Prior training and recommendations for future training of clinical research faculty members.

A survey was conducted of the first authors of half of the research papers published in 18 leading peer-reviewed medical journals over a six-month period in 1986. The first authors of every other article (that is, alternate articles) published during this period were sent a questionnaire that assessed their previous research training and their recommendations for training of clinical research faculty members. Of the 772 physicians sent the questionnaire, 482 responded. Of these, 57% had had some research training in medical school, 52% had received such training during residency, and 87% had received it during their fellowships. Fifty-six percent had taken a formal statistics course, 31.4% had taken some computer training, and 87% had received research supervision in a nonstructured format (that is, no formal course work). Introductory biostatistics was the only formal course in any of the seven subject areas listed in the questionnaire that more than one-third of the respondents had taken, yet at least two-thirds recommended that researchers take formal courses in all of the seven areas listed. The investigators recommend for medical students and postgraduate trainees a graduated experience of research training and exposure that includes formal courses.