Responsibilities, resources and results: practicalities of testing in-house developed clinical software

Regulatory authorities place obligations on the producers of pharmaceutical computer software to ensure that it performs in the manner that it was designed to do. The reliability of data processed by computer systems is a key concern both for the regulatory authorities and pharmaceutical companies, which use it as evidence of the safety and efficacy of pharmaceutical products. Therefore, testing of software used in the pharmaceutical industry faces a tough challenge to satisfy the reliability criteria of its two customers: the regulatory authorities and the pharmaceutical company. Further challenges for software are posed by regulatory authorities, who see the widespread use of a particular computer system as a benefit. The more widely used and understood the software is, the lower the risk of unexpected failure and compromise of the data. Pharmaceutical companies, on the other hand, often wish to use innovative computer solutions which make the most of new technology and give performance improvements over other systems. An off-the-shelf word processing package presents a very low risk in the eyes of a regulatory authority. However, a one-off state-of-the-art computer system, especially if it is only used by one company, naturally demands a greater degree of control and scrutiny to ensure that the risk is contained