Certainly the most striking finding was that one in every ten Canadian children under 12 is living with a parent who has some form of psychiatric disorder. Furthermore, the vast majority of these parents report no mental health care in the previous 12 months. In addition, 1 in 6 children exposed children come from single parent homes – two factors which are cause for significant concern. While parental psychiatric disorders convey a risk to children in and of itself, it may also serve as an identifier for a series of adversities that also increase risk to offspring such as exposure to trauma, high-risk neighbourhoods, downward social mobility and poor social and economic support.
Whereas there has been no previous work of this scope to date, the estimate of exposure to past year alcohol abuse and dependence in American households, according to the 1992 National Longitudinal Alcohol Epidemiological Survey, is 10.25% for children under 12. Of these children, 70.4% were directly exposed to parental alcohol abuse or dependence, yielding a prevalence of exposure of 7.2% . Our numbers for Canada indicate that the prevalence of exposure to parental substance use disorders, and alcohol abuse and dependence (i.e. excluding illicit substances) for children under 12 is 11.4% and 8.3%, respectively (data not shown).
Substance use disorders were the most common psychiatric disorder experienced by parents of children under 12. Research indicates that these children are at a higher risk of developing substance use disorders themselves, as well as non-substance related psychopathologies . This may be due to the fact that parents who abuse alcohol are more likely to expose their children to a number of adverse events. Specifically, these children are at an increased risk of encountering emotional, sexual, and physical abuse, domestic violence, parental separation, incarceration, illicit drug use, witnessing suicide attempts, as well as a combination of more than one of the above adverse experiences . These have been shown as strong predictors of future alcohol abuse and depression in children [46–50].
Similarly, population-based data shows that children (interviewed when adults) of parents with psychiatric symptoms appear to be at higher risk of not only the same disorder that the parent experienced, but also of most other disorders . Furthermore, a recent longitudinal study has shown that children of parents with MI are at higher risk of mortality, which remains elevated from birth to early adulthood . It has been suggested that the transmission – given the absence of better wording – of psychiatric disorders from parents to children can be categorized in two broad classes: anxiety and depression – or chronic dysphoric disorders – and 'acting-out' disorders, represented mainly by harmful substance use . Although initial findings suggested that children of parents with disorders from one of these groups were only at higher risk of developing a disorder from the same group [19, 51], this hypothesis has been questioned [52, 53]. Thus, future strategies should perhaps be less focused on prevention or identification of risk factors for any specific diagnosis, but on broader arenas that may likely encompass improvement of parenting skills, child protection and follow-up. Furthermore, the finding that most children living with a parent affected by MI are also in single parent families, indicates the need for supportive strategies for these parents and children. Children from such families are at higher risk of MI, substance abuse, death due to suicide, and drug overdose , as compared to two-parent families with one mentally ill parent. Additionally, children from families where both parents have a history of psychiatric disorders, compared to cases where only one parent experienced psychopathology [14, 34], are also at higher risk, indicating that there may be some shielding effect exerted by the parent without MI. This is confirmed by empirical observations of families where the presence of a father with no history of MI may buffer the effects of maternal psychopathology, and lower the children's risk of developing possible MI .
Forecasting for a better future
Although the mechanisms through which parental MI influences children's mental health and development are not clearly understood, the presence of the association is well documented. However, the use of such evidence to generate policy and planning strategies aimed at reducing the burden carried by these children has been limited. Also, as it is estimated that only half of the burden of mental disorders can be reduced through currently available treatment modalities , the development of new preventive strategies has been suggested as a possible alternative [55, 56]. Certainly, our findings suggest a significant population of children for whom such prevention programs should be targeted at in the hopes of reducing future burden.
Most methodological limitations of the study indicate we may be underestimating the number of children exposed to parental psychiatric disorders. The CCHS 1.2 collected information on one adult respondent per household and may miss individuals that were homeless, hospitalized or living in institutions at the time of the survey. These individuals are more likely to have psychiatric disorders and if they were missed by the survey but do live in the household, we may be underestimating the proportion of children exposed to parental psychiatric disorders.
The fact that the survey did not collect information on the mental health status of other family members may also result in an underestimate of the number of children exposed to parental psychiatric disorders. The confidence intervals around the estimates were calculated using the weights that take into account non-response, probability of selection and the complex sampling scheme adopted by CCHS 1.2.
As mentioned earlier, the survey covers the most prevalent psychiatric disorders, but no all of them. For example, the survey did not include certain diagnosis such as psychosis and personality disorders. In addition, the criteria for the diagnosis of substance abuse outlined in the 4th edition of the Diagnostic and Statistical Manual of Mental Disorders' (DSM-IV) were also not included in the survey by Statistics Canada. These decisions were beyond our control but may arguably lead to an underestimation of the burden of mental disorders in Canada's adult population  and as a consequence, to an underestimate of the number of children exposed to parental psychiatric disorders.
The need for further studies
Family-oriented interventions to prevent adverse outcomes among children of parents experiencing MI are rare. However, it is encouraging to see that there is a growing body of literature evaluating the effectiveness of such strategies – geared towards various age groups – in reducing the incidence of MI [56–58]. Prevention programs such as the Incredible Years Program, a behavioural training program targeting parents in high risk families, have been well studied [6, 7, 59–64] and are known to improve parenting skills and parental interaction with the child. Interventions such as the Nurse-Family Partnership, an evidence-based, nurse home-visiting program for low-income, first-time parents and their children, have been able to reduce exposure of high-risk children to adverse events, and to prevent a series of developmental problems among these children, as well as in the overall target population [3–5, 65–71].
Identifying these children has important implications. Child psychiatric disorders usually persist into adulthood [72, 73], and prevention represents an opportunity to reduce health expenditures and promote sustainability of the health care system . Identifying which children are at risk is also one of the possible keys to the success of the Nurse-Family Partnership, and its possible cost-effectiveness, once it has targeted these children. Other preventive strategies could be implemented by psychiatrists and other mental health professionals if the identification of the patients that are parenting small children was part of the routine of mental health service providers.
Finally, none of the adverse exposures are chosen by the children themselves, and neither are family composition or background. Coupled with the fact that these factors play an important role in child development and have far reaching effects into adulthood [1, 75, 76], this issue should raise awareness and promote action in child advocacy at the level of health professionals, as well as policy. Improving support for children and families with parental MI may be the key to enhancing protective factors and reducing risk of future morbidity. We hope that documenting the significant number of children exposed to parental psychiatric disorders serves as a stimulus for action that will foster safer and healthier development for them.