The IMG strategy is not completely clear about the opponents from whom the confidentiality of personal health information is to be protected. According to the IBM taxonomy, these opponents could be naive outsiders (such as undergraduate hackers at American universities), sophisticated insiders (such as the IT staff of an NHS supplier attacking a system to obtain competitive information) and funded organisations (such as a foreign national intelligence organisation seeking to obtain personal health information on MPs and senior officials for the purposes of blackmail) [1]. If the aim to to prevent `anybody' getting hold of information (as stated on page 12), then one must assume that the opponents include funded organisations.
The best assessment of national intelligence agency capabilities that we can make is as follows:
So an assumption that opponents could include national intelligence agencies is not compatible with a 64 bit key length. Assuming a well designed algorithm, a 64 bit key could probably not at present be found by an average individual attacker using keysearch. Such an attacker might just be able to find a 56 bit key if he were to sell his house to pay for the equipment, so the strategy's claim that 56 bits are too few and 64 bits are enough, is consistent with a threat model consisting of average individuals only.
The NHS Executive appears to be already aware of this problem. According to the NWCS pilot board minutes, it was estimated that five years was the operational life of Red Pike [61]. It does not seem prudent to deploy a system that will on its own advocates' admission be obsolete by the time it is fully fielded.