-Medical records serve many functions but their primary purpose is to support patient care. The RCP Health Informatics Unit (HIU) has found variability in the quality of records and discharge summaries in England and Wales. There is currently a major drive to computerise medical records across the NHS, but without improvement in the quality of paper records the full benefits of computerisation are unlikely to be realised.The onus for improving records lies with individual health professionals. Structuring the record can bring direct benefits to patients by improving patient outcomes and doctors' performance.The HIU has reviewed the literature and is developing evidence-based standards for record keeping including the structure of the record. The first draft of these standards has been released for consultation purposes. This article is the first of a series that will describe the standards, and the evidence behind them.

The validity of hospital episode statistics was questioned by Körner in 1982. Recent publications have shown that problems persist in England and Wales, and that the quality of the data is inadequate for the task. The lack of involvement of clinicians in the process of data collection and validation is no longer acceptable. To rectify the situation there should be a change of process and culture, supported by education and investment. NHS data definitions of terms such as 'spells' , 'episodes' and 'diagnoses' need to be reviewed. The development of separate data processes to monitor national service frameworks is regrettable.

-Hospital episode statistics were originally designed to monitor activity and allocate resources in the NHS. Recently their uses have widened to include analysis of individuals' activity, to inform appraisal and revalidation, and monitor performance. This study investigated physician attitudes to the validity and usefulness of these data for such purposes, and the effect of supporting individuals in data interpretation. A randomised study was conducted with consultant physicians in England, Wales and Scotland. The intervention group was supported by a clinician and an information analyst in obtaining and analysing their own data. The control group was unsupported. Attitudes to the data and confidence in their ability to reflect clinical practice were examined before and after the intervention. It was concluded that hospital episode statistics are not presently fit for monitoring the performance of individual physicians. A more comprehensive description of activity is required for these purposes. Improvements in the quality of existing data through clinical engagement at a local level, however, are possible.

Balancing service commitments with educational development is never easy. The European Working Time Directive and the Chief Medical Officer's report on senior house officer training bring new challenges. The Royal College of Physicians has developed ‘Laying the Foundations for Good Medical Practice’, a resource that enables tutors to teach generic skills to small groups of trainees in a fun and interactive way.