To our knowledge, no other studies have evaluated the economic impact of vaccination with PCV7 during an influenza pandemic. Studies have examined the cost-effectiveness of PCV7 against pneumococcal disease [21
]. However, these studies did not examine PCV7 in the context of influenza. Results of our study are consistent with studies examining other interventions to reduce the epidemiologic and economic impact of an influenza pandemic. For example, studies in Israel [64
], Singapore [65
] and the UK [66
] found stock-piling antiviral drugs to be either cost-effective or cost saving in a pandemic. One study assessed vaccination with a 23-valent pneumococcal vaccine in the context of an influenza pandemic in the Netherlands. The study found that in the absence of an available influenza vaccine at the start of a pandemic, pneumococcal vaccination should be administered to the elderly and groups at a high risk for influenza, in order to limit the number of resulting hospitalizations and deaths [67
Several conservative assumptions regarding the effects of pneumococcal vaccination and disease epidemiology were made. We assumed no direct protection against pneumococcal disease after 5 years of age. While it is unlikely that the vaccine will wane to zero when children turn age 5, data are not available to characterize the degree of waning vaccine efficacy in this age group over time. We therefore made assumptions consistent with previously published cost-effectiveness models of PCV7 [21
]. The pneumonia (bacteremic and non-bacteremic) case-fatality rate of 10% during a pandemic assumes access to antibiotics would not be disrupted; however, it is likely that access may be restricted due to increased demand and subsequent shortages. We also included conservative estimates of indirect (herd) effects against AOM and pneumonia in the vaccine-eligible population, taking the midpoint of observed vaccine efficacy in the NCKP trial and the observed reductions in hospital admissions from a community-based study [18
]. Even under these conservative assumptions, the base-case outputs predict a large positive public health and economic impact of vaccination.
Our study is subject to a number of limitations inherent in the study design. First, the decision-analytic model is necessarily a highly simplified representation of the disease transmission and outcomes of pneumococcal disease. Although we accounted for some differences in treatment and outcomes using age stratification, we recognize that the US population and health-care delivery system is highly heterogeneous and may not be well represented by the relatively simple structure of this model. We also note that data used to estimate vaccine effectiveness and outcomes were derived and synthesized from a variety of sources, and this process of interpretation and decision-making is subject to bias. Although extensive sensitivity analyses to evaluate the effect of alternative parameter choices on our outcomes showed no change in the overall conclusions, we recognize that different assumptions may have yielded different results. Costs used in this analysis were taken from published data and standard sources; the extent to which they reflect the true costs of administering medical care is unknown. Furthermore, this study was conducted from a third-party payer perspective rather than a societal perspective, and as such does not include costs of pneumococcal disease related to lost productivity, caregiver time, transportation, or other unreimbursed expenses. Inclusion of indirect costs presumably would have added substantially to the total cost burden of pneumococcal disease and the potential cost savings with PCV7. In addition, the model was estimated using US data, and care should be used in generalizing our results to other settings and populations.
Because the disease and mortality burden have differed both in magnitude and age distribution during the four pandemics that have occurred in the past century, it is difficult to characterize a "typical" pandemic. Our assumptions derived from the 1918 pandemic likely reflect a relatively severe influenza pandemic scenario. Preliminary novel influenza A (H1N1) pandemic data indicate that it is less virulent than previous pandemic strains, with fewer required hospitalizations, and deaths more likely to be concentrated in those with underlying medical conditions [68
]. There is also evidence that the presence of secondary bacterial infections in persons with influenza is lower in the 2009 pandemic relative to the 1918 pandemic [10
], indicating that the novel H1N1 disease burden will likely fall between that of a normal season and a severe pandemic. Because PCV7 vaccination is beneficial under both the severe pandemic and seasonal epidemic scenario, we can infer that PCV7 would also result in cost savings and public health benefits in a less severe pandemic. Future analyses assessing the impact of pneumococcal vaccination in the context of the novel influenza A (H1N1) pandemic will be undertaken as more data become available.
We acknowledge that access to care and antibiotics, as well as volume and means of travel have changed dramatically since 1918. However, under various assumptions meant to capture how access to care, antibiotic use, and differences in disease transmission may alter disease dynamics (i.e., reducing incidence and mortality during a pandemic), pneumococcal vaccination still remains highly cost saving. We also acknowledge that this 1-year model estimated from 2006 data includes the indirect (herd) effects of a national immunization program with PCV7 that has been established over 6 years, and only accounts for the costs of vaccination of 4 birth cohorts (children aged 1 year, 2 years, 3 years, and 4 years). In order to quantify the current value of the ongoing vaccination program, we chose to include the full effects of vaccination. If we had accounted for the costs and benefits over each year since the launch of PCV7, the economic argument likely would be diminished, as the indirect benefit grew over the 4 - 6 years after the launch of the vaccine. However, even if we assume that only half of the indirect benefit had accrued after 4 years of vaccination, the model predicts that vaccination would avoid >700,000 cases of pneumonia and save >$3.5 billion in a severe influenza pandemic year.
We also note that the model includes neither the cost nor potential benefits of influenza vaccination at the onset of a pandemic. Assumed pandemic-level rates of influenza in the model are consistent with no vaccination or a vaccine miss-match. If, however, we assume the influenza vaccine is well matched (80% efficacy) with coverage equal to that reported in 2006 [70
], severe pandemic incidence would decline from 30% to 24 percent [71
]. Under these conditions, the model predicts that vaccination with PCV7 would result in 681,000 cases of pneumonia avoided (compared to 715,000 cases in the base-case) and $6.2 billion in cost savings. Non-pandemic model results are based on recent influenza incidence data and reflect current rates of influenza vaccination.
Finally, it is important to note that S. pneumoniae is not the only pathogen known to complicate influenza during a pandemic. In a recent autopsy study of 22 H1N1 victims with bacterial co-infection, 45% were co-infected with S. pneumoniae, 32% were co-infected with S. aureus, and 27% were co-infected with Group A streptococcus (GAS). It is possible that S. aureus or GAS could be the primary cause of bacterial co-infections in a future influenza pandemic, and S. pneumonaie could play a less prominent role than in past pandemics. In that case, the benefits of pneumococcal vaccination would be less than those reported in this analysis. Vaccination programs against other pathogens such as S. aureus or GAS should be considered as other possibilities for influenza pandemic preparedness.
Our model predicts that routine pneumococcal vaccination is a proactive approach to mitigate effects of a future influenza pandemic, preventing numerous cases of pneumococcal disease and averting >100,000 deaths. The model also highlights the potential to save billions of dollars in health-care costs during an influenza pandemic by avoiding cases of pneumococcal disease and associated health-care utilization. This analysis details additional potential public health and economic benefits of universal vaccination with PCV7, which has already been estimated to be highly cost-effective or cost saving in many countries from various regions of the world [50
]. Countries that have not yet implemented a pneumococcal vaccination program may want to consider it as part of influenza pandemic preparedness.