J Coid1, N Kahtan, A Cook, S Gault, B Jarman. 1. Academic Section of Forensic Psychiatry, St Bartholomew's and the Royal London School of Medicine and Dentistry.
Abstract
BACKGROUND: The planning and development of secure forensic psychiatry services for mentally disordered offenders in England and Wales has proceeded independently within different regional areas. However, certain mental disorders, offenders, and offending behaviour are all more prevalent in geographical areas characterized by socio-economic deprivation and social disorganization. Failure to consider these factors has led to inadequate service provision in some areas and inequity in funding. A new model is required to predict admissions to these services as an aid to resource allocation. METHOD: Actual admissions (N = 3155) to high and medium secure psychiatric services for seven of 14 (pre-reorganization) Regional Health Authorities, 1988-94. Expected admissions were calculated for each district using 1991 census data adjusted for under-enumeration. Standardized psychiatric admission ratios were calculated and a range of social, health status, and service provision data were used as explanatory variables in a regression analysis to determine variation between districts. RESULTS: Actual psychiatric admissions varied from 160% above to 62% below expected for age, sex, and marital status, according to patients' catchment area of origin, measured according to deciles of the distribution of underprivileged area scores at ward level. The most powerful explanatory variables included a composite measure of social deprivation, ethnicity and availability of low secure beds at regional level. CONCLUSION: Admission rates to secure forensic psychiatry services demonstrate a linear correlation with measures of socio-economic deprivation in patients catchment area of origin. A model was developed to predict admissions from District Health Authorities and is recommended for future use in resource allocation. Identification of factors that explain higher admission rates of serious offenders with mental disorder from deprived areas is a priority for future research.
BACKGROUND: The planning and development of secure forensic psychiatry services for mentally disordered offenders in England and Wales has proceeded independently within different regional areas. However, certain mental disorders, offenders, and offending behaviour are all more prevalent in geographical areas characterized by socio-economic deprivation and social disorganization. Failure to consider these factors has led to inadequate service provision in some areas and inequity in funding. A new model is required to predict admissions to these services as an aid to resource allocation. METHOD: Actual admissions (N = 3155) to high and medium secure psychiatric services for seven of 14 (pre-reorganization) Regional Health Authorities, 1988-94. Expected admissions were calculated for each district using 1991 census data adjusted for under-enumeration. Standardized psychiatric admission ratios were calculated and a range of social, health status, and service provision data were used as explanatory variables in a regression analysis to determine variation between districts. RESULTS:Actual psychiatric admissions varied from 160% above to 62% below expected for age, sex, and marital status, according to patients' catchment area of origin, measured according to deciles of the distribution of underprivileged area scores at ward level. The most powerful explanatory variables included a composite measure of social deprivation, ethnicity and availability of low secure beds at regional level. CONCLUSION: Admission rates to secure forensic psychiatry services demonstrate a linear correlation with measures of socio-economic deprivation in patients catchment area of origin. A model was developed to predict admissions from District Health Authorities and is recommended for future use in resource allocation. Identification of factors that explain higher admission rates of serious offenders with mental disorder from deprived areas is a priority for future research.