1 The sensitivity of long-term toxicity tests is impaired due to the 'background noise' of spontaneous lesions which are unrelated to treatment.
2 The need for a comprehensive source of computerized information concerning the occurrence and incidence of spontaneous lesions in control animals has been highlighted by initiatives in Europe and the USA. It is, however, essential to identify the potential users, and the type of information required for such a database to be of value.
3 This information has been acquired following an international survey of the pharmaceutical industry in Europe, Japan and the USA, including responses from 48 toxicologists and toxicopathologists representing 38 company groups.
4 Thirty-eight respondents indicated that they would use a historical control database that was regularly updated with the majority of respondents suggesting that they currently use external sources (Breeder's data, the literature, other companies) occasionally to acquire information on control animal pathology data.
5 The majority (94%) of the respondents indicated that a control animal database should contain information on both neoplastic and non-neoplastic lesions for use in evaluating long-term studies, in particular carcinogenicity studies.
6 The survey confirms the need for a historical control animal pathology database wider then those currently available.