Associations between dietary fatty acid intakes and risk of coronary heart disease (CHD) are not entirely consistent in prospective studies in the U.S. and Europe. Such studies in Japan are rare. The objective of this study was to examine the association between dietary total, saturated (SFA), and polyunsaturated (PUFA) fatty acids, cholesterol intake and CHD mortality using the dataset of NIPPON DATA90. At the baseline in 1990, we performed blood biochemical measurements and a nutritional survey on participants from 300 randomly selected districts. After exclusion of participants with a history of CHD and/or stroke at the baseline, we followed 7,819 community residents (3,254 men and 4,565 women, age ≥ 30) for 15 years. We estimated individual nutrient intakes among family members by weighed food records in three consecutive representative days. During the follow-up, there were 42 CHD deaths in men and 30 in women. Mean daily SFA and PUFA % calorie intakes were 5.90±1.36%, and 5.59±1.30%, respectively in men, and 6.48±1.53%, and 6.08±1.14% in women. A Cox analysis adjusted for age, vegetable and fruit intakes, and other confounders in women found that SFA intake was significantly associated with CHD mortality (hazards ratio per one quintile increment=1.34, 95% confidence intervals: 1.02-1.74, P=0.03), while no such association was noted in men. No associations were found between other fat intakes and CHD mortality in men or women. In conclusion, SFA intake was positively associated with CHD mortality independent of confounders in women, but not in men.