Seasonal influenza is a major cause of mortality resulting in an estimated 36,000 deaths annually in the USA alone [24
]. The annual vaccination policy against seasonal influenza in most countries has been focused on protecting groups at high risk for complications of influenza, including the elderly, pregnant women, young children, and individuals with chronic diseases.
However, vaccinating high-risk populations is unlikely to reduce the burden of seasonal epidemics, because these groups represent only a fraction of the population among whom the virus spreads [25
]. In addition, the attack rates in these groups are relatively low (8.8–13.5 per 100 persons for age ≥ 65 y). However, the attack rate is 25 per 100 persons in children aged 5–9 y and can reach 40 per 100 persons during pandemics, as experienced during 1918–19 [25
Another challenge with only vaccinating high-risk groups is that the vaccines may not work as well in these at-risk populations. The efficacy of the influenza vaccine is dependent on the immunological status of the specific population being vaccinated and on the type of vaccine. For example, in healthy children, pooled estimates suggest that the live attenuated vaccine leads to a 79% efficacy (absolute risk reduction (ARR) 152.4 per 1000, number needed to treat (NNT) 6.6 persons) for reducing laboratory-confirmed influenza with 38% efficacy for reducing symptoms [26
]. In contrast, in the elderly, no significant direct benefit of routine inactivated trivalent influenza vaccine was observed against influenza (RR 1.04, 95% CI 0.43–2.51) [26
]. Of note, however, well-matched vaccines prevented complications in residents of long-term care facilities (vaccine efficacy (VE) of hospital admission 45%, 95% CI 0.16-0.64; all-cause mortality 60%, 95% CI 0.23–0.79) [27
]. It is likely that the relative ineffectiveness of inactivated influenza vaccine in the elderly population is due to immune senescence, a waning of the immune system with age [28
The fact that groups at the highest risk of complications from infection often benefit the least from the vaccine is an important public health and scientific challenge. Providing an indirect benefit to these groups by vaccinating those who respond well to vaccines is one way to mitigate this public health challenge. Data exist that a herd effect due to vaccination in children may help protect high-risk groups [30
]. Recently, a cluster randomized study of trivalent inactivated influenza vaccination administered to 947 children and adolescents in Hutterites colonies in Canada showed a dramatic herd effect.The protective effectiveness in non-recipients of study vaccine was 61% (95% CI 0.08–0.83; p
= 0.03) for reducing laboratory-confirmed influenza (3.1% in unvacci-nated adults of vaccinated colonies vs 7.6% in unvaccinated colonies; ARR 40.0 per 1000, NNT 25.0 persons; ) [32
]. Such data lead to other important questions that still need to be answered, including the cost-effectiveness of vaccinating healthy children, the minimum uptake of vaccine in children needed to show a herd effect, and whether universal influenza vaccination is cost-effective. As has been demonstrated by Finnish researchers, the influenza vaccine is cost-effective when administered to children aged 6–13 y when considering the direct benefits of the vaccine [33
]. Since there is now evidence that a benefit of up to 60% effectiveness may be seen in unvaccinated individuals due to the herd effect [32
], this argues for an even greater cost-effectiveness of immunizing children.
Figure 4 The vaccine herd effect on influenza in a randomized control trial in Canada (adapted from Loeb et al. ).
The Society for Healthcare Epidemiology of America recently released a statement that influenza vaccination of healthcare personnel is a core patient and healthcare personnel safety practice [34
]. Whether vaccination of healthcare personnel can lead to a herd effect reducing laboratory-confirmed influenza among patients is still inconclusive. Pooled data from a Cochrane review of 3 cluster randomized controlled trials showed no reduction of laboratory-confirmed influenza (OR 0.86, 95% CI 0.44–1.68; p
= 0.66), lower respiratory tract infections, admission to hospital (OR 0.89, 95% CI 0.75–1.06), and deaths from pneumonia (OR 0.82, 95% CI 0.45–1.49) in patients when healthcare personnel were vaccinated [35
]. However, given that mathematical models suggest a herd effect [36
], more rigorous studies need to be conducted.