vaccine candidates would be uniform across target populations. (Actual decisions about whether to incorporate specific vaccines into EPI should be based on local assessments of disease burdens, resources, and other considerations.) Thus, utilization is not used to differentiate among vaccine candidates in this analysis. Methods described in the committee’s previous report can be adapted if utilization is expected to differ among vaccines.
Adeniyi, J.D. 1972. Cholera control: Problems of beliefs and attitudes. Int. J. Health Educ. 15:238–245.
Azurin, J.C., and M.Alvero. 1971. Cholera incidence in a population offered cholera vaccination: Comparison of cooperative and uncooperative groups. Bull. WHO 44:815–819.
Becker, M.H., ed. 1974. The Health Belief Model and Personal Health Behavior. Thorofare, N.J.: Charles B.Slack.
Henderson, R.H. 1985. Personal communication. World Health Organization, Geneva.
Hingson, R. 1974. The impact of health beliefs on behavior during an immunization program in rural Haiti, 1972. Health Educ. Mono. 2:505–508.
Hingson, R., and N.Lin. 1972. Communication participation and individual receptivity to health innovations. Int. J. Health Educ. 16:5–14.
Hsu, F.L.K. 1955. A cholera epidemic in a Chinese town. Pp. 135–154in Health Culture and Community, B.Paul, ed. New York: Russell Sage Foundation.
Institute of Medicine. 1985a. New Vaccine Development: Establishing Priorities, Volume I. Diseases of Importance in the United States. Washington, D.C.: National Academy Press.
Institute of Medicine. 1985b. Vaccine Supply and Innovation. Washington, D.C.: National Academy Press.
Janz, N.K., and M.H. Becker. 1984. The health belief model: A decade later. Health Educ. Quart. 11:1–47.
Lin, N., R. Hingson, and J.Allwood-Paredes. 1971. Mass immunization campaign in E1 Salvador, 1969. Evaluation of receptivity and recommendation for future campaigns. HSMHA Health Rept. 86:1112–1121.
Ogionowo, W. 1973. Socio-psychological factors in health behavior: An experimental study on methods and attitude change. Int. J. Health Educ. 16:1–15.
Ristori, C. 1969. Selection of work procedures and establishment of objectives in vaccination programs. Bull. PAHO 66:32–42.
World Health Organization. 1985. Reappraisal of early OPV and DPT immunization. Weekly Epidemiol. Rec. 60(37)284–286.