This paper examines how qualitative research can contribute to the evaluation of medical information systems. Most qualitative studies of the use of medical computer systems adopt either an interactionist or, less commonly, an ethnomethodological perspective. The paper compares and contrasts the two approaches through the detailed discussion of two case studies, one rooted in each tradition. It identifies the implications of using these different analytical approaches and assesses their strengths and weaknessess. The paper argues that the preference for interactionism has led qualitative researchers to overlook important aspects of the social processess which surround the use of computer systems and that, consequently, a shift in emphasis towards ethnomethodological research is necessary. Nonetheless, it concludes by asserting that both strands of qualitative research can illuminate the organizational impact of medical computer systems.