The differences in survival rates of colon cancer patients between hospital districts were used as an indicator of equity in cancer treatment in Finland. The merits of two different assessment methods were evaluated. The large variation was mainly due to confounding by age and the extent of the disease and to random variation.