The purpose of the present study was to examine the relationship of a food frequency pattern and survivorship in a cohort of community elderly residents (259 men, 445 women), aged 65 years old and over, based on a 7-year follow up study. Baseline survey was undertaken in July 1988 by door-to-door method. The cohort was followed up until August 1995, during which time survivorship of subjects was investigated periodically each year. Eighty-one men (31.3%) and 71 women (19.1%) died during the follow-up period. The food frequency pattern was examined by factor analysis based on a food frequency questionnaire for 15 food-groups at the baseline survey. Results obtained were as follows: 1)The first factor was expressive of frequent intake pattern of plant foods, the second factor was frequent intake pattern of meat, fat, and oil, the third factor was frequent intake pattern of bread and rice(negative), and the fourth factor was frequent intake pattern of salty pickled vegetables, fish, and shellfish. 2)The first factor as represented by frequent intake pattern of plant foods had significant negative effect on all-cause mortality for both the whole 7 years and the last 4 years surveys, adjusted for age, sex, education level, drinking habit, smoking habit, exercise habit, history of hypertension, and instrumental activities of daily living.