The choice of diagnostic criteria for biological research.

The need for operational definitions of psychiatric syndromes is now widely accepted, and their use has improved reliability considerably. However, the existence of numerous competing definitions of several of the major syndromes and political pressures to use particular definitions highlight our lack of adequate criteria of validity. One reason why it is difficult to decide where the boundaries of individual syndromes should best be drawn is that we have largely failed to demonstrate discontinuities or "points of rarity" between related syndromes. The considerations that ought to govern the choice between competing definitions in this situation are discussed, and it is suggested that the best research strategy at present may be to use two or three alternative definitions simultaneously.