Unemployment rate as an updatable health needs indicator for small areas

Background The objective of the study was to compare updatable unemployment rates with the unemployment rate and composite deprivation indices from the 1991 Census as health needs indicators for small areas. Methods Townsend, Carstairs and Jarman indices and male unemployment rates were calculated from the 1991 Census, for 275 wards of the former East Anglian health region with unchanged boundaries between 1981 and 1991. Male unemployment rates were also derived from April 1991 unemployment benefit claimant figures, using both Office of Population Censuses and Surveys mid-year estimates of population and estimates derived from Family Health Services Authority patient registers as the denominator. Ward values were compared using Pearson product moment correlation. Results All three unemployment measures were closely related to each other and all were broadly as effective in predicting ward variations in mortality and long-term illness in 1991 as the compound deprivation indices of Jarman, Townsend and Carstairs. Conclusion Updatable unemployment rates were as suitable as the composite indices as an indicator of relative health needs for small areas in the year of the Census and might be expected to be superior in inter-censal years.