There is now strong evidence that vaccines have very substantial nonspecific (heterologous) effects in children in high-mortality regions.1,2 The hypothesis states that, until the next vaccine is given, BCG vaccine approximately halves mortality from diseases other than tuberculosis and, provided vitamin A has not been given, measles vaccine approximately halves mortality from infections other than measles. Conversely, until the next vaccine is given, whole-cell diphtheria-tetanus-pertussis vaccine (DTP) increases mortality in girls from diseases other than diphtheria, tetanus, and pertussis.1
The first major study suggesting that vaccines have important nonspecific effects was published more than 11 years ago.3 It was a large cohort study in Guinea-Bissau in West Africa, and the suggestion that DTP may increase mortality caused considerable controversy. During the subsequent decade, many more observational studies were published, with conflicting results.1,4 However, analysis of observational studies of this issue is particularly difficult due to the potential for survival bias (which favors immunization), and because unvaccinated children in observational studies in high-mortality regions tend to be disadvantaged children with an increased risk of dying (which causes selection bias in favor of immunization).2,4,5 Consequently, WHO stated in 2008 that conclusive evidence about the nonspecific effects of vaccines is unlikely to be obtained from observational studies.6
Fortunately, we now have a substantial body of evidence from randomized trials, and this supports the hypothesis that BCG and measles vaccines have potent beneficial nonspecific effects (Table 1).1,7–11 In this issue of the Journal, Dr. Biering-Sorensen and her colleagues from Guinea-Bissau report the results of a small randomized trial of BCG8 that had the exactly same design as a large trial that was published in 2011.7 The 2 trials combined show that the neonatal mortality rate ratio was 0.52 (95% confidence interval: 0.33–0.82) when BCG was given at birth to low birth weight babies (study 1 in Table 1).7,8 Given that >50% of infant deaths occur in the neonatal period,12 it is very good news indeed that giving BCG at birth approximately halves neonatal mortality. In the Guinea-Bissau trials, BCG reduced mortality within 3 days of vaccination, with a total of 9 deaths in the BCG group and 21 deaths in the control group. This very rapid onset of protection may seem surprising. However, a study in mice has shown that BCG induces resistance to vaccinia virus within 1–2 days by an effect on CD4 T-cells.13
Other randomized trials have shown that BCG reduces mortality from infections other than tuberculosis. A randomized trial in Guinea-Bissau found that, among children who had received a booster dose of DTP at 18 months of age, the mortality rate ratio after BCG was 0.36 (0.13–0.99) compared with controls (study 2 in Table 1).9 In 6 controlled trials performed in the United States and the United Kingdom in the 1940s and 1950s, the mortality rate ratio for diseases other than tuberculosis was 0.75 (0.59–0.94) for BCG versus control (study 3 in Table 1).1
There is also evidence from randomized trials about the nonspecific effects of measles vaccine. Provided they had not been given vitamin A, children in Guinea-Bissau who were randomized to receive an extra dose of measles vaccine at 4.5 months of age had a mortality rate ratio of 0.59 (0.39–0.89) compared with children who received only 1 dose of measles vaccine at 9 months of age.10 When measles cases were censored, the mortality rate ratio was 0.65 (0.43–0.99) (study 4 in Table 1). In Gambia, Guinea-Bissau, Senegal, and Sudan, the mortality rate ratio was 0.53 (0.37–0.77) for girls randomized to receive measles vaccine at 9 months of age (study 5 in Table 1)1; all the girls were immunized against measles at either 5 or 9–10 months of age, therefore the difference was not due to infection by measles virus.
There have been no randomized trials of the effect of DTP on all-cause mortality. However, subset analyses of the randomized trials of BCG in Guinea-Bissau strongly suggest that DTP increases mortality.2 Among low birth weight neonates randomized to receive BCG at birth,7 the infants who had received DTP by 2 months of age had a mortality rate ratio of 4.30 (1.50–12.2) (study 6 in Table 1); this is very unlikely to have been due to selection bias because the infants given DTP were heavier babies who should have had a lower mortality rate than the smaller infants who did not receive DTP by 2 months of age. In study 7 in Table 1, among children randomized to receive BCG at 19 months of age, those given BCG then DTP had a mortality rate ratio of 5.12 (2.01–16.7) compared with children given DTP then BCG.9 This finding is also very unlikely to be due to selection bias because the age at which DTP was administered did not affect the mortality rate in children randomized to the control group (who did not receive BCG at 19 months of age). Study 8 in Table 1 is the only study of mortality after the introduction of DTP into a community; the mortality rate ratio was 1.92 (1.04–3.52) for children who received DTP compared with those who did not receive DTP.11
Less than 50% of neonates in high-mortality regions are given BCG during the neonatal period,14–16 and yet the evidence presented here suggests that BCG halves neonatal mortality. BCG is usually supplied in 10 or 20 dose vials, and many health clinics delay opening a vial until they have a large number of infants to vaccinate.8 Economic modeling could help us decide whether to supply small clinics with single-dose syringes of BCG, or just say that BCG from a multidose vial should be given even if it is needed by only 1 child.17
A key determinant of child mortality is the most recent vaccine administered; BCG and measles vaccines reduce mortality, but DTP increases mortality. The current Expanded Program on Immunization schedule is BCG-polio at birth; DTP-polio at 6, 10, and 14 weeks; measles vaccine at 9 months; and, in many countries, a booster dose of DTP at 18 months. With this schedule, DTP is the most recent vaccine for 50 of the 60 months between birth and 5 years of age. DTP would be the most recent vaccine for only 4 of the 60 months if the schedule were changed to BCG-polio at birth; DTP-polio at 6, 10, and 14 weeks; measles vaccine at 18 weeks; DTP booster at 18 months; and measles vaccine at 19 months.
If all neonates in high-mortality regions were given BCG at birth and the revised schedule of immunization were adopted with an extra dose of measles vaccine at 18 weeks (at a cost of only US$0.60 per dose delivered), this might prevent up to 1 million (30%) of the 3.2 million neonatal deaths each year in developing countries, and 1.5 million (30%) of the 4.8 million deaths between 1 month and 5 years of age. This would be a huge reduction in under 5 year mortality achieved at very low cost, using only vaccines that are already in the routine Expanded Program on Immunization schedule.
Unfortunately, policy makers have been all too ready to dismiss the nonspecific effects of vaccines as quirky findings based on flimsy evidence.6,18 Now, that we have substantial evidence from randomized trials and immunologic studies that these effects are extremely important, it is time for strong international support for the conduct of randomized trials to test the effects of BCG, measles vaccine, and DTP on all-cause mortality in children in other high-mortality countries.19 The benefits are likely to be spectacular.
1. Shann F. The non-specific effects of vaccines. Arch Dis Child. 2010;95:662–667.
2. Shann F. The nonspecific effects of vaccines and the expanded program on immunization. J Infect Dis. 2011;204:182–184.
3. Kristensen I, Aaby P, Jensen H. Routine vaccinations and child survival: follow up study in Guinea-Bissau, West Africa. BMJ. 2000;321:1435–1438.
4. Aaby P, Benn CS, Nielsen J, et al.. DTP vaccination and child survival in observational studies with incomplete vaccination data. Trop Med Int Health. 2007;12:15–24.
5. Farrington CP, Firth MJ, Moulton LH, et al.. Epidemiological studies of the non-specific effects of vaccines: II–methodological issues in the design and analysis of cohort studies. Trop Med Int Health. 2009;14:977–985.
6. World Health Organization. Meeting of Global Advisory Committee on Vaccine Safety, 18–19 June 2008. Wkly Epidemiol Rec. 2008;83:287–292.
7. Aaby P, Roth A, Ravn H, et al.. Randomized trial of BCG vaccination at birth to low-birth-weight children: beneficial nonspecific effects in the neonatal period? J Infect Dis. 2011;204:245–252.
8. Biering-Sorensen S, Aaby P, Napirna B, et al.. Small randomized trial among low-birth-weight children of Bacillus Calmette-Guerin vaccination at first health center contact. Pediatr Infect Dis J. 2012;31:306–309.
9. Roth A, Benn C, Ravn H, et al.. Effect of revaccination with BCG in early childhood on mortality: randomised trial in Guinea-Bissau. BMJ. 2010;340:c671.
10. Aaby P, Martins CL, Garly M-L, et al.. Non-specific effects of standard measles vaccine at 4.5 and 9 months of age on childhood mortality: randomised controlled trial. BMJ. 2010;341:c6495.
11. Aaby P, Jensen H, Gomes J, et al.. The introduction of diphtheria-tetanus-pertussis vaccine and child mortality in rural Guinea-Bissau: an observational study. Int J Epidemiol. 2004;33:374–380.
12. UNICEF. The State of the World's Children 2011: Adolescence—An Age of Opportunity. United Nations; 2011.
13. Mathurin KS, Martens GW, Kornfeld H, et al.. CD4 T-cell-mediated heterologous immunity between mycobacteria and poxviruses. J Virol. 2009;83:3528–3539.
14. Breiman RF, Streatfield PK, Phelan M, et al.. Effect of infant immunisation on childhood mortality in rural Bangladesh: analysis of health and demographic surveillance data. Lancet. 2004;364:2204–2211.
15. Elguero E, Simondon KB, Vaugelade J, et al.. Non-specific effects of vaccination on child survival? A prospective study in Senegal. Trop Med Int Health. 2005;10:956–960.
16. Clark A, Sanderson C. Timing of children's vaccinations in 45 low-income and middle-income countries: an analysis of survey data. Lancet. 2009;373:1543–1549.
17. Lee BY, Norman BA, Assi T-M, et al.. Single versus multi-dose vaccine vials: an economic computational model. Vaccine. 2010;28:5292–5300.
18. Fine P, Elliman D. Non-specific effects of vaccines: in context. Arch Dis Child. 2010;95:661.
19. Shann F, Nohynek H, Scott JA, et al.. Randomized trials to study the nonspecific effects of vaccines in children in low-income countries. Pediatr Infect Dis J. 2010;29:457–461.