Risk Factors for Disability Pension in Former Child Welfare Clients: Same or Different as for Peers?

Background: Childhood related risk factors associated with receipt of disability pension are well documented in longitudinal studies. But it is not known if the patterns and strengths of those risk factors differ between former child welfare clients and their majority population peers. Methods: Twelve Swedish national registries were used to longitudinally follow the entire Swedish population born between 1973 and 1978 to age 30-35 (N=530500), including more than 18.000 former child welfare clients. Linear probability models were used to calculate risk differences between five groups of child welfare clients and their majority population peers, complemented with sex stratified analysis. In addition, the hypothetical reductions, that is, population attributable fractions (PAF), were estimated, taking the prevalence of the risk factors into account. Results: The pattern were the same for both the majority population and the former child welfare clients: the most potent risk factors for having been awarded disability pension at age 30-35 were poor school performance (measured at age 15-16) and low educational attainment in adulthood. In terms of strength, however, were these associations even stronger for former child welfare clients than for their majority population peers. Likewise, the hypothetical reductions in the prevalence of disability pension attributed to these risk factors were larger among the child welfare clients. Other risk factors, such as household poverty and parental psychopathology, were not as potent. The pattern was similar for both sexes, but boys yielded a bit higher PAFs. Conclusion: The pattern of risk factors for receiving disability pension later in life is the same for former child welfare clients and their majority population peers, but the strength of these links are stronger for former child welfare clients regardless of sex. Henceforth, reductions in the prevalence of disability pension among child welfare clients may be influenced by prevention strategies targeting poor school performance in this very vulnerable population.