Objective-To determine the incidence of drug related problems that fail to be noted on casualty cards in patients subsequently admitted, and to compare medication histories as recorded by accident and emergency (A&E) senior house officers (SHOs) and a pharmacist. Methods-An initial retrospective survey of 1459 acute inpatient admissions through A&E over a three month period was followed by a prospective study of 33 elderly patients. Results-In the retrospective survey, 52 medication related problems were confirmed after examination of the medical records, of which only 16 were identified in A&E. In In the first part of this study we investigated drug related problems retrospectively among people presenting to an A&E department and subsequently admitted as inpatients, by examining the drug histories taken by the A&E senior house officer (SHO). In a second part we attempted further quantification of potential drug related problems which may have been missed due to incomplete history taking, by comparing drug histories taken by A&E SHOs to those taken by a pharmacist. MethodsFirst, a retrospective analysis of A&E cards was undertaken at Charing Cross Hospital, London, over a three month period for all patients attending the department who were subsequently admitted as medical inpatients. Patients admitted because of intentional or accidental overdose, poisonings, or substance abuse were not included. The cards were screened for potential drug related problems by examining the drug history as recorded by the A&E SHO. The identification process was performed to a list of specific criteria shown in table 1. The medical records of these patients were then screened to identify those drug related problems which were of clinical significance. This was judged to be the case if specific mention was made in the notes, or as in a few cases, if the clinical evidence was strong enough to confirm the problem. A note was made of the point in the patient's admission and by whom the drug related problem was identified. The findings of the pharmacist surveying the records were validated by a second, independent, clinical pharmacist.For the second part of the study elderly patients were interviewed by a pharmacist using a structured questionnaire to ascertain