It is a truth not universally acknowledged that no sooner the introduction of a new regulation in possession of a logical rationale than it is wont, rapidly and progressively, to become more bureaucratic and completely divorced from the original intent. (With Apologies to Jane Austen)

And what better example of the truth of this dictum than the way in which GCP is applied. The objectives, when GCP started to be introduced in earnest in the 1980s, were clear and rational: to ensure the safety of clinical trial subjects and the integrity of the data. Since then the requirements have been expanded, codified and harmonised, but the underlying objectives are unchanged.

Why do clinical trial sponsors accept dirty data and what can be done about it?

Imagine that this is the process for buying a custom-finished car.

You agree the specification with the dealer, sign a contract and pay a substantial deposit. In addition you may have to pay a fee of 20% to 40% of the cost of the car for ‘overheads’. In anticipation of your purchase you have hired a mechanic, who makes regular visits to the car manufacturer to ensure that the work is being carried out according to specification. Frequently the visits are futile, as essential staff or information is missing. Finally the day comes to take possession of the car. Your mechanic goes to collect the car, only to find that certain bits and pieces (such as the steering wheel, the odd spark plug and left front door) are either missing or have not yet been fitted. Gathering up all the pieces available, your mechanic pays in full for the car, brings everything home and completes the less-than-perfect assembly.