Objective This presentation will report the findings of a study addressing the temporal relationship of social disadvantage and the onset of childhood limiting long-term illness or disability (LLTI/D) in a large UK census-based longitudinal study, the Office of National Statistics Longitudinal Study (ONSLS) and discuss the contribution of these findings to resolving the issue of social disadvantage as cause, as well as consequence, of childhood LLTI/D.
Methods Children born between the 1981 and 1991 UK censuses who became ONSLS members and were present at the 2001 census were identified. Data were extracted on those children who didn't have LLTI/D in 1991 but did in 2001 (index) and those with no LLTI/D on either occasion (comparison). A social disadvantage index (SDI) comprising low household social class, rented housing and no car ownership was constructed for 1991. Logistic regression models were fitted on LLTI/D with the SDI as the independent variable of interest adjusted for age, gender, lone parenthood, and ethnicity.
Results 60,000+ children became ONSLS members between 1981 and 1991 and were present at the 1991 and 2001 censuses. 52,438 in the index and comparison groups had complete data. In bivariate analysis, social disadvantage, age, gender and lone parenthood but not ethnicity were significantly associated with the onset of LLTI/D in the index group. After adjustment for confounding, the SDI showed a finely graded association with onset of LLTI/D in the index group (most disadvantaged OR 2.12 (1.77,2.54); disadvantaged in two domains OR 1.45 (1.20, 1.75);disadvantaged in one domain OR 1.14 (0.94, 1.40)).
Discussion Our findings support the possible role of social disadvantage in the aetiology of childhood LLTI/D. The presentation will discuss the limitations of the study and additional research that will be needed to confirm the causal role of social disadvantage.