Methods: Data from 10,217 participants of the 1958 National Child Development Survey were used for this analysis. Social class at ages 7, 16 and 33 years was chosen to represent socioeconomic circumstances during childhood, adolescence and adulthood, respectively. Two subjective oral health indicators (lifetime and past-year prevalence of persistent trouble with gums or mouth) were measured at age 33 years. The critical period, accumulation and social trajectories models were tested in logistic regression models and the most appropriate lifecourse model was identified using the structured modelling approach (SMA). In SMA, the three models are formulated in terms of alternative specifications of regression models of the outcome Y on the time points S1, S2, S3 and their two-way interactions, which are then contrasted in a series of nested models.
Results: Overall, 67%, 57% and 40% of participants were in manual social class at ages 7, 16 and 33 years, respectively. The lifetime and past-year prevalence of persistent trouble with gums or mouth was 5.2% and 4.3%, respectively. The critical period model showed that only adulthood social class was significantly associated with oral health. For the accumulation model, a monotonic gradient was found between the number of periods in manual social class and oral health; and four out of eight social trajectories were found to be distinctive. The social trajectories model was no significantly different from the saturated model indicating that it provided good fit to the data. The specifications for the critical period and accumulation models showed poorer fit to data than the saturated model.
Conclusion: This study shows the social trajectories model was the most appropriate, in terms of model fit, to describe the association between social class and oral health.