Background Teenage-pregnancy is a known social problem and has remained at the forefront of policy within the UK. The social aetiology of teenage pregnancy is known to be linked with deprivation. However, deprivation is a multifaceted term making it difficult to quantify, this is shown by the large number of measures that have been created in the UK. This study compared the discriminating ability of these measures with respect to teenage pregnancy. Furthermore, this project also examined the temporal-spatial patterns of teenage-conceptions within Aberdeen City.

Methods This was a population based study from 1950–2010, using registry based data from the Aberdeen Maternity Neonatal Databank (AMND). The AMND holds information on fertility and obstetric related events since 1951 to the present. The main outcome variable was a conception age of less than twenty years. This outcome was compared by a number of deprivation measures, these were the Scottish Index of Multiple Deprivation (SIMD), Carstairs Index, and Registrar General’s Occupation based social class. Each measure was commenced at different times resulting in a varying sample size for each of the measures. Logistic regression models showing the strength of the association and receiver operating characteristic (ROC) curves showing the discriminating ability of the measure were constructed as part of the analysis. Spatial analysis comprised of creating post code based maps of teenage pregnancies.

Results There was a decline in teenage-conceptions over the sixty year period. All the measures of socio-economic status were highly associated with teenage-pregnancy and given as Odd Ratios. For example the adjusted association between SIMD and teenage conception had an OR of 10.36 (95% CI 8.46, 12.68) which compared the most deprived decile to the least deprived decile. The ROC curves showed that socioeconomic measures performed better than chance at determining teenage conceptions (X² = 158.7, P < 0.0001). They further showed that the SIMD performed best at predicting teenage pregnancies by having the largest area under the curve with a value of 0.73.

Conclusion Despite a slight decline in teenage pregnancies over past decades, there is still an evident association between deprivation and teenage pregnancy. Moreover, this study shows that the SIMD measure is the most suitable measure at determining teenage-conceptions. As the SIMD is the most recent and multifaceted measure of socioeconomic status, this supports its future use in health inequality research.