How do physicians construct medical disposals? In this paper it is argued that insights developed in the sociology of science (in particular so-called laboratory studies) can be of help in addressing this relatively unexplored question. At the level of clinical action, medical sociologists have until recently generally separated the 'content' of medical action from the 'social' aspects, the former constituting a domain inaccessible to sociological investigation. This asymmetrical treatment of 'cognitive' and 'social' elements is stated to be based on two assumptions: (I) historical and examination data are seen as 'facts' which the physician only needs to 'reveal' and (II) medical criteria and disposal options are regarded as scientific, fixed 'givens'. A 'laboratory study' of medical problem solving in clinical practice, however, shows that these assumptions do not hold. It is argued that the physician, in transforming a patient's problem to a solvable problem, does not just combine some 'cognitive' items together, but actively articulates an array of heterogeneous elements with the transformation. In this construction process, both the elements and the transformation take shape. Finally, the importance of the concept of 'routines' for a sociological answer to the main question of this paper is shown.
