In issue no. 7/2012 of this journal, Director Geir Stene-Larsen of the National Institute of Public Health claims that the swine-flu vaccine Pandemrix had been clinically tested and that my assertion in an editorial in issue 5/2012 that it had not been clinically tested therefore is wrong (1, 2). Ishould of course have used different wording, since «not clinically tested» is a very imprecise expression. Instead, I should have written «insufficiently clinically tested». In particular, insufficiently clinically tested to recommend vaccination of the entire Norwegian population. That was my point.
Clinical tests of a drug can be very diverse: From the very earliest stages, with small trials on humans – often performed by the manufacturer of the drug – to see whether the drug is harmful, i.e. has frequent and serious side effects (phase 1 studies) to comprehensive studies that balance the drug’s effects against side effects that can be of a serious as well as a less serious nature (phase 3 studies). Solid phase 3 studies are commonly required before a drug can be approved by the authorities. Since these studies also may fail to detect rare, but serious side effects, the authorities increasingly often require systematic follow-up studies (phase 4 studies). In other words, more than ordinary reporting of side effects.
How the potential effect of a drug can be balanced against potential side effects depends on the group of patients that will use the drug. For seriously ill patients who have few other treatment alternatives, even quite serious side effects may be deemed acceptable if an effect is possible. For a healthy population facing a disease which is harmless to the vast majority, there must be extremely good documentation of effects in order to allow any side effects – frequent or rare – whatsoever.
Stene-Larsen and the Institute of Public Health claim that the vaccine had been clinically tested to a «sufficient» extent, even though the tests had been performed on a similar vaccine against another virus, the bird-flu virus H5N1. However, the response to the virus itself is the decisive factor for effects as well as side effects, so the studies of the bird-flu vaccine had a limited transferability to the swine flu.
Stene-Larsen further says that the Institute of Public Health delayed the decision to recommend the swine-flu vaccine to everybody until 23 October 2009, since «on that date we had evidence that Pandemrix did not produce frequent, serious side effects». This logic is flawed. When recommending a vaccine to an entire, healthy population and knowing that the disease is harmless for the vast majority, it is insufficient to ensure that the vaccine does not produce «frequent, serious side effects». It must also be ascertained that it is effective to an extent that offsets the risk of the very rare, but serious side effects that tend to occur – unfortunately also in this case.