Histologic definitions: In interpreting the Brookhaven medical program data on thyroid nodules the histopathological classification of thyroid nodules used by the expert panel of pathologists needs to be reviewed. This classification, based on diagnostic categories recommendedby the World Health Organization (Hedinger and Sobin, 1974) and modified in 1981 by Dr. Donald Paglia of the Department of Surgical Pathology, University of California, Los Angeles, for the panel’s use, has been applied to all thyroid specimens obtained at surgery since the beginning of the program: Adenomatous nodule: a focal proliferative lesion consisting of changes typical of adenomatous goiter; the lesions are hyperplastic and do not fulfill criteria of true neoplasms. Adenoma: an encapsulated proliferative lesion with a uniform internal growth pattern and benign clinical course. Occult papillary carcinoma: a small nonencapsulated sclerosing carcinoma; considered to be clinically benign even if associated with positive regional lymph nodes. Papillary carcinoma:larger,infiltrating carcinoma, usually containing both papillary and follicular components. The four pathologists on the panel that review the Marshallese specimens are: Dr. L.V. Ackerman, Health Sciences Center, SUNY, Stony Brook, NY; Dr. W.A. Meissner, formerly with New England Deaconess Hospital, Boston, MA; Dr. A.L. Vickery, Massachusetts General Hospital, Boston, MA; Dr. L.B. Woolner, Mayo Clinic, Rochester MN. Histologic sections of all surgically removed thyroid tissue have been examined by these authorities. Although most diagnoses have been unanimous, some were controversial. In the following analysis and discussion of Marshallese nodules, the "most neoplastic" diagnosis has been selected when there has been a split decision, with the "least neoplastic” being the adenomatous nodule, next being the adenoma, the third being occult papillary cancer, and the most neoplastic being the carcinoma. Appendix D lists all exposed persons had thyroid surgery which confirmed a lesion. Thyroid nodules in the Comparison grou The examination of the Comparison gr been invaluable in interpreting the thyroi risk data among the exposed population though they do not constitute an ideal " group, it is not likely that a better co population could have been obtained initiating a formal, prospective researcheff appropriate that the voluntary cooperatio membersof this group be gratefully ackno by all who have relied on the Marshallese nodule data to interpret the role of radidgti causing thyroid disease. In that the un Comparison group comprises persons of R ancestry and was quite closely age- and matched whenselected in 1957, this group representative of the exposed Rongelap po than any other Marshallese community and more so than a population of non-Marshal The development of thyroid nodules |i Comparison population is similar spontaneous thyroid nodule incidence r elsewhere. Maxonet al. (1977) concluded rate of development of benign thyroid nod thyroid carcinomas in western countries i and 0.01% of the population per year, resp and that the incidenceis linear with respec In 1990 the number of person-years of obs of the Comparison group was 10,400. Th based on the conclusions of Maxonetal. (1 expected number of thyroid nodules, ben malignant, would be 8.3, of which 6 or 7 w benign and 1 or 2 would be carcinomas.* I nodular thyroids were detected, of which benign and 2 were carcinomas (see T Possible sources of inaccuracy include the fo (1). Only surgically confirmed nodulesare i Therefore, since several unoperated nodul been diagnosed in the Comparison gro "observed" number may underestimate t numberof thyroid nodules. However, the nodules may have been lipomas or neuro example, and therefore appropriately exclu One of the two occult papillary car diagnosed in this group was not detected surgery. This “nodule” is therefore exclude the total number of nodules is given as than 9, as listed in Table 2. (3) 1984 is thf latest 16 .