Accepted: September 9,1991

The authors analyze the distribution of suicides according to the variables of sex, age and marital status in the Canary Islands, during the period 1977-1983, by means of a register that they themselves created in order to correct serious deficiencies in the official data. There were notable differences between men and women, and the tendencies observed in each case are also very different. The authors argue that this makes it necessary to sep-arate the sexes in the epidemiological studies on suicide. The advantages and disadvantages of the direct and indirect methods in the standardization of the specific suicide rates are discussed. The specific rates related to marital status are standardized according to age and, likewise, the specific rates related to age are standardized according to marital status, for each sex. Once confounding factors are controlled, it becomes clear that there is a direct relationship between age and suicide. After adjusting for age, it is noted that the pattern of risk for different marital status categories varies by sex.