The quest to predict which A-level students will make good doctors could be helped by the discovery of a link between how they study when they are 18 and their later clinical abilities, writes Aisling Irwin.

Chris McManus, a psychiatrist at Imperial College School of Medicine at St Mary's, has been studying the issue as a result of worries that medical students are being selected on the wrong criteria - for example with too much emphasis given to A level results. Although their drop-out rates are not high, drop-outs are expensive and there are fears that many doctors are lost to the profession after they graduate.

Professor McManus had previously shown that medical students with lower A level grades do worse in their finals and are more likely to drop out. But the association is not strong. "If you pushed up entry by a grade you would barely affect the drop-out rates," he said.

Since finals generally do not measure practical ability, he asssessed the amount of clinical experience students have had by the time they take their finals. Much of this depends on the keenness of the student. He counted how many patients, operations and conditions the students had seen; and how many practical procedures they had done. He found clinical participation was unrelated to finals results, which means finals are failing to measure practical ability. He reasoned that, since finals results are related to A level entry, possibly A level results do not predict practical ability either.

ÖйúAƬ

Professor McManus then sent questionnaires on study methods to 6,000 medical school applicants. He categorised them as: skimmers, who memorise facts in order to pass exams; deep studiers, who really want to know the answers; and strategic students, who do a bit of each because of a drive for success.

He then found the vital connection - study method predicts later depth of clinical involvement. He said medical finals should be better designed to test for other more practical factors. A level entry grades should be reduced to BCC to make room for other criteria.

ÖйúAƬ

Professor McManus thinks his research has also solved a perplexing finding from his comparisons of universities that interview prospective students and those that do not.

He now believes universities who interview may select students who work in a certain way and are therefore better clinically. The conference, Choosing Tomorrow's Doctors, held by St George's Hospital Medical School, heard calls for students to be selected according to a new raft of characteristics.