Readers of Clinical Evidence [a database from the British Medical Journal Evidence Centre] who are interested in influenza will have been struck by the disparity between policy recommendations and the clinical evidence of the performance of inactivated influenza vaccines.
For example, there are few RCTs assessing the effectiveness of inactivated vaccines in children and the elderly. Only five RCTs have been carried out in elderly people, of which only one was carried out in the past 2 decades using vaccines available today.
Although the evidence is more robust in healthy adults, and partly supports the use of vaccines, this is the population who are universally considered to need them least.
The reasons for the contradictions between policy and evidence… are complex and include: the relative rarity of influenza; the current confusion between influenza-like illness (ILI) and influenza… the inability of vaccines to protect populations from an ever-mutating agent; and the difficulty of conducting meaningful prospective studies to assess vaccine efficacy.
In addition, the powerful image of influenza depicted by the media is not proportional to the actual threat.
Here, I examine the evidence for and the impact of the first two factors listed above — the incidence of influenza, and the masking of its rarity by the systematic failure to distinguish between influenza (a disease) and influenza-like illness (a syndrome, caused also in minor part by influenza viruses).
The causal relationship between the two is scarcely investigated and is frequently overlooked…
I realised the importance of incidence only after having carried out scores of Cochrane reviews and updates on influenza vaccines and antivirals.
The incidence statistic for influenza… is estimated from virological testing of symptomatic people… the patient presenting to a physician typically has a syndrome (influenza-like illness, or ILI) that can be caused by various agents. Only a proportion of these syndromes is caused by influenza A and B viruses…
To determine (not estimate) the incidence of influenza at any one time, virological testing of a truly random sample of people with ILI is needed. At the same time, testing for all other major causal agents should be carried out, but this is not typically done.
Based on studies in the Cochrane database, incidence of influenza is estimated at around 7%.
[However,] 7% is not the absolute incidence of influenza in the general population, but is rather the portion of ILI that is caused by influenza, making the incidence of influenza itself in the general population much smaller (approximately 0.5%).
A brief review of pie studies published in the past decade and available in the Cochrane database paints a remarkably similar picture… with an incidence of influenza of 0.5% to 1% of ILIs.
[O]ur aim here is to discuss why influenza inactivated vaccine performance is poor, and why most studies rely on non-specific outcomes, such as death from all causes, and hospitalisations for pneumonia and influenza (which are not usually based on virological testing).
One possible answer is that seasonal influenza is a relatively rare and benign condition, with an incidence not exceeding 1% in the general population during autumn and winter months.
Therefore, if the incidence in the unvaccinated population is low, then the ratio will be close to 1 and effectiveness will be low. So, vaccines seem to be less effective in illnesses with low incidence.
[E]vidence presented here points to influenza being a relatively rare cause of ILI and a relatively rare disease. It follows that vaccines may not be appropriate preventive interventions for either influenza or ILI.