Impact of Influenza Vaccination on Seasonal Mortality in the US Elderly Population. Simonsen L, Reichert TA, Viboud C, Blackwelder WC, Taylor RJ, and Miller MA. Archives of Internal Medicine, 2005;165:265-272.
Explanatory note: Assessing the true impact of influenza viruses on mortality is difficult. For decades the Centers for Disease Control and Prevention has estimated influenza mortality by a surrogate marker: the winter season excess pneumonia and influenza (P&I) related deaths in people over 65 years of age above a baseline of the statistically ‘expected’ numbers of deaths. The data are collected from US National multiple cause of death certificates. Of course infections other than influenza can contribute to this ‘excess mortality’.
Trivalent inactivated influenza vaccine (TIV) has been repetitively shown in clinical trials to protect against influenza. For this reason, immunization has been encouraged for all persons at increased risk for complications of influenza. As a consequence, influenza vaccination in the United States for all persons 65 years or older increased from between 15% and 20% before 1980 to 65% in 2001. There are many reasons that affect whether high risk groups receive the vaccine, however.
Despite increasing TIV coverage in those 65 years of age and older, estimates of influenza- related mortality in the elderly have increased over time.
Taking into account the aging of the population and the different types of influenza viruses that have caused epidemics, has there been a reduction in mortality as influenza vaccine coverage has increased among the elderly?
The researchers collected P&I and all causes of deaths for the United States for the years 1968-2001 and compared these to the annual population estimates. TIV immunization coverage was estimated for 1989-1998 for the various groups from the National Health Interview Survey. Influenza virus strains that were circulating were determined by review of the annual Morbidity and Mortality Weekly Report from CDC.
Using a statistical model, the researchers calculated age-specific excess P&I and all cause mortality for the 33 seasons. They then examined whether the changing age distribution of the US could account for the increasing trend.
Influenza A(H3N2) viruses dominated influenza seasons in the 1990s. This study analyzed mortality trends over time for seasons dominated by influenza A(H3N2) viruses separately from those dominated by influenza A(H1N1) or B viruses.
After adjusting for an aging population and whether the seasons were influenza A (H3N2), the investigators found no evidence to suggest that there has been a reduction in either excess P&I or all cause mortality during the 3 decades of study.
The study found that for people 65 to 74 years of age, excess mortality rates during influenza A(H3N2) seasons fell between 1968 and the early 1980s but remained approximately constant thereafter. The mortality rate remained constant throughout the 3 decades for persons 85 years of age or older. Excess mortality in A(H1N1) and B seasons did not change.
The study estimated that influenza-related mortality was always less than 10% of the total number of winter deaths among the elderly during the 33 seasons studied.
After controlling for the aging population and influenza A (H3N2), this study was unable to demonstrate that the increased TIV immunization coverage of the elderly had reduced the calculated influenza-related deaths.
Studies with designs similar to this one have demonstrated the remarkable decline in cases of measles, chickenpox, and many of the childhood vaccine preventable diseases. Negative studies of this type are difficult to interpret. For example, the authors do not have data on which of the deaths, calculated to be due to influenza, are actually due to the influenza virus because the deaths are not laboratory confirmed. Thus some of the deaths could be due to other causes which would tend to minimize impact of vaccine. Further, there could be differences between uptake of vaccine between populations most at risk of influenza deaths and other more healthy elderly. To the extent that the more healthy elderly receive a disproportionately larger number of the influenza vaccine doses and the persons most at risk are less vaccinated, an impact of vaccination on overall calculated mortality may be masked.
The best studies of influenza vaccine effectiveness are those determined in randomized, double blind placebo controlled trials which have demonstrated the effectiveness of TIV vaccine in the elderly.
Nevertheless, the available data suggest effectiveness is diminished in elderly populations compared to younger healthier populations. Low levels of protection against death due to influenza in the most elderly populations may well relate to their lesser response to TIV than younger individuals. That is why the Advisory Committee on Immunization Practices has recommended that healthy contacts of persons at high risk of complications be vaccinated annually to reduce the likelihood that high risk persons will be exposed.