Progress and challenges in psychosocial and behavioral research in cancer in the twentieth century.

Research in the psychosocial and behavioral aspects of cancer has shown steady growth since the 1950s, and its course of development has paralleled the history of medical techniques in treating cancer. Table 1 outlines this parallel evolution from the 1850s to the 1960s. The roles of the American Cancer Society and the National Cancer Institute (NCI) in spearheading and nurturing research in this area are documented. Interest in psychooncologic questions can be traced back for centuries to the search for etiologic factors and psychologic variables that would explain individual vulnerability to cancer. The first psychologic studies of cancer patients were reported in 1951 and 1952 from the Massachusetts General Hospital and Memorial Sloan-Kettering Cancer Center, respectively. The 1970s saw new interest in psychosocial and behavioral research with many issues being addressed for the first time: better care of the terminally ill through more humanistic approaches including better means of pain control; ethical concerns related to patient rights and their status as subjects in experimental protocols; trying to measure quality of life for cancer patients on protocols; seeing the need for multidisciplinary collaborative groups to make up for the absence of formal training in this area; and the need to design valid, accurate measuring scales specific to the symptomology of patients with cancer. Table 4 outlines how the 1980s gave increasing recognition and support to the psychosocial dimensions of cancer. This period produced a series of key conferences that examined a broad research and education perspective and produced recommendations that remain a benchmark in regard to instrumentation, conceptual models, pitfalls of psychosocial research, training, and education, and the organization of research efforts. New precision has been added to the field in the past 6 years: studies measuring concurrent psychologic, endocrine, and immune function; use of statistical modeling to incorporate quality of life data as a correction factor in survival data (TWiST and QALY); and broadened definitions of medical outcome to include functional status, thus allowing advances in psychiatric measurements to help answer questions in cancer. The challenges for the 1990s are identified in a summary in Table 9. Especially noteworthy is the observation that the comprehensive research needed today cannot be carried out by any one discipline alone. New approaches call for areas of the social sciences formerly inactive in cancer research (e.g., anthropology) to contribute the tools and expertise required to address the problems.