ObjectiveTo examine incidence of cancer and cancer mortality in relation to high blood pressure.DesignA longitudinal study of middle-aged men from a random population sample.SettingGöteborg, Sweden.SubjectsWe studied 7396 men aged 47–55 years without diagnosed cancer at baseline (1970–1973).Mainoutcome measures Incidence of cancer and mortality from cancer.ResultsBy the end of December 1992, 1401 men had been diagnosed with cancer at any site and 651 had died from cancer. Of the men in the highest fifth of the systolic blood pressure distribution (above 166 mmHg) 126 per 10 000 observation years were diagnosed with cancer at any site, compared with 91 per 10 000 in the lowest fifth [below 130 mmHg; relative risk after adjustment for age, smoking, body mass index, treatment for hypertension and leisure time physical activity 1.41 (95% confidence interval 1.19–1.68); P for trend 0.0001]. Of men in the highest fifth, 55 per 10 000 observation years died from cancer, compared with 42 in the lowest [adjusted relative risk 1.41 (1.09–1.82); P for trend 0.01]. Several types of tumour tended to be more common among men with hypertension, but only genito-urinary cancers [age-adjusted relative risk 1.39 (1.04–1.85)] and non-melanoma skin cancer [age-adjusted relative risk 1.98 (1.12–3.51)] were significantly so. Findings were similar for diastolic blood pressure and if data for the first 5 years of follow-up were excluded. There was an increase in risk of cancer also during the first 5 years [adjusted relative risk 1.80 (1.10–2.92) for systolic blood pressure and 1.77 (1.05–2.99) for diastolic blood pressure].ConclusionsWe found an excess risk of cancer and of death from cancer for men with high blood pressure. Although the increase in risk was comparatively modest, the findings are of public health importance, insofar as both hypertension and cancer are common conditions.