Pneumococcal polysaccharide vaccines do not protect the elderly from pneumococcal infections.

33 The decision by the United States Food and Drug Administration (FDA) in 1977 to license a pneumococcal vaccine containing 14 of the 90 known serotypes of Streptococcus pneumoniae was based on little evidence. 1 The only published trials of this product then available involved healthy people with unusual risks of pneumococcal infection: South African gold miners and people living in the New Guinea highlands. In more industrialised countries, however, individuals at highest risk of pneumococcal infection are the elderly and those with certain chronic illnesses. Although the vaccine had not been studied in these populations, the government-sponsored group that formulates national immunisation guidelines in the United States, the Advisory Committee on Immunisation Practices (ACIP), recommended that they be vaccinated to prevent pneumococcal pneumonia. In doing so, they committed the grave scientific error of taking information obtained from certain populations and applying it to other, very different ones. Subsequent prospective controlled and blinded trials of the 14-valent vaccine, or a later one containing 23 serotypes, included more than 100,000 patient-years of observation in trials in the USA, Finland and Sweden. 2-8 These investigations demonstrated that the ACIP's recommendations were unjustified: whether examined individually or in aggregate, these studies showed that the vaccine did not reduce pneumococcal pneumonias specifically, pneumonias from any cause, or overall mortality in the elderly or the chronically ill. In fact, combining the results of these trials shows that the frequency of each of these adverse outcomes was actually higher in those receiving the vaccine. At least seven meta-analyses (including two not cited by Assendelft et al.) in this issue have been published that collectively reviewed at least 16 randomised controlled trials – both blinded and unblinded – comprising almost 50,000 patients. The analyses have differed in their methods, the kinds of studies included, and their classification of the information. 9-15 Nevertheless, they agree in concluding that in industrialised nations the pneumococcal vaccine is ineffective in the elderly and the chronically ill. A recent large retrospective cohort study of the vaccine in the elderly in the USA (not available to Assendelft et al.) that evaluated 47,365 patients 65 years of age or older for three years also showed that it was ineffectual. 16,17 As with several other studies, the authors dwelled on the nonsignificant reduction in pneumococcal bacteraemias in the vaccine group, while downplaying the larger failure of the vaccine to prevent pneumonia or deaths. The reasons …