Chapter 20:  Vaccine-preventable Diseases

Causes and Epidemiology of Diseases Prevented by Vaccines Used in National Immunization Programs

The epidemiology and burden of vaccine-preventable diseases vary by country and region partly because of differences in vaccine uptake. Numerous other factors that contribute to the disease burden include geography, seasonal patterns, crowding, nutritional status, travel to and from other countries, and possibly genetic differences in populations that affect disease severity. Table 20.1 summarizes the features of selected vaccines in use in childhood immunization programs throughout the world.

Estimates of the Current Burden of Vaccine-preventable Diseases and of the Burden Averted by Vaccination

Table 20.2 provides WHO estimates of deaths from selected vaccine-preventable diseases for 2001, taking immunization coverage rates into account. The greatest burden of disease is in Sub-Saharan Africa, which accounts for 58 percent of pertussis deaths, 41 percent of tetanus deaths, 59 percent of measles deaths, and 80 percent of yellow fever deaths. East Asia and the Pacific has the greatest burden from hepatitis B, with 62 percent of deaths worldwide. South Asia also experienced a high disease burden, particularly for tetanus and measles.

Table 20.2 also shows the extent of mortality in the absence of immunization and the estimated number of deaths averted by vaccination. In 2001, vaccination averted up to 52 percent of yellow fever deaths, 61 percent of measles deaths, 69 percent of tetanus deaths, 78 percent of pertussis deaths, 94 percent of diphtheria deaths, and 98 percent of polio deaths that would have occurred in the absence of vaccination. These results demonstrate the significant effect that vaccination programs have had on worldwide disease burden. The figures also show that vaccination programs have been less successful in reducing the disease burden in Sub-Saharan Africa, where coverage rates are lower.

Table 20.3 reports WHO estimates of disability-adjusted life years (DALYs) lost from vaccine-preventable diseases by region for 2001, demonstrating the high burden of disease worldwide from disability associated with sequelae of hepatitis B (liver cancer and cirrhosis), pertussis, and tetanus.²

Expanded Program on Immunization

WHO initiated the EPI in 1974 to provide countries with guidance and support to improve vaccine delivery and to help make vaccines available for all children (Hadler and others 2004; Turk 1982; WHO 1974). A standard immunization schedule was established in 1984 on the basis of a review of immunological data for the original EPI vaccines: BCG, diphtheria-tetanus-pertussis (DTP), oral polio, and measles vaccines (Halsey and Galazka 1985).

Today, national immunization programs in developing countries are responsible for improving access to the traditional EPI antigens and introducing new vaccines. In 2002, the EPI introduced the Reaching Every District strategy, which focused on achieving an 80 percent coverage rate of DTP3 in 80 percent of districts and using immunization contacts to deliver other high-priority child health interventions. In addition to delivering vaccinations, national immunization programs are concerned with the quality and safety of immunization through the adoption of safe injection technologies (autodisabled syringes, storage boxes, and incinerators) and proper cold chain and vaccine stock maintenance.

In most developing countries, immunizations are provided through a system of fixed facilities at different levels of the health system. Immunization campaigns are discrete, time-limited efforts at national or subnational levels that usually focus on specific antigens (for example, polio). Mobile strategies rely on the use of specialized vehicles to transport health professionals and vaccines to deliver services to remote or migrating populations. Outreach is a strategy by which staff members from a health facility travel to villages and surrounding areas to administer vaccines. Extended outreach refers to more targeted and intensive efforts.

In 1999, the major international development partners involved in immunization (for example, WHO, UNICEF, the World Bank, and bilateral donors) joined the Bill & Melinda Gates and Rockefeller Foundations, the vaccine industry, and nongovernmental organizations to create GAVI (http://www.vaccinealliance.org) to increase access to new and underused vaccines in the world's poorest countries, improve access to basic immunization services, and accelerate research and development pertaining to new vaccines and delivery technology. Through the Vaccine Fund, GAVI raised more than US$1.3 billion to strengthen immunization systems, introduce new vaccines, and support safe injection practices. More than US$3 billion has been pledged for the next 10 years. Between 2000 and 2003, an additional 4 million children were vaccinated with DTP3, 42 million with hepatitis B, nearly 5 million with Hib, and more than 3 million with yellow fever vaccine.

Costs and Cost-Effectiveness of Existing Vaccination Programs

Brenzel and Claquin (1994) and GAVI (2004) estimate the cost per fully immunized child (FIC) for the traditional six EPI antigens as approximately US$20.³ We evaluated the cost per FIC for the childhood EPI cluster antigens by World Bank region on the basis of published and unpublished data. These studies used a standard costing approach that estimated the costs of labor, vaccines, supplies, transportation, communication, training, maintenance, and overhead and included the annualized value of equipment, vehicles, and building space (Khaleghian 2001; USAID, Asia–Near East Region 1988; WHO 1988). The number of FICs in these studies was measured using community-based sample surveys (Henderson and Sundaresen 1982).

Our literature review found 102 estimates of total and unit immunization program costs from 27 countries between 1979 and 2003 for different immunization delivery strategies (Berman and others 1991;1 Beutels 1998;1 Beutels 2001; Brenzel 2005; Brenzel and Claquin 1994; Brinsmead, Hill, and Walker 2004; Creese 1986; Creese and Domínguez-Ugá 1987; Domínguez-Ugá 1988; Edmunds and others 2000; Griffiths and others 2004; Levin and others 2001; Pegurri, Fox-Rushby, and Walker 2005; Robertson and others 1992; Soucat and others 1997; Steinglass, Brenzel, and Percy 1993). All costs were converted to 2001 U.S. dollar equivalents. Because total and unit costs are related to population size, table 20.4 reports population-weighted results only. National immunization program refers to total national costs for all strategies.

The population-weighted mean cost per FIC for all regions and all strategies is approximately US$17, with a range of US$3 to US$31. The lowest mean population-weighted cost per FIC was for extended outreach services (US$5.81), perhaps because the strategy is a more targeted approach. Routine facility-based strategies had lower average costs (US$13.65 per FIC) than campaigns (US$26.82 per FIC) or mobile strategies (US$25.84 per FIC). Higher unit costs associated with these strategies are possibly attributable to a different mix of inputs as well as greater expenses for per diems, fuel, and social mobilization. The results also vary by World Bank region, with East Asia and the Pacific (US$13.25) and Sub-Saharan Africa (US$14.21) having lower estimates of cost per FIC than Europe and Central Asia (US$24.12) and the Middle East and North Africa (US$22.15).

The findings of our analysis are generally supported by the literature (Creese 1986; Brenzel and Claquin 1994; Khaleghian 2001), which has shown that variation in the cost per FIC is related to the mix of delivery strategies, the prices of key inputs such as vaccines, and the overall scale of programs. In addition, an analysis of 13 national financial sustainability plans for immunization reveals a wide range in the cost per FIC by region and strategy.⁴

Recurrent costs are the lion's share of total immunization costs (80 percent for fixed facility strategy and 92 percent for campaigns), which has implications for the need for continuous and predictable program financing. Labor costs account for the largest share (roughly 30 to 46 percent of total cost) for all strategies except extended outreach. Vaccine costs range from 8 percent for mobile strategies to 29 percent for extended outreach strategies. Transportation costs account for the second-largest share of EPI costs for mobile strategies, while building costs account for a greater share of fixed facility strategies.

Using data from table 20.4 on costs per FIC and multiplying by the size of the population covered, we estimate US$1.17 billion for the total cost of immunization programs in developing countries in 2001, with a range of US$717 million to US$1.48 billion. At US$20 per FIC, the cost of the six traditional vaccines in developing countries would have amounted to US$1.57 billion in 2001. Table 20.5 shows that the estimated cost per death averted ranges from US$205 in South Asia and Sub-Saharan Africa to US$3,540 in Europe and Central Asia. These results suggest that the cost per death averted rises with coverage rates. Europe and Central Asia, Latin America and the Caribbean, and the Middle East and North Africa had higher coverage rates in 2001, resulting in fewer deaths that could be averted. The table also shows that the cost per DALY from the traditional EPI vaccines ranges from US$7 to US$438, depending on region, mix of strategy, and levels of scale.

Our analysis highlights the variation in cost per FIC by region and strategy and demonstrates the value of more disaggregated results for making policy decisions. However, given the limited sample of estimates available for the regions and strategies, the results should be used as an indicative guide for policy making and not as a substitute for country-specific cost-effectiveness evaluations of strategies. In addition, our estimates do not take into account household costs, such as time spent seeking services, and other social costs. Our estimates also do not consider the direct and indirect costs of acute illnesses prevented by vaccination or the costs of long-term complications from disease and of adverse events associated with vaccination (though the latter are unlikely to have a significant impact on costs because rates of serious complications are extremely low). Furthermore, the analysis focuses on FICs and underemphasizes the benefits of partial immunization. Future economic evaluations of immunization program alternatives could consider these factors as a critical step in determining the allocation of scarce resources among high-priority health interventions.

Cost-Effectiveness of Increasing Immunization Coverage for the Traditional EPI

WHO (2004) estimates that in 2001, 30 million children were inadequately immunized with DTP. Achieving higher coverage rates by improving access for remote populations, accelerating immunization delivery strategies, and introducing new vaccines will mean increasing the level of investment (Batt, Fox-Rushby, and Castillo-Riquelme 2004).

We estimated the costs of scaling up EPI coverage for a hypothetical population of 1 million in each region between 2002 and 2011. Costs were reported in 2001 dollars, and a 3 percent discount rate was applied. Brenzel (2005) provides details on the methods. The costs of scaling up coverage are based on vaccine and delivery costs per dose. We derived vaccine costs from the unit price of each vaccine (provided by WHO, UNICEF, and the Vaccine Fund); wastage rates for vaccines and injection supplies by strategy; the required injection supplies; and the number of doses per FIC. A 2 percent adjustment was made for inflation. We used data on the cost per FIC generated earlier to derive delivery costs per dose by strategy and region by subtracting the costs of vaccines, injection supplies, and fixed costs.

Fixed costs were excluded from the scaling-up exercise because they were assumed to remain constant during the projection period.

• First, the largest projected coverage increase of 9 percentage points (figure 20.1) may not require additional infrastructure investments.

• Second, how and to what extent fixed costs would change by region is uncertain, and a conservative approach would be to exclude them.

• Third, because most immunization costs are recurrent costs, the analysis focuses on these.

• Finally, because the scale factor is derived from the unit costs of a health center visit, the assumption of constant fixed costs in the short run appears reasonable.

Previous studies have found that the main cost drivers of immunization costs are the mix of strategies and the scale of immunization programs (Brenzel and Claquin 1994; Domínguez-Ugá 1988; Robertson and others 1992; Soucat and others 1997). Countries are unlikely to achieve 90 percent or more coverage relying on fixed facilities alone because of limited population access. We estimate the proportion of FICs obtained for each strategy by region.⁵ The best mix of strategies for increasing coverage will vary by country depending on the dispersion of the population, the access to health facilities, the vaccines being delivered, and the effectiveness of various strategies in reaching target populations. Estimates are also adjusted for the level of scale by a factor derived from the unit cost per health center contact by coverage level (Mulligan and others 2003), and details are provided in Brenzel (2005).

The total additional cost of reaching higher coverage levels was divided by the number of deaths averted. Coverage projections for 2002–11 were based on statistical modeling of official WHO and UNICEF estimates for the period between 1995 and 2002 for all developing countries. The model relates coverage in future years to that in the previous year, with the relationship between past and future coverage differing for each region and economic status combination.

Figure 20.1 shows historical and projected coverage rates by region. The figure shows that coverage increased in all the regions during the late 1980s under universal childhood immunization. After 1990, when funding for universal childhood immunization waned, the figure indicates the subsequent stagnation and, in some cases, the declines in coverage rates. For the scaling-up period, we project that the coverage of FICs will increase from 78 to 79 percent in East Asia and the Pacific, from 92 to 95 percent in Europe and Central Asia, from 88 to 90 percent in Latin America and the Caribbean, from 91 to 95 percent in the Middle East and North Africa, from 70 to 79 percent in South Asia, and from 52 to 61 percent in Sub-Saharan Africa. The projections show that three of the six regions are expected to achieve 90 percent FIC by 2011. East Asia and the Pacific, South Asia, and Sub-Saharan Africa will require additional intensive efforts to achieve higher coverage rates.

Table 20.6 reports the results of the scaling-up analysis for the traditional EPI vaccines, for tetanus toxoid vaccination for women of reproductive age, and for selected new vaccines. The discounted incremental cost per child vaccinated with the traditional EPI vaccines ranges from US$10.89 in Latin America and the Caribbean to US$12.84 in the Middle East and North Africa. The number of discounted deaths averted because of full immunization depends on incremental coverage rates and varies from 747 in Europe and Central Asia to 14,584 in Sub-Saharan Africa, resulting in regional variations in the discounted incremental cost per death averted from US$169 in Sub-Saharan Africa to US$1,754 in Europe and Central Asia.⁶

DALYs were estimated indirectly based on the ratio of deaths to DALYs for each disease in 2001. This ratio is applied to the hypothetical population in each World Bank region over the projection period. Calculated this way, the number of DALYs averted will not account for changes in the average age of infection that ordinarily results from expanding immunization coverage. This method over-estimates the number of DALYs and thus under-estimates the cost/DALY. Cost-effectiveness ratios should be treated as indicative only. The cost/DALY ranges from $2 to $20 for scaling up traditional immunizations.

For tetanus toxoid immunization, the additional discounted cost per person vaccinated ranges from US$3.28 to US$4.06. The cost per death averted varies from US$271 to more than US$190,000. The results of this analysis fall within the range of estimates reported in the literature (Berman and others 1991; Steinglass, Brenzel, and Percy 1993). Differences in coverage levels and in protection against neonatal tetanus through skilled delivery contribute to the variation in results across regions.

The analysis shows both an increase in costs and potential benefits from scaling up immunization programs. In practice, the costs and benefits related to scaling up in any one region will be highly dependent on a few countries or subnational areas within countries. Aggregate country- or region-level data do not reveal the efficiency that could best be obtained by targeting immunization efforts on specific countries or geographic areas rather than making diffuse investments across regions. For instance, Miller and others (1998) show that India and Nigeria contribute the most to estimates of global measles deaths; therefore, reducing transmission in those countries would contribute the most to reducing the global disease burden caused by measles.

Despite its importance for policy, empirical and country-specific evidence on how immunization program costs change as coverage increases is lacking. Because scaling up immunization coverage will require more intensive efforts to find unvaccinated children, an extra cost for vaccinating each additional child is generally expected. Nevertheless, most cost-effectiveness studies assume constant returns to scale (Elbasha and Messonnier 2004; Karlsson and Johannesson 1998) even when emerging evidence suggests that the cost of vaccinating each additional child may rise with the size of delivery unit (Valdmanis, Walker, and Fox-Rushby 2003). Box 20.1, which focuses on scaling up traditional immunization coverage, and box 20.2, which focuses on new antigens, summarize the results of two studies that shed more light on this subject.

Costs and Cost-Effectiveness of Adding New Antigens to the Current Immunization Schedule

We also estimated the additional costs per person vaccinated and cost per death averted of introducing new and underused vaccines into the traditional EPI in a hypothetical population of 1 million in each region between 2002 and 2011. The new vaccines considered protect against hepatitis B, yellow fever, Hib, measles, rubella, Japanese encephalitis, and meningococcal A, as well as inactivated polio vaccine (IPV). For comparison purposes, we assumed that new vaccines were introduced in 2002.

The additional cost of combination vaccines is net of the original cost of DTP vaccination to avoid duplication. The delivery cost per FIC was apportioned to individual antigens on the basis of the share of number of doses per FIC for that antigen (Brenzel 2005). Cost estimates are based on the number of doses required for full immunity (that is, hepatitis B, Hib, and IPV vaccines require three doses for full immunity, and meningococcal A requires two doses for full immunity). Results are reported in table 20.6.

The analysis also assumes that an additional visit to a health facility is required for new doses (depending on timing in the EPI schedule).Combination vaccines may be more cost-efficient because of potential savings in supplies, syringes, and health workers' time, in addition to the overall health benefits of reducing the number of required injections. However, if the combination vaccine does not reduce the number of visits a child would ordinarily need to make to a health facility, any cost savings may be subsumed by the higher costs of increasing coverage.

The discounted incremental cost per person ranges from less than US$1 to US$16.23, depending on the unit price of vaccine, the type of vaccine, the delivery strategy, and the coverage levels. The results lead to several conclusions:

• First, the additional incremental cost per person vaccinated is relatively small for some new vaccines.

• Second, because fixed costs are excluded, the results represent conservative estimates of additional costs.

• Third, because of price uncertainty, cost variations are greatest for newer vaccines, such as the DTP-IPV combination.

The second opportunity for measles has the lowest cost per death averted, ranging from US$23 to US$1,906 for fixed facility strategies, and from US$65 to US$1,363 for campaigns These results are consistent with the literature. Foster, McFarland, and John (1993) find an incremental cost per death averted ranging from US$335 to US$552 in urban areas and from US$327 to US$706 in rural areas. The Africa Measles Partnership (2004) estimates a cost per death averted of US$131 to US$393 in the African context, but these figures include the costs of infrastructure.

In the hypothetical populations, the incremental cost per death averted for the pentavalent vaccine ranged from US$1,433 to more than US$85,000, depending mostly on the number of potential deaths that could be averted. Although a wide range of results was found, these estimates are supported by the literature. Miller (1998) estimated between US$3,127 and US$3.2 million per life saved for Hib vaccine. Brinsmead, Hill, and Walker's (2004) systematic review of the literature on the cost-effectiveness of Hib vaccine finds wide variations in results because of methodological differences and epidemiological and health system characteristics. The discounted incremental cost of introducing the pentavalent (DTP–hepatitis B–Hib) vaccine is roughly equal to the total mean cost of the traditional vaccine package estimated earlier. This finding implies that introducing this combination vaccine may double the financial requirements, an implication that is supported by data from national financial sustainability plans for immunization (Lydon 2004).

The incremental discounted cost per person vaccinated with a birth dose of hepatitis B is approximately US$2, and that for the tetravalent vaccine was between US$7 and US$8. The 10-year time period for our analysis is too short to accumulate deaths averted resulting from hepatitis B vaccination because deaths from liver cancer occur at older ages. Beutels's (1998, 2001) reviews of studies of the cost-effectiveness of introducing hepatitis B vaccine indicate that results vary depending on assumptions of endemicity and the methodology used, with a cost per death averted ranging from US$3,500 to US$271,800.

Rubella vaccination had a low additional cost per person vaccinated, at slightly more than US$1. Golden and Shapiro (1984) found that vaccinating all prepubertal children with rubella vaccine had the highest benefit-cost ratio (ranging from US$1.70 to US$1.96). Most benefits were future cost savings from long-term institutional care. When rubella was delivered in combination with measles and mumps, the benefit-cost ratios varied from US$4.70 to US$38.80 (Hinman and others 2002).

The additional cost per person vaccinated with one dose of Japanese encephalitis vaccine was between US$4.37 and US$4.56. A study in Thailand using two doses showed a cost per child ranging from US$2.31 to US$4.20, depending on the mode of delivery (Siraprapasiri, Sawaddiwudhipong, and Rojanasuphot 1997). Ding and others (2003) estimate a cost per case averted of US$258 and a cost per DALY averted of US$16.80 for a five-dose inactivated Japanese encephalitis vaccine.

Our analysis suggests an additional discounted cost per person vaccinated for injectable polio vaccine of between US$6.60 and US$7.32, depending on coverage levels and mix of delivery strategy. The additional discounted unit cost of the combination DTP-IPV vaccine was higher, ranging from US$13.88 to US$15.28. These results are also sensitive to the current prices of the vaccine, which will probably decline in coming years. Brenzel (1995) finds that in India the cost per case prevented for the combination DTP-polio vaccine was much lower than for oral polio vaccine (OPV), primarily because the combination vaccine was associated with a greater reduction in the number of polio cases. Miller and others (1996) suggest that introducing IPV into routine vaccination in the United States would cost an additional US$15 million to US$28 million depending on the type of schedule adopted, resulting in a cost per vaccine-associated paralytic poliomyelitis case prevented of approximately US$3 million. Sangrugee, Caceres, and Cochi (2004) found that the least costly option would be for programs to stop providing OPV after postpolio eradication and certification and that optionally introducing IPV with universal IPV had the highest costs and the lowest expected number of vaccine-associated paralytic poliomyelitis cases. If the unit price of IPV fell to US$0.47, switching to IPV from OPV would be economically worthwhile.

Financial Sustainability of Immunization Programs

Even though research has demonstrated that vaccination against childhood diseases is one of the most cost-effective health interventions, governments in many developing countries are considering how to meet the financing requirements of immunization programs, particularly as new vaccines are introduced and programs are scaled up. GAVI is working with countries to prepare for the transition from grant funding and to secure the overall financial sustainability of national programs. Approximately 55 countries have prepared national financial sustainability plans for immunization. These plans help countries evaluate the current and future costs and financing of national immunization programs and identify strategies to address future funding gaps (GAVI 2004; http://www.who.int/immunization_financing/en).

According to a recent analysis of financial sustainability plans, specific costs for immunization programs represent an average of 2 percent of total health spending and 6 percent of government health spending and are equivalent to less than 0.2 percent of gross domestic product on average. However, this profile changes after new and more expensive vaccines are introduced. In some countries, program-specific costs for immunization can reach as high as 20 percent of government health spending with introduction of combination vaccines (Lydon 2004). This share is related to the current unit price of the vaccine, which is expected to decline.

Governments and their development partners are challenged to find ways to finance and sustain immunization programs. In countries that are implementing reforms to achieve greater transparency and fiscal discipline through sectorwide approaches and medium-term expenditure frameworks, the additional financing requirements are compounded by the need to operate within a fixed budget for the health sector, so that increased funding needs for one program may necessitate budget cuts for others. This example illustrates the potential tradeoffs that exist at the country level, which create both opportunities for more open policy dialogue in relation to priority setting for the use of scarce public funds and risks that the cost of new vaccines may not be readily integrated into national plans and budgets. Because of the financial implications of reaching higher coverage levels and simultaneously introducing new vaccines, policy makers will not only have to weigh the cost-worthiness of alternative investments but also have to understand their long-term budgetary implications.

Improving the Costs and Cost-Effectiveness of Immunization Programs

The cost-effectiveness of immunization programs could be improved by either reducing costs or improving programs' health benefits. Programs could reduce costs by using a more efficient mix of delivery strategies, reducing vaccine wastage, and using lower-cost inputs while maintaining the same quality of service. Reductions in the price of vaccines in the near future will also reduce costs. Innovations in vaccine technology may result in more widespread use of vaccine vial monitors, and increased use of heat-stable vaccines could potentially reduce the cost of the cold chain, although these innovations may themselves add to costs. The number of children and adults immunized can be increased by creating additional demand for vaccination; reducing missed opportunities; and reducing the dropout rate between the first and third doses of DTP, hepatitis B, and other vaccines. Finally, changes in the EPI schedule could affect total costs by reducing the number of doses required to achieve immunity and thereby reducing the number of visits, resulting in savings in the costs of labor, supplies, transport, and perhaps overhead.

The EPI schedule was established in 1984 based on a review of immune responses to diphtheria, tetanus, pertussis, polio, and measles vaccines starting at different ages and with varying intervals between doses (Halsey and Galazka 1985). The EPI schedule administers three doses of DTP at the shortest possible intervals to complete the immunization series as early in life as possible. However, if the primary series could be reduced to two doses with a booster dose at 12 to 15 months of age, the cost savings from reduced visits and one fewer dose of DTP in countries that administer a fourth dose of DTP would be considerable. Additional serological studies would be needed to compare the existing EPI schedule with the theoretical schedule before a new schedule could be adopted. Also, other vaccines to be introduced into immunization programs would need to be revaluated in this schedule. Two doses of IPV administered beginning at two months of age induce protective levels of antibodies between 95 and 100 percent for each of the three polio types (Halsey and others 1997; Plotkin and Vidor 2004).

Some countries with a low incidence of tuberculosis (such as those of Eastern Europe) are considering the discontinuation of routine BCG vaccination, given the low risk of acquiring tuberculosis in early childhood. If the BCG were not administered during the first month of life, program costs would be reduced by the value of one visit and by the costs associated with vaccine purchase, shipping, storage, and administration.

Research Agenda

Private and public sector investment in research and development pertaining to new vaccines and improved use of existing vaccines is considerable. Most research and development is focusing on vaccines likely to have the greatest effect in the developed world and the best financial return; however, by means of public-private partnerships for product development, foundations have stepped in to support vaccine research and development for diseases for which the greatest burden occurs in developing countries.

New vaccines are being developed that could be incorporated into EPI schedules, including vaccines that protect against rotavirus, S. pneumoniae, malaria, cervical cancer associated with human papilloma virus, HIV/AIDS, and dengue. New and improved vaccines are also being developed to protect against meningococcal infections in infancy and Japanese encephalitis (NIH 2000). WHO recently created the Initiative for Vaccine Research Department to facilitate global coordination of research and development efforts for these and other vaccines.

In compiling data for this chapter, we noted a number of key gaps in knowledge that could usefully drive a research agenda and contribute to more evidence-based policy making in the future (Fox-Rushby and others 2004). First, little is currently known about how and why delivery costs change with increasing numbers of vaccinations and at higher coverage rates and whether economies of scale can be achieved. Little is known about the relative cost-effectiveness of different strategies to increase coverage given different baseline coverage rates. This issue relates to other questions of the optimal timing for introducing new vaccines and of how decisions should vary given different epidemiological and economic settings. Future research should therefore consider the extent to which cost-effectiveness analyses need to be repeated for every country or context or whether (and how) estimating and validating relationships across countries and accounting for uncertainty in estimates of costs and effects are possible.

Second, more attention needs to be given to measuring effect. For example, even though the coverage of single antigens required to reach particular levels of FICs should be accounted for, economic evaluations need to move beyond such indicators of output to measuring effect on the quantity and quality of life. In evaluating different schedules, methodological research needs to focus on how to incorporate the combined effects of multiple vaccinations in this respect. Remarkably few studies have considered the effect on nonhealth benefits, such as economic growth and welfare. The larger the package of vaccinations considered, the more important this question becomes.

Conclusions

This chapter confirms that vaccination of children and women with the traditional EPI vaccines is a highly cost-effective public health intervention, although cost-effectiveness ratios vary by region, delivery strategy, and level of scale. Overall, vaccination has had a significant effect on reducing mortality and morbidity from childhood diseases and will be a priority intervention for achieving the child health Millennium Development Goals. Improving and sustaining measles control are among the most cost-effective interventions in high-mortality regions.

Establishing and maintaining high immunization coverage rates in many of the poorest developing countries have proven challenging for those with high population growth rates, limited infrastructure and resources, and fluctuating demand for services. According to historical coverage rate trends, Europe and Central Asia, Latin America and the Caribbean, and the Middle East and North Africa are expected to achieve 90 percent coverage of FIC by 2011, with East Asia and the Pacific, South Asia, and Sub-Saharan Africa lagging behind.

Increasing and sustaining higher immunization coverage rates will require further efforts so that disease control can be maintained, particularly when a perception exists at the community level that vaccine-preventable diseases are no longer a major public health issue. At higher coverage rates, further disease burden reductions will be smaller, which will affect relative cost-effectiveness. Targeted approaches in countries or at subnational levels could potentially yield high returns, especially in those areas with poor control of vaccine-preventable diseases.

Our analysis shows that the cost per FIC will increase as countries scale up immunization coverage and introduce new vaccines. Adding more antigens to traditional EPIs has been successfully accomplished in many countries, especially for Hib and hepatitis B vaccines. Although many of the new vaccines under consideration are more expensive than those for the original six targeted EPI diseases, they may still be relatively cost-effective compared with other interventions and with treatment costs. Our analysis shows a wide range of cost-effectiveness estimates depending on the type of vaccine, vaccine prices, coverage levels, and delivery strategy, with the additional incremental cost per person being relatively small for some new vaccines. Declines in unit prices of new vaccines also will affect cost-effectiveness results.

Financing and sustaining immunization programs are challenges that governments in developing countries and their development partners will face. The financial implications of reaching higher coverage levels and the simultaneous desire to introduce new vaccines will require policy makers to consider both the relative cost-effectiveness of interventions and the long-term budgetary implications.

Although global and regional estimates of cost-effectiveness of interventions are useful guides, further analytical work will be needed to evaluate the relative benefits (deaths and cases averted and DALYs) and costs (delivery and treatment) of vaccines for different delivery strategies and higher coverage rates, particularly at the country level.

Acknowledgments

The authors thank Tina Proveaux for editorial and technical assistance and Howard Barnum, Mariam Claeson, Felicity Cutts, Tony Measham, and Philip Musgrove for reviewing drafts of the manuscript. Santiago Cornejo and Ravi Cheerukupalli provided valuable inputs.