It is unknown whether influences of midlife whole diet on the long-term CHD mortality risk are independent of genetic and common environmental factors or familial predisposition. We addressed this question prospectively using data from the National Heart, Lung, and Blood Institute Twin Study. We included 910 male twins who were middle-aged and had usual diet assessed with nutritionist-administered, cross-checked dietary history interview at baseline (1969–1973). Moderation-quantified healthy diet (MQHD), a dietary pattern, was created to evaluate a whole diet. Primary outcome was time-to-CHD death. Hazard ratios (HR) were estimated using frailty survival model. Known CHD risk factors were controlled. During the follow-up of 40 years through 31 December 2009, 113 CHD deaths, 198 total cardiovascular deaths and 610 all-cause deaths occurred. In the entire cohort, the multivariable-adjusted HR for the overall association (equivalent to a general population association) was 0·76 (95 % CI 0·66, 0·88) per 10-unit increment in the MQHD score for CHD, and the multivariable-adjusted HR for a twin with a MQHD score ten units higher than his co-twin brother was 0·79 (95 % CI 0·64, 0·96, P=0·02) for CHD independent of familial predisposition. Similar results were found for a slightly more food-specified alternative moderation-quantified healthy diet (aMQHD). The between-pair association (reflecting familial influence) was significant for CHD for both MQHD and aMQHD. It is concluded that associations of MQHD and aMQHD with a lower long-term CHD mortality risk are both nutritionally and familially affected, supporting their use for dietary planning to prevent CHD mortality.