Leading article: Another tragedy of errors

In almost any enterprise, the news that management has plans to install a new computer system tends to be greeted as a threat rather than a promise. Nowhere, though, are there more grounds for IT apprehension than in the public sector, where one project after another has been dogged by severe overruns in time and budget, and the end product invariably falls some way short of being, in that celebrated phrase, fit for purpose.

The latest example, or rather the latest episode, in the long-running tragedy of errors that is the NHS electronic database concerns the computerising of patient records for the whole of southern England. Government negotiations with the supplier, Fujitsu, have broken down and the contract has now been terminated. The overall project, currently running more than four years late, is likely to be further delayed. What with penalties for terminating the contract and the extra expense of finding a new supplier, the final cost to the taxpayer, at present estimated at almost £13bn, looks set to rise another notch.

Why should it be that ministers stumble so predictably over big computer projects? Other countries have experienced problems, to be sure, but our government seems to have a particularly lethal touch. From the ill-fated Child Support Agency onwards, it is hard to think of any major public-sector IT project that has been delivered on time and on budget and done everything it was required to do.

One explanation may be that too many people and interests are involved in drawing up the specifications; another is that because the Government came to computerisation relatively late, it was unrealistic about what a single system could accomplish and its aspirations were over-centralised. It has also been suggested that the qualities of senior civil servants are not necessarily those required to commission, or oversee the commissioning of, IT systems. Yet, surely, in the time since the CSA debacle, something should have been learnt – if not from mistakes here, then from successful projects overseas.

The only flicker of hope is that something does seem to have been learnt, if not nearly enough. The huge NHS project was separated, for the purposes of suppliers' contracts, into geographical areas, so the termination of one contract does not halt the whole project. Contracts also stipulate payment only after services are up and running. In other words, the consequences of the disagreement with Fujitsu could have been a whole lot worse.

This might be some consolation. But the NHS project should stand as a warning. If the Government is having so much trouble computerising patient records, the chances for a trouble-free introduction of ID cards must range from slender to nil.

Interesting? Click here to explore further