Comparing the ability of different area measures of socioeconomic status to segment a population according to caries prevalence.

Objective: to compare the ability of different area based indicators of socio-economic status to segment a population of 5-year-old children according to caries prevalence.

Basic Research Design: The study population consisted of all 5-year-old children in seven districts in the North West Region of England who were examined in whole population surveys during the 1995/6 NHS epidemiological survey. This population was segmented according to caries prevalence by market penetration analyses using Super Profiles and the ONS geodemographic classifications, Jarman and Townsend (at ward and ED level) deprivation indices, the single census variables of unemployment and percentage of households without a car, and also the school that the children attended. Lorenz curves were plotted from the outputs of the penetration analyses.

Main Outcome Measures: Overall caries prevalence for the population was 55.3%. All the indicators provided a very similar picture. Large differences were detected between the highest and lowest market penetration rankings for each indicator, ranging from 42.5% for the ONS geodemographic classification to 31.4% for the Townsend index at enumeration district level. However, for each indicator, the fall between these two extremes was gradual. This picture was represented by the similar effectiveness scores of around 10% for each indicator which were derived from the Lorenz curves. Each indicator could identify approximately 30% of those with disease in the topmost 25% of the total population in the penetration ranking. When the analysis was restricted to those children with severe disease (dmft> or =5) a similar picture was found, however the effectiveness score increased to 19%.

Conclusions: There was a remarkably consistent inequitable distribution of population disease prevalence found between deprived and affluent area types, irrespective of what measuring instrument was used. Although there was a large difference in prevalence (dmft>0 and dmft>5) found between deprived and affluent areas, there was a gradual fall between the two extremes. This gradual fall has implications for oral health improvement strategies.