A small area analysis of self-poisoning and self-injury in the region of Oxford

Abstract

The number of persons attempting suicide in 1972–3 was related to the 1971 Census data by enumeration district in a defined area centred on Oxford City. Variations among enumeration districts in the number of attempters was partly explined by the age and marital state distribution of the population, but a few areas were identified where the high rates were not accounted for in this way. These included areas on council estates, both in the City and outside it, with marked indications of social disadvantage, and a socially mixed and probably rapidly changing inner city area. Persons attempting suicide were predominantly young and included relatively more women and more single people, but otherwise appeared to resemble the population of their immediate areas in terms of distribution by socio-economic group, housing conditions and indicators of social disadvantage. A model was fitted to the 1972–3 data to ‘predict’ the number of attempters in an enumeration district from the Census information. The fit was not entirely satisfactory, and suggested that there was a sub-group of districts with a near zero probability of producing an attempter which was not identifiable from the Census data. When fitted retrospectively to attempted suicide data for 1969 the performance of the model was poor. It was not possible to distinguish the effect of random variation from that of systematic change, but there was some evidence that certain areas with high attempted suicide rates in the City also showed a rapid increase in rates.