- Top of page
Analysis of data on the incidence of typhoid fever and the costs of vaccination of travellers to the developing world indicates that vaccination may not be cost-effective for travel to countries with a moderate-to-high endemicity. It may be reasonable to selectively vaccinate travellers to countries with a very high incidence of typhoid fever, and particularly those who are visiting relatives or who will be in close contact with the local population. Vaccination of travellers on standard tourist itineraries is probably not necessary. The basic preventative measure for typhoid fever should be the avoidance of potentially contaminated food and drink.
Typhoid fever is a systemic infection caused by Salmonella enterica serovar Typhi that is acquired by ingestion of contaminated food or water . Humans comprise the only known reservoir of this bacillus. Typhoid fever is uncommon in industrialised regions such as the USA, Canada, Europe, Australia and Japan (which are all regions with low endemicity), but is still common in developing countries. The disease affects c. 16 million persons/year worldwide, most of whom reside in developing countries . The result is that new cases of typhoid fever in industrialised countries are related increasingly to travel to developing countries. For example, in the USA, the number of cases of typhoid fever reported annually has remained relatively stable at about 450, but the proportion of travel-related cases increased from 33% in 1967–1972 to 62% in 1975–1984, and to 72% in 1985–1994 [3–5]. Based on such data, the public health authorities in most industrialised countries recommend vaccination against typhoid fever for travellers to the developing world (i.e., countries in Asia, Africa, and Central and South America) [6–8].
Three typhoid fever vaccines are available currently: (1) a parenteral heat–phenol-inactivated whole-cell vaccine (although this is not used widely because of its high toxicity); (2) an oral live-attenuated vaccine manufactured from the Ty21a strain of S. enterica Typhi; and (3) the purified capsular polysaccharide parenteral vaccine Vi [9–14]. However, the protective efficacy (51–73%) of these three vaccines, which was evaluated in populations living in endemic areas and not in travellers to foreign countries, is unsatisfactory [15–18]. In addition, data regarding the cost-effectiveness of these vaccines are scarce. Two reports have suggested that immunisation against typhoid fever of all travellers to the developing world is not cost-beneficial [12,14].
In order to examine the issue of cost-effectiveness, we collated data regarding the incidence of typhoid fever in travellers from the USA to the developing world in the period 1974–1994 [3–5]. Relevant information was collected from the list of cases of reportable disease published in Morbidity and Mortality Weekly Reports (MMWR) by the Centers for Disease Control and Prevention (Atlanta, GA, USA), and typhoid fever case report forms submitted by state and local health departments. Data on the efficacy and toxicity of typhoid fever vaccines (parenteral whole cell, oral Ty21a, and parenteral Vi) were obtained from a meta-analysis of 17 efficacy trials and 20 toxicity studies . These trials examined the efficacy of vaccines in people living in endemic areas only, and not in travellers. Subsequently, data from these studies were extrapolated to travellers, although this is an unproved assumption. Data regarding the costs of the vaccines were obtained from the unit costs published in the UK . Data were analysed for the two vaccines that are currently used widely, namely the oral Ty21a and parenteral Vi vaccines, and not for the parenteral whole-cell vaccine, which is no longer used widely because of its high toxicity.
According to this information, the cost of typhoid fever vaccination was calculated for travellers in terms of every case prevented. If the risk of travel-associated typhoid fever in unimmunised travellers to a particular region is one case/n persons travelling, the sum of travellers that must be immunised to prevent one case of typhoid fever is n ÷ the efficacy of the vaccine used. So, the cost for every case prevented is the sum of vaccination (n ÷ efficacy) × unit cost of the vaccine.
Table 1 summarises the available epidemiological data on the incidence of typhoid fever in international travellers visiting various countries during the period 1974–1994. Based on these data, the countries were classified into three subgroups according to the incidence of typhoid fever: (1) countries with a moderate endemicity of typhoid fever (Middle East), with about ten cases/million travellers; (2) countries with a high endemicity (Central America, South America, North Africa, South Africa and Southeast Asia), with about 20 cases/million travellers; and (3) countries with a very high endemicity (Indian subcontinent, and probably countries of Central Africa for which there are no good epidemiological data), with about 200 cases/million travellers.
Table 1. Incidence of typhoid fever/million travellers for each country visited
|Area visited||Number of cases of typhoid fever/106 travellers|
|1977–1979 ||1974–1975 ||1980–1990 ||1985–1994 |
| Central America||6.5–13||8.5||ND||5–21|
| Other Caribbean countries||0.7||2.0||5.3||< 1–3|
| Other South America||8–24||13.2||12.7||2|
| Other Near Eastern||14.4||9.4||ND||< 1–2|
| Far East and Pacific||6.5||4.8||43.8||5–12|
| North Africa||94||ND||ND||ND|
| Sub-Saharan Africa||12.3||7.2||ND||ND|
According to the meta-analysis of Engels et al., the efficacy of two doses of parenteral whole-cell vaccine is 73% (95% CI, 65–80%), that of three doses of oral Ty21a vaccine is 51% (95% CI, 35–63%), and that of one dose of parenteral Vi vaccine is 55% (95% CI, 30–71%). The whole-cell vaccine is associated more frequently with adverse events (mainly fever and local reactions) and was not considered further in the present analysis. The unit cost for oral Ty21a vaccine (Vivotif) is Euro 21, and that of Vi vaccine (Typhim Vi) is Euro 15. Table 2 lists the corresponding calculated costs for each case of typhoid fever prevented among travellers to each of the country subgroups.
Table 2. Cost of a typhoid fever vaccination programme in travellers for every case of typhoid fever prevented
|Subgroups of developing countries||Cost of vaccination/case of typhoid fever prevented |
|Countries with moderate endemicity (c. 10 cases/million travellers)||Euro 4076||Euro 2718|
|Countries with high endemicity (c. 20 cases/million travellers)||Euro 2038||Euro 1359|
|Countries with very high endemicity (c. 200 cases/million travellers)||Euro 204||Euro 136|
According to this analysis, vaccination against typhoid fever for travellers is probably not cost-effective in countries with moderate or high endemicity. From the public health point of view, the cost figures in Table 2 are high compared to other usually non-funded health care needs. It seems more reasonable to selectively vaccinate travellers to countries with a very high endemicity of typhoid fever, and those at high risk because of special circumstances. Travellers who are visiting relatives, or who will be in close contact with the local population, need to be immunised. In contrast, the vaccination of travellers on standard tourist itineraries is probably not necessary [14,20,21].
The above analysis is not without its limitations. The most important is the fact that none of the efficacy trials examined the travelling population; rather, they examined the population living in endemic areas, whose baseline immunity probably differs from that of travellers. Therefore, the extrapolation of the results can be questioned, and further research is needed to clarify the efficacy and cost-effectiveness of typhoid fever vaccination programmes in travellers . Nevertheless, guidelines for preventing typhoid fever in travellers should be clarified. Specifically, it should be stressed that vaccination against typhoid fever is not the main preventative measure. Indeed, avoidance of potentially contaminated food and drink should be the basic advice given to travellers, because the ingestion of a large inoculum of S. enterica Typhi may result in infection despite immunisation.