Salonen JT [North Karelia Project, University of Kuopio, Box 40, 70101 Kuopio 10, Finland], Heinonen OP, Kottke TE and Puska P. Change in health behaviour in relation to estimated coronary heart disease risk during a community-based cardiovascular disease prevention programme. International Journal of Epidemiology 1981, 10: 343-354.The relationship between change in one coronary heart disease risk factor and levels of other risk factors was investigated in independent pre- and post-intervention samples and in a cohort surveyed both before and after a community-based cardiovascular disease prevention programme. A risk score that did not include the factor being analysed for change was developed and used to estimate risk. In the independent samples generally neither change in smoking nor change in saturated fat intake was related to coronary heart disease risk estimated from other factors in either men or women. Only the intake of saturated fats reduced slightly more among the hypertensive than other men (p< 0.05). In the cohort, only change in smoking by women was related to initial risk estimated from other factors (p< 0.05). Since change in behaviour had no consistent relation to the preprogramme risk level, it was concluded that health behaviour change in the population was based on common lifestyle changes in the intervention community, and that either face-to-face counselling or new techniques of mass communication are needed to induce high-risk individuals to do more to change their risk-inducing behavio.