For the outcome of symptoms of depression, 3.9% of the data was missing and it is possible that the worse social characteristics of the women without information on their Malaise Inventory score may have led to a small increase in the estimation of the effect of being in care in the adjusted model. A major limitation of this study is that a large number of women with a history of being in selleck inhibitor care may not be included in the MCS due to their not agreeing to take part or being ineligible because their own children had been taken into care. Furthermore, information on the childhood socioeconomic

status of the mothers was not available. However, although we can compare our prevalence to estimates of children currently in care, it is not possible to obtain estimates of how many women of child-bearing age, at the time of our cohort, may have been in care during their childhoods without prevalence data of children in care from the 1960s to the 1990s. Therefore, we are unable to say what the likely proportion of women who would have been excluded would be. However, it is possible that women who were excluded due to having their own children taken into care may be more likely to have had worse social outcomes than those who entered the cohort. We would assume that if the data on these women had been captured, the results of

this study would have been more extreme. A systematic review of the characteristics of families whose children were taken into care showed that a low-socioeconomic status was the factor most associated with this outcome.43 This systematic review noted a large variation by country in the factors associated with families

requiring children to be placed in care, and that only one study was found from the UK. This and subsequent studies suggest that in the UK, children who have been taken into care are more likely to have had mothers who were younger, to have a history of substance misuse or mental illness, to live in a deprived neighbourhood, to be from a lower social class, and to live in overcrowded or rented accommodation.3 44 Potential mechanisms and policy implications The population of this study is women born between the 1960s and mid-1980s, and their children who are now 13–14 years old. The findings of this study have relevance to these Drug_discovery children as they enter adolescence and adulthood, as evidence suggests that activity in the early years can have lasting effects on health and psychosocial functioning.45–47 Unfortunately, the outcomes for looked-after children in the UK remain poor, during their childhood and also when they enter adult life.5 48 49 The increased policy focus on the early years, education and integrated care in recent times may have helped to improve outcomes for these children.