A randomised controlled trial of smoking cessation is reported in 1445 male smokers aged 40-59 at high risk of cardiorespiratory disease. The 714 men in the intervention group were recalled for a series of personal interviews with a doctor. After one year, 51% of the intervention group reported that they were not smoking any cigarettes, and most of the others reported a reduction. Compared with the "normal care" group, the men in the intervention group showed a decline in the prevalence of sputum production and dyspnoea; ventilatory function did not improve but its rate of decline was significantly slowed. There were no evident effects on blood pressure levels, nor on electrocardiographic findings over three years, nor on sickness absence over one year. Mortality follow-up has continued for an average of 7.9 years; 98 (13.7%) of the intervention group have died, compared with 94 (12.9%) of the "normal care" group. The 95% confidence limits on mortality range from 2.63% in favour of intervention to 4.37% in favour of normal care. The power of the trial has been reduced by smoking cessation in the "normal care" group. It is concluded that smoking cessation in these middle-aged men improved the symptoms and progress of chronic bronchitis; but the reversibility of the risk of cigarettes to the smoker's life may have been overestimated in observational studies.