This part of the panel is concerned with investigations in patients. The physician as clinical investigator has a dual role, and to quote Professor Emmerglick (1) "he is both healer and experimentalist, and from t h s dual role at times a conflict of interest can arise." This confhct must always be resolved by recommending that course of action most likely to result in the greatest benefit to the patient. To ensure that the patient's best interests are always served, the physician needs some type of fail-safe buffer, which will help him avoid any trend toward placing research first and the patient second. In the early days, when Castle, Minot, Loeb, Atchley and others were fashioning modern clinical investigation, the buffer was provided by the way in which their studies matured. They were primarily interested in patient care and the instruction of the young, but saw opportunities to provide new insights into disease mechanisms through clinical observation. There was sufficient time for full reflection and for extensive discussion with senior colleagues before undertaking each new step in the research. As time passed, and clinical investigation became a primary career ladder, the time for reflection and discussion was shortened, and almost disappeared in the haste to publish and to meet grant deadlines. The creation of human rights committees fashioned a buffer of sorts to help the investigator maintain his priorities toward the patient's welfare. Such standards, while essential for any clinical trial, do not always guarantee compliance during the conduct of the trial. The physicianinvestigator, while assured of a detached and objective study design, is placed back into his unbuffered dual role in the implementation of the study. Perhaps we have reached the need in clinical investigations for a patient advocate to monitor the operation of the trial in order to remove the physician investigator from his conflicting roles. Such a responsibility should not be given to a committee, but to another physician, a peer of the investigator, sympathetic to the research, but without any stake in the outcome of the investigation. agreed upon (1). They include at least two elements. First, the patient must concur in the decision on everything that is done to him or her. I would favor leaving to the discretion of the physician the circumstances and details of obtaining informed consent. Guidelines
The principles on which one would base such protective advocacy are generally Address reprint requests to C. Gordon Zubrod, M.D., Director, Comprehensive Cancer Center for the State of Florida at the