OBJECTIVES--To describe how coronary revascularisation rates in the United Kingdom (in the public and private sectors) vary by age and sex; how these relations have changed between 1987 and 1993; whether significant differences exist between geographical areas, public and private sectors, and hospitals; and to make comparisons with trends in North America. DESIGNS--Secondary analysis of data on the age, sex, procedure, NHS/private, and health district of residence of patients. SETTING--Resident population of South East Thames, East Anglian, and North Western health regions and Greater Glasgow, Lanarkshire, and Ayr/Arran health boards (11.6 million; 20% United Kingdom population). PATIENTS--All 19,665 residents who underwent either coronary artery bypass grafting or percutaneous coronary angioplasty without any concomitant procedure during 1987-8, 1989-90, 1991-2 and 1992-3 in either NHS or independent hospitals. MAIN MEASURES--Population based rates of revascularisation by age, sex, area of residence, and NHS/private treatment. Secular trends in the age (mean, standard deviation, range), and sex ratio (male to female) of patients. RESULTS--Revascularisation rates in men were about four times higher than in women (1992-3: 1340 v 362/10(6) aged 25 years or more). The highest rates were in those aged 55-64 years (for men) and 55-64 and 65-74 years (for women). In 1992-3 the mean age of female patients was three years older than that for men (61.2 v 58.3) and that for coronary artery grafting was over two years older than for angioplasty (59.4 v 56.9). Between 1987-8 and 1992-3 the male to female ratio decreased (4.2:1 to 3.55:1) and the mean age of patients increased steadily by about six months each year. Intervention rates for the older groups increased faster than those for the younger, particularly in high rate regions. The age and sex mix of patients varied between regions and districts/boards. The mean age of patients varied by nine years and the sex ratio varied twofold between NHS hospitals. The male to female ratio was higher in private than NHS patients (1992-3: 5.5:1 v 3.6:1), suggesting greater access to care for men than women in the private sector. The trends observed in the United Kingdom are similar to those that have occurred in North America, with the exception of a decrease in the male to female ratio, which has not previously been reported. CONCLUSION--The increase in the revascularisation rate has been accompanied by an increasing proportion of women and older people. The extent of these changes varies between geographical areas. The change in the sex ratio has occurred despite an increasing contribution by the private sector, to which women have less access than men.