Document text
Machine-read from the scan, so expect some errors. Lines that couldn’t be read are left out.
Page 1
UN ITED STATES ATOMIC ENERGY COMMISSION Medical Research Branch, DBM MEETING SCIENTIFIC ADVISORY COMMITTEE OF THE HFAL!I!H AND MORTALITY STUDY - OAK RID03, OCTOBER 22, 1971 My general impression is that the meeting, as a meeting of an Advisory Committee, was unproductive. One reason for failure was the confusion with regard t o the purpose of t h i s meeting. It was not u n t i l sometime during the afternoon that Dr. Mancuso explained the specific purpose of the meeting: workers whether the data available a t this time were of a type s u i t - able for presentation a t the meeting of the Health Physics Society. t o advise him and his coDuring the morning session, the presentations were of a general nature on the history of the project and s t a t u s and f'uture expansion plans of the Hanford p a r t of the study. What I consider strong criticism aimed in general a t the nature and quality of data collection came from Dr. Alice Stewart. She compared the data &.rs" and strongly urged perfecting of data quality and the introduction of refinements on the basis of need as demonstrated by results of data manipulation instead of attempting t o construct a perfect study. Dr. Sanders' reply to this was that preliminary Stewart questioned a160 the validity of some of the endpoints selected t o determine the health experience of AEC workers. "he consensus was that cause of death and life-shortening were critical endpoints and that any "hidden delayed effects other than cancer (such as the one causing e a r l y withdrawal from the work force) were highly improbable." D r . Elston also pointed out t o Dr. Sanders that he i s not looking a t mortality but tries to determine when people i n the study die and questioned the value of siblings as a control.
Page 2
Dr. Elstm discussed some of the statistical aspects of the presentations by Drs. Msncuso, Sanders and Brodsky. He presented some calculations made t o determine the expected sensitivity of data manipulation with regard to mortality and l i f e shortening and expressed the opinion that such calculations should have been made before the study was started and should have formed the basis for decisions on data collection, sample size and size of control little justification for some of the tests applied and urged many more analyses w i t h "valid and more powerfhl tests" (cost of data analysis is trivial compared t o collection costs). He also made the general statement that ''before you see any data, you better know you have some good methods t o analyze them." I a l s o had the impression that he could find very The remainder of the meeting was very uninspiring and concerned mainly a discussion of the suitability of presenting data a t the members I got the impression that they looked with great favor on an expansion of the study t o cosmic proportions. From some of the remarks made by the other committee I n general terms, one overall impression of the meeting was the inability of the investigators (Drs. Mancuso, Sanders and Brodsky) t o engage i n meaningfil. i n depth discussions with their advisors and their obvious high degree of insecurity as t o what t o do with Medical Research Branch Division of Biology and Medicine