A quarter of intimate examinations performed by medical students on anaesthetised or sedated patients are carried out without adequate consent, a survey published in the British Medical Journal suggested yesterday.

Medical schools are being urged to review the way they teach students to perform intimate examinations. The Bristol University medical school, which compiled the study, has already strengthened its ethical and legal rules in response to shortcomings in local hospitals.

The school believes its findings - which relate to vaginal and rectal examinations, especially in earlier years of training - would be similar elsewhere.

A survey of students at Bristol suggested the more junior ones often felt uncomfortable about performing such examinations but did not dare to show it.

The study was conducted after third-year students expressed worries that they had been asked to act inappropriately. The results have been presented to other medical schools in Britain and abroad.

Yvette Coldicott and her fellow authors conclude: "Trust and respect are essential to the doctor-patient relationship, yet this study suggests they are missing from students' experiences of learning to do intimate examinations. Medical schools have a duty to deliver ethically informed training programmes that develop doctors' skills and are acceptable to the patient volunteers."

The report concedes that some patients may have given consent to trained doctors without students' knowledge. "However, in absence of specific consent, the student is liable to a legal charge of assault, so it is disturbing if any examinations are done without the student knowing whether the patient has consented."

Students must take responsibility for their conduct and can not merely suggest coercion or helplessness before supervising staff. A survey of other medical schools in Britain suggested many had formal policies on teaching vaginal examination but most had none for rectal examinations.

The Bristol school has since made sure students have to obtain consent for examinations from patients before they are prepared for surgery or given anaesthetic.

Clive Roberts, its medical clinical dean, said: "Medical ethics have moved a long way over the last 20 or 30 years. Doctors take on board very much that they have to respect the wishes of their patients."

The issue was whether the consent hospitals obtained from patients as part of their treatment extended to students conducting intimate examinations, he said. The view was that it did not.

Britt-Ingjerd Nesheim, a medical educator from Oslo, said: "The patient must be treated as the student's teacher, not a training tool. If we can succeed in this, patients will feel they have not only helped a nice young person to learn something important but also have received much more attention than elsewhere."

Peter Singer, a medical ethicist at Toronto University, said the report, which would generate "a firestorm of controversy", was a triumph of academic freedom: "The value of any guideline is not in how artfully it is crafted but in how well it is implemented and what effect it has. This is particularly critical in medical education, where the informal curriculum reigns, and deep cultural change is needed."

He added: "Rather than seeking to fix blame, we should find systemic solutions to the ethical challenges of medical education."