Objective—To determine the prevalence of, and attitudes towards, observed and personal research misconduct among newly appointed medical consultants. Design—Questionnaire study.Setting—Mersey region, United Kingdom.Participants—Medical consultants appointed between Jan 1995 and Jan 2000 in seven different hospital trusts (from lists provided by each hospital's personnel department). Main outcome measures—Reported observed misconduct, reported past personal misconduct and reported possible future misconduct.Results—One hundred and ninety-four replies were received (a response rate of 63.6%); 55.7% of respondents had observed some form of research misconduct; 5.7% of (...) respondents admitted to past personal misconduct; 18% of respondents were either willing to commit or unsure about possible future research misconduct. Only 17% of the respondents reported having received any training in research ethics. Anaesthetists reported a lower incidence of observed research misconduct (33.3%) than the rest of the respondents (61.5%) (p<0.05). Conclusion—There is a higher prevalence of observed and possible future misconduct among newly appointed consultants in the UK than in the comparable study of biomedical trainees in California. Although there is a need for more extensive studies, this survey suggests that there is a real and potential problem of research misconduct in the UK. (shrink)