This study was designed to determine the extent to which the medical record contained evidence of coordination of care. Coordination of care was defined as the recognition of information (problems, therapies, intervening visits and tests) about patients from one visit to a follow-up visit. Overall, there was concordance between the medical record and independent observation of the physician-patient interaction in 70-85 per cent of instances. When there was clear indication that the practitioner recognized the information, the chart contained evidence of this recognition 68-85 per cent of the time, depending on the type of information. However, if the information was highly salient, the record contained evidence of recognition in a much greater percentage of intances: 95 per cent for distinctly identified problems, 83 per cent for problems which were contained within the text of progress notes, 96 per cent for major drugs, and 94 per cent for abnormal tests. The data from this study support the less direct evidence of others that the medical record adequately reflects extent of recognition of important information about patients by practitioners.